Clever Geek Handbook

How to implement bank card recognition with HUAWEI ML Kit

general information



In previous articles, we talked about how to use the HUAWEI ML Kit to create a smile detection function and an applet for document photos. In this article, I will show you how to implement the bank card recognition feature so that users can link a bank card with a minimal investment of time.







Purpose of the function of recognition of bank cards



Before we start developing, let's look at what the bank card recognition function is for. It is most relevant for applications with payment functions such as banking applications and online shopping. These applications often have a number of common requirements:



  • Linking a bank card


Users can link their bank cards to make fast online payments.



  • Bank transfers


Users can transfer money between accounts at the same bank or between different banks.



  • Real name authentication and identity verification


Users can quickly authenticate with their real name and verify their identity based on their bank card information.



To use each of these options, the user must enter their bank card details, including the card number and expiration date. You know, of course, that it is easy to make mistakes and waste a lot of time when entering data manually. But thanks to the HUAWEI ML Kit bank card recognition service, users will be able to enter their data quickly and without errors.



Application of the bank card recognition service

The bank card recognition service allows the camera to recognize bank cards and read key data such as card number and expiration date. Supported by an ID recognition service, it offers a number of popular features such as identity verification and credit card number entry.







We offer a bank card recognition plugin that you can integrate to make this service available in your application. The plugin will handle the camera data for you.



Development start

1. Preparation

Add 1.1 Huawei's Maven repository to a file on the project level build.gradle

Open the file build.gradle in the root folder of your Android Studio project and add the address of the Maven repository. 



buildscript {
    repositories {        
        maven {url 'http://developer.huawei.com/repo/'}
    }    }allprojects {
    repositories {       
        maven { url 'http://developer.huawei.com/repo/'}
    }}


1.2 Add SDK dependencies to app-level file build.gradle 



dependencies{  
  implementation 'com.huawei.hms:ml-computer-vision-bcr:1.0.3.303' 
  implementation 'com.huawei.hms:ml-computer-card-bcr-plugin:1.0.3.300' 
  implementation 'com.huawei.hms:ml-computer-card-bcr-model:1.0.3.300' }


1.3 Enable automatic app update to the latest model



To have your app automatically update the machine learning model after downloading from the HUAWEI AppGallery, add the following data to the AndroidManifest.xml file: 



<manifest 
   ... 
   <meta-data              
       android:name="com.huawei.hms.ml.DEPENDENCY"   
       android:value= "bcr"/> 
       <!--If multiple models are required,set the parameter as follows:        
       android:value="object,ocr,face,label,icr,bcr,imgseg"-->  
   ... </manifest>


1.4 Request access rights to Camera and Memory in AndroidManifest.xml file 



 <uses-permission android:name="android.permission.CAMERA" />
 <uses-permission android:name="android.permission.WRITE_EXTERNAL_STORAGE" />


2. Code development

2.1 Create callback functions that are called after receiving the recognition result



Override the onSuccess, onCanceled, onFailure and onDenied functions :



  • onSuccess : Called after recognition completes. MLBcrCaptureResult displays the recognition result.


  • onCanceled : Called if the user canceled the recognition.


  • onFailure : Called when recognition fails.


  • onDenied : Called if the recognition request was denied for any reason (for example, if the camera is not available).


private MLBcrCapture.Callback callback = new MLBcrCapture.Callback() { 
    @Override 
    public void onSuccess(MLBcrCaptureResult bankCardResult){ 
    }  
    @Override 
    public void onCanceled(){ 
    }   
    @Override 
    public void onFailure(int retCode, Bitmap bitmap){ 
    } 
    @Override 
    public void onDenied(){ 
    }   };


2.2 Set the recognition parameters for the captureFrame API call of the recognizer Recognition

data is returned by the callback function created in 2.1. 



private void startCaptureActivity(MLBcrCapture.Callback callback) { 
    MLBcrCaptureConfig config = new MLBcrCaptureConfig.Factory()  
        .setOrientation(MLBcrCaptureConfig.ORIENTATION_AUTO) 
        .create(); 
    MLBcrCapture bankCapture = MLBcrCaptureFactory.getInstance().getBcrCapture(config); 
    bankCapture.captureFrame(this, callback); }


2.3 Call the method specified in clause 2.2 with the callback of the recognition button 



@Override 
public void onClick(View v) { 
    switch (v.getId()) {  
        case R.id.detect: 
            startCaptureActivity(callback); 
            break; 
        default: 
            break; 
    } }


Try it yourself



Let's take a look at how the bank card recognition feature works.







Source

We've uploaded the source to GitHub, so feel free to make changes.

github.com/HMS-Core/hms-ml-demo

More details you can visit



See the sample code for recognizing bank cards in MLKit-Sample \ module-text \ src \ main \ java \ com \ mlkit \ sample \ activity \ BankCardRecognitionActivity.java.

Please visit our official website for details.



New examples



We will share with you other useful features of the HUAWEI ML Kit. Stay tuned!


More articles:


All Articles