摘要
本文提出了一种基于Android平台的立木胸径快速测量系统.首先,采用张正友标定法实现了相机自动标定功能,并根据标定结果完成立木图像畸变的矫正;然后,通过YCbCr色彩空间对MeanShift算法进行改进,缩短了图像处理耗时,并减弱了光照的影响;通过改进后的MeanShift算法增强了立木图像纹理与边缘信息,并结合GrabCut算法实现了立木图像的准确分割,提取了立木图像前景;最后,根据立木图像前景提取结果与摄影测量原理构建了立木胸径测量模型和深度提取模型,无需参照物即可实现基于移动端的立木胸径快速测量.本系统比算法改进前有效提高了立木胸径测量的速度,耗时缩短不低于80%,测量结果的最大误差不超过6%,可实现立木胸径的快速测量.
A fast DBH measurement system based on Android platform was proposed.Firstly,tree image was automatically calibrated using Zhang Zhengyou calibration method to correct distortion.Then the MeanShift algorithm improved by YCbCr color space was used to amplify the texture and edge information of the standing tree image,which can reduce the time consumption of image processing and the interference of light.After that the foreground image was obtained based on the segmentation of the standing tree image generated by GrabCut algorithm.Finally,based on foreground extraction and photogrammetry principle,the improved DBH measurement model and depth extraction model of standing tree were constructed,yielding a fast DBH measurement method without reference on the mobile terminal.Verification tests showed that the system can effectively improve the speed of DBH measurement,over 80%reduction on time,with the highest error less than 6%.
作者
吴铮
吴晓柯
徐爱俊
陶忠良
WU Zheng;WU Xiaoke;XU Aijun;TAO Zhongliang(School of Information Engineering,Zhejiang Agriculture and Forestry University,Hangzhou,Zhejiang 311300,China;Zhejiang Big Data Development Center for Agricultural and Rural Affairs,Hangzhou,Zhejiang 310020,China)
出处
《福建农林大学学报（自然科学版）》
CSCD
北大核心
2021年第5期649-654,共6页
Journal of Fujian Agriculture and Forestry University:Natural Science Edition
基金
国家自然科学基金项目(31670641)。
作者简介
吴铮(1997-),男.研究方向:农林业智能设备与信息系统.Email:769737584@qq.com;通信作者:陶忠良(1972-),男,正高级农艺师,硕士.研究方向:农业信息技术应用.Email:zisung@qq.com.
