The Research And Implementation Of Background Modeling And Updating Algorithm Based On Mixture Gaussian Model

In computer vision and intelligent video surveillance community, the higher levels such as moving object classification, tracking and behavior understanding heavily depend on the results of moving object detection. Moving object detection is one of the fundamental tasks for video analysis, which is active topic widely researched in the last decade around the world. In this thesis, we focus on the background modeling of complex environment and moving object detection, and place emphasis on mixture Gaussian model(GMM) and updating background especially. The main contributions are as follows:1. This paper researches GMM through systematical theoretical derivation and experimental analysis.2. Several background modeling and updating algorithms are widely researched and analyzed based on extensive experiments. We place emphasis on single Gaussian model(GSM) and mixture Gaussian model(GMM) and updating background especially.3. A new method about the initialization of mixture Gaussian parameters is proposed to solve the problem of inaccurate background model. And considering the different characteristics of the mean and variance, the confidence of each point to the background, the relevance of the new obtained frame and the background model, we improve the classical mixture Gaussian background updating algorithm.4. The moving object detection is implemented after a series of post-processing such as morphological processing, connected component labeling, filter processing, positioning, and objectives of the merger movement of regional identity, merging moving regions and moving object labeling.5. The practical applications of moving object detection based on mixture Gaussian background modeling and updating algorithm are implemented such as moving object detection and object detection around.