Moving objects detection algorithm based on three-dimensional Gaussian mixture codebook model

作者：Huang Jin^*; Jin Weidong; Qin Na

来源：Journal of Southwest Jiaotong University, 2012, 47(4): 662-668.

DOI：10.3969/j.issn.0258-2724.2012.04.020

• Adjusting parameters
• Bayes models
• Code-words
• Codebook models
• Codebooks
• Comparative experiments
• Frame rate
• Frames per seconds
• Gaussian Mixture Model
• Gaussian mixtures
• Gaussian model
• Moving objects detection
• Object Detection
• RGB space
• Visual surveillance

In order to solve the difficulty of adjusting parameters for the codebook model and the computational complexity of probability distribution for the Gaussian mixture model in intelligent visual surveillance, a moving objects detection algorithm based on the three-dimensional Gaussian mixture codebook model was proposed. In this algorithm a codebook model based on RGB space is built, and then a three-dimensional Gaussian model based on R, G and B components in codewords is established. In this way, the characteristic of the three-dimensional Gaussian mixture model for the codebook model can be obtained. The experimental results show that the average frame rate of the proposed algorithm is about 23.0 frames per second, while it is about 9.0 frames per second for the iGMM (improved Gaussian mixture model) algorithm, about 6.2 frames per second for the BM (Bayes model) algorithm, and about 10.7 frames per second for CBM (codebook model) algorithm in the comparative experiments. Furthermore, the proposed algorithm can obtain a good detection quantity.

• 出版日期2012
• 单位西南交通大学