中文摘要：

提出了一种基于证据融合的视频语义概念检测方法。提取了镜头关键帧的分块颜色矩、小波纹理特征和视觉词汇直方图，利用SVM对3种特征数据分别进行训练，建立模型；对各SVM模型泛化误差进行分析，采用折扣系数法对不同SVM模型输出的分类结果进行修正；采用基于min—max算子的证据融合公式对修正后的输出进行融合，把融合结果作为最终的概念检测结果。实验结果表明，新方法提高了概念检测的准确率，优于传统的线性分类器融合方法。

英文摘要：

A video semantic concept detecting method based on evidence fusion is proposed. Features including grid color moment, wavelet texture and visual word histogram are extracted from key frames of video shots, and for each type of feature, a SVM model is trained. By analyzing the generalization error of each SVM model, a discounting coefficient method is implemented to modify classification results of these models. Then these modified results are fused with an evidence fusion equa- tion based on min-max operator, and the fused result is regarded as the final semantic concept detection result. Experimental results show that the new method has improved the detection accuracy and outperforms the traditional linear classifier fusion method.