监控图像中的前挡风车窗定位方法与实现.docVIP

PAGE PAGE 25 监控图像中的前挡风车窗定位方法与实现 摘要：随着社会进步和计算机与网络技术的发展，智能交通系统的应用日益普及。智能交通系统综合运用各种先进的计算机技术来解决车辆管理、交通优化和车辆安全监控等方面的难题，可以提高车辆管理效率，缓解交通压力，使车辆得到更好的监控管理。 目前，机器视觉方法在智能交通系统中得到普遍运用，通过采集监控视频与图像并进行智能分析，获得有用的交通状况和车辆信息，以供进一步科学管理与决策。针对特定目标的图像分割方法是机器视觉的基础技术之一，本文即研究对车辆监控图像中的前挡风车窗进行定位分割。 本文提出一种基于边缘提取和组合优化的车窗定位方法，通过改进Canny算子来提取车辆前挡风玻璃图像的边缘，再通过搜索横线和左右竖线并组合为基本轮廓库，进而通过优化筛选出最佳基本轮廓，最后通过边缘拟合优化得到车窗形状。这种汽车前挡风车窗定位方法可以帮助人们在监控图像中定位车窗、划定感兴趣区域并提取汽车特征，以帮助进行车辆定位与识别，为智能交通系统提供必要信息，保障系统的顺利运行。 关键词：图像分割，阈值分割，matlab，Sobel The positioning method and implementation of the front windshield window in image monitoring Abstract: With the development of social progress and computer and network technology, the application of intelligent transportation system is becoming more and more popular. Transportation System uses all kinds of advanced computer technology to solve the vehicle management, transport optimization and vehicle safety monitoring and other aspects of the problem. It can improve vehicle management efficiency, ease traffic pressure, and make the vehicle get better monitoring. At present, machine vision methods widely used in Intelligent Transportation Systems. Through the acquisition of surveillance video and images and intelligent analysis, it can obtain useful traffic conditions and vehicle information to further the scientific management and decision-making. Image segmentation method for a particular target is one of the basic technologies of machine vision; this paper is the study of the segmentation of front windshield window in vehicle monitoring image. It is proposed a window location method based on edge extraction and combination optimization in this paper. Through improved canny operator to extract the edge of the front windshield of the vehicle image, and then buy searching through the combination of horizontal and vertical search about basic contour, and then select the best basic outline by optimization, finally obtained by fitting window shape optimization. The front windshield posit