Scene Image Classification Descriptors

Abstract

This paper mainly focused on scene image classification which is one of the challenging topics in computer vision and which is also important in terms of image search and image retrieval applications. this paper mainly emphasizes on the three approaches, first one is a 3-Dimensional Local Binary Pattern (3DLBP) descriptor is proposed for color image local feature extraction. Second, a new shape descriptor (HaarHOG) is introduced by combining Haar wavelet transformation and Histogram of Oriented Gradients (HOG). Third, these descriptors are fused using an optimal feature representation technique to generate a robust 3-Dimensional LBP-HaarHOG (3DLH) descriptor that can perform well on different scene image categories. This paper describes in detail the three approaches and also shows effect on combining both approaches. Also we explain the different novel image descriptors with their role.

Scene classification, HaarHOG descriptor, 3D-LBP descriptor, image search