Monitoring System Based On Face Target

Monitoring system technology is a frontier research field involving pattern recognition,image processing,computer vision and artificial intelligence,etc.It has great economic value and broad application prospect,which is widely studied by researchers at home and abroad so that monitoring system technology is one of the hot topics in domestic scholars.The article is used the method of software and hardware combination to avoid the algorithm complexity caused by the over reliance on soft technology in the monitoring system.The research of monitoring system is carried out.In order to be more accurate in the face location,the results of the human face detection were analyzed.In order to be able to track more human face information,different localization algorithms were adopted for different results.On the basis of analyzing the theory of intelligent monitoring system,the composition of monitoring system is determined.According to the function and the connection of each part of the system,the structure of the system is established.Based on the VS2008 environment and Open CV auxiliary library,the scene image is collected by the camera.Using Haar-like features to find the image of the human face feature rectangle,the cascade classifier is generated by Ada Boost algorithm face detection.According to different detection results,different face location algorithm is used for face location.The image display window center is selected as the coordinate origin.The face position error is obtained by the method of linear transformation.Using the mapping method to convert the twodimensional pixel deviation into two-dimensional angle deviation,it is sent to the lower machine Mega2560.According to the received angle deviation information,the single chip controls the camera to make the human face always in the position of the coordinate origin and track face real-time.The 3D model of monitoring device is built on the basis of Open GL.And it is cascaded with the device body to make them unify action so that we can observe the rotation angle at any time according to the three-dimensional model of device body.Finally,according to the information received by the microcontroller,using controlling rudder unit to control the camera to make camera to track face realtime.The external voice control circuit is built based on MCU for controlling the system of the state by the voice chip.Sampling and analysis of face detection and tracking,results show that the system not only can complete single face and multi face detection but also track face accurately.