基于图像的前车状态警示系统设计与实现

发布时间：2018-04-22 07:28

本文选题：图像采集 + 车牌定位　；参考：《大连海事大学》2017年硕士论文

【摘要】：我国汽车保有量逐年增加给道路交通安全带来了巨大压力,每年追尾交通事故均造成大量的人员伤亡和巨大的经济损失,因此,降低或避免此类事故发生将对改善国内交通安全环境起到重要作用。为此,本课题研究一种基于图像的前车状态警示系统,辅助驾驶员感知前车状态。该系统通过图像传感器采集本车前方车辆和路面情况,经图像处理后可识别前车车牌及前车车灯状态。本文完成的主要工作包括:1.通过对机器视觉在智能交通领域的应用分析,给出了现有主流机器视觉方法在道路交通领域应用时的优缺点,确定了课题的研究内容和方案。2.从前车状态检测需求出发,经过可行性及成本等分析后给出了基于图像的前车状态警示系统的解决方案,明确了系统中图像采集单元、图像处理单元和警示单元的功能及其他们之间的关系。3.针对前车状态警示系统的图像采集和处理单元硬件,给出了基于海思HI3518E处理器平台和OV9712图像传感器的图像采集系统的组成和开发环境。4.针对前车状态警示系统的车灯状态识别算法中车灯位置,本文提出了一种简易的车牌算法,可快速有效的识别前车车牌区域,从而可进一步获得前车车灯位置,并通过实验验证了该方法的有效性。5.针对前车状态警示系统的车灯状态识别算法中车灯状态,给出了一种在图像HSI空间根据车灯亮度信息判断车灯明暗的方法,并通过实验验证了该方法的有效性。综上所述,通过本文给出的前车状态警示系统,可在一定程度上有效检测前车状态并告知本车驾驶员,从而能够有效降低或避免追尾事故的发生。
[Abstract]:The increase in the number of cars in our country has brought great pressure to road traffic safety year by year. Every year, rear-end traffic accidents cause a large number of casualties and huge economic losses. Reducing or avoiding such accidents will play an important role in improving the domestic traffic safety environment. Therefore, this paper studies an image-based warning system to assist drivers to perceive the state of the front car. The system collects the vehicle and road conditions in front of the vehicle through the image sensor, and after image processing, it can recognize the status of the vehicle license plate and the vehicle lamp in front of the vehicle. The main work accomplished in this paper includes 1: 1. Based on the analysis of the application of machine vision in the field of intelligent transportation, the advantages and disadvantages of the existing mainstream machine vision methods in the field of road traffic are given, and the research content and scheme of the subject are determined. Based on the analysis of the feasibility and cost of the former vehicle status detection system, the solution of the vehicle status warning system based on the image is given, and the image acquisition unit in the system is defined. The function of image processing unit and warning unit and their relationship. Aiming at the hardware of the image acquisition and processing unit of the vehicle status warning system, the composition and development environment of the image acquisition system based on Hayes HI3518E processor platform and OV9712 image sensor are presented. In this paper, a simple license plate algorithm is proposed to recognize the position of vehicle lights in the vehicle lamp status recognition algorithm of the warning system, which can quickly and effectively identify the front car license plate area, and thus further obtain the position of the headlight. The effectiveness of the method is verified by experiments. Aiming at the vehicle lamp state recognition algorithm of the vehicle state warning system in front, this paper presents a method to judge the light and shade of the vehicle lamp in the image HSI space according to the luminance information of the vehicle lamp, and the validity of the method is verified by experiments. To sum up, the warning system can effectively detect the state of the front car and inform the driver of the vehicle to a certain extent, so as to effectively reduce or avoid the occurrence of rear-end accident.
【学位授予单位】：大连海事大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】