Blog

Is your online identity on the line?

Posted on | by Signzian

Among the questions that can potentially induce an existential crisis is, “Who are you?”. Among the phrases that can answer this question are, “You are who you are”, “You are what you do”, “You are what you eat”, and the optimistic “You are what you want to be”. This blog isn’t a metaphysical exploration of who we all are. But, as residents of the Internet Age, it is an attempt to figure out what constitutes our online identities, how they’re threatened, verified, and protected.

The sum of an individual’s characteristics and interactions on the internet can be said to be their online identity. Every minuscule piece of information about a person on the internet adds to this virtual identity. Since our interaction with different online sites would be different, each site has its own understanding of who a person is.

Amazon knows what products you like to buy. Zomato knows all about your midnight cravings. Google has seen the panicky medical searches made at the slightest hint of a cough. Uber probably knows where you live. Facebook knows your friends just as much as it knows you. Netflix knows what you binge after work, and Spotify knows your workout jam.

This may not give a clear picture of the individual to these platforms, but defines a way to identify them. All of these are an individual’s partial identities, the aggregate of which make up who we are online.

Our online identities are related to a number of facets in the virtual world. The prominent ones are discussed ahead:

Profiles, Cookies, and Footprints

Site Profiles:
For many of the apps and sites we sign up on, we construct profiles. When it is not required to feed some data for the creation of a profile, unbeknownst to us, the site creates a profile for us. This is to distinguish an individual, maintain a record for them, and secure their information. An attribute called an “identifier” is needed to create this automatic profile. The identifier is a way of referring to a set of characteristics. It can be something like a number given by the site, your email ID or a username.

Browser Cookies:
A cookie can be understood as a piece of information sent by a web server for the browser to store. The browser returns the cookie to the server the next time the page is opened. Cookies usually contain at least two pieces of data:

• a unique user identifier
• some information about that user

It’s cookies that preserve the state of a user’s interaction on a site across browser sessions and page reloads. They help in the optimum functioning of the website, and although seemingly innocuous and invisible, cookies can store various data points. If you allow your browser to accept cookies, you are being tracked. For example, sites using embedded Google tools such as the search bar, trace your activity via cookies whose data Google will have access to.

Cookies fall under the purview of data protection regulations such as the GDPR, which is a testament to the personal nature of the data they contain. Explicit consent is now required from the user to allow the cookies to attach themselves to an individual’s browser.

Digital Footprints:
We are constantly leaving behind a digital footprint on the internet. This refers to the traces of data we leave behind on the internet, the primary constituents of which are website cookies and social media activity. This footprint is what is commonly commercialized. Third parties such as advertisers pay for this data, and digital footprints are thus monetized with companies having access to our data. This may lead to:

  • Data deduced from footprints used for big data analytics
  • Loss of privacy and anonymity
  • Information shared with advertisers without our explicit knowledge for targeted advertising
  • Malicious activity such as identity theft

Identity Providers

In the technical sense, identity provision can be distilled into three forms:

a) Traditional or retrospective identities: Individuals receive a credential from a third party after a trusted enrollment process. It can then be used to authenticate oneself.

b) Low trust or self-asserted identities: The third party merely issues an identifier to the individual which it can confirm when asked.

c) Behavioral identities: When enough data about an individual is collected by the service providers to decipher that the same person is visiting multiple times.

Google or Facebook have the capacity to act as a trusted identity provider (IdP) by authenticating an individual on behalf of some other online entity that is being signed in to. These are called social IdPs and are accepted across platforms for their convenience, bypassing re-authentication processes. With this possibility, the digital world is shifting from siloed credentials to those that are accepted across platforms. For example, your employee ID cannot be used to identify you at an airport, but your Google account can be used to make a booking through MakeMyTrip.

While this type of identity aggregation makes the onboarding process efficient for individuals, it is also beneficial for advertisers. As opposed to using subsets of data from different platforms, referring to an IdP allows them to:

  • Personalize experience
  • Provide recommendations
  • Preserve customer histories
  • Prepare and proliferate hyper targeted ads

However, this gives rise to the problem of panopticism. It is a concept by French philosopher Michel Foucault used to explain a kind of internal surveillance. In this case, the IdP will be able to keep track of each place an individual is authenticating, without them being overtly aware of the data being subliminally collected.

It is evident that social logins are shaping the future of our digital identities. The European Commission has even proposed the idea of using national ID cards to access online services, such as Facebook, Uber and Twitter. A Facebook profile could thus be lobbied as a border-less digital identity. On the flip side, KYC for social media users has been proposed in India as a way to combat trolling online.

The proliferation of aggregated identities reveals the difficulty and need to remain anonymous in the 21st century. This has paved the way for the existence of alternatives like:

  • Sign in with Apple, which can authenticate a user using Face ID on their iPhone without turning over any of their personal data to a third-party company.
  • Anonymous social apps such as Whisper which functions as any other social media site, except for they are supposed to be completely anonymous. Users are issued a random nickname upon joining which cannot be searched.

Threats to Online Identity

With our use of the internet entwined with our online identity, the aim isn’t to be anonymous anymore, but to control the degree to which subsets of our data are revealed to public and private entities. While it is difficult to be in full control, it is important to be familiar with the existing threats to our online identity.

Data breaches

A data breach is when confidential information is accessed by an entity not having the authorization to do so. It is common for breaches to go undetected, or companies to not reveal to customers that there has been a breach. “HaveIBeenPwned” is a website that lets netizens check if their personal data has been leaked in data breaches. It reveals the company, year, and constituents of the data breaches where one’s data was compromised from an online account. You can then take remedial measures to safeguard your data and be vary of unsafe logins.

Hackers

Hackers use a multitude of ways to reel in victims of identity theft. Two of the most common ones are phishing and keylogging.

  1. Phishing is the process of deceiving an individual into sharing sensitive information. It involves attaching links or malicious software codes/bits to a non-suspicious medium like e-mail or a Facebook attachment (pictures/audio clips etc). Clicking on the link/attachment will redirect you to an imitation of a trusted website where you would need to provide your credentials. For example, when being phished a careful observation of the URL space in your browser will show that you are not truly on facebook.com, but instead a sly imitation like fac3book.com. An individual’s login credentials, passwords, bank details etc, can then be sold on the blackmarket, used to steal identities or commit bank fraud.
  2. Keylogging is the retrieval of information through the act of covertly recording keystrokes of a device user. An attacker can use keylogging to intercept sensitive information such as passwords and credit card numbers. A preventive measure is using a virtual keyboard when logging in or carrying out transactions online.

