他的回复:
可以试试用python的ddddocr ,你附件里的20个,查不到识别了14个出来,70%的成功率。代码如下:import ddddocr image_path = r"C:/Users/admin/Desktop/验证码" ocr = ddddocr.DdddOcr() img_bytes = None for var in range(20): path = image_path+"/" + str(var)+".png" with open(path,'rb') as f: img_bytes = f.read() strCode = ocr.classification(img_bytes) print(strCode)