Python3使用腾讯云文字识别(腾讯OCR)提取图片中的文字内容实例详解

百度OCR体验地址：

https://ai.baidu.com/tech/imagerecognition/general

腾讯OCR体验地址：

https://cloud.tencent.com/act/event/ocrdemo

测试结果是：腾讯的效果要比百度的好

腾讯云目前额度是：

每个接口 1,000次/月免费，有6个文字识别的接口，一共是6,000次/月

百度接口调用之前写过文章

python实现百度OCR图片识别过程解析

使用步骤

1、注册账号： https://cloud.tencent.com/

2、开通服务：https://console.cloud.tencent.com/ocr/general

3、申请访问秘钥：https://console.cloud.tencent.com/cam/capi

4、通过 API 或 SDK 或命令行来使用服务

具体参考《操作指南》：https://cloud.tencent.com/document/product/866/17622

接口使用

1、安装SDK

https://github.com/TencentCloud/tencentcloud-sdk-python

pip3 install tencentcloud-sdk-python

2、代码实例

# -*- coding: utf-8 -*-

import json

from tencentcloud.common import credential
from tencentcloud.common.profile.client_profile import ClientProfile
from tencentcloud.common.profile.http_profile import HttpProfile
from tencentcloud.ocr.v20181119 import ocr_client
from tencentcloud.ocr.v20181119.models import (
 GeneralAccurateOCRRequest,
 EnglishOCRRequest,
 GeneralBasicOCRRequest,
 GeneralEfficientOCRRequest,
 GeneralFastOCRRequest,
 GeneralHandwritingOCRRequest
)

class TencentOcr(object):
 """
 计费说明：1,000次/月免费
 https://cloud.tencent.com/document/product/866/17619
 """
 SECRET_ID = "你的秘钥 SECRET_ID"

SECRET_KEY = "你的秘钥 SECRET_KEY"

# 地域列表
# https://cloud.tencent.com/document/api/866/33518#.E5.9C.B0.E5.9F.9F.E5.88.97.E8.A1.A8
 Region = "ap-beijing"

endpoint = "ocr.tencentcloudapi.com"

# 通用文字识别相关接口
 # https://cloud.tencent.com/document/api/866/37173
 mapping = {
   # 通用印刷体识别（高精度版） ok
   "GeneralAccurateOCR": GeneralAccurateOCRRequest,

# 英文识别 ok
   "EnglishOCR": EnglishOCRRequest,

# 通用印刷体识别 一般
   "GeneralBasicOCR": GeneralBasicOCRRequest,

# 通用印刷体识别（精简版）（免费公测版）no
   "GeneralEfficientOCR": GeneralEfficientOCRRequest,

# 通用印刷体识别（高速版）一般
   "GeneralFastOCR": GeneralFastOCRRequest,

# 通用手写体识别 ok
   "GeneralHandwritingOCR": GeneralHandwritingOCRRequest,

}

def __init__(self):
   cred = credential.Credential(self.SECRET_ID, self.SECRET_KEY)

httpProfile = HttpProfile()
   httpProfile.endpoint = self.endpoint

clientProfile = ClientProfile()
   clientProfile.httpProfile = httpProfile
   self.client = ocr_client.OcrClient(cred, self.Region, clientProfile)

def get_image_text(self, image_url, ocr="GeneralAccurateOCR"):
   req = self.mapping[ocr]()
   req.ImageUrl = image_url
   resp = getattr(self.client, ocr)(req)
   return json.loads(resp.to_json_string())['TextDetections'][0]['DetectedText']

def main():
 tencentOcr = TencentOcr()
 url = "https://ocr-demo-1254418846.cos.ap-guangzhou.myqcloud.com/general/GeneralBasicOCR/GeneralBasicOCR3.jpg"
 print(tencentOcr.get_image_text(url, ocr="GeneralHandwritingOCR"))

if __name__ == '__main__':
 main()

来源：https://blog.csdn.net/mouday/article/details/104346484