Detection of Disaster Affected Regions based on Change Detection using Deep Architecture
Naga Pavan Srivathsav C1, Anitha K2, Anvitha K3, Maneesha B4, Sagar Imambi S5
1Naga Pavan Srivathsav Chevuru, Department of CSE, Koneru Lakshmaiah Education Foundation, Guntur (Andhra Pradesh), India.
2Anitha Katta, Department of CSE, Koneru Lakshmaiah Education Foundation, Guntur (Andhra Pradesh), India.
3Anvitha Kona, Department of CSE, Koneru Lakshmaiah Education Foundation, Guntur (Andhra Pradesh), India.
4Maneesha Bommineni, Department of CSE, Koneru Lakshmaiah Education Foundation, Guntur (Andhra Pradesh), India.
5Dr. Sagar Imambi Shaik, Professor, Department of CSE, Koneru Lakshmaiah Education Foundation, Guntur (Andhra Pradesh), India.
Manuscript received on 07 March 2019 | Revised Manuscript received on 20 March 2019 | Manuscript published on 30 March 2019 | PP: 124-128 | Volume-8 Issue-5, March 2019 | Retrieval Number: E2890038519/19©BEIESP
Open Access | Ethics and 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: Natural disasters pose a serious threat to national economy, human lives and can disturb the social fabric of the society, although we can not entirely prevent these natural disasters from happening but with the advancements in the satellite imagery, remote sensing and machine learning it has become possible to minimize the damage caused by them. Satellite images are very useful because they can give you a huge amount of information from a single picture. Since it is becoming easy to get these satellite images the climate and environmental detection systems are in high demand. In this paper, we propose a post disaster system which we have named, Automatic Disaster Detection System (ADDS) which is designed to detect the disaster affected areas and help in the relief operations. The existing methods for detection of disaster affected regions are mostly dependent on manpower where people use the drone technology to see which area is affected by flying that drone over a large area which takes a lot of time. A new approach of Convolution Neural Network towards detection of disaster affected areas through their satellite images is examined in this paper which is comparatively better than previous image processing techniques. This method is based on deep learning which has been a widely popular technique for image processing in recent past. This technique can help save lives by reducing the response time and increasing the efficiency of the relief operations.
Keyword: Convolution Neural Network, Deep Learning, Image Processing, Flood Detection, Satellite Imagery, Remote Sensing, Machine Learning.
Scope of the Article: Agent Architectures, Ontologies, Languages and Protocols