结合模糊分割和遗传算法的航空影像车辆提取方法

发布时间：2018-01-31 04:02

本文关键词： 目标提取车辆边缘遗传算法航摄影像模糊分割　出处：《测绘通报》2017年08期 　论文类型：期刊论文

【摘要】：利用高分辨率遥感影像实现道路车辆监测在交通管理系统中占有重要地位,然而受成像条件及车辆本身颜色多样性的影响,很难从遥感影像中直接提取道路车辆。为此,本文提出了一种结合遗传算法和数学形态学的遥感影像道路车辆提取方法。首先利用模糊聚类方法建立影像分割模型,并利用遗传算法对其进行求解,从而实现影像中道路、车辆及环境的精确分割;然后对分割结果中不同层进行赋值,实现分割结果的二值化;最后利用数学形态学操作从分割的道路—车辆的二值结果中剔除道路,从而实现道路车辆的提取。利用本文方法对航空影像进行了道路车辆提取试验。试验结果表明,本文方法不仅能够准确识别地面车辆位置,还能精确提取其轮廓线。
[Abstract]:Using high-resolution remote sensing image to realize road vehicle monitoring plays an important role in traffic management system, but it is affected by imaging conditions and color diversity of vehicle itself. It is difficult to extract road vehicles directly from remote sensing images. In this paper, a method of road vehicle extraction from remote sensing image based on genetic algorithm and mathematical morphology is proposed. Firstly, the image segmentation model is established by fuzzy clustering method and solved by genetic algorithm. In order to achieve accurate segmentation of road, vehicle and environment in the image; Then the different layers in the segmentation results are assigned to realize the binary segmentation results. Finally, the mathematical morphology is used to remove the road from the binary result of the segmented road-vehicle. In order to achieve the road vehicle extraction. Using the method in this paper to carry out the road vehicle extraction test. The experimental results show that this method can not only accurately identify the position of the vehicle on the ground. The contour line can also be accurately extracted.
【作者单位】：辽宁工程技术大学测绘与地理科学学院遥感科学与应用研究所;
【基金】：国家自然科学基金(41301479) 大学生创新创业训练项目(201510147005) 辽宁省自然科学基金(2015020090)
【分类号】：TP18;TP751
【正文快照】： 作为图像处理的重要内容,目标提取已有广泛的应用,如房屋[1]、道路[2-3]、植被[4]等的提取。道路车辆提取作为目标提取中的重要任务之一,在监测道路车流量、车流密度等交通管理方面具有独特的潜在应用价值。近年来,随着无人机航测技术的发展,获取高分辨率遥感影像的成本越来越

【相似文献】