基于图像分析的车祸现场识别技术研究

Research on traffic accident scene recognition technology based on image analysis

由于车祸现场环境复杂,科研组织曾提出的各类车祸现场识别方法不具备较高的识别精度和较短的识别时间。为了合理管理城市交通,提出一种基于图像分析的车祸现场识别方法。使用光敏传感器对城市交通视频中的车祸现场图像进行采集,并提取出图像的灰度阈值。图像定位工作以车祸现场图像的灰度阈值为输入,对其进行分割和边缘删除,获取其中重要信息的定位图像并进行输出。图像识别工作以定位图像为输入,采用模式匹配技术识别出其中的目标车祸参数并发出警报,以实现对车祸现场图像中目标车祸参数的精准、高效定位。实验结果表明,所提方法具有较高的识别精度和较短的识别时间。

The various traffic accident scene recognition methods proposed by a scientific research organization can not accurately recognize the traffic accident in complex traffic accident scene environment in short time.In order to manage the urban traffic reasonably,a traffic accident scene recognition method based on image analysis is put forward.The photosensitive sensor is used in the method to acquire the traffic accident scene image in urban traffic video,and extract the gray?level threshold of the image.The gray?level threshold of the traffic accident scene image is taken as the input of image positioning to conduct its segmentation and edge deletion,and acquire the positioning image with important information for image output.The positioning image is taken as the input of image recognition.The pattern matching technology is used to recognize the parameters of target vehicle accident and send an alarm to locate the parameters of target vehicle accident in vehicle accident scene image accurately and efficiently.The experimental results show that the proposed method has higher recognition accuracy and shorter recognition time.