ISP Tracking

Internet Service Providers store the logs of IP addresses and session timings for billing and legal purposes. However, it can be used for questionable purposes as well:

  • Data retention, whistleblower Edward Snowden revealed the National Security Agency requested information from ISPs in the US for surveillance purposes
  • Data monetization, selling of personal data (this is legal in some countries)
  • Bandwidth throttling, in areas where net neutrality is threatened
  • Monitoring, for torrents and illegal file shares, copyright infringements

Surveillance

While citizens have a right to privacy in India, constitutional provisions are not yet in effect to question government surveillance of personal data. Excesses of government surveillance, and exceptions in personal data protection laws can lead to aspects of one’s online identity being used against them. It is imperative to hold our governments accountable to privacy demands. The epitome of surveillance is China’s social credit system which tracks individual, corporate, and government behavior across the country in real time to build a database on its citizens.

Verifying Online Identity

While the privacy of our identity is a concern, on the flip side banks and fintechs are concerned with verifying this identity. With a multitude of transactions happening online, verification of our digital identities is imperative.

A digital identity comprises of two forms of information:

  1. Digital attributes: Email address, date of birth, government issued ID, biometrics, login credentials etc.
  2. Digital activities: Likes and comments of social sites, purchase history, photos on Instagram etc.

For the most part, the verification of an identity is done by authenticating digital attributes.

The classic method to gain an acceptable level of assurance that the identity of an online customer matches their real-world identity is a three part paradigm which includes verifying:

  1. Something the individual knows (eg. password/ security question)
  2. Something the customer has (eg. identity card)
  3. Something the customer is (eg. biometrics, such as a fingerprint)

Banks may require more information for security reasons. One way is to observe and an individual’s behavioral data such as login habits. When there is an anomaly, the bank can then alert the customer and verify activity to prevent fraud.Digital identity verification service can encompass social media identity as a layer of verification. For example, BlaBlaCar and Ola request sharing of social media profiles as an additional layer for a quicker KYC process.

More than seventy financial institutions including 7 major banks in India trust Signzy’s RealKYC and VideoKYC solutions to make the entire process simple, secure, and compliant.

Protecting Online Identity

While it seems as if the virtual walls of the internet have eyes and ears, it is not difficult to protect your data. Although it appears as if the government and private companies alike are after your personal data and online identity, with data protection regulations in effect, no one can access your data without your knowledge and in some places, your consent. (To read more on the regulations in place in the EU and India, you can take a look at our article comparing the GDPR and PDP Bill)

Here are a few ideas on how to safeguard your private information:

  • Inspect privacy policies before granting permissions
  • Change passwords often
  • Avoid unprotected or public Wi-Fi networks
  • Have a primary and secondary email. When logging in to a new site you do not trust, use a secondary email which is not linked to any other accounts with personal information
  • Use a Virtual Private Network (VPN) to access the internet. This masks your IP and ensures your trail is encrypted, dissuading any malware to follow into your device.
  • Try not to save passwords on your browser. This can protect you from malicious cookies that may get access to the rest of your saved passwords.

Whether you know who exactly you are or not, you’re now adept to protect who you are in the digital world.

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Signzy

Written by an insightful Signzian intent on learning and sharing knowledge.

 

Data privacy: the debacle & the debate (GDPR vs PDP)

Posted on | by Signzian

In an increasingly data driven digital economy, Big Tech companies have an eye, ear, and finger on the pulse of billions.

Depending on how deep you’ve let Amazon, Facebook, and Google sync into your life (pun intended), the data these companies have access to has reached an increasing level of detail. The digital era has molded us into great liars when it comes to signing up to online sites. While complaining about how ridiculous it seems to identify traffic lights to prove we’re not robots, we mechanically lie about reading all the Terms and Conditions. By agreeing to the T&C we may have inadvertently let the company use and sell our data for reasons we weren’t aware of.

Contextualizing the need for personal data protection

In the past few years, the headlines have been replete with worrying instances from the digital world. From large scale data breaches to controversial targeted political ad policies and inconclusive investigative hearings on privacy. The Facebook–Cambridge Analytica data scandal of 2018 exposed how unethically sourced personal data could be used for thought manipulation. Data of about 87 million Facebook users was inappropriately harvested by the political consulting firm, Cambridge Analytica, and was used for electoral advertising.

The mammoth scale and global repercussions of this scandal altered the history of the privacy debate. It revealed the imperative need to have wide-scale legal mechanisms. A system needed to be enforced to regulate what data will be collected, what it will be used for, and how permission should be sought from its owners. Organizations would have to be held accountable to such provisions through a transparent legal process. These regulations were to be designed to protect the privacy and personal data of netizens and perhaps rein in the power and influence of giant tech companies.

Introducing EU’s GDPR and India’s PDP

The European Union set precedence with the European General Data Protection Regulation (GDPR). The GDPR was adopted in 2016 and enforced on 25 May 2018. It is not a mere directive, but a regulation. This implies that it is directly binding and applicable although it does allow for some flexibility to individual member nations to adjust the provisions. The GDPR is also not an Act, which means that its members have passed their own legislations based on the regulation.

In India, a regulation governing data privacy and data protection is set to be passed this year. The need stemmed from the 2017 Supreme Court judgement on the Right to Privacy. (Read our article on how the judgment impacted the digital world and the financial sector here.) A draft data protection bill was then composed by a committee headed by Justice B. N. Srikrishna. After about 2 years of contentious debate on the bill, during which it was floated for public feedback from stakeholders, it was tabled in the Indian Parliament on 11 December 2019. Currently, a joint parliamentary committee is scrutinizing the revised draft of the bill, codified as the Personal Data Protection Bill (PDP Bill). Post this, it will be debated in the Indian Parliament and finally passed.

It is yet to be determined whether the Indian PDP Bill is closer to the EU’s progressive GDPR or to China’s policy of control. Either way, it has managed to irk both Big Tech companies and privacy advocates alike. Companies with data banks aren’t happy with the cost and hassle of compliance. They deem the bill as isolationist due to its restrictive certification requirements to operate in India. Privacy advocates highlight how the exceptions in the bill can lead to State excesses of control over our data. They warn of government mission creep. Mission creep is the gradual expansion of an intervention, here, it implies the dangerous possibility of the State having access to all our data in the absence of a Privacy Law.

This blog is an exploration of how the GDPR and PDP Bill are similar, yet different in various ways.

Coming to terms with the terminology

Before delving into specifics, it’s important to be acquainted with the terminology used in the legal mechanisms for data privacy. The two regulations also use different terms for the same entity:

 

  • Data processor: Any person or legal entity including the State who processes the data. This may consist of the data controller or data fiduciary itself or a third party.
  • Interestingly, the PDP Bill’s definition of personal data differs from the international definition in the GDPR.

