Extensive Content Feature based Image Classification and Retrieval using SVM

Abstract

Abstract: The classification and retrieval of picture advances in the field of image retrieval, particularly content-based image retrieval, are presented in this work. Scaling Invariant Feature Transform (SIFT) and developed K-Means clustering approach can be used to first arrange the features extracted based on the bag of visual words (BOW). The two stages of our retrieval method are retrieval and classification. The k-Nearest Neighbor (kNN) and Support Vector Machine (SVM) techniques were used to classify the photos based on their attributes and results were compared. This will categorize the images into different groups to improve the precision and recall rate. Following image classification, similar images matching the query image are pulled from the appropriate class.

Index Terms: Bag of visual Words, Support Vector Machines, k-Nearest Neighbor, Scaling Invariant Feature Transform, classification, retrieval, classification