基于图像纹理分析的动态车辆识别方法研究

发布时间：2018-01-15 17:36

本文关键词：基于图像纹理分析的动态车辆识别方法研究　出处：《公路交通科技》2017年10期 　论文类型：期刊论文

【摘要】：在智能交通系统中,动态图像识别技术是系统应用的基础核心技术之一。以应用于交通监控、智能驾驶系统等场景的HSV空间动态车辆识别为基础,研究并论证提出了新的检测识别方法,实现对运动车辆的检测识别、目标追踪、驾驶辅助等功能。研究问题的难点是,如何从复杂的背景中分割运动物体,是检测方法能否有效的至关重要的一步,在研究了目前存在的各种方法之后,提出了一种新的基于阴影检测的HSV空间自适应背景模型的车辆追踪检测算法,算法基于HSV空间图像处理,采用最大类间方差法获取相邻帧二值化阈值,利用纹理信息进一步确定动态图像以及确认图像范围。通过截取由监控系统获取的视频信息,并对其进行图像处理检测车辆移动轨迹。从监控视频信息中获取两帧不同时刻的图像信息,在HSV空间进行相邻帧检测。由于阈值的选择将直接影响判断精度,本研究将固定阈值法进行了改进,该阈值是通过统计模型对整幅图像上灰度值进行计算,并通过最大类间方差法确定阈值。最后经过实际视频图像验证,仿真试验流程清晰,试验结果达到预期设想。
[Abstract]:In the intelligent transportation system, dynamic image recognition is one of the core technologies in the application of the system. It is based on the dynamic vehicle recognition in HSV space which is applied to traffic monitoring and intelligent driving system. This paper studies and demonstrates a new detection and recognition method to realize the detection and recognition of moving vehicles, target tracking, driving assistance and other functions. The difficulty of the research is how to segment moving objects from the complex background. It is a very important step whether the detection method can be effective or not. After studying the existing methods, a new vehicle tracking detection algorithm based on HSV space adaptive background model based on shadow detection is proposed. The algorithm is based on HSV spatial image processing and uses the maximum inter-class variance method to obtain the binarization threshold of adjacent frames. Using texture information to further determine the dynamic image and confirm the range of the image by intercepting the video information obtained by the monitoring system. The vehicle moving track is detected by image processing, and the image information of two frames and different times is obtained from the monitoring video information. The detection of adjacent frames in HSV space. Because the choice of threshold will directly affect the accuracy of judgment, the fixed threshold method is improved in this study. The threshold is calculated by statistical model on the gray value of the whole image. The threshold value is determined by the method of maximum inter-class variance. Finally, the simulation process is clear and the experimental results reach the expectation after the actual video image verification.
【作者单位】：中国公路车辆机械有限公司;
【基金】：交通运输部标准研究项目(201611109)
【分类号】：TP391.41;U495
【正文快照】： 0引言智能交通系统(Intelligent Transportation System,简称ITS)是未来交通系统的发展方向,它是将先进的信息技术、数据通讯传输技术、电子传感技术、控制技术及计算机技术等有效地集成运用于整个地面交通管理系统而建立的一种在大范围内、全方位发挥作用的,实时、准确、高效

【相似文献】