An Improved Combination of Spectral and Spatial Features for Vegetation Classification in Hyperspectral Images

Abstract

Due to the advances in hyperspectral sensor technology, hyperspectral images have gained great attention in precision agriculture. In practical applications, vegetation classification is usually required to be conducted first and then the vegetation of interest is discriminated from the others. This study proposes an integrated scheme (SpeSpaVS_ClassPair_ScatterMatrix) for vegetation classification by simultaneously exploiting image spectral and spatial information to improve vegetation classification accuracy. In the scheme, spectral features are selected by the proposed scatter-matrix-based feature selection method (ClassPair_ScatterMatrix). In this method, the scatter-matrix-based class separability measure is calculated for each pair of classes and then averaged as final selection criterion to alleviate the problem of mutual redundancy among the selected features, based on the conventional scatter-matrix-based class separability measure (AllClass_ScatterMatrix). The feature subset search is performed by the sequential floating forward search method. Considering the high spectral similarity among different green vegetation types, Gabor features are extracted from the top two principal components to provide complementary spatial features for spectral features. The spectral features and Gabor features are stacked into a feature vector and then the ClassPair_ScatterMatrix method is used on the formed vector to overcome the over-dimensionality problem and select discriminative features for vegetation classification. The final features are fed into support vector machine classifier for classification. To verify whether the ClassPair_ScatterMatrix method could well avoid selecting mutually redundant features, the mean square correlation coefficients were calculated for the ClassPair_ScatterMatrix method and AllClass_ScatterMatrix method. The experiments were conducted on a widely used agricultural hyperspectral image. The experimental results showed that (1) the The proposed ClassPair_ScatterMatrix method could better alleviate the problem of selecting mutually redundant features, compared to the AllClass_ScatterMatrix method; (2) compared with the representative mutual information-based feature selection methods, the scatter-matrix-based feature selection methods generally achieved higher classification accuracies, and the ClassPair_ScatterMatrix method especially, produced the highest classification accuracies with respect to both data sets (87.2% and 90.1%); and (3) the proposed integrated scheme produced higher classification accuracy, compared with the decision fusion of spectral and spatial features and the methods only involving spectral

Joint Sparse and Low-Rank Multitask Learning with Laplacian-Like Regularization for Hyperspectral Classification

“…The normalized residues of the same sample from class 4 (i.e., corn) located at (35,5) are plotted in Figure 13, from which one can observe that, by utilizing the 3D-MPs features, the sllMTL 2 and sllMTL 3 yield more accurate classification results than the MTL and sllMTL 1 . Moreover, it is shown in Tables 4-6 that the sllMTL 2 and sllMTL 3 outperform the MTL and sllMTL 1 , and the sllMTL 3 performs much better than the sllMTL 2 since a Laplacian-like regularization is added in the sllMTL 3 to take full advantage of the label information of training samples. In addition, as displayed in Figures 9-11, the classification maps of sllMTL 3 are more close to the ground truth (see Figures 4a, 5b and 6b) than other methods.…”

“…It is observed from Table 5 that the classification performance of 3D-MPs is better than the EMP with the same classifier. The coefficient matrices of a test sample from class 4 (i.e., corn) located at (35,5) in the Indian Pines data for (b) MTL, (c) sllMTL 1 , (d) sllMTL 2 , and (e) sllMTL 3 . The x-axis labels the task number and the y-axis labels the representation number.…”

“…Another important issue is to discuss the sensitivity of the parameters in the proposed 3D-MPs-sllMTL 3 . There are two key parameters in the 3D-MPs-sllMTL 3 : λ and β.…”

“…Thanks to the high spectral and spatial resolution, hyperspectral images (HSIs) have opened up new avenues for the remote sensing applications, including earth observation [1], environmental science [2], agriculture [3], mineral identification [4], etc. Supervised classification [5][6][7][8], which aims at assigning each pixel in the HSI into an accurate class label with representative training samples, plays an important role in those applications.…”

Hashing Based Hierarchical Feature Representation for Hyperspectral Imagery Classification

“…Hyperspectral sensors can provide images with hundreds of continuous spectral bands, which has attracted a number of applications such as environmental monitoring and mineral prospecting [1][2][3][4]. Among many surveys about hyperspectral imagery (HSI) analysis, land cover accurate classification is an important research topic.…”