Deep Convolutional Neural Networks for the Classification of Snapshot Mosaic Hyperspectral Imagery

Abstract

Spectral information obtained by hyperspectral sensors enables better characterization, identification and classification of the objects in a scene of interest. Unfortunately, several factors have to be addressed in the classification of hyperspectral data, including the acquisition process, the high dimensionality of spectral samples, and the limited availability of labeled data. Consequently, it is of great importance to design hyperspectral image classification schemes able to deal with the issues of the curse of dimensionality, and simultaneously produce accurate classification results, even from a limited number of training data. To that end, we propose a novel machine learning technique that addresses the hyperspectral image classification problem by employing the state-of-the-art scheme of Convolutional Neural Networks (CNNs). The formal approach introduced in this work exploits the fact that the spatio-spectral information of an input scene can be encoded via CNNs and combined with multi-class classifiers. We apply the proposed method on novel dataset acquired by a snapshot mosaic spectral camera and demonstrate the potential of the proposed approach for accurate classification.

Comparative Analysis of Generalized Intersection over Union and Error Matrix for Vegetation Cover Classification Assessment

“…Over the last few years, CNNs have been successfully applied to many classical image processing problems, such as denoising, (1) super-resolution, (2) pansharpening, (3,4) segmentation, (5,6) object detection, (7,8) change detection, (9) and classification. (10)(11)(12)(13) The main strengths of CNNs are (i) an extreme versatility that allows them to approximate any sort of linear or nonlinear transformation, including scaling or hard thresholding; (ii) no need to design handcrafted filters, replaced by machine learning; and (iii) high-speed processing, due to parallel computing. On the downside, for correct training, CNNs require the availability of a large amount of data with the ground truth (examples).…”

Plant disease identification using explainable 3D deep learning on hyperspectral images

“…In contrast, hyperspectral images can be considered as a stack of 2D images, exhibiting correlations both in space as well as in the spectral directions. To extend DCNN’s applicability to hyperspectral images, a 3D analogue of the convolutional filter was proposed and such 3D-CNN models have been used in classification of hyperspectral images for some interesting engineering applications [33–35]. This is a promising approach to use for hyperspectral image based classification of plant diseases.…”

A CNN-Based Fusion Method for Feature Extraction from Sentinel Data

“…In the last few years, CNNs have been successfully applied to many classical image processing problems, such as denoising [50], super-resolution [51], pansharpening [8,24], segmentation [52], object detection [53,54], change detection [27] and classification [17,[55][56][57]. The main strengths of CNNs are (i) an extreme versatility that allows them to approximate any sort of linear or non-linear transformation, including scaling or hard thresholding; (ii) no need to design handcrafted filters, replaced by machine learning; (iii) high-speed processing, thanks to parallel computing.…”