需求：读取图片内的文字，图片包含url形式的和image形式的
实现思路：python调用腾讯api，参考腾讯官方文档:https://cloud.tencent.com/document/product/866/17596
步骤：调用api需要配置header请求头，请求头需要鉴权签名，鉴权签名需要api密钥。
鉴权签名：https://cloud.tencent.com/document/product/866/17734
api密钥的获取：登陆腾讯云https://console.cloud.tencent.com/cam/capi
代码如下
import time
import base64
import hmac
import hashlib
from hmac import new as hmac
import requests
import json
import datetime
import random
\'\'\'通过登陆腾讯云获取自己的api密钥\'\'\'
appid=\'12******25\'
SecretID=\'AKIDGS************************NPpyNp\'
SecretKey=\'Xt*************************iwybH\'
\'\'\'初始化除api外的其他参数\'\'\'
currentTime = int(time.time()) #当前时间戳
expiredTime = currentTime+2592000 #签名的有效期 此处定义为当前时间+30天
bucket=\'\'# 图片资源的组织管理单元，历史遗留字段，可不填
rand=7648353324 #随机串，通过函数 random.randint(1,9999999999) 生成
fileid=\'\' #资源存储的唯一标识，单次签名必填；多次签名选填，如填写则会验证与当前操作的文件路径是否一致
\'\'\'配置 拼接签名串\'\'\'
src_str=\'a=\'+appid+\'&b=\'+bucket+\'&k=\'+SecretID+\'&e=\'+str(expiredTime)+\'&t=\'+str(currentTime)+\'&r=\'+str(rand)+\'&f=\'+fileid
\'\'\'定义生成签名的函数\'\'\'
def hash_hmac(ac_key,orignal):
SignTmp = hmac(bytes(ac_key,\'utf-8\'),bytes(orignal,\'utf-8\'), hashlib.sha1).digest()#+\'.\'+bytes(orignal,\'utf-8\')
Sign = base64.b64encode(SignTmp+orignal.encode())
return Sign
\'\'\'生成签名\'\'\'
authorization=hash_hmac(SecretKey,src_str)
\'\'\'定义请求协议\'\'\'
url=\'https://recognition.image.myqcloud.com/ocr/handwriting\'
\'\'\'配置请求头\'\'\'
headers={
\'Authorization\':authorization,
\'Host\':\'recognition.image.myqcloud.com\'
}
# 使用 url 的请求示例
url_img=\'https://images.jiandaoyun.com/Fm0I5jLH9zGFpYn5SLoEP-EhWOmC\'#+\'.png\'
data_img={\'appid\':appid,\'url\':url_img}
\'\'\'
#使用 image 的请求示例
url_img=\'kuaiji-5-243.jpg\',open(\'D:\\python\\kuaiji\\kuaiji-5-243.jpg\',\'rb\'),\'image/jpeg\'
data_img={\'appid\':appid,\'image\':(url_img)}
\'\'\'
r = requests.post(url,files=data_img,headers=headers)
data=json.loads(r.content.decode(\'utf-8\'))
count=len(data[\'data\'][\'items\'])
try:
for i in range(count):
x=data[\'data\'][\'items\'][i][\'itemcoord\'][\'x\']
y=data[\'data\'][\'items\'][i][\'itemcoord\'][\'y\']
content=data[\'data\'][\'items\'][i][\'itemstring\']
print(content)
except:
print(\'wrong\')
代码执行结果