向量机的视频检索应用

提出一种基于支持向量机的视频检索方法,用于复杂背景灰度图像的识别。算法首先用相似度分析方法进行粗筛选,滤去大量非目标窗口,之后用支持向量机对通过的窗口进行分类。由于在通过相似度分析方法所限定的子空间内训练SVM,有效地降低了训练的难度。实验对比数据表明,该方法降低了分类器的训练难度,计算复杂度较低,大大提高了检测速度。

An efficient method of image detection based on Support Vector Machine(SVM)is proposed in this paper.Firstly,a similar coarse filter with relatively lower computational complexity is applied to the whole input image to iflter out target of the non-target,then fol ows the SVM classiifer to make the ifnal decision,so the detection process is speeded up.The experiment results show that the method can effectively detect image under complicated background,and the processing time is shorter than using SVM alone.