Bank Card Recognition with ML Kit
The ability to scan and recognize a bank card is essential for financial and shopping apps. ML Kit aims to make the process as simple as possible for developers. Using ML Kit, your app can feature a simple and secure bank card recognition feature. You'll be able to use the camera on a phone to scan and capture the details off of a bank card.
Users of your app will be able to point their camera at their card, and their information will be read quickly. Once the information is gathered, the app will return to a screen where their card number and info have been accurately recorded. This is where the user has an opportunity to review their information and make sure everything is correct.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Using the ML Kit demo APK, you can see this technology in action. Use the BCR feature to scan a credit or debit card and see the information show up on your phone. You can even see the security features in action, like how screenshots and screen recording are disabled to prevent saving sensitive information.
How Developers are Implementing Bank Card Recognition Recognition
Developers that are offering apps that host financial services will use ML Kit to make this a secure and easy process. This can also be used as identity verification if this is what your app requires. For example, in the past, when binding a bank card for online payment, users have to manually input their card numbers. It is easy to make mistakes. At present, with the bank card recognition service, inputting bank card numbers is automatic, thereby quick and accurate, greatly improving user experience.
The bank card recognition service can be implemented into an app using this HMS plug-in. Your app can directly call the bank card recognition plug-in to obtain the recognition result through the callback function without processing camera streams.
Developer Resources
Huawei provides plenty of documentation and guides to help you get started with ML Kit's bank card recognition. You can get started with this guide here.
For all of the functions of ML Kit, refer to their service portal here.
For an overview of their APIs, browse the comprehensive resource library here.
You can also look at different ways that ML Kit can be implemented, by seeing a collection of sample codes here.
Related
Imagine that one of the players for your game has spent a lot of time and effort becoming an advanced player, but is then unable to use his/her props, level, or points in the game on a new device, simply because they are not signed in to their account. The user would need to start from scratch, which could be so frustrating that they'd opt to uninstall the game. Account Kit can help you avoid this nightmare scenario, by enabling your game to share its digital assets across devices.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Gameloft, an acclaimed and award-winning game publisher, has chosen to integrate Account Kit. Now Gameloft game apps allow for sign-in via one-click authorization on new devices, after which digital assets owned by the user, and linked to the user's HUAWEI ID, can be accessed on the new device. This has substantially enhanced user experience and user loyalty, by integrating Account Kit and other services like In-App Purchases, the game apps not only bring high-quality gaming experience to AppGallery users, but also see a great increase of their own downloads.
In addition, Account Kit incorporates two-factor authentication and facial or fingerprint sign-in technologies, and complies with international standards such as OAuth 2.0 and OpenID Connect, to keep user data and privacy under lock-and-key. A lot of financial apps from around the globe have integrated with Account Kit after their rigorous internal evaluation and have released the new version, due to the kit's formidable security and privacy protection capabilities.
More than 900 million users in 190+ countries and regions use HUAWEI IDs, a massive, global user base that can expand the reach and audience for your app to unexpected new heights. Account Kit will continue to upgrade to serve as an intuitive and accessible method of entry to your apps, helping you attract and retain users with unprecedented success.
For more information, please visit:
Account Kit Development Guide:
https://developer.huawei.com/consum.../introduction-0000001050048870?ha_source=hms1
Account Kit Codelab:
https://developer.huawei.com/consumer/en/codelab/HMSAccounts/index.html#0?ha_source=hms1
KBZPay is Myanmar's fastest growing mobile wallet app, enabling millions of people to store, transfer, and spend money directly from their smartphones. KBZPay is powered by KBZ Bank, a bank with a 40% market share in the domestic retail and commercial banking sectors. Moving into the digital age, KBZ Bank has worked with Huawei for years to build digital financial infrastructure and services for its users nationwide.
The Challenges
Mobile banking is balanced on three main demands: performance, convenience, and security. To move with future trends, KBZPay wants to provide the ultimate user experience built on trust and loyalty. This app is dedicated to delivering convenience to users, and ensuring that users know their private financial information is secure.
Specifically, users want hardened security for services like changing PIN or applying for a loan, and a streamlined process for verification, which was inconvenient. In most cases, users needed to call or even go to their bank in person for account verification.
In addition, KBZ Bank wanted to better leverage its internal resources, preventing them from being restrained by any limits.
Why HMS Core ML Kit
To improve their product portfolio, KBZPay browsed the offerings on HMS Core ML Kit, a toolkit with various machine learning capabilities. KBZPay settled on the liveness detection function, which captures and verifies user face data to determine whether a face is real or is a fake.
This function offers a range of features, including:
● Accurate verification: During the testing and implementation phases, liveness detection proved to be 99% accurate in identifying and verifying faces, helping to protect user accounts.
● Integrate once, use everywhere: Liveness detection enables users to change pins and passwords without calling or visiting KBZ Bank, ensuring higher UX.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
The Benefits
The liveness detection function makes verification much easier, allowing users to complete verification swiftly. KBZPay users can now verify their identity anywhere, anytime through the app which is secure against fake face attacks and does not require the user to take additional actions.
This cooperation between KBZPay and Huawei signals the first banking app in Myanmar to implement liveness detection from ML Kit. Looking forward, KBZPay plans to work with Huawei into other key scenarios, like login and loan applications.
Discover more Developer Stories and how you can grow with Huawei.
