Research of Image Retrieval Technique Based on Integrated Moments
Yuanhui Yu · Journal of Shenyang University of Chemical Technology · 2011
The image retrieval technique based on content becomes a hot research topic.In order to solve the problem of global color moments lacking the spatial distribution information of color and low retrieval efficiency of the traditional integrated features,the energy moment based on discrete cosine transform is presented as an efficient feature descriptor.Based on this,a new approach for image retrieval using combined annular color moments and energy moments features is proposed.The experimental results show that the accuracy and speed of the proposed image retrieval method based on annular color moments and energy moments features outperform those of the traditional integrated color-texture method,and also keep the characteristic of rotation and scaling invariance.