Thematic classification of differences

The underlying principles and intent of the PDP Bill resemble the provisions enshrined in the GDPR. Aspects such as the need to have a clear purpose of processing personal data, consent requirements, personal rights, and the appointment of Data Protection Officers in organizations are closely adapted from the GDPR.

However, there are a range of differences between these two instruments of privacy. Here, the language and enforcement provisions aren’t compared, but the stance both mechanisms take on different issues.

These have been classified into the following themes:

1. Classification of data

 

Critical data has not yet been defined by the Indian government. Although the category resembles the list of “special categories” in the GDPR, the EU’s regulation has defined what the category entails while in India the government has the power to declare any data as critical data. The GDPR does not have separate localization rules for this type of data, unlike India. This is explained ahead.

2. Data localization and cross border data flows

Data localisation requires the collection, processing, or storage of certain types of data within the borders of the nation where the data was generated, before being internationally transferred.

GDPR stance

The aim of data protection frameworks is to protect the data while safeguarding its free flow. The GDPR has no hard data localization conditions. It allows for cross-border transfer of all types of data if the country of data transfer has an adequate framework of data protection.

PDP Bill stance

On the other hand, the Indian regulation’s requirements seem to restrict data’s free flow.

  • Sensitive personal data: This category of data when collected, shared or disclosed to the data fiduciary in India has to be stored only within the borders of the State. It may be transferred beyond the territory of India for processing, subject to explicit consent and conditions.
  • Critical personal data: Strict data localization norms exist for this category of data. It can only be processed within the borders of India. The problem arises since this type of data has not even been defined yet.

Due to firm opposition, the 2018 draft was amended to dilute data localisation requirements (such as storing a mirror copy of all personal data in India). Yet, the GDPR’s approach to handling data is considered more pragmatic since it ensures data gets similar protection once it moves out of the jurisdiction of the regulation.

3. Right to restrict processing

The GDPR grants the data subject the right to limit the processing of their data. This means that the processing of personal data can be stalled at an intermittent stage. This can be requested on the grounds of unlawful processing, data inaccuracy etc. The PDP Bill doesn’t enshrine any such right to the data subject.

4. Right to not be subjected to automated decisions

The GDPR grants the right to not be subjected to automated decision-making, such as profiling. Profiling is the automated processing of personal data to assess certain things about an individual. This right gives the data subject the recourse of obtaining human intervention. This is when such data is solely automatically processed to make an important decision, has legal consequences or significantly affects the individual.

For example, automated processing can be used to profile potential behaviour of an individual in a faster way. It is possible that the individual will not behave in the manner the results project. In that case, if such profiling affects the legal rights of the individual, the person can legally request human intervention.

The PDP Bill does not ascertain this right. While it encourages individuals to seek remedy through courts in case of such discrimination, it does not empower an individual to decide how their data should be processed.

5. Storage limitation

The GDPR lays down specific exceptions for increasing the storage period of collected data. These exceptions include public interest, historical, scientific, and statistical reasons.

On the other hand, the PDP Bill mandates the explicit consent of the data principal to store data for a longer duration of time than is needed to satisfy the purpose for which it is collected. The GDPR does not necessitate this consent.

What does this mean for your organization?

The most contentious question is whether GDPR compliance implies PDP compliance. It is briefly addressed in this section to understand how these bills affect an organization’s compliance needs.

  • Areas such as the anonymization standards differ between the PDP Bill and the GDPR.
  • With no parallel of ‘critical personal data’ in the GDPR, companies will have to be careful with their processing of this classification for India.
  • Unlike the GDPR, the PDP Bill also mandates the explicit consent of the data principal to store data for a longer duration of time.

Such differences and more, warrant that companies pay close attention to the compliance needs of the PDP Bill, even if they meet the requirements of the GDPR.

Other interesting follow-up questions will be explored in our next blog in the PDP Bill series.

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Signzy

Written by an insightful Signzian intent on learning and sharing knowledge.

 

Collaboration is the Key for Banks and Fintech Startups

Posted on | by Signzian

The banking ecosystem is in a continuous state of disruption. Traditional banks are trying to navigate the reality with legacy systems and hoping to switch to digital banking. Banking institutions are facing pressure due to rising customer demands for better customer experience.

For so long, banks and fintech companies stood on either side, competing against each other. However, they are fast learning that collaboration might be the key to their success. A joint venture between banking and fintech is the path to long-term growth, increased revenue, and satisfied customers.

Collaboration is the Key for Banks and Fintech Startups

Challenges for Banks on the Offline Road and the need for digital banking

The banking sector has traditionally relied on paper-based processes. Here are a few challenges banks face due to lack of innovation and digital banking:

  • Improving customer loyalty — Without a tech-enabled experience, it is challenging to provide convenience to the customer. When banks fail to innovate, they compromise with the services they offer. This, in turn, negatively impacts the loyalty of the customer.
  • Resource optimization — Without efficiency in operations, it is difficult to optimize resource usage. Both time and money are assets to any financial organization and a lack of technology hampers a bank’s ability to optimize its resources.
  • Personalization — Without data and analytics, personalization is hard to achieve. Therefore, banks face the constant challenge of learning about their customers’ varied interests and behavior when they don’t leverage technology.
  • Transparency — Trust and transparency get lost under piles of paper forms and applications. Therefore, banks that don’t opt for digital banking, fail to achieve process transparency and compromise with both employees and customers on the trust front.
  • Omni-channel — When customers want to make fewer visits to the branch, it is critical that their interactions with the bank be seamless across other channels. Digital banking can create an omnichannel experience through social media, website and mobile app platforms.

These challenges are not recent events, but banks have been facing them since forever. There was no way to resolve these issues, except by taking the help of fin-tech companies.

Why Legacy Financial Institutions are Under Urgent Pressure

The World Fintech Report 2018 by Capgemini outlines the following reasons why even the biggest banks face pressure today:

  • New business models — New trends such as Peer-to-Peer payments and lending, social network scoring solutions, and crowdsourced solutions are pushing legacy financial institutions to innovate.
  • Speed and efficiency — As customers demand better experiences, banks are forced to rethink their processes. Fintech startups are making digital banking accessible and convenient. Banks are expected to follow suit. Real-time updates, proactive notifications, alerts, and agile innovation are a part of the enhanced customer experience.
  • Transparency — That is something banks cannot achieve with traditional systems. Digital banking is needed to bring in transparency into the various processes. Fintech firms are leading the way by showing cost upfront and offering services at lower costs.
  • Personalization — Digital banking is better positioned to offer a personalized experience to their customers by leveraging data and analytics. If banks want to retain customers, they will have to level up their personalization.
  • Operational efficiency — Streamlines delivery and product development provide a significant competitive advantage. With a digitally-enabled solution, fintech firms are improving operational efficiency and pushing legacy organizations to innovate and reinvent.

