录播系统中图像分割与轮廓提取算法的研究

在基于图像识别技术的录播系统中,通常使用背景差分法检测运动目标,在当前帧与背景帧做差之后,需要进一步选取合适的阈值进行图像分割,从而最终提取出运动目标的轮廓.研究了基于最大类间方差准则的图像阈值分割,在此基础上进一步研究了掏空内部点的轮廓提取算法,并且使用轮廓缺陷修补技术改进轮廓提取算法,从而得到封闭的轮廓,最后使用Freeman链码表示提取的轮廓.实验证明,该算法为录播系统的实现提供了良好的技术保障.

Research on Image Segmentation and Contour Extraction in Video Recording and Broadcasting System

The?recording and broadcasting system?based on?image recognition technology usually?uses the background?subtraction algorithm to detect the moving object.?After making the difference between the current?frame and the background frame, we need to select the appropriate threshold for image segmentation in order to extract the moving object`s contour .This paper studies the image threshold segmentation based on Otsu criterion, further studies the contour extraction algorithm based on the hollowed interior points and uses the technology of contour defect repair to improve the contour extraction algorithm, then gets the closed contour. And finally uses the Freeman chain code represent the contour. Experimental results show that the proposed algorithm provides a good technical support to realize the recording and broadcasting system.