作者: Thammasak Thianniwet , Satidchoke Phosaard , Wasan Pattara-Atikom
DOI: 10.1007/978-90-481-8776-8_23
关键词:
摘要: We proposed a technique to identify road traffic congestion levels from velocity of mobile sensors with high accuracy and consistent motorists’ judgments. The data collection utilized GPS device, webcam, an opinion survey. Human perceptions were used rate the into three levels: light, heavy, jam. successfully extracted vehicle’s moving patterns using sliding windows technique. Then fed ANN J48 algorithms. comparison between two learning algorithms yielded that model shown best result which achieved as 91.29%. By implementing on existing report systems, reports will cover comprehensive areas. method can be applied any parts world.