基于混合高斯模型和模板匹配的目标测速
    点此下载全文
引用本文:安勃翰 1 ,裴 亮 1 ,杨文将 2.基于混合高斯模型和模板匹配的目标测速[J].计算技术与自动化,2022,(3):1-5
摘要点击次数: 281
全文下载次数: 0
作者单位
安勃翰 1 ,裴 亮 1 ,杨文将 2 (1.辽宁工程技术大学 测绘与地理科学学院,辽宁 阜新 1230002.北京航空航天大学 宇航学院北京 100000) 
中文摘要:针对目前视觉测速方法存在检测精度低、鲁棒性差的问题,研究了一种结合模板匹配和混合高斯模型的测速方法。首先,利用混合高斯模型进行背景建模,提取出仅包含目标轮廓信息的前景区域图像。其次,使用参数优化的模板匹配法获取目标的高精度像素位移。最后,使用改进的二维测量模型法精确获取目标对应的像素尺寸。利用目标物理尺寸与图像中目标像素尺寸的比例关系来计算实际速度。实验结果显示,该方法对不同形状目标在不同速度下的测速结果的相对误差均在5%以内,准确度较高。
中文关键词:单目视觉  目标测速  混合高斯模型  模板匹配法  二维测量模型
 
Target Velocity Measurement Based on Gaussian Mixture Model and Template Matching
Abstract:Aiming at the problems of low detection accuracy and poor robustness of current visual velocity measurement methods, a velocity measurement method combining template matching and Gaussian mixture model is studied. Firstly, Gaussian mixture model is used to model the background, and the foreground image containing only the target contour information is extracted. Secondly, the template matching method with parameter optimization is used to obtain the high-precision pixel displacement of the target. Finally, the pixel size corresponding to the target is obtained by using the improved two-dimensional measurement model method. The actual speed is calculated by using the proportional relationship between the physical size of the target and the size of the target pixel in the image. The experimental results show that the relative error of the speed measurement results of the algorithm for targets with different shapes and speeds is less than 5%, and the accuracy is high.
keywords:monocular vision  target velocity measurement  Gaussian mixture model  template matching method  two dimensional measurement model
查看全文   查看/发表评论   下载pdf阅读器