Smart Detection of Vehicle Accidents using Object Identification Sensors with Artificial Intelligent Systems
P. Amrith1, E. Umamaheswari2, R.U. Anitha3, Devi Mani4, D M Ajay5
1P. Amrith, VIT University Chennai, (Tamil Nadu), India.
2Dr. E. Umamaheswari, VIT University Chennai, (Tamil Nadu), India.
3Dr. R.U.Anitha, Assistant Professor, Department of Computer Science, King Khalid University, Abha, Saudi Arabia.
4Dr. Devi Mani, Assistant Professor, Department of Computer Science, King Khalid University, Abha, Saudi Arabia.
5D M Ajay, VIT University Chennai, (Tamil Nadu), India.
Manuscript received on 24 January 2019 | Revised Manuscript received on 30 March 2019 | Manuscript published on 30 January 2019 | PP: 375-379 | Volume-7 Issue-6, March 2019 | Retrieval Number: E1929017519©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: According to Government of India, around 1,46,000 people [20] lost their lives in five hundred thousand of accidents, where 7% of lives could have been saved if they would have got medical attention before-hand [21]. This can be achieved by intimating the accident to the nearby emergency unit in minimal time by using artificial intelligence based on the severity of accidents. In existing methods, the accidents are detected using On-based unit, and transmitted to the control unit using nearby antennas, where the severity of the accidents are classified using data-mining. Then the fetched data is compared with existing accident dataset which it is retrieved from previous accidents, the analysed results are then transmitted to the nearby emergency unit [1], [2]. This will lead to ambiguous prediction of data because if the data doesn’t exist in the database, intensity of accident must be analysed manually in which it leads to increase in time complexity for transmitting the data to the nearby emergency unit due to intermediate infrastructure. To overcome these drawbacks, in this proposed system the accidents are detected using sensors and the severity of accident will be calculated using machine learning algorithms like k-means clustering and Support vector machine (SVM) classification under reinforcement learning with help of force and impact obtain while vehicle crashes, then the values are transmitted to the nearby emergency unit using Breadth-first-search in the form of A* Search algorithm
Keywords: Vehicle Accident Detection, Smart Sensors, Support Vector Machine (SVM), K-means, Reinforcement Learning, Breadth-First-Search (BFS)
Scope of the Article: Deep Learning