Relief-F筛选波段的小麦白粉病早期诊断研究

为了准确监测小麦白粉病染病早期病情,给喷药防治提供技术指导,论文将染病初期的小麦叶片作为研究对象。首先,利用高光谱图像数据,通过图像特征分割出叶片区域和病斑区域,定量计算病情严重度;其次引入Relief-F算法提取染病早期最敏感波段和波段差,计算出白粉病病害指数PMDI （Powdery mildew disease index）;并通过分析病情指数DI （Disease index）与11种植被指数（含PMDI指数）的相关性及线性模型,得出PMDI模型有最高的决定系数（R2=0.839 9）和最低的均方根误差（RMSE=4.522 0）,效果优于其他病害植被指数的结果（其中,Normalized Difference Vegetation Index,NDVI的模型决定系数最高,R2=0.777 1,RMSE=5.336 4）;最后,选择PMDI和NDVI植被指数分别构建小麦白粉病染病早期病情严重度的支持向量回归模型。结果表明:经敏感波段筛选构建的PMDI指数的预测结果更好,预测模型的R2=0.886 3,RMSE=3.553 2,可以实现小麦白粉病早期无损诊断,这为指导作物病害喷药防治提供重要的技术支撑。     

基于多光谱成像选取四季豆叶片的特征波段

在400~720nm波段范围,基于液晶可调谐滤波器(LCTF)和CMOS相机组合的多光谱成像系统,以四季豆叶片为研究对象每隔5 nm进行成像。根据图像亮度信息法和波段指数法的相关原理,首先分别计算得到各波段四季豆叶片的波段指数值和可识别度;然后对四季豆叶片的波段指数值和可识别度进行排序,综合图像的灰度离散、亮度信息丰富和波段的相关性小等特点,得出545、630、645、720、650和570 nm波段有较大的波段指数值和较好的识别度;最后根据最小欧氏距离法和光谱角度匹配法分别对四季豆叶片的特征波段的分类精度予以计算,两种方法的分类精度分别为100.00%和83.33%,得出选取的特征波段对四季豆叶片具有较好的分类精度。因此,545、630、645、720、650和570 nm波段可作为四季豆叶片的特征波段。     

独立成分分析的多光谱图像融合算法

为解决图像空间信息的问题,文章提出了一种独立成分分析的多光谱图像融合算法,将多光谱图像的RGB 3个波段和近红外图像共4个波段进行独立成分分析变化,并对其做加权平均得到主图像信息,将主图像信息与全色图像加权求平均得到一副新的图像,然后将这幅图像还原到4个波段得到融合后的结果图像。     

基于模糊C-均值聚类的作物叶部病斑图像分割

农作物病害的发生对我国农业生产危害较大,运用机器识别技术对农作物病害图像进行自动识别有着重要的意义.对于玉米叶部病斑,大多数分割算法不能很好地分割出来,采用快速模糊C-均值聚类算法,对玉米染病叶片图像进行分割,并通过实验验证了这种算法在聚类优化性能不变的前提下,病斑和背景的区分很明显,分割效果较好.     

基于室内高光谱激光雷达的典型树种叶片光谱观测和分类

高光谱激光雷达综合了高光谱和激光雷达特征,可为植被生理生化参数提取提供更加精确的遥感探测,但其应用潜力尚未得到充分挖掘。以北京10个典型树种的单叶为样本,开展室内高光谱激光雷达的叶片观测试验,并进行树种分类研究,为未来高光谱激光雷达的林业应用提供基础。首先进行可调谐高光谱激光雷达(Hyperspectral LiDAR,HSL)叶片高光谱测量,并完成与ASD地物光谱仪所测数据对比实验;其次,应用随机森林方法实现10种叶片的分类研究,其输入的特征指数为融合全部波段、部分敏感波段的光谱指数。结果表明:(a)HSL在波段650～1000 nm (71个通道)内观测的叶片高光谱和ASD光谱一致(R~2=0.9525～0.9932,RMSE=0.0587);(b)只用原始波段反射率分类精度为78.31%,其中分类贡献率最大波段的是650～750 nm,使用此波段进行分类精度为94.18%,表明利用红边波段(650～750nm)进行树种分类是十分有效的;(c)对树种敏感的波段为680 nm、685 nm、690 nm、715 nm、720 nm、725 nm、730 nm;(d)结合敏感波段光谱指数与植被指数分类精度82.65%。该研究结果表明在单叶级别,利用高光谱激光雷达能够准确地反映目标叶片的光谱特征并且能有效进行树种分类;未来将可能在野外应用中精确提取目标的生理生化参数。     

