基于图像处理的水稻剑叶夹角测量  

Measurement of blade angle in rice based on image processing

作　　者：李家超 罗斌 周亚男 黄硕 侯佩臣 Li Jiachao;Luo Bin;Zhou Yanan;Huang Shuo;Hou Peichen(College of Mechanical and Electrical Engineering,Xinjiang Agricultural University,Urumqi,830052,China;Information Technology Research Center,Beijing Academy of Agriculture and Forestry Sciences,Beijing,100097,China;Intelligent Equipment Research Center,Beijing Academy of Agriculture and Forestry Sciences,Beijing,100097,China)

机构地区：[1]新疆农业大学机电工程学院,乌鲁木齐市830052 [2]北京市农林科学院信息技术研究中心,北京市100097 [3]北京市农林科学院智能装备技术研究中心,北京市100097

出　　处：《中国农机化学报》2024年第10期228-232,共5页Journal of Chinese Agricultural Mechanization

基　　金：广东省科技计划项目(2022B0202110003)。

摘　　要：水稻剑叶夹角对于水稻理想株型的选育具有重要意义。为实时、准确地实现水稻剑叶夹角在体测量,采用颜色空间转换、骨架提取、Hough直线提取和K-means聚类等图像处理技术,设计水稻剑叶夹角测量算法,并将算法置于服务器,结合背景板与手机端的应用程序,实现水稻剑叶夹角的实时测量。利用该算法在田间对160株水稻样本进行剑叶夹角测量,将该算法得到的数据与人工测量的数据进行对比。结果表明,该算法与人工测量结果一致性高,决定系数为0.9996,平均绝对误差为0.32°,能够实现水稻剑叶夹角田间在体精准测量。Rice blade angle is of great significance for the breeding of ideal plant type of rice.In order to achieve real‑time and accurate measurement of the angle between rice sword leaves in the field,image processing techniques such as color space conversion,skeleton extraction,Hough line extraction,and K-means clustering were used to design a rice sword leaf angle measurement algorithm.The algorithm was placed on a server and combined with a background board and a mobile application to achieve real‑time measurement of the angle between rice sword leaves.By using this algorithm,160 samples of rice with grains were measured in the field for the angle between the sword leaves,and the data obtained by the algorithm was compared with the manually measured data.The results show that the proposed algorithm is in high agreement with the manual measurement results,the coefficient of determination is 0.9996,the average absolute error is 0.32°.It can achieve accurate measurement of blade angle of rice in the field.

关 键 词：水稻 剑叶夹角 图像处理 在体测量 

分 类 号：S511[农业科学—作物学] TP391.4[自动化与计算机技术—计算机应用技术]