基于边缘特征的自动运动对象分割

为了准确分割出视频场景中的运动对象，该文提出了一种基于边缘特征的运动对象分割及跟踪算法。首先对相邻帧进行自适应变化检测，得到相邻帧二值差分图像。结合当前帧Canny算子检测的边缘图像，获得运动对象的初始边缘模板。其次对运动对象的运动分为快变和慢变两部分进行跟踪并更新运动对象的边缘模板。最后对运动对象的边缘模板进行数学形态学处理得到运动对象的外轮廓，使用梯度向量流场作为外力的改进活动轮廓算法收缩获得运动对象准确的闭合轮廓曲线。该算法对运动对象的整体运动和局部形变都有很强的鲁棒性, 能够得到运动对象准确的轮廓，并且对复杂背景有很好的适应性。

Automatic Moving Object Segmentation Based on Edge Features

In order to segment the moving objects accurately, an edge based moving objects segmentation algorithm is proposed in this paper. First of all, the initial edge model of thc moving object is obtained on the basis of the adaptive change detection method and Canny edge detection. Secondly, the binary edge model is tracked and updated to accommodate rotation and changes in shape of the tracked object. Finally, the outer contour of the moving object is determined based on the mathematical morphological operation and an improved active contour which uses the gradient vector as the external force guides the initial outer contour moving to the actual contour. The proposed algorithm is robust to the entire motion and local deformation of object, adaptable to the complex background and can obtain the accurate moving object contour.