摘要
在农作物生长自动化观测方面,图像获取及处理技术能自动识别农作物的生长特征。为获取大田中农作物的生长状况和气象数据,将农作物从采集到的大田实景图像中准确分割出来是基础且关键的一步。大田玉米生长环境和生长状态的复杂性,导致很多算法不能准确和完整地分割出图像中的群体玉米,从而使分割误差较大,以致后续的识别和分析工作无法进行。因此,从最基本的三基色原理出发,利用颜色特征,采用统计学方法,分析大田玉米的生长特征及其在图像中的特点,给出了一种有效的阈值分割方法,可得到准确且完整的玉米图像分割结果。
In the field of crop automation observation,the main technology of obtaining corn growing conditions information is image acquisition and processing.In order to obtain the growth conditions and meteorological data of crops,extracting crops accurately from the field scene is a critical step.Many existing algorithms cannot segment the maize in the filed accurately and completely because of the complexity of field corn growth environment and the growth status.In this paper,starting from the basic principle of three primary colors,using color characteristics and statistical methods,the growth characteristics of field maize and its features in the image are analyzed,so as to give an effective threshold segmentation method to obtain accurate and complete corn segmentation results.
作者
李涛
翟秀丽
贾红宾
LI Tao;ZHAI Xiuli;JIA Hongbin(College of Information Engineering,Xinxiang Vocational and Technical College,Xinxiang 453006,China)
出处
《河南工程学院学报（自然科学版）》
2024年第1期60-64,共5页
Journal of Henan University of Engineering：Natural Science Edition
关键词
作物自动化观测
图像分割
大田玉米
阈值分割
crop automation observation
image segmentation
field maize
threshold segmentation
作者简介
李涛(1989—),男,河南新乡人,讲师,主要研究方向为图像处理。
