Research on target intrusion detection algorithm based on improved ViBe using infrared thermal imaging

Abstract

The petrochemical industry plays an active role in driving the growth and structural upgrading of the entire national economy. In the storage process of refined oil, personnel theft is an important factor causing economic losses. Using infrared thermal imaging technology to monitor the perimeter of the oil depot can effectively improve the level of security monitoring. According to the application requirements of personnel intrusion detection in oil storage areas, this paper studies the moving target detection method under the static platform, and adopts the improved ViBe moving foreground target detection method to effectively extract the moving foreground and effectively eliminate the small interfering targets. Kalman filter combined with Hungarian algorithm is used to track the moving target. The simulation results show that the algorithm can effectively achieve the effective trajectory prediction and tracking of the moving target. Finally, it is transplanted on the hisilic 3519v101 embedded platform to achieve the requirements of real-time detection.