1.导入需要的库,包括百度的api接口跟cv2图像截图图库
import cv2 from aip import AipOcr # 读取图片,利用imshow显示图片 pic = cv2.imread(r'Y:\cut\img1.png') pic = cv2.resize(pic,None,fx = 0.5, fy = 0.5) cv2.imshow('img',pic) cv2.waitKey(0)
2.截取图片,获取需要的信息,包括以下信息
# 删除不必要的部分 img = pic[210:500, 100:580] # 截取各部分的文字 time = pic[400:430, 100:580] business = pic[370:400, 100:580] goods = pic[350:380, 100:580] money = pic[210:300, 100:580] num = pic[460:500, 100:580] # 查看截取的部分是否合适 gener_name = ['time','business','goods','money','num'] excel_data = {} pd_columns = ["a","b","c","d","e"] # 标题
3.定义函数将截取好的图片另存到文件夹
def shotcut_image(args): for index in gener: cv2.imwrite('image/{}.png'.format(args), img)
4.调用百度api接口,实现文字识别
# 导入api AppID = '24177719' API_Key = 'p8skmRYfHGoVGR4UU03Q5jiM' Secret_Key = 'dyM0tzSILBZu9CFqZ7IkjWwECGaws4xo' cilent = AipOcr(AppID,API_Key,Secret_Key) def get_words(img_name): with open('image/{}.png'.format(img_name), 'rb') as f: result = cilent.basicAccurate(f.read()) return result
5.最后将信息转为Dataframe,利用pandas的to_exccel功能,将数据放到excel里面
def convert_to_dataframe(words): # 构建dataframe result = words['words_result'] for word in result: excel_data.setdefault('a', []).append(word['words']) # 将所有words读取后,取出语句存入excel def convert_to_excel(): frame = DataFrame(excel_data, columns=pd_columns) # todo 表头需要额外处理,这里指定不设置表头 frame.to_excel('out.xls',index=False, header=False)
免责声明:本站发布的内容(图片、视频和文字)以原创、转载和分享为主,文章观点不代表本网站立场,如果涉及侵权请联系站长邮箱:mmqy2019@163.com进行举报,并提供相关证据,查实之后,将立刻删除涉嫌侵权内容。
长按识别二维码并关注微信
更方便到期提醒、手机管理