Call baidu-aip-SDK through node.js to implement the ID card recognition function

This article mainly introduces the function of calling baidu-aip-SDK through node.js to realize ID card recognition. It has a certain reference value. Now I share it with everyone. Friends in need can refer to it

In a recent project, the customer needed to implement the ID card recognition function. After reasonable planning, he decided to use Baidu AI's ID card recognition.

ID card recognition is a type of text recognition. There are many similar functions, such as driver's license recognition, etc. The principles are the same.

For front-end beginners, if you want to achieve this function by relying on the page, js cannot be achieved. It can be said with certainty that the background must cooperate with the implementation. The implementation process is to upload a picture to the backend. The backend uses server-side languages ​​​​such as Java and PHP to call Baidu AI's SDK. The obtained parameters are then returned to the front-end page for display. The call to this interface must be on the server side. If the front-end wants to implement it, it can only be achieved through node.js.

Then let’s explain the whole process in detail:

1.node

First, you need to download and install node. The installation tutorial of node is clear on the official website or the novice tutorial. instance.

After installation, enter node -v in the cmd command line. If the version number is returned, the installation is successful.

Also enter npm -v in the command line to check the cmd version. Now node integrates cmd. If you follow the tutorial to install node, node will be successful. npm should be successful.

2. Project construction

First find the project root directory or create a new folder as the root directory

　

There are two ways to install Baidu components. One is to download baidu-aip-SDK from Baidu ai, and the other is to directly install the baidu-aip-SDK package through the command line. Everything is simple here, through the command line. Installation:

Run the cmd command line in this directory. If you are not very good at command line operations, you can right-click on the current folder Shift and it will be added in the right-click menu. Open the options of the command window

##Enter npm install baidu-aip-sdk

## After the installation is completed, such a folder and a .json format file will be added to the directory

You can open the node_modules folder and confirm whether there is baidu-aip-sdk Folder

If this is the case, the function will be built.

3. Function code

Create a new js file in the root directory, as shown in the figure

This js is the node client for image recognition Terminal, a place used to provide a series of interaction methods.

Add the following code to js:

var AipOcrClient = require("baidu-aip-sdk").ocr;
// 设置APPID/AK/SK
var APP_ID = "你的 App ID";
var API_KEY = "你的 Api Key";
var SECRET_KEY = "你的 Secret Key";
// 新建一个对象，建议只保存一个对象调用服务接口
var client = new AipOcrClient(APP_ID, API_KEY, SECRET_KEY);

The APP ID / Api Key / Secret Key here needs to be applied for https://cloud in the Baidu Cloud Management Center. baidu.com/

After that, continue to add function codes to this js. What we are doing now is ID card recognition, so copy the relevant code for ID card recognition into js.

var fs = require('fs');
var image = fs.readFileSync("你的身份证路径").toString("base64");
var idCardSide = "front";
// 调用身份证识别
client.idcard(image, idCardSide).then(function(result) {
    console.log(JSON.stringify(result));
}).catch(function(err) {    
// 如果发生网络错误    
console.log(err);
});

Put the ID card picture you need to identify in the root directory, as shown in the picture

Change the ID card path in js to the image path, for example

var image = fs.readFileSync("./png1.png").toString("base64");

4. Call

to open the command line in the root directory and enter

node start.js

Wait a few seconds, and the parameters in the ID card image will be returned in the command line.

Okay, here we just provide an idea for node to call Baidu AI and ID card recognition. During the project implementation process, if the entire project does not use node, it will be left to the back-end personnel to implement this function. , even if node is used, it is recommended to let the backend staff implement it... After all, the time of the front-end bosses is very precious.

Baidu ai is very powerful, covering basically all the technologies currently on the market. ID recognition and even text recognition are just a small part of it. The principles of the recognition functions are the same. You need to be familiar with node. Next, take a look at the api sdk documentation and you will be able to write it down.

The above is the entire content of this article. I hope it will be helpful to everyone's study. For more related content, please pay attention to the PHP Chinese website!

Related recommendations:

A simple analysis of the debugging method of node.js

Responsive React Native Echarts component introduce

The above is the detailed content of Call baidu-aip-SDK through node.js to implement the ID card recognition function. For more information, please follow other related articles on the PHP Chinese website!