Linux System Design For Motion Detection

Motion detection research plays a significant role in computer vision technology.It requires instantaneity and analysis of enormous data.Therefore the speed of processing images is the key factor.Taking advantages of System on Chip,such as small size,low power consumption and fast speed as well as its compatibility with OpenCV Base,ARM chips have a bright prospect to serve the development of image processing.Firstly,the research background of motion detection at home and abroad is analyzed,and the current motion detection algorithm is compared in this thesis.Because the ViBe algorithm has good performance on the detection of moving objects,and the algorithm itself is less time-intensive,it is very suitable for running under the ARM architecture.Therefore,the ViBe algorithm is used as the motion detection algorithm in this thesis.The traditional image processing runs under the x86 architecture,and the corresponding processor chip has the disadvantages such as high cost and high power consumption,therefore,the motion detection system based on ARM architecture and Linux system is proposed and implemented in this thesis.In the aspect of hardware design,the image sensor and image data caching scheme are compared and analyzed.Since the description of the details of the image is less demanding,the low cost CMOS sensor OV9650 is used as the image sensor chip in this thesis.And the driver for OV9650 is coded.At the same time,in order to improve the operating efficiency of CPU,the data filling mechanism of existing ring buffer area is improved.Although the Vi Be algorithm has good detecting performance and has the advantages of low time complexity,there are problems of void phenomena and ghost phenomena when detecting moving objects.In order to further improve the accuracy of motion detection,an improved algorithm based on ViBe algorithm is proposed.In order to solve the problem of ghosting phenomenon in the early detection using ViBe algorithm,a joint optimization algorithm of ViBe algorithm and interframe difference algorithm is proposed.The three-frame difference method is used to update the background model of the ViBe algorithm in order to speed up the elimination of the background miscarriage of justice,thus the problem of ghost phenomenon is solved.At the same time,in order to solve the void phenomenon in the actual detection environment using ViBe algorithm,the operation of expansion and corrosion in morphology is employed in this thesis.The proposed algorithm based on Vi Be algorithm solves the problems of void phenomenon and ghost phenomenon,therefor the accuracy of target detection is greatly improved.Finally,the large number of experimental verification and data analysis show that the improved algorithm based on ViBe algorithm under Linux system greatly improves the accuracy of moving object detection,and the application scene of this system has been extended.