Abstract

Content-Based Image Retrieval is a technique used to retrieve similar images where the most challenging aspect is to bridge the gap between low level feature layout and high level semantic concepts. Efficient and effective retrieval techniques of images are desired to work out a certain image on the condition that the result would be more suitable than the input image. A novel method is proposed, which links various images as threads; where the texture features and shape features of the images are extracted and stored. The minimum distance between the query image and the thread provides the output. These threads are based upon the query result. The resultant retrieval system is found to be beneficial and interactive.