Hyperspectral Image Classification Using Deep Genome Graph-Based Approach

Abstract

Recently developed hybrid models that stack 3D with 2D CNN in their structure have enjoyed high popularity due to their appealing performance in hyperspectral image classification tasks. On the other hand, biological genome graphs have demonstrated their effectiveness in enhancing the scalability and accuracy of genomic analysis. We propose an innovative deep genome graph-based network (GGBN) for hyperspectral image classification to tap the potential of hybrid models and genome graphs. The GGBN model utilizes 3D-CNN at the bottom layers and 2D-CNNs at the top layers to process spectral–spatial features vital to enhancing the scalability and accuracy of hyperspectral image classification. To verify the effectiveness of the GGBN model, we conducted classification experiments on Indian Pines (IP), University of Pavia (UP), and Salinas Scene (SA) datasets. Using only 5% of the labeled data for training over the SA, IP, and UP datasets, the classification accuracy of GGBN is 99.97%, 96.85%, and 99.74%, respectively, which is better than the compared state-of-the-art methods.

Mutual Information-Driven Feature Reduction for Hyperspectral Image Classification

“…A hyperspectral image (HSI), which is acquired at a contiguous spectral wavelength of the electromagnetic spectrum (EM), is a rich data source for a wide range of real-world remote sensing applications, including agriculture, geology, mining, military surveillance, and others [ 1 , 2 ]. Moreover, an HSI is set up as a hypercube and often has hundreds of contiguous, narrow bands in the spectral image [ 3 , 4 ]. Due to the fact that each of these image bands contains varying intensities for the ground cover, they are each referred to as individual features [ 5 , 6 , 7 ].…”

Improving Feature Learning in Remote Sensing Images Using an Integrated Deep Multi-Scale 3D/2D Convolutional Network

“…Remote sensing involves the use of sophisticated camera sensors to remotely (from satellite or aircraft) detect and monitor the physical characteristics of a given portion of Earth's surface area using the reflected and emitted radiation [1]. The rapid technological innovation in remote sensing has resulted in the development of complex hyperspectral image (HSI) sensors that capture both voluminous (hundreds) spectral bands and highresolution spatial information of the earth's surface to produce a three-dimensional (3D) HSI data cube [2,3], as shown in Figure 1.…”

“…Wu et al [21] designed the 3D ResNeXt structure using feature fusion and label-smoothing strategies [21]. Tinega et al [1], developed a GGBN model that used the biological genome concept to combine 3D- The structural layout of the pre-processing-based methods is depicted in Figure 2. These methods separately acquire the raw HSI data's low-level spectral and spatial features.…”

“…Wu et al [21] designed the 3D ResNeXt structure using feature fusion and label-smoothing strategies [21]. Tinega et al [1], developed a GGBN model that used the biological genome concept to combine 3D-CNN and 2D-CNN and residual connections in a network structure [1]. Among other researchers who utilized a mixture of 3D-CNN and 2D-CNN in their network structure, and residual connections to produce the state-of-the-art models include zhao et al [22], and Tinega et al [23].…”

“…The remainder of this work is structured as follows: Section 2 explains the research technique; Section 3 examines the experimental setup; Section 4 discusses the experimental results and discussion; and Section 5 concludes this research. CNN and 2D-CNN and residual connections in a network structure [1]. Among other researchers who utilized a mixture of 3D-CNN and 2D-CNN in their network structure, and residual connections to produce the state-of-the-art models include zhao et al [22], and Tinega et al [23].…”