该系统包括：识别前端设备（含显示器、摄像头、RS485、NFC读卡器、补光灯、POE供电模块等）；人脸注册系统（含图片注册、视频注册）；人脸信息管理系统（含用户管理、人脸信息管理、设备管理、设备授权）；API管理系统（含用户注册、人脸注册、人脸信息获取接口）。

模式可选择在线实时识别或离线识别，需要对前端设备进行定制，前端可对门禁、闸机等设备进行控制，后端进行数据处理和信息推送。

单机版识别套件可以通过左图中的设备完成人脸的注册、识别和数据回传功能，体积小巧，功耗低，适合桌面级别的人脸应用场景。

服务器版包括完整的用户管理、人脸管理单元，系统提供标准的b/s架构API接口（RESTFUL）使得二次开发更容易，对各种前端设备和语言兼容性极好，灵活部署，最多支持65536个设备在线。

• 完整的使用说明
• 协助部署该项目
• 完整的知识产权保护和转移手续
• 客户可根据需要定制页面和功能
• 有限的售后服务
• 第三方接入文档

Developing documents

Title:Face recognition system documents
website: http://www.upare.com
Author: Hallwann
Email: webmaster@howwant.com

Codes:
0:Success
1:Unknown error
2:invalid parameter
3:Engine unsuppored
4:Memery error
5:Bad state
6:User cancled
7:Expired
8:Pause
9:Buffer Overflow
10:Buffer Underflow
11:Disk space low
12:Compoent not exsit
13:Member global data not exist
28672:SDK error
28673:Invalid APPID
28674:Invalid DETECTKEY
28675:APPID & DETECTKEY do not match
28676:Invalid SDKKey
28677:System version unsupported
28678:Licence expired
69632:PhotoStyling error
69633:Invalid engine handle
69634:Invalid memery manager handle
69635:Invalid device id
69636:Deivce id unsupported
69637:Invalid model handle
69638:Invalid model size
69639:Invalid image handle
69640:Image format unsupported
69641:Invalid image parameters
69642:Image size too large
69643:CPU unsupport AVX2
73728:Face recognition base error
73729:Invalid memory infomation
73730:Invalid face recognition image information
73731:Invalid face information
73732:No GPU avilable
73733:Feature level mismatched
9001:Device permission denied
9002:You need post parameters
9003:Serial number or device name post error
9004:Device name and the serial number mismatched
9005:Incomplete parameters submitted
9006:No faces detected
9007:Internal function submission error
9111:The image can to be coverted to BGR24
9112:The image can to be coverted to NV21
9113:Fail to read file
9114:Memery error
9115:Function parameter acquisition error
9116:Fail to initialize face detector engine
9117:Fail to initialize face feature engine
9118:Fail to initialize face age engine
9119:Fail to initialize face gender engine
9120:Fail to initialize face match engine
9121:Static face detected Error
9122:Static feature caught Error
9123:Static age estimated Error
9124:Static gender estimated Error
9125:Static face matched Error

APIS:

API:/faces/api/token
Flag:token
Method:post
Usage:Post devices names and serial numbers to get the current token
parameters:
(string)deviceName
(string)sn
return:(json){“flag”:”token”,”code”:0,”token”:”1f3568b85a11dbc4ae0eddf2b6c88c237fbb2d37”}
remarks:Every time you post to server need to get a new token

API:/faces/api/faceDetect
Flag:facedetect
Method:post
Usage:Get the face and its coordinates & angle in the photos you posted
parameters:
(string)token
(file)image
return:(json){“flag”:”facedetect”,”code”:0,”faces”:1,”faceContent”:[{“id”:0,”left”:132,”top”:702,”right”:876,”bottom”:1446,”angle”:1}]}
remarks:angle=1: 0 degree; angle=2: 90 degree; angle=3: 90 degree; angle=4: 180 degree; angle=5: 30 degree; angle=6: 60 degree; angle=7: 120 degree; angle=8: 150 degree; angle=9: 210 degree; angle=10: 240 degree; angle=11: 300 degree; angle=12: 330 degree

API:/faces/api/faceAge
Flag:faceage
Method:post
Usage:Estimate age by face in image
parameters:
(string)token
(file)image
(int)left
(int)top
(int)right
(int)bottom
return:(json){“flag”:”faceage”,”code”:0,”age”:22}

API:/faces/api/faceGender
Flag:facegender
Method:post
Usage:Estimate gender by face in image
parameters:
(string)token
(file)image
(int)left
(int)top
(int)right
(int)bottom
return:(json){“flag”:”facegender”,”code”:0,”gender”:0}
remarks:gender 0=male,gender 1=female, gender -1=unknow

API:/faces/api/faceFeature
Flag:facefeature
Method:post
Usage:Get the picture’s faces feature data(base64)
parameters:
(string)token
(file)image
(int)left
(int)top
(int)right
(int)bottom
return:(json){“flag”:”facefeature”,”code”:0,”feature”:”ry6D1Oh5CldK2naH2v8”,”featureSize”:123}

API:/faces/api/faceRecognize
Flag:facerecognize
Method:post
Usage:Matching two features(base64)’ data to obtain confidence
parameters:
(string)token
(string)pending
(string)original
return:(json){“flag”:”facerecognize”,”code”:0,”score”:0.848038}
remarks:Confidence level >=0.6:match,else not match

API:/faces/api/getDbs
Flag:getDbs
Method:post
Usage:Get array of current device’s faces database
parameters:
(string)deviceName
(string)deviceSN
(string)token
return:(json){‘Flag’:’getdbs’,’code’:’codes’,’message’:’messages’,names:[{‘name’:’name1’,’username’:’username1’,’faceData’:’facedata1’,’picUrl’,’picurl1’},{‘name’:’name2’,’username’:’username2’,’faceData’:’facedata2’,’picUrl’,’picurl2’},…]}

API:/faces/api/regFace
Flag:regFace
Method:post
Usage:Put the uploaded picture’s characteristic points in to datebase (For Mobile phone applications)
parameters:
(string)deviceName
(string)deviceSN
(string)username(unique)
(string)faceName
(string)picBase64(base64)
(string)faceData(base64):The face’s characteristic points to base64 type
return:(json){“flag”:”regFace”,”code”:0}

API:/faces/api/clinetFeedback
Flag:clinetfeedback
Method:post
Usage:Clients/devices upload the recognized face inforamtion to the server
parameters:
(string)deviceName
(string)deviceSN
(string)faceName

return:(json){“flag”:”clinetfeedback”,”code”:0}

API:/faces/api/clinetWebResult
Flag:clinetwebresult
Method:post
Usage:Get clients/devices recognzition results
parameters:
(string)deviceName
(string)deviceSN

return:(json){“flag”:”clinetwebresult”,”code”:0,”faceName”:”admin”,”time”:1550563890}

• Complete instructions for use
• Assist in the deployment of the project
• Complete intellectual property protection and transfer procedures
• Customers can customize pages and functions according to their needs
• Limited after-sales service