摘要
针对传统混合高斯背景算法在进行运动目标检测时出现拖影和碎化、算法性能差等缺点,融合背景差分、自适应阈值与高斯成分个数自适应算法提出基于区域训练的自适应背景建模进行改进.检测时,通过待检帧与构建的背景模型进行差分及对差分图像自适应阈值分割获得运动目标与背景区域.进行背景模型训练时,运动目标区域像素保持其混合高斯模型不变,背景区域的像素正常训练并实现高斯成分个数自适应算法,使混合高斯模型高斯成分仅由实际背景像素构建和产生,提高算法的性能与背景模型构建的有效性.实验表明,该算法对有诸多不确定性的视频序列构建的背景模型都有较好的适应性,能消除拖影和碎化,计算速度有一定提高,能快速响应实际场景的变化.
The detection result of traditional GMM algorithm easily becomes fragmentary and exists shadow, the fixed number of Gaussian component leads to bad performance. Aiming at improving these defects, using the background difference method and self-adaptive threshold segmentation, self-adaptive Gaussian mixture background modeling based on area training is proposed to improve it. The background difference method and self-adaptive threshold segmentation classify the pixels in each frames into moving targets and background area. When training the background model, this new algorithm keeps the Gaussian mixture background model of the pixels in moving target area unchanged and never build new Gaussian component for this area. Background area is updated in regular way, make the number of Gaussian component for each pixel in this area to be self-adaptive, keep the Gaussian component of background model only updated by the real background pixels, improved the performance of the algorithm and validity for background construcring. Experiments show that the background model built based on the proposed algorithm for video sequences with uncertainties has good adaptability, it can eliminate the shadows and quickly response to the change of the actual scene, the computing speed of this model improves a lot as well.
出处
《小型微型计算机系统》
CSCD
北大核心
2015年第5期1104-1108,共5页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61170277)资助
上海市教委科研创新重点项目(12zz137)资助
沪江基金项目(C14002)资助
关键词
GMM
背景差分法
自适应阈值分割
高斯成分个数自适应
区域训练
Gaussian mixture model
the background difference method
the adaptive threshold segmentation
self-adaptive of Gaussian component number
area training
