Three registration methods, bid farewell to the rush hour in the economic peak, often encounter too many visitors, registration and reception is not timely.
The front desk should not only register information, but also inform the cooperative employees in the company to come to the reception in a hurry.
During this period, visitors may have to wait for a long time, affecting the visit experience and reducing the corporate image score.
Different from the traditional handwritten registration, the Yunmai visitor management system adopts three modes of Registration: business card recognition registration, ID card recognition registration, and face recognition registration.
When registering, the visitor only needs to provide a business card or ID card. The system will automatically scan and read the ID information and save it. At the same time, the camera at the front desk will not capture the visitor's face and perform additional face registration.
Face recognition re-registration, one-touch notification cooperators use business card recognition and ID card recognition technology to assist in registration information, not only faster, but also more complete registration information, plus face registration, effectively overcome the problems of incomplete registration information, scribbled registration and inconsistent certificates.
In addition, there is an additional benefit of face registration.
Visitors will leave face information after * visits. When visitors visit again, they do not need to show their ID cards and business cards. The camera detects past Face Records, the system will automatically notify the remark partners to come out for reception through mail, short messages, mobile phones, chat tools and other means, and reduce the visitor registration and reception work in one step. 1. Face detection face detection (Face Detection)
It is a technology to detect the position of the face in the image.
The input of the face detection algorithm is a picture, and the output is a face frame coordinate sequence (
0 face frame or 1 face frame or multiple face frames).
In general, the output face coordinate frame is a square facing up, but there are also some face detection techniques that output a rectangle facing up or a rectangle with a rotation direction.
The common face detection algorithm is basically a process of scanning and discrimination, that is, the algorithm scans within the image range and then determines whether the candidate regions are faces one by one.
Therefore, the calculation speed of face detection algorithm will be related to image size and image content.
During the development process, we can speed up the algorithm by setting the input image size, or * small face size limit, or the upper limit of the number of faces. 2. Face registration face registration (Face Alignment)
It is a technique to locate the coordinates of the key points of the facial features.
The input of the face registration algorithm is a face picture plus a face coordinate frame, and the coordinate sequence of the key points of the facial features is output.
The number of key points of facial features is a preset fixed value, which can be defined according to different semantics (
Common are 5 points, 68 points, 90 points, etc).
Some face registration technologies with better results at present are basically realized through the deep learning framework. These methods are all based on the coordinate frame of face detection and deduct the face area according to some preset rules, scale the fixed size, and then calculate the position of the key point.
Therefore, if the time consumption of the image scaling process is not taken into account, the face registration algorithm can be a process with a fixed amount of computation.
In addition, compared with face detection or The Face feature extraction process mentioned later, the calculation time of face registration algorithm is much less.