Opportunities for Banks and Fintech Companies

According to a survey by PwC [1], 82 percent of insurers, asset managers, and banks plan to increase the number of collaborations they have with fintech startups over the next three to five years. And we see various ongoing acceptance from banks to collaborate with new innovation offered by fintech startup players.

Taking example from our own journey, presenting here case in point, Signzy, a platform that helps financial institutions:

  • Onboard customers through a seamless process that reduces hassle and friction.
  • Scale faster with an AI and ML-based regulatory engine.
  • Reduce costs.
  • Cut turnaround time.
  • Use advanced cryptography to create robust security and data protection infrastructure.
  • Leverage a range of white-labeled solutions to drive faster digital transformation.

Signzy is trusted by several large banking corporations such as ICICI Bank, SBI, Aditya Birla Financial Services, Mahindra Finance, Edelweiss, and so on.

Here are a few challenges Signzy’s fintech solutions help banks solve:

  • KYC — Banks need their customers to fill out KYC forms with their details. Traditionally, the process used to be paper-based. This meant that any mistake in one form would compel customers to start all over again. Signzy enables bank-grade digital KYC in real-time. An API matches the biometrics of the customer, checks the data in government records, and warns the user of potential document forgery as the customer fills in details.
  • Background check — Traditionally, the bank staff usually gather all identity documents from each customer and manually does a background check of the customer information. Signzy offers a simple and digital way to accomplish this. Algorithmic Risk Intelligence allows banks to do a holistic background check, discovering any court cases and legal lists, fetching anti-money laundering related data, checking the UN CFT List and NIA Most Wanted list.
  • Contracts — Signzy is replacing physical contracts with digital ones. These come with video and voice verification, blockchain implementation, biometrics, and high performance. Smart contracts are the way to go.
  • SME Onboarding — Merchant onboarding is a seamless task with Signzy’s offering that helps clients cut down onboarding time from 2 weeks to a few hours. Features include a mobile link with in-built regulatory rules, real-time document verification, Aadhar-backed contract signing, and AML background check.
  • Transaction Banking with Corporates — Signzy’s offerings can help banks automate complex regulatory procedures with AI and robotics to significantly reduce TAT and enhance the customer experience. Comprehensive risk and regulatory checks can help banks mitigate risk in dealing with large enterprises on both liability and asset products.
  • Insurance — The insurance process can be simplified and digitized with Signzy’s individual onboarding system. It simplifies the onboarding journey using advanced biometrics and fraud detection capabilities reducing risk and enhancing user experience.

There’s an entire list of solutions that Signzy offers to bank institutions to catalyze their digital transformation journey.

Roadblocks in Banking and Fintech Collaboration

Understanding the opportunity might not be equal to taking action for banks and fintech start-ups. Between varying cultures, different infrastructures, and ever-changing compliances, a collaboration between banking institutions and fintech companies is far from simple. Many opportunities and proposed deals get derailed as a result.

Here are top three things both sides must consider before finalizing a proposed partnership:

  • Consider any cultural gap — Make sure that the cultural match is not too challenging. There must be a willingness to adapt by both sides in a partnership.
  • Understand challenges — Collaboration with fintech startups can help financial institutions alleviate the major challenges they face.
  • Leverage data and innovation — The massive data that financial institutions have by their side is by far their most underused and critical asset. Fintech startups should leverage this data and innovate around it.

The success of collaboration rests with the organizations that can understand each other’s pain points and strengths, and work to improve the customer experience while reducing operational costs.

However, few fintech startups can offer the level of personalization, contextuality, speed, and seamless delivery to help financial institutions achieve digital transformation.

Signzy is capable of doing that. Want to collaborate with a forward-thinking fintech startup that can help you leverage digital processes?

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Moni Gupta

 

AI-Based Digital Onboarding Transforming Banking Organizations

Posted on | by Signzian

Customers are increasingly demanding better digital banking experiences to match the integrated experiences they receive in other industries. Financial institutions continuously find themselves under pressure to deliver a mobile-first, customer-centric digital experience to customers throughout their journey starting right from account opening. Artificial Intelligence has the potential to drive digital transformation in banking and redefine the way financial institutions interact with customers during the digital customer onboarding process.

The onboarding system is generally the first experience a customer has with any banking institution and that in itself sets the tone of the ongoing experience the bank can deliver. The client onboarding process is also critical by the fact that it gathers Know Your Customer Data for continuing maintenance and management of the account.

AI in banking has become a priority, and bankers have identified two areas worthy of digital transformation in banking – KYC and digital onboarding.

 

How Can Banks Offer a Seamless Onboarding Experience?

There are four critical areas in which banks can focus to revolutionize the onboarding experience they offer:

  1. Digitize processes, indeed

     Often financial institutions fail to understand that digital is more than online and mobile. Their customers demand sophisticated interfaces, but banks continue to work through manual and paper-based workflows. This way, bankers are rendered unable to answer customer queries about process status and task completed. Enabling a single point of truth of each customer and eliminating manual, disconnected processes will empower bankers to quickly onboard customers and collect and manage their information seamlessly. AI in banking will also increase the quality of data and help minimize errors.

  2. Allow customers to bank anytime, anywhere

     Digital transformation in banking will help customers to carry out functions 24×7, from anywhere using their preferred device or channel. In turn, financial institutions will be able to market products across multiple channels, improving sales and customer retention. By introducing an Omni channel experience through responsive web design, banks will be able to offer frictionless user experiences allowing customers to undertake the digital onboarding process on one platform and finish it on a different one.

  3. Collect data once

     Customers often face frustration when they have to provide the same information over and over again for different banking tasks. Digital transformation in banking can change that. Banks have started experimenting with end-to-end interfaces that allow customer details to be entered once and reflect everywhere else. This cuts down a lot of hassle for both customers and banking officials. In conjunction with AI in banking, institutions are also considering the applications of blockchain to consolidate data across a financial organization.

  4. Personalize experiences

     Predictive analytics can be leveraged to learn more about customer behavior- where customers spend their time, where they abandon sessions, and so on. With this data, banking institutions can personalize experiences for customers and expedite the digital onboarding process.

Banking organizations are ripe with data ingested by intelligent capture only waiting to introduce processing intelligence.

Digitize the Onboarding System & Opt for Online ID Verification

AI is helping banks set up a seamless digital onboarding experience. Consider this: when a customer opens a new bank account or applies for a loan, they have to provide a number of documents and ids to their banks such as employment proof, identity proof, and proof of address.

Additionally, the bank staff needs to physically scan each document to verify specific clauses, values, and statements in their process. Then, bankers need to verify particular intentions or assumptions with the customer before deciding each customer’s creditworthiness for a loan or a product.

These manual activities are tiresome and can very well induce error anywhere in the end-to-end process.

