IMAGE RANKING BASED ON TEXTURE FEATURE USING CONTENT-BASED INFORMATION RETRIEVAL TECHNIQUES

Abstract

Now a day, extracting texture features from image are widely used techniques for image processing and computer vision. Texture feature are the characteristics of the texture that is present in an image. These features are used to identify the structures and patterns in an image that are not captured by some other traditional methods of feature extraction like color and shape. There are many methods for extracting the texture features from the image, but in this paper, an experimental analysis of texture feature extraction by using techniques like LDRP, SED, MSD and DWT are used. These features can be used for a variety of tasks, such as texture feature extraction, object recognition, image segmentation, and classification. The main ideation of the paper is to extract the texture features of images, and rank the images based on their similiarity score and evaluate the performance of algorithms. For obtaining these ranks the images with similar types of images are compared and most matching image are considered. For this process a comparative analysis of CBIR Texture feature techniques is carried out to show the most efficient techniques and most similar image. Ranking of images is done by comparing their texture features, the images with more similarity index is ranked first ,then next image with reducing similarity of texture feature is ranked next.