python sdk for 腾讯优图开放平台
- Requests,获取更新版本
http://docs.python-requests.org/en/latest/
1. 下载sdk到您的目录${python_sdk}
git clone https://github.com/Tencent-YouTu/Python_sdk.git
2. 在您需要使用sdk的文件中import TencentYoutuyun包
AppId
平台添加应用后分配的AppIdSecretId
平台添加应用后分配的SecretIdSecretKey
平台添加应用后分配的SecretKey签名
接口鉴权凭证,由AppId、SecretId、SecretKey等生成,详见http://open.youtu.qq.com/#/develop/tool-authentication
# -*- coding: utf-8 -*-
import time
import TencentYoutuyun
appid = 'xxx'
secret_id = 'xxxxxxx'
secret_key = 'xxxxxxxx'
userid= 'xxx'
#end_point = TencentYoutuyun.conf.API_TENCENTYUN_END_POINT // 腾讯云
#end_point = TencentYoutuyun.conf.API_YOUTU_VIP_END_POINT // 人脸核身服务(需联系腾讯优图商务开通权限,否则无法使用)
end_point = TencentYoutuyun.conf.API_YOUTU_END_POINT // 优图开放平台
youtu = TencentYoutuyun.YouTu(appid, secret_id, secret_key, userid, end_point)
ret = youtu.FaceCompare('you_path_one.jpg','you_path_two.jpg')
print ret
IMAGE_FILE_NOT_EXISTS -1 //文件不存在
IMAGE_NETWORK_ERROR -2 //网络错误
IMAGE_PARAMS_ERROR -3 //图片参数错误
PERSON_ID_EMPTY -4 //参数person_id 为空
GROUP_ID_EMPTY -5 //参数group_id 为空
GROUP_IDS_EMPTY -6 //参数group_ids 为空
IMAGES_EMPTY -7 //参数images 集合为空
FACE_IDS_EMPTY -8 //参数face_ids 集合为空
FACE_ID_EMPTY -9 //参数face_id为空
LIST_TYPE_INVALID -10 //不是list类型
IMAGE_PATH_EMPTY -11 //传入的image_path为空
OCR_NOT_ENOUGH_TEXTLINES -5201 //名片无足够的文本
OCR_TEXTLINES_SKEWED -5202 //名片文本行倾斜角度太大
OCR_TEXTLINES_FUZZY -5203 //名片模糊
OCR_UNRECOG_NAME -5204 //名片姓名识别失败
OCR_UNRECOG_TEL -5205 //名片电话识别失败
- 示例
- `youtu = TencentYoutuyun.YouTu(appid, secret_id, secret_key, userid, end_point)`
- 参数`ytopen_sdk::AppSign`
- `appid` 业务中的应用标识AppId
- `secret_id` 秘钥SecretId
- `secret_key` 秘钥SecretKey
- `userid` 用户id
- `end_point` 服务后台路径,默认是优图开放平台,支持腾讯云,人脸核身(核身服务需联系腾讯优图商务开通权限)
- API分为开放平台API和人脸核身API,人脸核身API访问权限需要联系商务开通;
- 开放平台API访问end_point为TencentYoutuyun.conf.API_YOUTU_END_POINT
- 人脸核身API访问end_point为TencentYoutuyun.conf.API_YOUTU_VIP_END_POINT
- 所有接口,对于编码问题,涉及到中文的请使用encode('iso8859-1').decode('utf-8')进行转码
接口调用统一返回值说明
- 返回值
`Json`格式的返回结果,具体字段参考API文档
- 人脸核身服务接口列表(需联系腾讯优图商务开通权限,否则无法使用)
- 接口end_point选择: TencentYoutuyun.conf.API_YOUTU_VIP_END_POINT
- livegetfour(self, seq = '')
- livedetectfour(self, validate_data, video_path, seq = '', card_path = '', compare_flag = False)
- idcardlivedetectfour(self, idcard_number, idcard_name, validate_data, video_path, seq = '')
- idcardfacecompare(self, idcard_number, idcard_name, image_path, data_type = 0 , session_id = '')
- FaceCompare(self, image_pathA, image_pathB, data_type = 0)
- idcardocr(self, image_path, data_type = 0, card_type = 1 ,seq = '')
ValidateIdcard(self, idcard_number, idcard_name, seq = '')
- 腾讯优图开放平台接口列表
- 接口end_point选择: TencentYoutuyun.conf.API_YOUTU_END_POINT
- DetectFace(self, image_path, mode = 0, data_type = 0)
- FaceShape(self, image_path, mode = 0, data_type = 0)
- FaceCompare(self, image_pathA, image_pathB, data_type = 0)
- FaceVerify(self, person_id, image_path, data_type = 0)
- FaceIdentify(self, group_id, image_path, data_type = 0)
