基于高级视觉特性的感兴趣区域判定算法

提出了一种快速确定视频场景中感兴趣区域的算法，其基本思想是通过对人类视觉系统HVS的某些高级视觉特性进行分析，对图像分段后获得的区域做出相应的视觉重要性估计，从而确定每个区域在主观意义上的视觉感兴趣值，该值的大小反映了其重要性程度。实验结果表明，利用该算法所获得的感兴趣区域基本符合人眼的主观判断，而且算法的复杂度较低，具有较高的实用价值。

ROI decision algorithm based on advanced visual properties

A rapid algorithm to evaluate the importance of different regions in the video scenes was proposed. The principle of the algorithm is to make corresponding visual importance estimates of segmented regions by analyzing the advanced visual properties. As a result, the subjective visual ROI values were ascertained, which reflected the importance of each region. The experimental results show that the ROI values produced by the proposed algorithm accord with human subjective conclusions well. Besides, the computation cost of the algorithm is not high, and it has high utility value.