## MATLAB单目视觉下的行人测距技术研究

1.采用帧间差分法将视频序列的相邻帧进行差分，然后通过阈值分割得到运动行人的二值图像。对二值图像经过数学形态学处理，将运动行人分离出来，完成运动行人检测。
2.建立行人测距模型，通过实验的方法标定摄像机参数。根据测距模型，推导出距离公式，在行人高度假定的条件下求解行人的相对距离，并对测距结果进行误差分析。
3.利用MATLAB开发平台，实现对运动行人的检测，并根据测距公式求解出行人的相对距离。
最终通过实验验证，本文设计的测距模型可实现对20米至100米运动行人距离进行精确测量。17267

Title    Pedestrian Ranging Technology Based on Monocular   Vision
Abstract
Distance detection in passive mode can not only save energy , but also enhance the secrecy of detection equipment .So more and more attention has been paid to passive detection technology. At the same time, we can get some very important information about the measured objects and track them by analyzing videos. Monocular vision pedestrian measuring method uses models of the distance measured process and calculates the distance by analyzing the previous videos. The main studying steps are as follows: 源自六/维\论]文[网!加7位QQ324.9114 www.lwfree.cn
Firstly, we subtract the adjacent frames by using frame difference method and use threshold segmentation to achieve the pedestrian’s binary image. After morphological processing, we can detect the pedestrians.
Secondly, we establish pedestrian ranging model and demarcate the camera parameters through the experimental method. Then we can calculate the distance under the conditions of assuming the pedestrian’s height.
Thirdly, we use MATLAB development platform to achieve the detection of pedestrians and to solve the relative distance by using ranging formula.
As a conclusion, the designed model can measure moving pedestrians at a distance of 20 to 100 meters accurately.
Keywords  Pedestrian Detection  Model  Camera Calibration  MATLAB

1 绪论 1
1.1  研究意义和背景1
1.2  国内外研究现状1
1.3  论文主要内容和章节安排 2
2  行人图像基本处理  4
2.1  图像的数字化和灰度化 4
2.2  图像滤波 5
2.3  图像分割 6
2.4  形态学处理6
2.5  行人特征的提取7
3  运动行人检测 9
3.1  背景差分法 9
3.2  光流法9
3.3  帧间差分法10
3.4  帧间差分法具体实现步骤10
4  单目测距模型建立 15
4.1  测距模型的建立 15
4.2  摄像机参数标定 15
4.3  公式推导16
4.4  实验数据记录 17
4.5  参数验证18
4.6  行人距离计算 19
5  系统设计与实现 21
5.1  软件介绍21
5.2  系统需求分析与模块实现21
5.3  实测结果23 MATLAB单目视觉下的行人测距技术研究:http://www.lwfree.cn/zidonghua/20180109/18873.html
------分隔线----------------------------