基于高光谱技术的长枣内外品质同时检测

采用近红外高光谱成像技术,对长枣表面轻微损伤 和果肉硬度进行无损检测。在970～ 1630nm波 长范围内对高光谱图像数据进行主成分分析(PCA),得到第3主成分图像最适合检测长枣表面 损伤。波段比(BR)算法中, 选取1387nm和1 229nm两个波段的图像进行比值运算, 采用 1455nm单波 段图像构建掩膜作用于比值图像, 最后对图像进行阈值分割和形态学变换完成损伤区域的特征提取。BR算法检测长枣轻微损 伤的准确率达 到91.5%。对反射光谱进行多元散射校正(MSC)后与长枣果肉硬度值进 行回归分析,选择相关系数较大的5个特 征波长作为BP神经网络输 入,建立果肉硬度预测模型。预测集相关系数R和均 方根误差(RMSEP)分别为0.904和15.163。研究结果表明,利用高光谱成像技术可以实现长枣内外品质同时检 测。     

基于分段主成分分析与波段比的鸡胴体表面粪便污染物检测

以鸡胴体为研究对象,应用高光谱图像技术结合分段主成分分析和波段比等数据处理方法来检测鸡胴体表面粪便污染物。首先采集400~1000 nm的鸡胴体表面高光谱图像;然后应用分段主成分分析获得7个特征波长(520.64,542.12,561.61,577.04,595.6,703.82和852.1 nm),并以577.04/520.64 nm波段比图像和852.1/703.82 nm波段比图像进行一次波段加运算后的图像作为检测鸡胴体表面粪便污染物的特征图像;最后运用阈值分割和数学形态学完成粪便污染物的提取。实验结果表明,对60个鸡胴体样本进行检测,盲肠、直肠和十二指肠粪便污染物检测正确率分别为100%,100%和96%,检测总正确率为93.3%。     

基于粒子群优化聚类的高光谱图像异常目标检测

高光谱图像的高维特性增加了图像的信息量,但 是同时也带来了“维数灾难”问题 。在高光谱图像异常目标检测 过程中,如何更好降低维数,去除波段冗余性和最大程度抑制背景干扰成为亟需要解决的 问题。针对此,本文提出了基于粒子群 优化(PSO)聚类的高光谱图像异常目标检测算法。算法首先利用粒子群方法对传统的k-均 值聚类进行优化,在不改变高光谱图像 波段特征的基础上用新的聚类方法对图像进行了波段子集类划分,使得具有相似特性的波段 归为一类;然后,通过主成分分析(PCA)变换使 得聚类后的图像数据中含有的异常目标变得突出,同时抑制背景干扰;最后,提取各子集主 成分中含有最大四阶累积量值的波 段,构成最优波段子集,并与核RX算法结合进行异常检测。利用真实的AVIRIS高光谱图像对 算法进行仿真,结果表明,算法检测精度高,虚警率低。     

用于高光谱图像分类的归一化光谱指数的构建与应用

成像高光谱的近地田间应用为农业定量遥感的发展提供了新的契机。如何发挥其图谱合一的数据优势,尤其在解析土壤、阴影等背景地物对作物养分反演模型的影响需要关注。该研究借助可见/近红外成像高光仪,在近地田间采集小麦群体的成像立方体,根据影像中光照裸土、阴影裸土、光照叶片和阴影叶片的反射光谱特征建立了归一化光谱分类指数,并应用该指数提取大豆影像中不同类型地物的光谱,分析了背景土壤剔除前后的大豆植被归一化光谱与叶绿素密度的决定系数变化情况。结果表明:土壤和阴影叶片光谱去除后,反演叶绿素密度的敏感波段由红-近红外区间（727 nm,922 nm）向蓝、绿,尤其是红波段（710 nm,711 nm）移动。对叶绿素密度敏感的波段区间表现为可见光增加,近红外减少,且红边波段决定系数最高。由此说明,基于归一化光谱指数的植被光谱提纯对定量遥感反演研究具有重要意义。     

