基于数字图像的小麦群体整齐度测量方法研究  

Study of Measurement Method of Population Uniformity of Wheat Based on Digital Image

作　　者：单成钢[1] 廖树华[2] 刘宾[3] 王志芬[1] 戴明宏[2] 刘明[2] 王璞[2] 

机构地区：[1]山东省农业科学院原子能农业应用研究所,山东济南250100 [2]中国农业大学农学与生物技术学院,北京100094 [3]山东省农业科学院中心实验室,山东济南250100

出　　处：《麦类作物学报》2008年第5期819-823,共5页Journal of Triticeae Crops

基　　金：国家“863”计划项目(2002AA2Z4021-1)

摘　　要：为了给小麦群体的快速诊断提供依据,利用数字图像技术对小麦群体整齐度进行了测量分析。从小麦田间水平拍摄的图像中提取了反映小麦行内整齐度的数字图像指标——群体整齐度系数(PUI),并构建了PUI与总茎数、株高整齐度、绿叶数整齐度以及目测整齐度的回归估测模型。经显著性检验,这些模型均达到显著水平,可用于估测相应的小麦群体农艺指标。同时,分析了图像纵切层数对提取PUI的影响,确定了在本研究条件下提取PUI的临界纵切层数是20层。该测量方法克服了传统整齐度测量的诸多缺陷,具有一定的可行性,对整齐度的快速获取具有参考价值。The population uniformity of wheat was an important index to measure the horizontal distribution in field and one of major contents of the comprehensive diagnosis for wheat.In this paper,the population uniformity index（PUI） was extracted from canopy image of wheat,then regression models between PUI and total stem number,plant uniformity,green leaves uniformity,subjective uniformity were developed,respectively.The significance test indicated that the models and parameters were significant,so they can be used in estimating agronomic traits of wheat well.Meanwhile,the effects of image layers number upon PUI were analyzed,and the results showed that the critical number of image layers was 20 under this research conditions.

关 键 词：小麦 群体整齐度 数字图像 测量方法 

分 类 号：S512.1[农业科学—作物学] S311