调用百度AI开放平台实现图片文字识别

版权声明：转载请说明来源，谢谢 https://blog.csdn.net/wsp_1138886114/article/details/83107004

闲话少说：
注册百度账号，获取APP_ID，API_KEY，SECRECT_KEY
查看SDK，下面是我调用识别图片

import os
import csv

from aip import AipOcr

APP_ID = '1180xxxx'
API_KEY = 'cChcRf45eLIRSQTqxxxxxxxx'
SECRECT_KEY = '12xG8GVcqS6I8doYwcSyGaCexxxxxxxx'
client = AipOcr(APP_ID, API_KEY, SECRECT_KEY)

input_dir = "./cut_picture"
list1 = []
for (path, dirnames, filenames) in os.walk(input_dir):
    list1.append(filenames)


def get_file_content(filePath):
    with open(filePath, 'rb') as fp:
        return fp.read()


Result = []
for j in list1[0]:
    path1 = input_dir + "/" + j
    print("药品图片文字：", path1)
    image = get_file_content(path1)

    """ 调用文字识别 """
    result = client.basicAccurate(image)
    print("识别输出：", result)
    Result.append(result)

    """ 如果有可选参数 """
    options = {}
    options["detect_direction"] = "true"
    options["probability"] = "true"

    """ 带参数调用通用文字识别（高精度版） """
    client.basicAccurate(image, options)
    print(result)

with open("./药品图片文字识别结果.csv", "w") as csvfile:
    write = csv.writer(csvfile)
    write.writerow(Result)