- MultiFaceIdentify(self, group_id, group_ids, image_path, data_type = 0, topn = 5, min_size = 40)
- NewPerson(self, person_id, image_path, group_ids, person_name= '', tag='', data_type = 0)
- DelPerson(self, person_id)
- AddFace(self, person_id, images, tag='', data_type = 0)
- DelFace(self, person_id, face_ids)
- GetInfo(self, person_id)
- SetInfo(self, person_id, person_name='', tag='')
- GetGroupIds(self)
- GetPersonIds(self, group_id)
- GetFaceIds(self, person_id)
- GetFaceInfo(self, face_id)
- fuzzydetect(self, image_path, data_type = 0, seq = '')
- fooddetect(self, image_path, data_type = 0, seq = '')
- imagetag(self, image_path, data_type = 0, seq = '')
- imageporn(self, image_path, data_type = 0, seq = '')
- imageterrorism(self, image_path, data_type = 0, seq = '')
- carclassify(self, image_path, data_type = 0, seq = '')
- idcardocr(self, image_path, data_type = 0, card_type = 1 ,seq = '')
- driverlicenseocr(self, image_path, data_type = 0, proc_type = 0, seq = '')
- bcocr(self, image_path, data_type = 0, seq = '')
- generalocr(self, image_path, data_type = 0, seq = '')
- creditcardocr(self, image_path, data_type = 0, seq = '')
- bizlicenseocr(self, image_path, data_type = 0, seq = '')
- plateocr(self, image_path, data_type = 0, seq = '')
- 接口end_point选择: TencentYoutuyun.conf.API_YOUTU_VIP_END_POINT
`livegetfour(self, seq = '')`
-参数
- `seq` 请求唯一标识
`livedetectfour(self, validate_data, video_path, seq = '', card_path = '', compare_flag = False)`
-参数
- `validate_date` livegetfour接口获取的四字唇语返回值
- `video_path` 用户录制视频本地路径
- `seq` 请求唯一标识
- `card_path` 当compare_flag字段为true时,用于与视频提取图像作人脸对比的图像本地路径
- `compare_flag` 决定是否需要作人脸对比,默认为false
`idcardlivedetectfour(self, idcard_number, idcard_name, validate_data, video_path, seq = '')`
-参数
- `idcard_number` 带网纹图像对应的身份证号码
- `idcard_name` 带网纹图像对应的身份证姓名
- `validate_data` livegetfour接口获取的四字唇语返回值
- `video_path` 用户录制视频本地路径
- `seq` 请求唯一标识
`idcardfacecompare(self, idcard_number, idcard_name, image_path, data_type = 0 , session_id = '')`
-参数
- `idcard_number` 带网纹图像对应的身份证号码
- `idcard_name` 带网纹图像对应的身份证姓名
- `data_type` 0表示本地图像,1表示url下载图像
- `image_path` 当data_type字段为0时,则image_path存放图像本地路径;data_type字段为1时,则image_path存放url路径
- `session_id` 请求唯一标识
- 接口
`FaceCompare(self, image_pathA, image_pathB, data_type = 0):`
- 参数
- `image_pathA` 待比对的A图片路径
- `image_pathB` 待比对的B图片路径
- `data_type` 用于表示image_pathA, image_pathB是图片还是url, 0代表图片,1代表url
`idcardocr(self, image_path, data_type = 0, card_type = 1 ,seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- `card_type` 0 代表输入图像是身份证正面, 1代表输入是身份证反面
- `seq` 请求唯一标识
`ValidateIdcard(self, idcard_number, idcard_name, seq = '')`
- 参数
- `idcard_number`身份证号
- `idcard_name` 身份证姓名
- `seq` 请求唯一标识
- 接口end_point选择: TencentYoutuyun.conf.API_YOUTU_END_POINT
- 接口
`DetectFace(self, image_path, mode = 0, data_type = 0)`
- 参数
- `image_path` 待检测的图片路径
- `mode` 是否大脸模式,默认非大脸模式
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- 接口
`FaceShape(self, image_path, mode = 0, data_type = 0)`
- 参数
- `image_path` 待检测的图片路径
- `mode` 是否大脸模式,默认非大脸模式
