National Changhua University of Education Institutional Repository : Item 987654321/16493

English | 正體中文 | 简体中文 | Items with full text/Total items : 6507/11669
Visitors : 30059379 Online Users : 546

RC Version 3.2 © Powered By DSPACE, MIT. Enhanced by NTU Library IR team.

Scope Adv. Search

Login ‧ Upload ‧ Help ‧ About ‧ Administer

NCUEIR > College of Technology > vr > Proceedings > Item 987654321/16493

Please use this identifier to cite or link to this item: http://ir.ncue.edu.tw/ir/handle/987654321/16493

Title:	The System for Appraisal of Vehicle Accident Based on Radial Basis Function Neural Networks
Authors:	Tseng, Wen-Kung;Lu, Chung-Sheng
Contributors:	車輛科技研究所
Keywords:	Accident appraisal;Appraisal basses;Environmental factor;Human factor;Radial basis function neural network
Date:	2011
Issue Date:	2013-05-06T04:45:03Z
Publisher:	IEEE
Abstract:	In Taiwan, there are hundreds of accidents every day recorded by government due to the human factor and environmental factor. The accident usually involved the money dispute; therefore the accident appraisal must indicate the bilateral parties' blame clearly: all blame; major blame; minor blame and none blame. Although the local police can give a preliminary analysis report at first, the report cannot be official evidence. If the people need a credible appraisal report, they have to apply for the Taiwan Provincial Government Traffic Accident Investigation Committee's accident appraisal report. However, applying for Committee's accident appraisal report will take long time. Therefore, this study employed radial basis function neural network to build an expert system for appraisal of bilateral vehicle accident. The database was built from 307 accident cases in Taiwan from the year of 2004 to 2008. According to Committee's analysis, there are 30 appraisal basses including 6 environmental basses and 24 vehicle basses chosen to be the input of the expert system. The training data includes three types: 70 cases training; 140 cases training; 207 cases training. Validation stage was carried out by using 100 fixed cases and the correctness was recorded. In the first stage, correctness rate is 76% for training with 70 cases. In the second stage, correctness rate is increased to 81% for training with 140 cases. In the third stage, correctness rate is increased to 89% for training with 207 cases. The training and validation processes were completed in one second. Therefore, the expert system proposed in this work is demonstrated to be an efficient system for the accident appraisal.
Relation:	Proceedings - 2011 7th International Conference on Natural Computation, ICNC 2011, 2: 869-872
Appears in Collections:	[vr] Proceedings

Files in This Item:

File	Size	Format
index.html	0Kb	HTML	647	View/Open

Loading...