An efficient similarity measure (PMM) for color-based image retrieval

Color histogram is an important Technical for color image database indexing and retrieving. However, the main problem with color histogram indexing is that it does not take the color spatial distribution into consideration. Previous researches have proved that the effectiveness of image retrieval increases when spatial feature of colors is included in image retrieval. In this paper, we introduce the local histogram to describe the spatial information of colors, to measure similarity between two images using the local histogram, the traditional approaches use the distance 1 L. To improve performance of the CBIR system, the permuto-metric measure (PMM) is used to measure similarity of images instead of classic distance 1 L. Experiment results prove that the CBIR using our new measure has better performance and perceptually relevant result.