Image retrieval using combination of color, texture and shape descriptor

摘要

Image retrieval is an active research area for the last two decades. This area is gaining more importance as the multimedia content over the internet is increasing. Color Texture and shape are the low level image descriptor in Content Based Image Retrieval. These low level image descriptors are used for image representation and retrieval in CBIR. This paper presents a Content Base Image Retrieval (CBIR) System using the image features extracted by color moments, wavelet and edge histogram. Combining the color, texture and shape feature leads to a more accurate result for image retrieval. Moreover the color moments are taken by partitioning the images into blocks hence it also gives spatial color information. Here SVM classifier is used to classify the images into different class and the similarity measure is taken only with the images in the same class. The retrieval is more accurate and time taken is less.