International Journal of Computer Science and Information Technologies
In this paper, the authors propose a new and efficient technique to retrieve the images based on Dominant Color, texture and Pseudo Zernike moments to achieve the higher retrieval efficiency and performance. First an image is uniformly divided into 8 partitions. After the above partition, the centroid of the each partition is selected as its quantized color. Texture of an image is obtained by using steerable filter. Shape feature of an image is obtained by using Pseudo Zernike moments. The combination of the Dominant color, Texture and Pseudo Zernike moments provide a robust feature set for image retrieval. Euclidean distance of color, texture and shape features is used in the retrieving the similar images.