选剔同步的高光谱遥感图像波段选择算法

为了降低高光谱遥感图像冗余度,减少后续的计算复杂度,提出了选剔同步的高光谱遥感图像波段选择算法。以主成分分析后的数据作为参考波段来源,以互信息作为选取波段的相似性度量,引入R-KL系数作为剔除波段的判别准则,利用边选取边剔除的方式进行波段选择。为了验证该算法的有效性,运用贝叶斯分类法对降维后波段进行分类,并与自适应波段选择和基于最大信息量的波段选择算法进行比较。结果显示当选取波段数目较少时,该算法的分类效果优于上述两种算法,当选取波段数目较多时,3种算法分类效果相当,故该算法是一种有效的波段选择算法。     

Application of the stochastic mixing model to hyperspectral resolution enhancement

A maximum a posteriori (MAP) estimation method is described for enhancing the spatial resolution of a hyperspectral image using a higher resolution coincident panchromatic image. The approach makes use of a stochastic mixing model (SMM) of the underlying spectral scene content to develop a cost function that simultaneously optimizes the estimated hyperspectral scene relative to the observed hyperspectral and panchromatic imagery, as well as the local statistics of the spectral mixing model. The incorporation of the stochastic mixing model is found to be the key ingredient for reconstructing subpixel spectral information in that it provides the necessary constraints that lead to a well-conditioned linear system of equations for the high-resolution hyperspectral image estimate. Here, the mathematical formulation of the proposed MAP method is described. Also, enhancement results using various hyperspectral image datasets are provided. In general, it is found that the MAP/SMM method is able to reconstruct subpixel information in several principal components of the high-resolution hyperspectral image estimate, while the enhancement for conventional methods, like those based on least squares estimation, is limited primarily to the first principal component (i.e., the intensity component).     

基于四阶累积量的波段子集高光谱图像异常检测

针对由于高光谱图像光谱和空间分布的复杂性导致核RX算法检测性能不高这一问题,提出了基于四阶累积量的波段子集非线性异常检测算法。首先先依据各相邻波段间的相关系数,将原始图像数据划分为多组波段子集;然后,利用主成分分析(PCA)构造的正交子空间对各波段子集进行背景抑制,得到图像误差数据;在此基础上,再次利用PCA提取各波段子集的特征信息,使异常目标信息集中于前面几个波段;最后,提取各子集主成分中含有最大四阶累积量值的波段,构成最优波段子集,并与核RX算法结合进行异常检测。利用真实的AVIRIS高光谱图像对算法进行仿真,结果表明,本文算法检测精度高,虚警率低,性能明显优于核RX算法。     

邻域谱概率协同表示的高光谱图像分类方法

为了提高高光谱遥感图像的分类精度, 通过结合像元邻域谱与概率协同表示方法, 提出了一种基于空间信息与光谱信息的分类方法。首先采用插值方法生成像元的邻域谱, 然后用概率协同表示方法将待测样本进行分类。用所提出的方法在AVIRIS Indian Pines和Salinas scene高光谱遥感数据库上进行分类实验, 并和主成分分析、支持向量机、稀疏表示分类器和协同表示分类器方法进行了比较。结果表明, 所提出的方法在AVIRIS Indian Pines数据库上识别精度比主成分分析法高约17%, 其识别精度和kappa系数都优于另外4种方法。该方法是一种较好的高光谱遥感图像分类方法。     

Fusion of hyperspectral data using segmented PCT for color representation and classification

