Abstract
In the present day to day life the retrieval of images from the larger database has become important like Google search. Presently, CBIR is trending method. In this paper images are retrieved in three stages with color, texture and shape feature extracted in each respective stage. The color feature extraction is done through finding different parameters for an RGB image, combined texture feature extraction through Gray level co-occurrence matrix GLCM and Gabor filter and shape feature extraction through region growing. The experiment is done on images present in coral database. The simulation results have been compared and it has been shown that the proposed method show high retrieval rate in terms of the average precision and average recall.
Highlights
Content based image retrieval deals with retrieval of images that rely on image contents
Gray level co-occurrence matrix (GLCM) features [3] used in existing retrieval systems are shown to be having low average recall and average precision rate
A hybrid method for image retrieval based on three different features is proposed
Summary
Content based image retrieval deals with retrieval of images that rely on image contents. GLCM features [3] used in existing retrieval systems are shown to be having low average recall and average precision rate. This can be used further improved using hybrid methods [4]. Content based image retrieval (CBIR) known as query by image and video content (QBIC) is basically used to search digital images present in large databases. In text based image retrieval, images are retrieved by giving name of the image By doing this the accurate results are not obtained whereas in content based image retrieval the features of the image are computed and are compared with database images which effectively retrieve the images. Gabor filters for texture defect detection is employed for locating accurate defect position [11]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
