应用数字图像技术估测冬小麦冠层生物量垂直分布特征的研究  

Application of Digital Image Processing for Determination of Vertical Distribution of Biomass in the Canopy of Winter Wheat

作　　者：单成钢[1] 廖树华[1] 龚宇[1] 梁振兴[1] 王璞[1] 

机构地区：[1]中国农业大学农学与生物技术学院

出　　处：《作物学报》2007年第3期419-424,共6页Acta Agronomica Sinica

基　　金：国家高技术研究发展计划(863计划)项目(2002AA2Z4021-1)

摘　　要：用数字图像技术研究了冬小麦冠层生物量的垂直分布。表明用一行小麦图像比多行小麦图像估测小麦生物量能更好地满足线性回归关系,估测效果更佳,以此为基础进一步研究了分层像素数估测小麦冠层分层现存生物量和有效生物量的方法。利用分层绿色像素数(LGPN)指标定性分析了不同栽培模式下冬小麦群体有效生物量的垂直分布和动态变化,并确定了基于图像特征的可用于定量分析的小麦群体垂直分布指数(I)。The vertical distribution of biomass, which indicates the characteristics of quantity and structure in canopy, is an important index of diagnosis and management for crop growth. However, traditional measurements are time-consuming and labor intensive and can result in significant mechanical damage to plants, such as layer upon layer cut method. The image processing technologies, which may provide more efficient and no-destructive methods for measurement, have recently appeared in the agronomic literature. This paper focused on the determination of biomass vertical distribution in the canopy of winter wheat, based on digital image processing technology. The digital images （2 272 by 1 706 pixels） were taken with digital camera of winter wheat rows in the field plots of an experiment during the 2004 to 2005 growing season. Excess Green image segmentation method was used to separate wheat population and background. At the same time, relative biomass weight was measured using layer upon layer cut method, and the distance between neighbor layers was 20 cm. The image pixels of a single row and biomass could fit regression relationship （ RD1^2 = 0.9682） much better than those of multiple rows and biomass. The regression relationship between the image pixels and dry matter weight much better than those and fresh matter weight. The estimating functions were established between biomass and pixel number （PN） in each layer （ Y = 6 × 10^-5X ＋ 24.72）, and between effective biomass and green pixel number （LGPN） in each layer （ Y = 7 × 10^-5X ＋ 25.09）, and meanwhile the vertical distribution of effective biomass and dynamic of winter wheat canopy under different management conditions were analyzed qualitatively with LGPN. Finally, an index, namely, image pixel vertical distribution index （I）, with which vertical distribution of canopy biomass of winter wheat could be analyzed quantitatively, was determined. The results show that the application of digital image processing has significant potentia

关 键 词：数字图像 冬小麦冠层 生物量 垂直分布 

分 类 号：S512.11[农业科学—作物学]