在移动终端下对基于内容的图像检索技术进行了研究与探讨,运用HSV颜色特征、灰度共生矩阵纹理特征的图像检索算法及查询点移动相关反馈等技术设计一种运行于手持移动终端的便携式图像检索系统.系统采用上位机运算处理,下位机与用户进行交互,并呈现检索结果的策略.结果表明,系统运行稳定,检索速度较快,准确率较高,对移动终端下的图像检索应用技术起到了推动作用.
This paper firstly researches on the technology of content-based image retrieval(CBIR).And then gives the Design and implementation of a portable CBIR system(CBIRS)on handheld mobile terminals.The CBIRS can successfully find similar images by their features including HSV(Hue,Saturation,and Value),gray level co-occurrence matrix and relevance feedback technology.The whole system consists of an upper computer and several lower computers.Retrieval algorithms are processed on the upper computer.Users can interact with the lower computer and check the result of image retrieval.Experiments show the CBIRS runs stably.It can find those similar images fleetly and accurately.The work of this paper will push forward the technology of CBIR on mobile terminals.