A Novel Hyperspectral Image Classification Pattern Using Random Patches Convolution and Local Covariance

Abstract

Today, more and more deep learning frameworks are being applied to hyperspectral image classification tasks and have achieved great results. However, such approaches are still hampered by long training times. Traditional spectral–spatial hyperspectral image classification only utilizes spectral features at the pixel level, without considering the correlation between local spectral signatures. Our article has tested a novel hyperspectral image classification pattern, using random-patches convolution and local covariance (RPCC). The RPCC is an effective two-branch method that, on the one hand, obtains a specified number of convolution kernels from the image space through a random strategy and, on the other hand, constructs a covariance matrix between different spectral bands by clustering local neighboring pixels. In our method, the spatial features come from multi-scale and multi-level convolutional layers. The spectral features represent the correlations between different bands. We use the support vector machine as well as spectral and spatial fusion matrices to obtain classification results. Through experiments, RPCC is tested with five excellent methods on three public data-sets. Quantitative and qualitative evaluation indicators indicate that the accuracy of our RPCC method can match or exceed the current state-of-the-art methods.

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

“…While adjusting a parameter to an optimum value in a specified range, the remaining parameters are kept constant. In order to validate the effectiveness of the GRR, we select some recent methods called JCRC-MTL [34], RPNet [38], RPCC [39], SSRPNet [41], GRPC [43], RPNet-RF [45] and MS-RPNet [46] for the comparison. In addition, an SVM-based version of GRR (GR-SVM) which classifies the features with SVM instead of RCR is also taken into account to prove the effectiveness of the GRR.…”

“…Over the years, different RPNet-based HSIC methods have been proposed. In random patches convolution and local covariance (RPCC) [39], the dimension of HSI was reduced by maximum noise fraction (MNF) [40]. Then RPNet was used to extract spatial features and local covariance matrices were utilized to extract spectral features.…”

Integrating MNF and HHT Transformations into Artificial Neural Networks for Hyperspectral Image Classification

“…MNF was applied for dimensionality reduction in this study. MNF segregates noise from bands through modified principal-component analysis (PCA) by ranking images on the basis of signal-to-noise ratio (SNR) [39,49]. MNF defines the noise of each band as follows:…”

“…Many state-of-the-art methods have been developed to solve the environmental noise and dimensionality problem of HSIs, such as statistical filters [27], feature-extraction algorithms [28][29][30][31], discrete Fourier transforms and wavelet estimation [32,33], rotation forests [34], morphological segmentation [35,36], support vector machine (SVM) [37], minimum noise fractions (MNFs) [38,39], and empirical mode decomposition (EMD). EMD is a one-dimensional signal-decomposition method that can decompose an input signal into several hierarchical components known as intrinsic mode functions (IMFs) and a residue signal [40][41][42][43].…”

A Multi-Kernel Mode Using a Local Binary Pattern and Random Patch Convolution for Hyperspectral Image Classification

“…Unlike in [38], a random batch classification method [39] used a random strategy to obtain the convolution kernel in the original HSI without training, which can greatly save running time. To utilize the correlation between local spectral features, Sun et al [40] proposed a random batch and local covariance classification framework that combined the covariance matrix with RPNet on the basis of [39], thereby greatly improving the classification accuracy.…”