基于机器学习算法的滴灌棉花地上部氮浓度估测模型研究  

Research on Estimation Model of Aboveground Nitrogen Concentration in Drip Irrigation Cotton Based on Machine Learning Algorithm

作　　者：王鹏翔 陈翔宇 魏春月 马怡茹 秦诗哲 周泽轩 张泽[1] WANG Pengxiang;CHEN Xiangyu;WEI Chunyue;MA Yiru;QIN Shizhe;ZHOU Zexuan;ZHANG Ze(Oasis Ecological Agriculture Key Laboratory of Xinjiang Production and Construction Corps,College of Agriculture,Shihezi University,Shihezi,Xinjiang 832003)

机构地区：[1]石河子大学农学院新疆生产建设兵团绿洲生态农业重点实验室,新疆石河子832003

出　　处：《中国农学通报》2024年第4期148-157,共10页Chinese Agricultural Science Bulletin

基　　金：兵团强青项目“棉花氮营养亏缺早期叶绿素荧光遥感监测技术研究”(2022CB002-01);天山英才培养计划;八师石河子市重点领域创新团队计划(2023TD01)。

摘　　要：地上部氮浓度是准确诊断作物氮素丰缺进而评价其生长状况的重要指标。通过构建基于高光谱的滴灌棉花地上部氮浓度估测模型,实现棉花氮素含量的实时、无损、精确获取,为精准施肥提供理论依据和技术支撑。以新疆北部主栽棉花‘新陆早45号’和‘新陆早53号’为供试品种,设置6个施纯氮处理(0、120、240、348、360、480 kg/hm^(2)),测量棉花冠层高光谱信息,利用函数变换去除冗余,一阶与二阶导数筛选结果相似,倒数的对数筛选结果较为分散。采用机器学习权重排序进行特征筛选,共选出359、371、751、752、746、739、755 nm等7个特征波段。同时遍历波长组合,优化前人研究与氮素高相关的植被指数,共选出RVI′_(810,460)、NDVI′_(811,856)、NDVI′_(750,705)、RVI′_(740,720)、RVI′_(851,852)、DVI′_(359,360)、NDVI′_(851,852)等7个光谱指数。将筛选得到的特征波段与植被指数分别利用岭回归、决策树、引导聚类、增强学习算法与棉花氮素建立养分估测模型,最终Adaboost迭代算法所建立滴灌棉花地上部氮浓度估测模型效果最优,模型精度R^(2)达到0.911,RMSE为1.362。利用光谱信息可以有效反演棉花氮营养状态,基于植被指数构建的模型估测精度较特征波段更为稳定;对现有的植被指数特征波段进行优化,可以有效提升模型的估测精度;对比分析不同建模方式下模型精度,集成学习相比单机器学习在进行棉花氮营养估测时更有优势。Aboveground nitrogen concentration is an important indicator for accurately diagnosing crop nitrogen abundance and deficiency and thus evaluating its growth status.By constructing a hyperspectral-based nitrogen concentration estimation model for aboveground nitrogen concentration in drip-irrigated cotton,the real-time,non-destructive and accurate acquisition of nitrogen content in cotton can be realized,providing theoretical basis and technical support for precise fertilization.Six N application treatments(0,120,240,348,360,480 kg/hm^(2) of pure N)were set up with the main cotton cultivars‘Xinluzao 45’and‘Xinluzao 53’in northern Xinjiang as the test varieties.Measuring the hyperspectral information of the cotton canopy and removing redundancy using functional transformations,the first-and second-order derivative screening results were similar,and the logarithmic screening of the inverse was more scattered.Machine learning weight ordering was used for feature screening,and a total of seven feature bands,including 359,371,751,752,746,739,and 755 nm,were selected.We also traversed the wavelength combinations to optimize the vegetation indices associated with high nitrogen in previous studies,and selected seven spectral indices including RVI′810,460,NDVI′_(811,856),NDVI′_(750,705),RVI′_(740,720),RVI′_(851,852),DVI′_(359,360),NDVI′_(851,852).The filtered feature bands and vegetation indices were used to establish a nutrient estimation model with cotton nitrogen using ridge regression,decision tree,bootstrap clustering,and enhancement learning algorithms,respectively.Finally,the aboveground nitrogen concentration estimation model of drip-irrigated cotton established by the Adaboost iterative algorithm was the most effective,with a model accuracy of R^(2) reaching 0.911 and an RMSE of 1.362.Spectral information can be used to effectively invert the nitrogen nutritional status of cotton,and the estimation accuracy of the model constructed based on the vegetation index is more stable than that of the

关 键 词：棉花冠层 高光谱 地上部氮浓度 植被指数 机器学习 

分 类 号：S562[农业科学—作物学]