Coding With Fun
Home Docker Django Node.js Articles Python pip guide FAQ Policy

WeChat gadget OCR printedText


May 20, 2021 WeChat Mini Program Development Document


Table of contents


ocr.printedText

This interface should be called on the server side, as detailed in the Service Side API.
This interface supports cloud calls. Developer tool version required . . . 1.02.1904090 (latest stable download)
wx-server-sdk >= 0.4.0

This interface provides universal print OCR identification based on small programs

How to call:

  • HTTPS call
  • Cloud calls

HTTPS call

The request address

POST https://api.weixin.qq.com/cv/ocr/comm?img_url=ENCODE_URL&access_token=ACCESS_TOCKEN

Request parameters

Property Type The default Required Description
access_token string Is The interface calls the credentials
img_url string Is To detect the picture url, pass this without passing the img parameter.
Img FormData Is Form-data media file identification, there is fileame, filelength, content-type and other information, pass this without passing img_url.

Returns a value

Object

The returned JSON packet

Property Type Description
errcode string Error code
errmsg string The error message
items string Identify the results
img_size string The size of the picture

Instructions for use

Interface Limits Subscription numbers, service numbers, enterprise numbers, and small programs that have been certified during the internal test can be called directly, with a limit of 500 times per day.

Using Tips this interface for the background interface, can be based on its own business hosting situation, with small programs of photography, photo selection, etc. together, you can complete the ID photo collection, upload, identification, information return and other processes, for the need to be based on identity cards, bank cards and other physical cards or certificates, collection of photos or text information and other business scenarios.

Image Description File size limit: less than 2m

The picture supports the use of the IMG parameter real-time upload, but also supports the use of the IMG_URL parameter to transfer the image address, and download the image by WeChat background.Universal print body OCR is suitable for screen screenshots, print photos and other scenes

Request data example

Example 1

curl http://api.weixin.qq.com/cv/ocr/comm?img_url=ENCODE_URL&access_token=ACCESS_TOCKEN

Example 2:

curl -F ‘[email protected]’“http://api.weixin.qq.com/cv/ocr/comm?access_token=ACCESS_TOCKEN”

Return an example of the data

{
    "errcode": 0,
    "errmsg": "ok",
    "items": [ //识别结果
        {
            "text": "腾讯",
            "pos": {
                "left_top": {
                    "x": 575,
                    "y": 519
                },
                "right_top": {
                    "x": 744,
                    "y": 519
                },
                "right_bottom": {
                    "x": 744,
                    "y": 532
                },
                "left_bottom": {
                    "x": 573,
                    "y": 532
                }
            }
        },
        {
            "text": "微信团队",
            "pos": {
                "left_top": {
                    "x": 670,
                    "y": 516
                },
                "right_top": {
                    "x": 762,
                    "y": 517
                },
                "right_bottom": {
                    "x": 762,
                    "y": 532
                },
                "left_bottom": {
                    "x": 670,
                    "y": 531
                }
            }
        }
    ],
    "img_size": { //图片大小
        "w": 1280,
        "h": 720
    }
}

Cloud calls

Cloud call is the ability provided by the small program Cloud Development to call WeChat open interfaces in cloud functions and need to be used in cloud functions via wx-server-sdk.

The interface method

openapi.ocr.printedText
Permissions for the ocr.printedText API need to be configured in config.json, details

Request parameters

Property Type The default Required Description
imgUrl string Is To detect the picture url, pass this without passing the img parameter.
Img FormData Is Form-data media file identification, there is fileame, filelength, content-type and other information, pass this without passing img_url.

The structure of img

Property Type The default Required Description
contentType string Is Data type, incoming MIME Type
value Buffer Is File Buffer

Returns a value

Object

The returned JSON packet

Property Type Description
errCode string Error code
errMsg string The error message
items string Identify the results
imgSize string The size of the picture

Abnormal

Object

The exception thrown

Property Type Description
errCode string Error code
errMsg string The error message

The legal value of errCode

Value Description The lowest version

Instructions for use

Interface Limits Subscription numbers, service numbers, enterprise numbers, and small programs that have been certified during the internal test can be called directly, with a limit of 500 times per day.

Using Tips this interface for the background interface, can be based on its own business hosting situation, with small programs of photography, photo selection, etc. together, you can complete the ID photo collection, upload, identification, information return and other processes, for the need to be based on identity cards, bank cards and other physical cards or certificates, collection of photos or text information and other business scenarios.

Image caption File size limit: less than 2M

Images support the use of img parameters for real-time upload, but also img_url parameters to transmit picture addresses, by WeChat background download pictures to identify. Universal print OCR is suitable for screens, printed photos, and other scenes

An example of requesting data

const cloud = require('wx-server-sdk')
cloud.init()
exports.main = async (event, context) => {
  try {
    const result = await cloud.openapi.ocr.printedText({
        type: 'photo',
        imgUrl: 'ENCODE_URL'
      })
    return result
  } catch (err) {
    return err
  }
}

Or

// cloud = require('wx-server-sdk')
// ...
// 方法返回 Promise
cloud.openapi.ocr.printedText({
  type: 'photo',
  img: {
    contentType: 'image/png',
    value: Buffer
  }
})

Return an example of the data

{
    "errcode": 0,
    "errmsg": "ok",
    "items": [ //识别结果
        {
            "text": "腾讯",
            "pos": {
                "left_top": {
                    "x": 575,
                    "y": 519
                },
                "right_top": {
                    "x": 744,
                    "y": 519
                },
                "right_bottom": {
                    "x": 744,
                    "y": 532
                },
                "left_bottom": {
                    "x": 573,
                    "y": 532
                }
            }
        },
        {
            "text": "微信团队",
            "pos": {
                "left_top": {
                    "x": 670,
                    "y": 516
                },
                "right_top": {
                    "x": 762,
                    "y": 517
                },
                "right_bottom": {
                    "x": 762,
                    "y": 532
                },
                "left_bottom": {
                    "x": 670,
                    "y": 531
                }
            }
        }
    ],
    "img_size": { //图片大小
        "w": 1280,
        "h": 720
    }
}