土壤含水量(soil water content, SWC)和土壤含盐量(soil salt content, SSC)是影响作物生长和农业生产力的重要因素。光学卫星图像已成为SWC和SSC估计的主要数据源。然而,在SWC或SSC变化较大地区,土壤水分和盐分会影响对方对光谱反射...土壤含水量(soil water content, SWC)和土壤含盐量(soil salt content, SSC)是影响作物生长和农业生产力的重要因素。光学卫星图像已成为SWC和SSC估计的主要数据源。然而,在SWC或SSC变化较大地区,土壤水分和盐分会影响对方对光谱反射率的响应,使得SSC和SWC的反演精度较差。对此,该研究提出了一个半解析性的反射率模型—RVS模型,来模拟植被光谱反射率(R_(v))对作物根区土壤含水量和含盐量的响应;并通过构建的RVS模型,对植被覆盖区域的土壤含水量和土壤含盐量进行同步监测。研究表明:RVS模型在反演研究区土壤含盐量和含水量时,精度较为可靠(水分:决定系数R^(2)为0.63~0.74,均方根误差为0.017~0.028;盐分:决定系数R^(2)为0.68~0.75,均方根误差为0.052 5~0.061 7)。在作物生长过程中,植被光谱反射率对深层土壤的含水量和含盐量的响应比对浅层土壤的含水量和含盐量的响应更加明显,而且随着作物的生长,影响光谱反射率的主导因素从土壤水分慢慢转向土壤盐分和水盐相互作用。该研究在一定程度上揭示了土壤水分、盐分、水盐交互作用对作物光谱反射率的干扰过程,实现土壤水分和盐分的同步监测,对实现区域尺度上土壤含盐量和含水量的精准监测具有一定的意义。展开更多
针对高光谱图像数据高维的特点,为进一步提高图像分类准确率,设计一种融合注意力机制的三维空洞卷积神经网络模型用于高光谱分类问题。该方法以3D卷积为基础,使用多尺寸卷积核策略,从不同尺度提取高光谱图像的特征信息;使用空洞结构卷积...针对高光谱图像数据高维的特点,为进一步提高图像分类准确率,设计一种融合注意力机制的三维空洞卷积神经网络模型用于高光谱分类问题。该方法以3D卷积为基础,使用多尺寸卷积核策略,从不同尺度提取高光谱图像的特征信息;使用空洞结构卷积核,可有效提取特征信息,同时增加网络的感受野。提出一种空间-光谱注意力模块,自适应聚焦信息,增加高光谱图像空间、光谱的特征表达能力。提出的方法在University of Pavia和Indian Pines等公开数据集上测试,分别取得99.61%、99.58%的总体分类准确率。与SVM、2D-CNN、3D-CNN、RES-3D-CNN算法进行比较,该文提出的算法在准确率和分类性能上优于其他算法。展开更多
In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonne...In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonnegative constrained principal component analysis method is proposed to construct a low-dimensional multi-spectral space and accomplish the conversion between the new constructed space and the multispectral space. First, the reason behind the negative data is analyzed and a nonnegative constraint is imposed on the classic PCA. Then a set of nonnegative linear independence weight vectors of principal components is obtained, by which a lowdimensional space is constructed. Finally, a nonlinear optimization technique is used to determine the projection vectors of the high-dimensional multi-spectral data in the constructed space. Experimental results show that the proposed method can keep the reconstructed spectral data in [ 0, 1 ]. The precision of the space created by the proposed method is equivalent to or even higher than that by the PCA.展开更多
文摘土壤含水量(soil water content, SWC)和土壤含盐量(soil salt content, SSC)是影响作物生长和农业生产力的重要因素。光学卫星图像已成为SWC和SSC估计的主要数据源。然而,在SWC或SSC变化较大地区,土壤水分和盐分会影响对方对光谱反射率的响应,使得SSC和SWC的反演精度较差。对此,该研究提出了一个半解析性的反射率模型—RVS模型,来模拟植被光谱反射率(R_(v))对作物根区土壤含水量和含盐量的响应;并通过构建的RVS模型,对植被覆盖区域的土壤含水量和土壤含盐量进行同步监测。研究表明:RVS模型在反演研究区土壤含盐量和含水量时,精度较为可靠(水分:决定系数R^(2)为0.63~0.74,均方根误差为0.017~0.028;盐分:决定系数R^(2)为0.68~0.75,均方根误差为0.052 5~0.061 7)。在作物生长过程中,植被光谱反射率对深层土壤的含水量和含盐量的响应比对浅层土壤的含水量和含盐量的响应更加明显,而且随着作物的生长,影响光谱反射率的主导因素从土壤水分慢慢转向土壤盐分和水盐相互作用。该研究在一定程度上揭示了土壤水分、盐分、水盐交互作用对作物光谱反射率的干扰过程,实现土壤水分和盐分的同步监测,对实现区域尺度上土壤含盐量和含水量的精准监测具有一定的意义。
基金Supported by the Foundation of Anhui Province Key Laboratory of Physical Geographic Environment(2022PGE010)The Fundamental Research Funds for the Central Universities,CHD(300102353508)+5 种基金the Key Laboratory of Southeast Coast Marine Information Intelligent Perception and Application,MNR(22101)National Natural Science Foundation of China(61801211)Natural Science Foundation of Jiangsu Province(BK20221478)Hong Kong Scholars Program(XJ2022043)S&T Program of Hebei(21567624H)Open Project Program of Key Laboratory of Meteorology and Ecological Environment of Hebei Province(Z202102YH)。
文摘针对高光谱图像数据高维的特点,为进一步提高图像分类准确率,设计一种融合注意力机制的三维空洞卷积神经网络模型用于高光谱分类问题。该方法以3D卷积为基础,使用多尺寸卷积核策略,从不同尺度提取高光谱图像的特征信息;使用空洞结构卷积核,可有效提取特征信息,同时增加网络的感受野。提出一种空间-光谱注意力模块,自适应聚焦信息,增加高光谱图像空间、光谱的特征表达能力。提出的方法在University of Pavia和Indian Pines等公开数据集上测试,分别取得99.61%、99.58%的总体分类准确率。与SVM、2D-CNN、3D-CNN、RES-3D-CNN算法进行比较,该文提出的算法在准确率和分类性能上优于其他算法。
基金The Pre-Research Foundation of National Ministries andCommissions (No9140A16050109DZ01)the Scientific Research Program of the Education Department of Shanxi Province (No09JK701)
文摘In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonnegative constrained principal component analysis method is proposed to construct a low-dimensional multi-spectral space and accomplish the conversion between the new constructed space and the multispectral space. First, the reason behind the negative data is analyzed and a nonnegative constraint is imposed on the classic PCA. Then a set of nonnegative linear independence weight vectors of principal components is obtained, by which a lowdimensional space is constructed. Finally, a nonlinear optimization technique is used to determine the projection vectors of the high-dimensional multi-spectral data in the constructed space. Experimental results show that the proposed method can keep the reconstructed spectral data in [ 0, 1 ]. The precision of the space created by the proposed method is equivalent to or even higher than that by the PCA.
