A novel approach for Content Based Image Retrieval from huge database sets

Advances in data storage and image acquisition technologies have enabled the creation of large image datasets. Due to the enormous increase in image database sizes, as well as its vast deployment in various applications, the need for Content Based Image Retrieval development arose. In this paper, an efficient Image Search engine has been developed which retrieves the relevant images in accordance to the query image from the huge miscellaneous database using the color and texture features. Color feature is represented by Dominant Color Descriptor (DCD) which defines the dominant colors within an image along with its percentages. Texture feature extraction is implemented using a novel approach called Localized Gabor Filtering. Ultimately, similarity score between query images and the images in database is calculated using the DCD feature vectors along with the mean and standard deviation of the output filtered images. The retrieved images are displayed in a sorted and ranked manner based using the obtained similarity scores. Experimental results show that the proposed method is more accurate and effective for large database with comparatively smaller feature vector dimensions.