With intelligent capture and online id verification, this can be done with a few clicks on a smartphone. Thus, saving bankers and customers a lot of time and effort.

Since customer satisfaction is a crucial differentiator for banking institutions today, automating and digitizing onboarding process can prove to be a game-changing strategy. In other words, the client onboarding process can determine customer experience, loyalty, referrals, sales, and profitability.

Hence, Digital transformation in banking will help financial institutions adapt to changing regulations quickly. With manual paperwork, turning a few regulations every month can seem a daunting task, while digital processes make the process easier.

How Signzy is Innovating in the Banking Space

Trusted by ICICI Bank, SBI, Birla Sun Life Mutual Fund, Edelweiss, MasterCard, PayU, HDFC Bank, and many other financial institutions, Signzy is a classic example of a tech-enabled solution disrupting the banking space.

Through Blockchain and AI solutions in banking and financial services, Signzy is reducing human interaction with customers at various levels, saving it only for the critical decision making aspects of banking.

Here are a few features Signzy offers across its product line:

  • Digital real-time KYC
  • Digital signature for KYC
  • Biometric signatures
  • Algorithmic Risk Intelligence to provide a satisfactory background check
  • Digital contracts

Signzy can help financial institutions decrease operational expenditure by 75 percent.

Case in Point: How Signzy led digital transformation in banking for its client

The existing process consisted of the following steps:

  • The customer manually fills in the application form.
  • The bank sales associate collects the forms and KYC documents without verifying.
  • The branch manager screens all documents to make sure everything is in place.
  • The central ops do a risk check on the documents collected.

If the documents have anything amiss, they go back a phase and the process restarts.

Here are the challenges the baking firm faced because of their manual processes:

  • Errors in filling out forms.
  • No real-time verification of the information submitted by the customer.
  • Missing documents or details.
  • Time-consuming process.
  • Manual documentation that’s hard to maintain.

The solution to this firm’s woes was a digitized platform to disrupt their inefficient process such as Signzy.

The benefits Signzy realized for this institution:

  • As customers filled out digital forms and submitted to bank associate, the information could be checked in real-time.
  • The verification process was automated as a result.
  • The auto-populating of form quickens the process.
  • Reduction in human typing errors by data extraction.
  • Company verification from government records.
  • The user experience was transformed into a smoother one.
  • The bank ops could now focus on key risk cases and undertake high-value tasks
  • Since the bank saw a dramatic reduction in their operational costs, their revenues surged.

Signzy can bring out massive measurable benefits for their customers through digital transformation in banking.

Signzy realized the following metrics for its client:

  1. 80% cost reduction in customer onboarding process.
  2. Reduction in the TAT from 3 days to 30 minutes- a huge time saving and a differentiator in customer experience.
  3. Three times more efficient sales.

As of the end of 2018, Signzy supports 51 customer accounts and answers half-a-million API calls each month. Signzy digitally transforms businesses focused on financial services.

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Ankit Ratan, CEO-Signzy

 

Defining & Measuring Software Quality Attributes at Scale

Posted on | by Signzian

Managing a largely decoupled microservices-based system is always a challenge. We all have maintained an unpleasant bug tracker at some point or the other. Around a year back, when we witnessed a growth spurt, we felt a need for systematically tracking our production systems to capture the live performance of our products and any issues our clients might be facing. It is important to know what your end-users will face when using your products.

The problem with microservices is that the point of failure is not easily detected. A simple solution to this is a tracking id associated with every request. But, if you have already built a production system with hundreds of fully functioning microservices, adding a tracking id requires an unfortunate amount of code changes that may or may not change the code logic.

Also, we noticed that a majority of issues in our systems were not being reported. End-users frequently decide that it is easier to repeat the software activity rather than identifying & reporting the issues. A user may first think — “Probably I made a mistake & did not use the product correctly!”, “It was a glitch, just retry!”.

“Our job as software engineers should be more than writing code, but proactively identifying customer needs & make their experience better.”

Our Solution

Initial Thoughts

If you don’t assign a quantitative value to a parameter, you can’t track it and you can’t improve on it. Evidently, any parameter talked about in qualitative terms is as good as nothing.

Quality improvement is a branch of the larger umbrella of Quality Assurance. QI emphasizes on setting up the right parametric measures during the development and release stages, which will eventually help you measure how your system performs in the real world. Getting the actual data from the real world gives you insights on how your software is getting used outside the lab conditions (your development center).

We wrote an interesting article about one of our large implementations replacing legacy banking systems, which is expected to work in rural areas of India. see section World is not the cozy laboratory, we know that!. The article emphasizes on understanding your last mile user data & build for that.

Software Quality Attributes

When we first thought of creating such a system at Signzy, the first thing we addressed is the identification of Software Quality Attributes relevant to our systems which we are going to capture and improve upon. It is also advisable to classify them based on their criticality in your systems & impact it would have on your business if the said metrics decline.

An exhaustive list of system quality attributes in Software engineering can be found here.

We chose the below metrics and created definitions for each. While they are mostly standard, it should be noted that input and measured parameters can be customized to fit your needs. For instance, we changed the calculation methodology (which is intentionally omitted in the below table) for 4 of the metrics to fit an API & SaaS software we create.

 

Continue reading →

Video KYC — The Banking future is here!

Posted on | by Signzian

At a macro level, India seems to be going through an “identity crisis”. Not in terms of whether she is a potential superpower or a grappling economy, but instead which papers and bills identify its constituents as Indian citizens.

Zooming in to the fintech ecosystem of the country, constantly identifying individuals through Know Your Customer (KYC) processes is imperative, but the latest developments in the sector are far from bleak. The past few years have seen rapid developments in ideas and technologies, with the regulatory space dishing out amendments to keep up.

With concepts like Artificial Intelligence (AI), face-matching, and Computer Vision now a practical reality instead of something fresh out of a sci-fi movie, the processes of authenticating customers have taken a step away from the physically daunting and expensive task of onboarding. Along the same tangent, the regulatory body RBI is also tasked with updating their KYC compliance norms. The fintech space is fast changing, and sometimes companies developing futuristic tech have solutions relegated to waiting in the wings until official norms give them the green light. This may require sitting back with a tub of popcorn for a few years.

The build up here is to introduce an esrtwhile non-compliant, yet simple, secure, and scalable method to establish the identity of an individual: Video KYC (VCIP).

Reaching Compliance: The past

  • In an earlier phase of “identity crisis”, the question was whether the unique identification card “Aadhaar” had constitutional validity itself. On 26 September 2018, the Supreme Court affirmed its constitutional validity but scrapped Section 57 of the Aadhaar Act that allowed private companies to use Aadhaar authentication and eKYC.

