Image retrieval system based on color layout descriptor and Gabor filters

Abstract

The current paper presents a content-based image retrieval (CBIR) system using the image features extracted by a color layout descriptor (CLD) and Gabor texture descriptor. CLD represents the spatial distribution of colors with a few nonlinear quantized DCT coefficients of grid-based average colors, whereas the Gabor filter works as a bandpass filter for the local spatial frequency distribution. These two descriptors are very powerful for CBIR systems. Furthermore, combining the color and texture features in CBIR systems leads to more accurate results for image retrieval. To compare the performance of image retrieval method, average precision and recall are computed for all queries. The results showed an improved performance (higher precision and recall values) compared with the performance using other CBIR methods.