An improved Content Based Image Retrieval using three region colour and straight line signatures of the image

In this work, we have performed Content Based Image Retrieval using combined colour histograms for three regions (top, bottom and middle) of the image. This will indirectly include the space information within it along with the colour information. Besides this, retrieval of images from the image database according to the user query image are done using straight line signatures of the images. The average precision is used as a metric to evaluate the performance of the system. Experiments were successfully implemented on 1000 images of Semantics Sensitive Integrated Matching for Picture Libraries using MAT Lab. The performance of our method is compared with four other previous methods such as Integrated Region Matching, Fuzzy Integrated Region Matching , CBIR using colour salient points and shape features and Interactive CBIR using Ripplet Transform and Fuzzy Relevance Feedback Mechanism which used the same data set and evaluation method. According to the arrived results, the performance of the proposed combined three region colour and straight line signatures based CBIR systems were significantly better than other compared methods.