With the 1,448-page judgment up for interpretation, a cloud of ambiguity loomed over India’s booming fintech industry; when was Aadhaar authentication to be stopped, and would the private space have to sacrifice the paperless, cashless and presence-less verification method it had adopted? Potential customers were seen on the opposite side of the regulations door as the industry suffered hiccups to onboard new customers after the judgement.

  • About six months later, on June 26, 2019, an expert committee on Micro, Small and Medium Enterprises (MSMEs), headed by UK Sinha, former chairman of the Securities and Exchange Board of India (SEBI) proposed the need for online video KYC. The panel recognized the drawbacks of physical presence and the sheer data handling required for even eKYC. Video-KYC was seen as a simple seamless process that could be done through a video chat where the customer can display documents. At that time the committee recommended it could be done through apps like Google Duo or Apple FaceTime.

Experts pointed out that considering these applications were of foreign origin, the RBI was unlikely to allow them. Under the Data Protection Bill, and the debate around data localization, the central bank was unwilling to let companies store customer data in foreign locations.

  • In the latest installment of updates, the RBI approved Aadhaar-based video authentication as an alternative to e-KYC on January 9, 2020. The amendment to the KYC norms allow banks and other lending institutions regulated by it to adopt a Video based Customer Identification Process (V-CIP) as a consent based alternate method of identity verification for customer onboarding.

Explaining Compliance: The present

Making sense of the latest amendments to regulations is not easy. We at Signzy have distilled it down to a 20-point cheat sheet to make sure it is. The changes due to the introduction of V-CIP are:

  1. Informed consent to be obtained from individual customer before the live V-CIP process
  2. RE (Regulated Entities) official to record video of the customer present for identification
  3. RE official is to capture a photograph of the customer during the session
  4. RE official to obtain identification information. This can be done through two methods depending on the entity type:
    Banks: OTP based Aadhaar eKYC authentication
    Non-bank RE: only Offline Verification of Aadhaar
  5. RE official to capture a clear image of PAN card which is to be displayed during the process
  6. Live location is to be recorded during the session
  7. RE official to ensure customer’s photograph matches them
  8. RE official to ensure provided identification details match the details on the Aadhaar/PAN
  9. Randomization of questions to ensure there is no pre-recording. This means that the sequence and/or type of questions during video interactions should be varied in order to establish that the interactions are in real-time
  10. The Aadhaar XML or Secure QR provided for offline verification should not be more than 3 days old
  11. Accounts opened through the V-CIP process will only be operational after a concurrent audit
  12. RE official to carry out a liveliness check
  13. The audiovisual interaction should be triggered from the domain of the RE itself
  14. An activity log along with the credentials of the official carrying out the process should be preserved
  15. Video to have a timestamp and be safely stored
  16. The amendment encourages the use of AI and face-matching technology
  17. RE official to redact/blackout Aadhaar number as per standard guidelines
  18. The interaction is to be necessarily done by a bank official and not an agent
  19. The process is to be operated only by specifically trained officials
  20. RE to ensure security, robustness and end to end encryption of the V-CIP application

This is a monumental step towards digitizing the authentication process for banks, lending startups and non-banking financial institutions.

Signzy: The future

Signzy’s video technology came into existence before the license to use it did. In 2016, bankers told us our tech was too futuristic and not practical, but now the future is here! Keeping up to its promise of delivering future ready digital onboarding solutions, Signzy is ready with a plug and play end-to-end digital Video KYC solution with V-CIP features.

Our systems are designed for banking grade technology which means they meet the strictest infosec regulations and data security requirements. Signzy’s video KYC is being used to onboard thousands of customers every month by SEBI regulated institutions. This solution has matured over dialects, browsers and low-internet scenarios. And also has one of the best facial recognition technology at the background (Can read more here)With RBI’s progressive move to bring Video KYC (Video Customer Identification Process) 2020, we look forward to onboarding RBI regulated institutes on our battle-tested solution!

If you would like to know more then look at the Video KYC section on our website

www.signzy.com

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Ankit Ratan, CEO-Signzy

 

India’s fastest growing Identity Verification company Signzy partners with Apriori for data aggregation services.

Posted on | by Signzian

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Paritosh Vatsal Tripathi

 

Removing blur from images

Posted on | by Signzian

Everyone misses perfect shots once in a while. Yeah, that’s a pretty shame (We all do that all the time!!!).

There are special moments which we want to capture to make them memorable for lifetime, but just because your camera shook or amount of noise in your camera can really hamper those special moment resulting in blurred images (Maybe your subject is on the move, the reason is not always bad cameras but bad timing as well!!!).

So, if you are also one of us who misses out their special moment, this post is just for you. In this post, you will get to know how you can restore blurred images. All the thanks and applause goes to Neural networks.

What are you going to learn?

From this blog post, you will learn how to make use of the neural network by image deblurring technique with the help of Scale-recurrent Networks. For more info on the technique, you can access this link. The network takes sequence of blurry images as input at different scales and produces a finite set of sharp images. The final output image is at the full resolution.

Figure 1: SRN architecture from the original paper

The method above uses end-to-end trainable networks for the images. Then it used multi-scale convolutional neural network with the approach of state-of-art.

These methods embark from an abrasive measure of the blurry images, and gradually try to recover the suppressed image at higher resolutions.

This Simple Recurrent Network aka SRN makes use of scale recurrent network for multi-scale deblurring. The solver and the corresponding parameter at each scale in a well-established multi-scale method are always same. This is a natural choice as it simply aims to solve the very same problem. If we vary scales in different parameters, then it may cause instability and the extra issue of unrestrictive solution space. Another concern to address here is the input images may have different motion scales and resolutions.

If you allow too much parameter tweaking in each scale, then this might result in creating a solution that is overfitted to a specific motion scale. There are people who believe that this method is also applied to CNN-based methods. Still, there are some recent cascaded networks that still prefer to use independent parameters for every single scale. They justify this method with a pointer which seems quite plausible. They proposed that sharing networks weights across different scales can significantly deteriorate training difficulty and also introduce stability benefit.

Their experiment shows that how with the help of recurrent structure and the combination of the above advantages, the end-to-end deep image deblurring framework can greatly mend training efficiency. They only use less than 1/3rd of the trainable parameters with faster testing time. Apart from this, their method is proven to produce high-quality results both qualitatively and quantitatively. Let’s not dive deep in the research paper for now. Allow me to present you our use-case of this deblurring technology.

We are well-established Global Digital Trust Company, which functions primarily in the domain of verification processes. For this verification process, our customers have to click photos of their documents and submit it for verification. There are probable chances where these photographs may be blurred either due to camera shake or any motion which causes difficulty in reading the document text.

