Video image moving object extraction method and system of segmented Gaussian mixture model

The invention provides a video image moving object extraction method and system of a segmented Gaussian mixture model. The method includes the steps of training and obtaining an effective Gaussian mixture model; using a frame difference method to segment frames in a video image; adjusting a local threshold of the Gaussian mixture model within each segmented frame; according to the adjusted local threshold of the Gaussian mixture model, classifying pixels within each segmented frame, and finally extracting a moving object. The video image moving object extraction system of the segmented Gaussian mixture model is used to implement the method. The method and system of the invention have better inhibition effect of dynamic disturbance when the moving object is extracted. 本发明提供了分段混合高斯模型的视频图像运动目标提取方法及系统，所述方法包括步骤：训练并得到有效的混合高斯模型；采用帧差分法对视频图像中的帧进行分段；在每段分好的帧内调整混合高斯模型的局部阈值；根据调整好的混合高斯模型的局部阈值，在每段分好的帧内将像素点分类，最终提取运动目标。分段混合高斯模型的视频图像运动目标提取系统，用来实现所述方法。本发明在提取运动目标时对动态干扰有较好的抑制作用。