iCome:基于多义性的图像检索系统

iCome:image retrieval system based on ambiguity

近年来,多媒体技术的发展使得图像的数量飞速增长,图像检索技术也越来越引起研究者的重视。经过研究人们发现语义鸿沟是导致图像检索系统效果不好的关键因素。以往的系统未能有效解决这一问题。因为多义性是产生语义鸿沟的一个关键因素,所以从多义性的角度构建了iCome图像检索系统。该系统考虑输出空间的多义性实现了基于文本标注的图像检索,考虑输入空间的多义性并结合用户反馈实现了基于内容的图像检索。

Recently,advances in multi-media techniques have greatly increased the number of images.Image retrieval has become a hot topic.It is well-known that a semantic gap is the major problem in content-based image retrieval（CBIR）,while the existing image retrieval systems cannot tackle this problem well.Since the ambiguity is one of the important reasons leading to semantic gap,iCome,an image retrieval system based on learning for ambiguity objects was built.Concerning the ambiguity in output space,iCome implemented text-based image retrieval.Concerning the ambiguity in input space besides relevance feedback,iCome implements content-based image retrieval.