Image Retrieval Using Gradient Operators

Abstract

The images are described by its content like color, texture, and shape information present in them.In this paper novel image retrieval methods discussed based on shape features extracted using gradient operators like Robert, Sobel, Prewitt and Canny. Masking of Gradient operators takes place for continuing the discontinue edges. Morphological operations like erosion and dilation are used along with canny. The proposed image retrieval techniques are tested on generic image database images spread across different categories. Gradient operators features are extracted using Figure of Merit (FOM). The average precision and recall of all queries are computed and considered for performance analysis. The performance ranking of the masks for proposed image retrieval methods can be listed as Robert, Canny, Prewitt, and Sobel.