视频检索中图像信息量度量

Image information measurement for video retrieval

综合考虑信息量度量的速度、性能要求,提出了相适应的显著图、多特征融合模型;基于区域划分融入空间关系,提出了分块信息熵的图像信息量度量方法(SEII);构建了信息量度量的标注数据集,并设计了性能验证方法。实验结果表明该度量方法符合人眼视觉的评价结果。度量方法在实际视频检索系统中进行对比应用测试,测试表明m AP提高4.4%,检索速度提高1.5倍。

To meet the speed and performance requirements, Sub-region entropy based image information measurement （SEII） method was proposed, which integrates the salient region detection, region division and features fusion. And, per- formance evaluation method was designed and many experiments were carried out, proving SEII coordinates with human vision evaluation. Also, SEII is evaluated in a real video retrieval system, which shows increase about 4.4% of mAP with 1.5 times speedup.