摘要
应用计算机数字图像技术研究了诊断棉花早衰程度的方法。在光照条件下采集了棉田群体图像,然后分别提取了红绿蓝(RGB)三色分量和色度、饱和度和亮度指标(HSI)。在RGB和HSI颜色模型下分析了各分量与棉花早衰程度的相关特性。然后得出诊断棉花早衰程度的指标。分析结果表明:颜色分量G/(R+B)和色度H分量与棉花早衰程度线性相关,可用作利用数字图像技术快速诊断作物长势的指标,而其它分量与棉花早衰程度没有明显的相关性;诊断模型G/(R+B)分量比色度H分量有更好的拟合优度。
Cotton premature senescence caused by inherent or environmental factors often produces an abnormal process in which the llfe of a partial or whole cotton plant terminates prematurely within the growing season. Premature senescence induced either pathologically or physiologically ,appears to be an abnormal termination in a total or partial plant before maturity , which usually causes depressed economical parameters in cotton. How to determine the premature senescence degree would be an important research area. Image processing is the best way to solve such problems. The method to diagnose the premature senescence grades of cotton was studied with the imaging process. The images of the cotton fields were taken under the sunlight conditions, and then the red, green, blue（RGB）, their relative ratios （r, g, b）, and the hues of the images were calculated. The correlations among premature senescence grades and their color parameters were analyzed using the RGB and the HSI model. The results showed that there were high linear correlations between the premature senescence grades and G/（R＋B） weight, and between the premature grades and the hue. So that the two parameters could be used as the indices of growth for the fast diagnosis using such image processing.
出处
《棉花学报》
CSCD
北大核心
2006年第3期160-163,共4页
Cotton Science
基金
科技部公益项目(2005-101588)"我国棉情监测预警信息系统研究与应用"
