Research And Implementation Of Facerecognition System Based On Labview

Face recognition technology system is applied in the field of biometrics. It is designed to be used as a facial image biometric to be encoded and identified with a unique identifier, now more widely used in the field of security. Due to the controllable rotation characteristics of PTZ(Pan-Tilt-Zoom) camera, PTZ camera based face recognition system can be applied to the public areas of the tracing monitoring or monitoring of the classroom.This article is based on the Labview graphical development environment to build the real-time image acquisition, face detection and recognition, PTZ camera tracking integration of face recognition system, the main research contents and development work is as follows:1. The driving method of PTZ cameras and the real-time image acquisition method, and the image pre-processing algorithm was in-depth studied in this paper. Focusing on analysis and make comparison between OpenCV and IMAQ(Image Acquisition System) which two are all real-time image acquisition method. Taking into account of the image acquisition efficiency and the image format conversion, we take IMAQ related image acquisition components for the method which was used for programming the real-time image acquisition and pre-processing module.2. On the basis of VDM (Vision Development Module) which was a visual development tool belong to Labview, a real-time face detection module has been designed and implemented in this paper. In the implementation process, the human face skin color based matching method and the template matching method have been combined to detect and locate the target human face region, which would be extract in real-time. The experimental results show that when the face matching threshold is set to400, the matching rate of human face detection can reach90%.3. Based on the in-depth research of the real-time tracking control method of PTZ camera, this article presents a kind of implementation and rotation angle calculation of the PTZ camera on two-dimensional plane. The experimental results show that for non-high speed of moving targets, it can realize real-time tracking of the target face image. Through the study of the design and algorithm of the face recognition module, this paper use hybrid programming method based on Labview and OpenCV, by calling the OpenCV correlation function to implement the core of the face recognition algorithm. The experimental results show that the accuracy rate of the face recognition can reach90%for the face belongs to the primitive face database, and then combined with SVM algorithm, the accuracy rate can reach92%or more.