基于压缩域显著点的图像检索方法  

Image retrieval based on salient points in compressed domain

作　　者：张问银[1] 聂真理[1] 

机构地区：[1]临沂师范学院信息学院,山东临沂276005

出　　处：《计算机应用》2005年第10期2354-2355,2361,共3页journal of Computer Applications

基　　金：国家自然科学基金资助项目(60373113)

摘　　要：给出了基于DCT压缩域显著点的快速图像检索方法,该方法直接利用少量DCT非零系数快速计算图像显著点,不需要完全解压缩,降低了计算复杂度,并且具有较强的鲁棒性。实验结果表明,基于显著点提取的图像特征表征能力强,适于图像检索。Compressed-domain image analysis and processing has been an active area in the fields of multimedia research. A fast image retrieval method based on salient points from DCT（Discrete Cosine Transform） compressed domain was presented in this paper, which direct computes the robust salient points from none-zero DCT coefficients without full decompression, so that a lot of computational cost was saved. Experimental results show that the image features obtained based on salient points are suited for image retrieval because of their powerful abilities to characterize the image.

关 键 词：压缩域图像检索 显著点 离散余弦变换 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]