论文部分内容阅读
为了探索运用数码照片中光谱(红、绿、蓝)的像素计算得到的冠层覆盖度(canopy cover,CC)对玉米长势及氮素营养状态进行非破坏性监测的技术。通过获取玉米冠层的数码照片图像,定量化数码照片色彩参数与作物叶面积指数(leaf area index,LAI)、冠层干重(shoot dry matter weight,DM)、叶片氮素含量(leaf nitrogen content percentage,N%)之间的关系。试验于2012年和2013年在中国农业科学院试验田进行,运用基于Visual Basic Version 6.0研发的玉米冠层图像分析系统,分析了玉米品种中单909在3个氮素水平条件下分别于9叶展时期、抽雄期和灌浆期的CC、11种色彩指数与植株LAI,DM,N%及产量之间的相关性,并对相关性显著的指标进行了拟合与建模。结果表明,CC与LAI(r=0.93,p<0.01),DM(r=0.94,p<0.01),N%(r=0.82,p<0.01)之间均达到了极显著水平;用CC估算LAI,DM和N%的模型均为幂函数,方程式分别是y=3.281 2x~(0.763 9),y=283.658 1x~(0.553 6),y=3.064 5x~(0.932 9);用与建模相独立的数据对模型验证,结果表明,CC估算LAI模型的实测值与模拟值基于1∶1直线的R~2,RMSE和RE分别是0.996,0.035和1.46%;CC估算DM模型的R~2,RMSE和RE分别是0.978,5.408g和2.43%;CC估算N%模型的R~2,RMSE和RE分别是0.990,0.054和2.62%。综上所述,模型能够较准确的通过CC估算不同氮肥水平条件下玉米9叶展时期、抽雄期和灌浆期的LAI,DM与N%,表明应用数码相机的光谱信息可实现对玉米生长过程中的生长状况及氮素营养状态进行实时无损快速监测与预测。
In order to explore techniques for non-destructive monitoring of maize growth and N status using canopy cover (CC) calculated from pixels in spectra (red, green and blue) of digital photographs. The digital photos of corn canopy were used to quantify the relationship between color parameters of digital photos and leaf area index (LAI), shoot dry matter weight (DM), leaf nitrogen content percentage, N%) the relationship between. The experiment was carried out in the experimental fields of Chinese Academy of Agricultural Sciences in 2012 and 2013. The corn canopy image analysis system developed based on Visual Basic Version 6.0 was used to analyze the effects of single 909 maize variety under the three nitrogen levels at 9 leaf stage , CC at tasseling stage and grain filling stage, correlation between 11 color indices and plant LAI, DM, N% and yield, and fitting and modeling of significant correlations. The results showed that there was a significant level between CC and LAI (r = 0.93, p <0.01), DM (r = 0.94, p <0.01) and N% The models of LAI, DM and N% are power functions, and the equations are respectively y = 3.281 2x ~ (0.763 9), y = 283.658 1x ~ (0.553 6), y = 3.064 5x ~ (0.932 9) The results show that the estimated RMSE and RE of the LAI model based on the 1: 1 line are 0.996, 0.035 and 1.46%, respectively. The CC estimates of the R ~ 2, RMSE and RE are 0.978, 5.408g and 2.43%, respectively. The R ~ 2, RMSE and RE of CC estimation N% model are 0.990, 0.054 and 2.62% respectively. In summary, the model can accurately estimate the LAI, DM and N% of maize at leaf-setting stage, tasseling stage and filling stage under different levels of nitrogen fertilizer by CC, indicating that the application of digital camera spectral information can achieve the growth of maize In the growth status and nitrogen status of non-destructive real-time rapid monitoring and prediction.