Classification of High-Resolution Images with Local Binary Pattern and Convolutional Neural Network
T. Gladima Nisia1, S. Rajesh2
1T. Gladima Nisia, Assistant Professor, Department of CSE, AAA College of Engineering & Technology, Sivakasi (Tamil Nadu), India.
2Dr. S. Rajesh, Associate Professor, Department of IT, Mepco Schlenk Engineering College, Sivakasi (Tamil Nadu), India.
Manuscript received on 27 March 2019 | Revised Manuscript received on 08 April 2019 | Manuscript Published on 18 April 2019 | PP: 885-887 | Volume-7 Issue-6S March 2019 | Retrieval Number: F03800376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In high-resolution satellite images, it is very important to classify the image accurately and classify each area of the image distinctly. However, it is not easy to identify complex patterns. To cope up this difficulty deep learning method is employed. Deep learning method is to automatically extract many features without any human intervention. Still the performance of the classification is enhanced by combining the deep features with texture features. The proposed system facilitates the deep feature learning strategy combined with texture-based classification. Here, texture features are extracted using Local Binary Pattern (LBP) and deep features by Convolutional Neural Network (CNN). The proposed system is implemented and the results are verified. Experimental results show that the efficiency of classification is improved when texture features are combined with deep learning approach.
Keywords: High-resolution Image; Deep Learning; Convolutional Neural Network (CNN); Image Classification; Local Binary Pattern (LBP).
Scope of the Article: Image Security