- `data_type` 用于表示image是图片还是url, 0代表图片,1代表url
- 接口
`FaceCompare(self, image_pathA, image_pathB, data_type = 0):`
- 参数
- `image_pathA` 待比对的A图片路径
- `image_pathB` 待比对的B图片路径
- `data_type` 用于表示image_pathA, image_pathB是图片还是url, 0代表图片,1代表url
- 接口
`FaceVerify(self, person_id, image_path, data_type = 0)`
- 参数
- `person_id` 待验证的个体id
- `image_path` 待验证的图片路径
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- 接口
`FaceIdentify(self, group_id, image_path, data_type = 0)`
- 参数
- `group_id` 识别的组id
- `image_path` 待识别的图片路径
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- 接口
`MultiFaceIdentify(self, group_id, group_ids, image_path, data_type = 0, topn = 5, min_size = 40)`
- 参数
- `group_id` 识别的组id
- `group_ids` 识别的个体存放的组id,可以指定多个组id,用户指定
- `topn` 候选人脸数量,一般使用默认值5
- `min_size` 人脸检测最小尺寸,一般使用默认值40
- 接口
`NewPerson(self, person_id, image_path, group_ids, person_name= '', tag='', data_type = 0)`
- 参数
- `person_id` 新建的个体id,用户指定,需要保证app_id下的唯一性
- `person_name` 个体对应的姓名
- `group_ids` 数组类型,用户指定(组默认创建)的个体存放的组id,可以指定多个组id
- `image_path` 包含个体人脸的图片路径
- `tag` 备注信息,用户自解释字段
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- 接口
`DelPerson(self, person_id)`
- 参数
- `person_id` 待删除的个体id
- 接口
`AddFace(self, person_id, images, tag='', data_type = 0)`
- 参数
- `person_id` 新增人脸的个体身份id
- `images` 数组类型,待增加的包含人脸的图片路径,可加入多张(包体大小<2m)
- `tag` 人脸备注信息,用户自解释字段
- `data_type` 用于表示images是图片还是url, 0代表图片,1代表url
- 接口
`DelFace(self, person_id, face_ids)`
- 参数
- `person_id` 待删除人脸的个体身份id
- `face_ids` 数组类型,待删除的人脸id
- 接口
`GetInfo(self, person_id)`
- 参数
- `person_id` 待查询的个体身份id
- 接口
`SetInfo(self, person_id, person_name='', tag='')`
- 参数
- `person_id` 待设置的个体身份id
- `person_name` 新设置的个体名字,为空无效
- `tag` 新设置的人脸备注信息,为空无效
- 接口
`GetGroupIds(self)`
- 参数
- 无
- 接口
`GetPersonIds(self, group_id)`
- 参数
- `group_id` 待查询的组id
- 接口
`GetFaceIds(self, person_id)`
- 参数
- `person_id` 待查询的个体id
- 接口
`GetFaceInfo(self, face_id)`
- 参数
- `face_id` 待查询的人脸id
`fuzzydetect(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`fooddetect(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`imagetag(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`imageporn(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`imageterrorism(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`carclassify(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`idcardocr(self, image_path, data_type = 0, card_type = 1 ,seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- `card_type` 0 代表输入图像是身份证正面, 1代表输入是身份证反面
`driverlicenseocr(self, image_path, data_type = 0, proc_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
- `proc_type` 0代表行驶证, 1代表驾驶证
`bcocr(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`generalocr(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`creditcardocr(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`bizlicenseocr(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
`plateocr(self, image_path, data_type = 0, seq = '')`
- 参数
- `image_path` 标识图片信息
- `data_type` 用于表示image_path是图片还是url, 0代表图片,1代表url
更多详情和文档说明参见 腾讯优图开放平台