基于混合高斯模型的运动目标检测算法

针对智能交通系统中运动目标检测阶段存在的不足,提出了一种基于自适应混合高斯模型(GMM)的改进算法。将隔帧差分的方法引入背景建模的初始判别阶段,从而迅速地检测出运动变化区域,提高了算法的灵敏度,同时也增强了对缓慢运行车辆的检测的适用性;将划分出的背景及运动区域赋予不同的更新率,使得背景显露区域得到迅速恢复,消去了运动车辆留下的"影子"。在此较为精确的背景模型下,结合灰度和canny边缘特征进行背景差分,有效地保留了与背景灰度相似的运动目标的轮廓。通过实验证明该检测算法取得了较好的效果。

Moving object detection algorithm based on Gaussian mixture model

An improved algorithm based on adaptive Gaussian mixture model is proposed to overcome the deficiency in the object detection phase of ITS(intelligent transportation system).At the beginning of background model the interval frame differencing method is put forward to make a judgment,which can rapidly detect the movement area.The sensitivity of the algorithm is improved and the adaptability for slow vehicles is enhanced.In order to recover background covered by paused objects when they start to move again,the different update rates in different areas are given,which can eliminate shadow that moving vehicles may leave.Lastly by combining gray and canny edge features background subtraction,moving target outline similar with the background’s gray is effectively retained.Experimental results confirm the good effort of the detection algorithm.