Fusion of hyperspectral data is proposed by means of partitioning the hyperspectral bands into subgroups, prior to principal components transformation (PCT). The first principal component of each subgroup is employed for image visualization. The proposed approach is general, with the number of bands in each subgroup being application dependent. Nevertheless, the paper focuses on partitions with three subgroups suitable for RGB representation. One of them employs matched-filtering based on the spectral characteristics of various materials and is very promising for classification purposes. The information content of the hyperspectral bands as well as the quality of the obtained RGB images are quantitatively assessed using measures such as the correlation coefficient, the entropy, and the maximum energy-minimum correlation index. The classification performance of the proposed partitioning approaches is tested using the K-means algorithm.     

Hyperspectral image quality based on convolutional network of multi-scale depth

Hyperspectral imagery has been widely used in military and civilian research fields such as crop yield estimation, mineral exploration, and military target detection. However, for the limited imaging equipment and the complex imaging environment of hyperspectral images, the spatial resolution of hyperspectral images is still relatively low, which limits the application of hyperspectral images. So, studying the data characteristics of hyperspectral images deeply and improving the spatial resolution of hyperspectral images is an important prerequisite for accurate interpretation and wide application of hyperspectral images. The purpose of this paper is to deal with super-resolution of the hyperspectral image quickly and accurately, and maintain the spectral characteristics of the hyperspectral image, makes the spectral separability of the substrate in the original image remains unchanged after super-resolution processing. This paper first learns the mapping relationship between the spectral difference of low-resolution hyperspectral image and the spectral difference of the corresponding high-resolution hyperspectral image based on multiple scale convolutional neural network, Thus, apply this mapping relationship to the input low-resolution hyperspectral image generally, getting the corresponding high resolution spectral difference. Constrained space by using the image of reconstructed spectral difference, this requires the low-resolution hyperspectral image generated by the reconstructed image is to be close to the input low-resolution hyperspectral image in space, so that the whole process becomes a closed circulation system where the low-resolution hyperspectral image generation of high-resolution hyperspectral images, then back to low-resolution hyperspectral images. This innovative design further enhances the super-resolution performance of the algorithm. The experimental results show that the hyperspectral image super-resolution method based on convolutional neural network improves the input image spatial information, and the super-resolution performance of the model is above 90%, which can maintain the spectral information well.     

Super-resolution reconstruction of hyperspectral images.

Hyperspectral images are used for aerial and space imagery applications, including target detection, tracking, agricultural, and natural resource exploration. Unfortunately, atmospheric scattering, secondary illumination, changing viewing angles, and sensor noise degrade the quality of these images. Improving their resolution has a high payoff, but applying super-resolution techniques separately to every spectral band is problematic for two main reasons. First, the number of spectral bands can be in the hundreds, which increases the computational load excessively. Second, considering the bands separately does not make use of the information that is present across them. Furthermore, separate band super-resolution does not make use of the inherent low dimensionality of the spectral data, which can effectively be used to improve the robustness against noise. In this paper, we introduce a novel super-resolution method for hyperspectral images. An integral part of our work is to model the hyperspectral image acquisition process. We propose a model that enables us to represent the hyperspectral observations from different wavelengths as weighted linear combinations of a small number of basis image planes. Then, a method for applying super resolution to hyperspectral images using this model is presented. The method fuses information from multiple observations and spectral bands to improve spatial resolution and reconstruct the spectrum of the observed scene as a combination of a small number of spectral basis functions.     

基于高光谱图像主成分分量的小目标检测算法研究

提出了一种基于图像主成分分量的高光谱小目标检测算法．作为一种多元数据集合,通常高光谱数据形成的几何体是一个超平面．主成分分析能有效估计这一几何体的本征维数．显著特征值对应的主成分体现了几何体大部分信息;而不显著特征值对应的主成分则代表了正交于几何体的信息,而这些信息中则包含了重要的内容,例如目标特性。文中提出的方法就是利用这些不显著的主成分分量来进行小目标检测．该方法减少了对先验光谱信息的依赖,提高了算法的实用性．