基于鱼眼镜头的树高测量方法  

Tree Height Measurement Method Based on Fisheye Lens

作　　者：宋佳音[1] 池志祥 张晓鹏 朱庆林 宋文龙[1] 杨柳松[1] SONG Jia-yin;CHI Zhi-xiang;ZHANG Xiao-peng;ZHU Qing-lin;SONG Wen-long;YANG Liu-song(Mechanical and Electrical Engineering,Northeast Forestry University,Harbin 150040,China;State Grid Heilongjiang Electric Power Transmission&Transformation Engineering Co.,Ltd.,Harbin 150040,China)

机构地区：[1]东北林业大学机电工程学院,哈尔滨150040 [2]国网黑龙江省送变电工程有限公司,哈尔滨150040

出　　处：《科学技术与工程》2021年第13期5424-5431,共8页Science Technology and Engineering

基　　金：中央高校基本科研业务专项(2572017CB13)。

摘　　要：为解决当前测量树高方法中存在测量误差大、测量不便等问题,提出一种基于鱼眼镜头的树高测量方法。该方法首先利用鱼眼镜头成像模型确立世界坐标系到图像像素坐标系的变换关系;然后以畸变系数的等距投影模型为基础,利用Scaramuzza鱼眼相机标定方法求解参数并建立测量系统模型;最后利用图像处理技术对实验图像进行轮廓提取、分割及极值点提取,实现对树高的测量。通过所提出的测量方法对树高进行测量,实验结果表明:测量方法的测量误差在-0.196~0.195 m,树高测量最高相对误差为2.75%,平均相对误差为1.505%,说明该方法有效减小了测量误差,具有很高的精确性,同时测量方法简便,可操作性强。In order to resolve the problems of large measurement error and inconvenient method in measuring tree height methods,a tree height measurement method based on fish-eye lens was proposed.Firstly,the fish-eye lens imaging model was used to establish the transformation relationship from the world coordinate system to the image pixel coordinate system.Secondly,the calibration method of Scaramuzza fisheye camera was used to solve the parameters and establish the measurement system model based on the isometric projection model with distortion coefficient.Finally,the image processing technology was used to extract and segment the experimental image contour and extract the extreme points measurement of tree height.The tree height was measured by the measurement method proposed.The experimental results show that the measurement error of the proposed method is between-0.196 m and 0.195 m,the maximum relative error of tree measurement is 2.75%,and the average relative error is 1.505%.It shows that the method can effectively reduce the measurement error and has high accuracy.At the same time,the measurement method is simple and easy to operate.

关 键 词：鱼眼镜头 等距投影 测量 极值点提取 树高 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术] S758[自动化与计算机技术—计算机科学与技术]