【摘要】 为了对海量视频数据进行有效的管理和快速浏览,急需对数字视频进行基于内容的视频检索。镜头分类是足球视频处理与检索的重要部分,针对目前现有足球镜头分类方法存在算法准确性不高或运算量过大的问题,提出了一种新的基于子窗口区域的镜头分类方法。该方法采用在HSV颜色空间中计算足球视频帧子窗口区域球场色像素比率,并辅以边缘信息的检测,对足球视频中的主镜头、中镜头、特写镜头和其他镜头进行了分类,实验结果表明该方法切实可行,具有很高的检出率和准确率。更多还原

【Abstract】 It is urgently needed to develop a solution to content-based video retrieval,in order to support efficient management and quick browsing mass video data.Shot classification is an important research content in soccer video processing and retrieval.Considering the shortcomings of the existed methods,we propose a novel approach which can classify soccer video shots into main,middle,close-up shots and other ones based on sub-window region.The key step is to calculate the proportion of pixels with field color in sub windows region of soccer video frame in HSV color space.Experimental results have shown that the proposed method has a very high precision and recall rate.更多还原