Captioning for Motion Detection for video surveillance Applications using Deep Learning
M. Nivedita1, Asnath Victy Phamila Y2, Harsh P.V3
1M Nivedita, Department of Computer Technology Anna University, Chennai, (Tamil Nadu), India.
2Asnath Victy Phamila, M.E. and Ph.D. degree, Department of Computer Science and Engineering from Anna University, India.
3Harsh PV, Department of Protecting your technology, Chennai, (Tamil Nadu), India.
Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 3180-3185 | Volume-8 Issue-8, June 2019 | Retrieval Number: H7326068819/19©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: Video surveillance has become a major tool in Security and has become sophisticated and fool-proof. Recent developments in the field of Image processing and object recognition have enabled us to integrate with the existing technology of surveillance. Today, the most popular type of video surveillance system is CCTV. But there are a lot of de-merits to this system which are mentioned in the below section. We are aiming at improving the existing technology by drastically increasing its reliability by using motion detection and image captioning to detect the moving object and alert the user by describing about that in the form of image captioning. We have developed motion detection using the build-in functions of OpenCV and Image processing and an Image captioning system using Neural Networks like Convolutional Neural Networks (CNN) and Recurrent Neural Network-Long Short-Term Memory (RNN- LSTM) etc. This generated caption is sent to the user for analysis of the situation.
Keyword: CNN, Image Captioning, LSTM, motion detection, Video Surveillance
Scope of the Article: Deep Learning.