Image Mining- Similar Image Retrieval Using Multi-Feature Extraction and Content Based Image Retrieval Technique

Abstract: In CBIR (Content-Based Image Retrieval), graphical features such as shape, color and texture are extracted to depict images. Each of the features is represented using one or more visual feature descriptors. An image mining using Content based Image Retrieval (CBIR) is an automatic process to search relevant images based on user input. A Global color Descriptor color distribution over an entire image. It is defined in the hue-saturation-value (HSV) color space and produces a 256 bin color histogram is a four-bit integer value, and then encoded by canny’s Edge detection. Similarity measure is also very essential part of CBIR to find the closeness of the query image with the database images. We have used two similarity measures namely Euclidean distance (ED) and sum of precision and recall. The overall performance of retrieval of the algorithm has been measured by average precision and recall performance.

Keywords: CBIR; Graphical Feature; Harr wavelet transform; Global color descriptor; Similarity measures