To solve the blurred image problem, we fed these images in the aforementioned Deblurring Model. The results were exhilarating. Below are some of the samples,

Concluding Remarks

What do you learn from this blog? You learn the use of scale recurrent network for deblurring images. With this technology, you can easily extract data from blurred identity card images. You don’t have to poke your customers again and again for the re-submission of the documents due to bad-quality or blurred images. Thanks for the read and do leave a comment to let me know what you feel about this technology. Adios for now fellas!!!

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Signzy

Written by an insightful Signzian intent on learning and sharing knowledge.

 

How we built a modern, state of the art OCR pipeline — PreciousDory

Posted on | by Signzian

Finally I am very happy writing this blog after a long wait. As the title suggests PreciousDory is a modern optical character recognition (OCR) engine which performs better than the engines from tech giants like Google, Microsoft, Abby in KYC use cases. We feel now it is time to tell the world how we built this strong OCR pipeline over the last couple of years.

We at Signzy are trying to build a global digital trust system. We solve various fascinating problems related to AI and computer vision. Of them, text extraction from document images was one of the critical problem we had to solve. In the initial phase of our journey we were using traditional rule based OCR pipeline to extract text data from document images. Those OCR engines were not that efficient to compete with global competitors. So In an urge to stay competitive with the global market we took an ambitious decision to build an inhouse modern OCR pipeline. We wanted to build an OCR engine which will surpass the global leaders in that segment.

 

The herculean challenge was out and our AI team accepted it with a bliss. We know building a production ready OCR engine and achieving best in class results is not an easy task. But we are a bunch of gallant people in our AI team. When we started researching about the problem we found very few resources to help us out. And we also stumbled upon the below meme ?

 

If You Can’t Measure It, You Can’t Improve It

The first task our team did was to create a test dataset that would represent all the real world scenarios we could encounter. The scenarios includes varying viewpoints, illumination, deformation, occlusion, background clutter, etc. Below are some samples of our test dataset.

Sample test data

When you have a big problem to solve, break it down into smaller ones

We spent a quite a lot of time in literature study trying to break the problem into sub-problem so that our individual team members could start working on it. We ended with the below macro level architecture.

Macro level architecture

After coming up with the basic architectures our team started exploring the individual entities. Our core OCR engine comprises of 4 key components.

  1. CropNET
  2. RotationNET
  3. Text localizer
  4. Word classifier

CropNET

This is the first step in the OCR pipeline. The input documents for our engine will have a lot of background noise. We needed an algorithm to exactly crop out the region of interest so that the job gets easier in the subsequent steps. In the initial phase we tried out lot of traditional image processing techniques like edge detection, color matching, Hough lines etc. None of them could withstand our test data. Then we took the deep learning approach. The idea was to build a regression model to predict the four edges of the document to be processed. The train data for this model was the ground truth containing the four coordinates of the document. We implemented a custom shallow architecture for predicting the outputs. We achieved good performance from the model.

RotationNET

This is the second stage in the pipeline. After cropping, the next problem to solve is rotation. It was estimated that 5% of the production documents would be rotated in arbitrary angles. But for the OCR pipeline to work properly the document should be at zero degree. To tackle the problem we built a classification model which predicts the angle of document. There are 360 classes corresponding to each degree of rotation. The challenge was in creating the training data. As we had only few real world samples for training each class, we had to build a custom exhaustive pipeline for preparing synthetic training data which closely matches with real world data. Upon training , we achieved impressive results from the model.

Text localizer

The third stage is localizing the text areas. This is the most challenging problem to solve. Given a document the algorithm must be able to localize the text regions for further processing. We knew building this algorithm from scratch is a mammoth task. We benchmarked various open source text detection models on our test datasets.

Text localization — Benchmark

After rigorous testing we decided to go with CTPN. Connectionist Text Proposal Network (CTPN) accurately localizes text lines in natural image. It detects a text line in a sequence of fine-scale text proposals directly in convolutional feature maps. It was developed with a vertical anchor mechanism that jointly predicts location and text/non-text score of each fixed-width proposal, considerably improving localization accuracy. The sequential proposals are naturally connected by a recurrent neural network, which is seamlessly incorporated into the convolutional network, resulting in an end-to-end trainable model. This allows the CTPN to explore rich context information of image, making it powerful to detect extremely ambiguous text.

 

Word classifier

This is the final stage and the most critical step in the OCR engine. This is the step where most of our efforts and time went into. After localizing the text regions in the document, the region of interest was cropped out of the document. Now the final challenge is predict the text from this. Upon rigorous literature study we arrived with two approaches for solving this problem.

  1. Character level classification
  2. Word level classification

Character level

This is one of the traditional approach. In this method the bounding box of individual characters are estimated and from them the characters are cropped out and presented for classification. Now what we have in hand is a MNIST kind of dataset. Building a classifier for this type of task is tried and tested method. But the real challenge in this approach was in building the character level bounding box predictor. Normal segmentation methods failed to perform on our test dataset. We thought of developing a FRCNN like object detection pipeline for localizing the individual characters. But creating the training data for this method was a tedious task and involves a lot of manual work. So we ended up dropping this method.

Word level classifier

This method is based on deep learning. Here we pass the full text localized region into a end to end pipeline and directly get the predicted text. The cropped text region is passed into a CNN for spatial feature extraction and then passed on to RNN for extracting temporal features. We are using CTC loss to train the architecture. CTC loss solves two problems: 1. You can train the network from pairs (Image, Text) without having to specify at which position a character occurs using the CTC loss. 2. You don’t have to postprocess the output, as a CTC decoder transforms the NN output into the final text.

The training data for this pipeline is cropped word image regions and their corresponding ground truth text. Since a large amount of training data was required to make the model converge, we made a separate data creation pipeline. In this we first get the cropped word regions from the document, secondly we feed it into third party OCR engine to get the corresponding text. We used this data to benchmark it against manually created human data. The manual data was again verified by a 2 stage human process to make sure the labels are right.

We achieved impressive results with the model. A sample output from the model.

 

Time for results

At Last we combined all the four key components into a single end to end pipeline. The algorithm now takes an input image of a document and gives the corresponding OCR text as output. Below is a sample input and output of a document.

 

Now the engine was ready to face our quality analysis team for validation. They benchmarked the pipeline against popular global third party OCR engines on our custom validation set. Below are the test results for certain important documents we were handling.

 

We tested our OCR engine against other top engines on different scenarios. It includes cases with no background, different background, high brightness and low brightness. The results shows that we are able to perform better than the popular known OCR engines in most scenarios.

Productionzation

The pipeline was built now and tested. But still it was not ready to face the real world. Some of the challenges in productionsing the system are listed below.

