| 基于连续小波变换定量反演冬小麦叶片含水量研究 |
| |
| 引用本文： | 王延仓,张萧誉,金永涛,顾晓鹤,冯 华,王 闯.基于连续小波变换定量反演冬小麦叶片含水量研究[J].麦类作物学报,2020,40(4):503-509. |
| |
| 作者姓名： | 王延仓 张萧誉 金永涛 顾晓鹤 冯 华 王 闯 |
| |
| 作者单位： | （1. 北华航天工业学院计算机与遥感信息技术学院，河北廊坊 065000;2. 北京农业信息技术研究中心，北京 100097;3. 河北省航天遥感信息处理与应用协同创新中心,河北廊坊 065000;4. 国家农业信息化工程技术研究中心，北京 100097） |
| |
| 基金项目： | 国家重点研发计划项目（2016YFD0300609）;北京市农林科学院科技创新能力建设专项（KJCX20170705）;国家自然科学基金项目（41401419）;河北省教育厅青年基金项目（QN2019213）;河北省青年科学基金项目（D2017409021） |
| |
| 摘 要： | 为了解连续小波转换对利用冬小麦冠层高光谱数据反演叶片含水量精度的提高效果,以河北省衡水市安平县为研究区,基于野外高光谱数据,提取、筛选其光谱特征敏感波段,应用光谱指数、连续小波变换进行光谱处理,并采用偏最小二乘法构建冬小麦叶片含水量的定量反演模型。结果表明,连续小波变换可明显凸显冬小麦冠层光谱特征,提升其对叶片含水量的敏感性。在连续小波变换下,基于1尺度构建的冬小麦叶片含水量的反演模型为最优模型,模型的决定系数(r~2)和RMSE分别为0.756和0.994%,独立样本验证时r~2和RMSE分别为0.766和1.713%,说明反演模型的拟合效果和预测精度均较高。因此,利用连续小波变换可将冠层光谱信息进行二次分配,能有效将有益信息与噪声信息进行分离,提升光谱信息对冬小麦叶片水含量的敏感性,增强冬小麦叶片水含量的预测能力与稳定性。
|
| 关 键 词： | 叶片含水量 高光谱 连续小波变换 偏最小二乘 |
Quantitative Retrieval of Water Content in Winter Wheat Leaves Based on Continuous Wavelet Transform |
| |
| WANG Yancang,ZHANG Xiaoyu,JIN Yongtao,GU Xiaohe,FENG Hu,WANG Chuang.Quantitative Retrieval of Water Content in Winter Wheat Leaves Based on Continuous Wavelet Transform[J].Journal of Triticeae Crops,2020,40(4):503-509. |
| |
| Authors: | WANG Yancang ZHANG Xiaoyu JIN Yongtao GU Xiaohe FENG Hu WANG Chuang |
| |
| Abstract: | In order to clarify the effect of continuous wavelet transform(CWT) on the accuracy of canopy hyperspectral data inversion in winter wheat,based on the field hyperspectral data,the sensitive band of spectral characteristics was extracted and screened in Anping county,Hengshui city,Hebei Province. Spectral index and CWT were used for spectral processing,and the quantitative inversion model of wheat water content was constructed by partial least squares method. The results showed that CWT could obviously highlight the spectral characteristics of winter wheat canopy and enhance its sensitivity to leaf water content. Under the CWT,the inversion model of winter wheat leaf water content based on the scale of 1 is the optimal model. The coefficients(r) and RMSE of the model are 0.756 and 0.994%,respectively. r and RMSE of the independent samples are 0.766 and 1.713%,respectively,which showed that the inversion model has high prediction accuracy. Therefore,CWT can be used to redistribute the spectral information of the canopy,effectively distinguish the useful information from the noise,improve the sensitivity of the spectral information to the water content of winter wheat leaves,and enhance the prediction ability and stability of the water content of winter wheat leaves. |
| |
| Keywords: | Water content of wheat leaves Hyperspectral Continuous wavelet transform Partial least squares |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《麦类作物学报》浏览原始摘要信息 |
| 点击此处可从《麦类作物学报》下载免费的PDF全文 |
|
