원문정보
Traffic Accident cause analysis from CNN image processing
초록
영어
Streets are one of the major interface of human activities in urban areas. Urban environment and urban dwellers interact each others in the urban streets. Analysing the urban environment is important for us to understand the potential impact of the urban built environment on urban dwellers. The publicly accessible Google Street View (GSV), which captures the streetscape appearances of cities around the world, presents a very good tool for urban studies at a fine level. In this study, we will use KoRoad’s ‘Frequent traffic accident location information’ for detailed spot location data and GSV for corresponding source of street-level image data for environmental component analysis.Convolutional neural network, also known as convnets or CNN, is a well-known method in computer vision applications to recognize objects from a picture or video. The purpose of the convolution is to extract the features of the object on the images locally so that the network can learn specific patterns within the picture and will be able to recognize it everywhere in the picture. In this paper, we describe our plan of using CNN classification method for analyzing the cause of traffic accidents with the data from GSV and KoRoad GPS information.
목차
I. Introduction
II. Data Preperation
III. Convolutional Neural Network Classification Plan
IV. Conclusion
References