Research On Key Technologies Of Intelligent Monitoring System Based On ZYNQ

With the rapid development of China's economy,the continuous improvement of living standards and the strengthening of safety awareness,people's awareness of security and security system is more in-depth.The existing security system mainly focus on video surveillance,usually save the camera's output information at first,then browse and check the video information manually.This manual processing method will inevitably lead to system anomalies,thus the system can't issue warning information in time.This makes a rapid,non-contact identification technology is an urgent need for the monitoring system,the technology can self-analyze the collected video images,confirm the identity of the information in an image,and send the abnormal information to the security personnel in time,to avoid the occurrence of non-security incidents.In this paper,combined with face recognition technology,study the key technology of intelligent monitoring systembased on ZYNQ,and lay the theoretical foundation for the achievement of intelligent monitoring.For the key technologies of intelligent monitoring system,the main contents of this paper are as follows:(1)A face detection algorithm based on LBP and AdaBoost is proposed.The problem of low detection rate caused by external factors such as complex background and uneven illumination is improved.Experimental results show that the algorithm has a strong adaptability to the external factors.(2)A facebook recognition algorithm based on LBP(Local Binary Pattern)and PCA(Principal Component Analysis)is proposed.In order to preserve the details as much as possible,the image should divide into blocks before the face feature is extracted.With the increase of the number of face segments,the dimension of the feature vector will gradually increase,which will lead to a very long time in the late training and recognition process.In this paper,the PCA algorithm is used to reduce the dimension vector,thus reducing the time of training and recognition.The LBP operator is used to extract the facial features,which makes the proposed algorithm robust to external factors such as illumination and complex background.(3)Aiming at the problem of poor real-time ability of the face detection algorithm in the embedded platform,a hardware acceleration method is proposed,which the hardware acceleration for preprocess and detection is actualized in the FPGA part of ZYNQ architecture.The result shows that after acceleration the average time of face detection can run up to 11 ms.In this paper,we mainly study the face detection and face recognition algorithms and its application in intelligent monitoring system,the problem of real-time difference and low recognition rate of face recognition algorithm is studied deeply,then propose a new method.In this paper,we implement the prototype system of intelligent monitoring system by using hardware and software co-design methods.Experiments show that the prototype system achieves the desired goal.