Explore more opportunities with Huawei at our Ecosystem Partners Website.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Many apps require users to verify their identity in order to use their services offline (such as checking into a hotel) and online (booking a train/air ticket, playing a game, for example). This requires identity document details to be manually entered, which can sometimes be let down by typos.
With the ID Card Recognition feature from HMS Core ML Kit, entering incorrect details will be a thing of the past.
OverviewThis feature leverages optical character recognition (OCR) technology to recognize formatted text and numbers of ID cards from images or camera streams. The service extracts key information (for example, name, gender, and card number) from the image of an ID card and then outputs the information in JSON format. This saves users from the trouble of manually entering such details, and significantly cuts the chances of errors occurring.
Supported Information
ID CardID NumberNameGenderValidity PeriodBirthdaySecond-generation ID card of Chinese mainland residents√√√√-Vietnam ID card√√√-√
When to UseApps in the mobile payment, traveling, accommodation, and other fields require an ID document image for identity verification purposes. This is where the ID Card Recognition service steps in, which recognizes and inputs formatted ID card information, for smooth, error-free input.
Take an e-commerce app for example. You can protect the security of your business by guaranteeing that all users shall verify their identity.
Service Features
All-round card recognition: Recognizes all eight fields on the front and back of a second-generation ID card of Chinese mainland residents.
Fast recognition: Quickly recognizes an ID card in just 545.9 milliseconds.
High robustness: Highly adapts to environments where the lighting is poor or conditions are complex. In such environments, this service can still deliver a high recognition accuracy of up to 99.53% for major fields.
After integrating this service, my demo app received very positive feedback from its testers, regarding its fantastic user experience, high accuracy, and great efficiency.
I recommend you try out this service yourself and hope to hear your thoughts in the comments section.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Developments in mobile network technology have facilitated numerous errands in our daily life, for example, shopping online and offline, paying utility bills, and transferring money. Such convenience is delivered by apps with the payment function that often requires a card to be bound to user's account. Users have to complete the whole binding process themselves, by manually inputting a long bank card number, which is both time-consuming and error-prone. The bank card recognition service from ML Kit overcomes these drawbacks by automatically recognizing a bank card's key details. With it, mobile apps can provide a better user experience and thus improve their competitiveness.
Service IntroductionThe service leverages the optical character recognition (OCR) algorithm to recognize a bank card from the image or camera streams (whose angle offset can be up to 15 degrees) captured from a mobile device. Then the service can extract key card details like the card number, validity period, and issuing bank. Extracted details are then automatically recorded into the app. It works with the ID card recognition service to provide a host of handy functions including identity verification and card number input, simplifying the overall user experience.
Demo
Use CasesCurrently, binding bank cards is required by apps in a range of industries, including banking, mobile payment, and e-commerce. This service can accurately extract bank card details for the purpose of performing identity verification for financial services. Take an e-commerce app as an example. With the service integrated, such an app can record the bank card information that is quickly and accurately output by the service. In this way, the app lets users spend less time proving who they are and more time shopping.
Features
Wide coverage of bank cards: This service supports mainstream bank cards such as China UnionPay, American Express, Mastercard, Visa, and JCB, from around the world.
Accurate and fast recognition: The service recognizes a card in just 566 milliseconds on average, delivering a recognition accuracy of over 95% for key details.
How to Integrate ML Kit?
For guidance about ML Kit integration, please refer to its official document. Also welcome to the HUAWEI Developers website, where you can find other resources for reference.
{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
Cards come in all shapes and sizes, which apps don't like. The different layout of details and varying number lengths make it difficult for an app to automatically recognize key details, meaning the user has to manually enter membership card details or driving license details to verify themselves before using an app or for other purposes.
Fortunately, the General Card Recognition service from HMS Core ML Kit can universally recognize any card. By customizing the post-processing logic of the service (such as determining the length of a card number, or whether the number follows some specific letters), you can enable your app to recognize and obtain details from any scanned card.
Service IntroductionThe general card recognition service is built upon text recognition technology, providing a universal development framework. It supports cards with a fixed format — such as the Exit-Entry Permit for Traveling to and from Hong Kong and Macao, Hong Kong identity card, Mainland Travel Permit for Hong Kong and Macao Residents, driver's licenses of many countries/regions, and more. For such documents, you can customize the post-processing logic so that the service extracts only the desired information.
The service now offers three types of APIs, meaning it can recognize scanned cards from the camera stream, photos taken with the device camera, and those stored in local images. It also supports customization of the recognition UI, for easy usability and flexibility.
The GIF below illustrates how the service works in an app.
Use CasesThe general card recognition service allows card information to be quickly collected, enabling a card to be smoothly bound to an app.
This is ideal when a user tries to book a hotel or air tickets for their journey, as they can quickly input their card details and complete their booking without the risk of losing out.
Service FeaturesMulti-card support: General card recognition covers a wider range of card types than those covered by the text recognition, ID card recognition, and bank card recognition services.
This service can recognize any card with a fixed format, including the membership card, employee card, pass, and more.
Multi-angle support: The service can recognize information from a card with a tilt angle of up to 30 degrees, and is able to recognize scanned cards that have a curved text with a bending angle of up to 45 degrees. Under ideal conditions, the service can deliver a recognition accuracy of as high as 90%.
I got to know how to integrate this service here. FYI, I also find other services of ML Kit intriguing and useful.
Look forward to seeing you in the comments section to know what ideas you've got for using the general card recognition service.