基于数码图像识别的棉花氮营养诊断研究

Cotton Nitrogen Nutrition Diagnosis Based on Digital Image

研究田间试验条件下不同施肥处理棉花不同叶位图像色彩参数(G、NRI、NGI、NBI、G/R和G/B)与硝态氮含量、叶绿素测量值(SPAD)、叶绿素含量等营养指标间的相关性,确立棉花氮素营养诊断的最佳色彩参数和曲线方程,以期为新型数码图像技术在棉花氮素营养诊断的应用研究提供理论基础。于2012—2013年在安徽省农业科学院棉花研究所安庆试验基地进行不同施肥处理的田间试验,供试品种为‘湘杂棉8号’F_1。设置8个施肥处理。分别在棉花蕾期、花铃期用Nikon D80数码相机获取棉花不同叶位图像并取样分析,研究数码相机进行棉花氮素营养诊断的最佳色彩参数,确定棉花氮素营养诊断的曲线方程。结果表明:(1)倒3叶硝态氮含量与红光标准化值NRI的相关性最好,R^2=0.8754。功能叶倒4叶次之,R^2=0.8013。(2)除倒1叶外,各叶位的SPAD值与数字化指标之间均有着良好的相关性。倒2叶与绿光标准化值NGI的相关性最好,相关系数为0.9591。(3)对于叶绿素含量,倒1叶与蓝光值B值相关性最好,为曲线正相关,R^2=0.9444。其次为倒3叶、倒4叶,相关系数分别为0.9294、0.931。因此,在进行棉花不同叶位氮素营养诊断时,应选择上部叶位倒1叶、倒2叶、倒3叶、倒4叶,并选择色彩参数B值、蓝光标准化值NBI、NRI进行相关性分析与诊断。

To provide a scientific basis for the application of digital image processing technique in nitrogen diagnosis of cotton, different fertilization treatments were set to explore relationship between the color parameters （G, NRI, NGI, NBI, G/R and G/B） of cotton leaves at different positions and stem sap nitrate concentration, SPAD readings and Chlorophyll content, to determine the best digital parameter and regression equation. A field experiment was conducted with different nitrogen application rates at the Experimental Farm of Anhui Academy of Agricultural Sciences from 2012-2013. Taking ＇Xiangza 8＇ as the test cultivar, the pictures of cotton leaves were obtained with Nikon D80 digital camera at bud stage and boll forming stage, meanwhile, cotton samples of N status were analyzed to determine the best color parameters and regression equations in cotton. The results showed that： （1） the correlation of stem sap nitrate content of inverse 3^rd leaf and NR/was the best, R^2=0.8754, the correlation of functional inverse 4^th leaf and NRI was also good, R^2=0.8013; （2） the SPAD readings and digital indicators had a good correlation on leaves at dllIerent posmons except inverse 1^st leaf, the best correlation was between inverse 2^th leaf and NGI, the correlation coefficient was 0.9591; （3） as to the chlorophyll content, the correlation of B and inverse 1^st leaf was the best, they had a positive curvilinear correlation, R2=0.9444, followed by the correlation of B and inverse 3^rd leaf and 4th leaf, the correlation coefficients was 0.9294 and 0.931, respectively. Therefore, in cotton nitrogen nutrition diagnosis of leaves at different positions, we should choose leaves of upper position which were inverse 1^st, inverse 2^th, inverse 3^rd and inverse 4^th, and also select the color index B, NBI and NR1.