识别效果

OCR（optical character recognition）文字识别服务模块，使用的google的开源OCR识别库Tesseract-OCR，支持多种语言，也可训练自己的语言库，结合jTessBoxEditorFX工具训练自定义识别类型，另提供通过百度AI的ORC识别方式

IMAGE_OCR

图像文字识别指令

使用google的开源OCR识别库Tesseract-OCR，通过上传图片，获取图片中的文字信息，支持图片格式 .jpg, .jpeg, .png, .bmp, .tif, .tiff, .webp

效果

BAIDU_OCR

图像文字识别指令

使用百度AI的OCR在线识别SDK，通过上传图片，获取图片中的文字信息，支持图片，url和PDF

返回数据

{
    "kind": 7002, 
    "data": {
        "words_result": [
            {"words": "江南"}, {"words": "汉乐府"}, {"words": "江南可采莲，"}, 
            {"words": "莲叶何田田。"}, {"words": "鱼戏莲叶间，"}, {"words": "鱼戏莲叶东,"}, 
            {"words": "鱼戏莲叶西,"}, {"words": "鱼戏莲叶南,"}, {"words": "鱼戏莲叶北。"}, 
            {"words": "长歌行"}, {"words": "汉乐府"}, 
            {"words": "青青园中葵,朝露待日晞。"}, 
            {"words": "阳春布德泽,万物生光辉。"}, {"words": "常恐秋节至,焜黄华叶衰。"}, 
            {"words": "百川东到海，何时复西归？"}, {"words": "少壮不努力，老大徒伤悲。"}
        ], 
        "words_result_num": 16, 
        "log_id": 1633339323063486109
    }
}