  1. Our OCR engine was using GPU for inference. But since we wanted the solution to be used by our clients without any change in their infrastructure, we removed all the GPU dependencies and rewrote the code to run in CPU.
  2. To serve large number of requests more efficiently we builded a queueing mechanism.
  3. For easier integration with existing client infrastructures, we provided the solution as a REST API
  4. Finally the whole pipeline was containerized to ease the deployment at enterprises.

Summary

Thus a mammoth of task building a modern OCR pipeline was accomplished. A special thanks to my team members Nishant and Harshit for making this project successful. One of the key take away from the project was that if you have an exciting problem and a passionate team in hand, you could make the impossible possible. And I could not explain a lot of steps in details since I had to keep the blog short. Do write to me if you have any queries.

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Signzy

Written by an insightful Signzian intent on learning and sharing knowledge.

 

Democratizing AI using Live Face Detection

Democratizing AI using Live Face Detection

Posted on | by Signzian

Democratizing AI using Live Face Detection!  Since the dawn of AI, facial recognition systems have been evolving rapidly to exceed our expectations at every turn. In a few years, you’ll be able to go through the airport basically just using your face. If you have bags to drop off, you’ll be able to use the self-service system and just have your face captured and matched. You’ll then go to security, the same thing happens just use your biometric. The big tech giants have proved this can be done on a massive scale. The world now needs higher adoption through the democratization of this technology, where even small organizations can use this advanced technology with a plug-and-play solution.

The answer to this is Deep Auth, Signzy’s in-house facial recognition system. This allows large-scale face authentication in real-time, using your everyday mobile device cameras in the real world.

Democratizing AI using Live Face Detection

Deep Auth, Facial Recognition System from Signzy

While a one-to-one face match is now very popular (thanks to the latest Apple iPhone X), it’s still not easy to authenticate people from larger datasets that identify you from thousands of other images. What is even more challenging is doing this in real-time. And just to add some bit of realism, sending images and videos over mobile internet slows this down even further.

This system can detect and recognize faces in real-time in any event, organization, office space without any special device. This makes Deep Auth an ideal candidate to use in real-world scenarios where it might be not possible to deploy a large human workforce or spend millions of dollars to monitor people and events. Workplaces, Education Institutes, Bank branches even large residential buildings are all valid areas of use.

Digital journeys can benefit from face-based authentication thus eliminating the friction of username, password, and adding the security of biometrics. There can also be hundreds of other use-cases which hopefully our customers will come up with, and help us improve our tech.

Democratizing AI using Live Face Detection

 

Deep Auth doing door access authorization.

Deep Auth is robust to appearance variations like sporting a beard,, or wearing eyeglasses. This is made possible by ensuring that Deep Auth learns the facial features dynamically (Online training).

Democratizing AI using Live Face Detection

 

Deep Auth working across different timelines

Technology

The technology behind face recognition is powered by a series of Convolution Neural Networks(CNN). Let’s divide the tech into two parts :

  • Face Detection
  • Face Recognition

Face Detection:

This part involves a 3 stage cascaded CNN network. This is to ensure the face is robustly detected. In the first stage, we propose regions (Objectablility score) and their regression boxes. In the second stage, we take these proposed regression boxes as the input and then re-propose them to reduce the number of false positives. Non-maximal suppression is applied after each stage to further reduce the number of false positives.

Democratizing AI using Live Face Detection

3 stage cascaded CNN for face detection.

In the final stage, we compute the facial landmarks with 5 point localization for both the eyes, nose, and the edges of the mouth. This stage is essential to ensure that the face is aligned before we pass it to the face recognizer. The loss function is an ensemble of the center loss and IoU (Intersection Over Union) loss. We trained the network for 150k iterations on the WIDER Face dataset.

Face Recognition:

The extracted faces are then passed to a siamese network to where we use a contrastive loss to converge the network. The siamese network is a 152 layer Resnet where the output is a 512-D vector depicting the encodings of the given face.

 

Democratizing AI using Live Face Detection

Resnet acts as the backbone for the siamese network.

We then use K- Nearest Neighbours(KNN) to classify each encoding to the nearest face encodings that were injected to KNN during the training phase. The 512-D vectorization used here compared to 128-D vectorization used in other face recognition systems helps in distinguishing fine details across each face. This provides high accuracy to the system even with a large number of non-discriminative faces. We are also working on extending the siamese network to extract 1024-D face encodings.

Benchmarks

Deep Auth poses impressive metrics on the FDDB database. We use 2 images to train each of 1678 distinct faces and then evaluate the faces with the rest of the test images. We then calculate the Precision and recall as 99.5629 and 91.2835 respectively, and with the F1 score of 95.2436.

Democratizing AI using Live Face Detection

 

Deep Auth’s Impressive scores!

We also showcase Deep Auth working in real-time, by face matching faces in a video.

Deep Auth in Action!

We tried something a little more cheeky and got our hands on a picture of our twin co-founders posing together, a rare sight indeed! And checked how good the Deep Auth really was. Was it able to distinguish between identical twins?

Democratizing AI using Live Face Detection

 

And Voila! It worked

Deep Auth is accessed using the REST API interface making it suitable for online training and real-time recognition. Deep Auth is self-servicing due to the fact it is robust to aging and appearance, which makes it an ideal solution to deploy in remote areas.

Conclusion

Hopefully, this blog was able to explain more about Deep Auth and the technology behind it. Ever since UIDAI made face recognition mandatory for Aadhaar authentication, face recognition will start to prevail in every nook and corner of the nation for biometric authentication. Thus democratization of face authentication allows even small companies to access this technology within their organizations. Hopefully, this should allow more fair play and give everyone a chance to use advanced technology to improve their lives and businesses.

In the next blog, we will explain how we have paired face recognition with spoof detection to make Deep Auth robust to spoof attacks. Please keep reading more on our AI section to understand how this is done.

About Signzy

Signzy is a market-leading platform redefining the speed, accuracy, and experience of how financial institutions are onboarding customers and businesses – using the digital medium. The company’s award-winning no-code GO platform delivers seamless, end-to-end, and multi-channel onboarding journeys while offering customizable workflows. In addition, it gives these players access to an aggregated marketplace of 240+ bespoke APIs that can be easily added to any workflow with simple widgets.

Signzy is enabling ten million+ end customer and business onboarding every month at a success rate of 99% while reducing the speed to market from 6 months to 3-4 weeks. It works with over 240+ FIs globally, including the 4 largest banks in India, a Top 3 acquiring Bank in the US, and has a robust global partnership with Mastercard and Microsoft. The company’s product team is based out of Bengaluru and has a strong presence in Mumbai, New York, and Dubai.

Visit www.signzy.com for more information about us.

You can reach out to our team at reachout@signzy.com

Written By:

Signzy

Written by an insightful Signzian intent on learning and sharing knowledge.

1 … 21 22 23 24 25 26