A Research on Deep Learning Advance for Landslide Classification using Convolutional Neural Networks
Jaishankar Bhatt1, Akash Gangwar2, Rahul Nijhawan3, Durgaprasad Gangodkar4
1Jaishankar Bhatt, Department of Computer Science and Engineering, Graphic Era University, Dehradun (Uttarakhand), India.
2Akash Gangwar, Department of Computer Science and Engineering, Graphic Era University, Dehradun (Uttarakhand), India.
3Rahul Nijhawan, Department of Computer Science and Engineering, Graphic Era University, Dehradun (Uttarakhand), India.
4Durgaprasad Gangodkar, Department of Computer Science and Engineering, Graphic Era University, Dehradun (Uttarakhand), India.
Manuscript received on 10 April 2019 | Revised Manuscript received on 17 April 2019 | Manuscript Published on 26 July 2019 | PP: 903-906 | Volume-8 Issue-6S4 April 2019 | Retrieval Number: F11840486S419/19©BEIESP | DOI: 10.35940/ijitee.F1184.0486S419
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: Landslides can easily be tragic to human life and property. Increase in the rate of human settlement in the mountains has resulted in safety concerns. Landslides have caused economic loss between 1-2% of the GDP in many developing countries. In this study, we discuss a deep learning approach to detect landslides. Convolutional Neural Networks are used for feature extraction for our proposed model. As there was no source of an exact and precise data set for feature extraction, therefore, a new data set was built for testing the model. We have tested and compared this work with our proposed model and with other machine-learning algorithms such as Logistic Regression, Random Forest, AdaBoost, K-Nearest Neighbors and Support Vector Machine. Our proposed deep learning model produces a classification accuracy of 96.90% outperforming the classical machine-learning algorithms.
Keywords: Convolutional Neural Network, Deep Learning, Re Lu, Pooling.
Scope of the Article: Classification