PROJECT TITLE :

PW-COG: An Effective Texture Descriptor for VHR Satellite Imagery Using a Pointwise Approach on Covariance Matrix of Oriented Gradients

ABSTRACT:

During this paper, a completely unique algorithm for textural feature description in terribly high resolution (VHR) satellite imagery is developed. It's primarily based on a pointwise (PW) approach on the feature covariance matrix. The most motivation of this work is to construct the covariance matrix of oriented gradients (COG) employing a nondense approach primarily based on characteristic points (i.e., keypoints) extracted from the image. The proposed descriptor, which is known as PW-COG, is predicted to be effective when applied to VHR pictures. Initial, a COG-based mostly descriptor is capable of not only capturing each radiometric and local geometric information (given by gradient options) from the image however also encoding their joint distribution and correlation, which are effectively relevant for texture characterization and discrimination. Second, by using a keypoint-primarily based approach, the proposed methodology is in a position to deal with large quantity of VHR image information, since we do not take into consideration all pixels of the image, without requiring the stationarity hypothesis. So as to demonstrate the effectiveness of the proposed descriptor, texture-based mostly image classification is disbursed. Experimental study on each Brodatz texture database and VHR satellite pictures using the proposed algorithm provides very competitive results, in terms of texture discrimination and algorithm complexity, compared to reference ways.