Classification of Heterogeneous Mining Areas Based on ResCapsNet and Gaofen-5 Imagery

Abstract

Land cover classification (LCC) of heterogeneous mining areas is important for understanding the influence of mining activities on regional geo-environments. Hyperspectral remote sensing images (HSI) provide spectral information and influence LCC. Convolutional neural networks (CNNs) improve the performance of hyperspectral image classification with their powerful feature learning ability. However, if pixel-wise spectra are used as inputs to CNNs, they are ineffective in solving spatial relationships. To address the issue of insufficient spatial information in CNNs, capsule networks adopt a vector to represent position transformation information. Herein, we combine a clustering-based band selection method and residual and capsule networks to create a deep model named ResCapsNet. We tested the robustness of ResCapsNet using Gaofen-5 Imagery. The images covered two heterogeneous study areas in Wuhan City and Xinjiang Province, with spatially weakly dependent and spatially basically independent datasets, respectively. Compared with other methods, the model achieved the best performances, with averaged overall accuracies of 98.45 and 82.80% for Wuhan study area, and 92.82 and 70.88% for Xinjiang study area. Four transfer learning methods were investigated for cross-training and prediction of those two areas and achieved good results. In summary, the proposed model can effectively improve the classification accuracy of HSI in heterogeneous environments.

Spectral Camouflage Characteristics and Recognition Ability of Targets Based on Visible/Near-Infrared Hyperspectral Images

“…Through the analysis and thinking of the experimental research and conclusions, this paper proposes the following prospects: (1) In terms of camouflage materials, the current camouflage materials cannot meet the requirements of visible light, near-infrared and other multi band compatible stealth, so the research and development of new spectral camouflage materials is an urgent problem to be solved. (2) In terms of imaging level, the development of near-infrared hyperspectral imaging technology is not yet mature, and the key lies in the limitation of near-infrared detectors. Therefore, it is necessary to develop near infrared detectors with wide band and high resolution.…”

“…Imaging spectrum technology, also known as hyperspectral imaging technology, is a new technology that integrates imaging technology and spectral analysis technology. It is widely used in agriculture [1], mineral detection [2], environmental detection [3] and other fields. Hyperspectral images can break the restriction of two-dimensional image space, expand object information to the spectral dimension, and effectively improve the accuracy and precision of target classification and detection.…”

Preface: Celebrating the 100th anniversary of chemical sciences in Nanjing University

“…Multi-scale features possess the capability to capture object representations at various scales in images, showcasing excellent performance across multiple tasks [21][22][23][24][25][26][27]. NAS-FPN [28] employs reinforcement learning to train a controller that identifies the optimal model architectures within a predefined search space.…”

SSMA-YOLO: A Lightweight YOLO Model with Enhanced Feature Extraction and Fusion Capabilities for Drone-Aerial Ship Image Detection

“…Manually extracted features [15] often fail to capture the complexity of data, resulting in the model lacking robustness in identifying vessels in complex and variable maritime environments. Consequently, deep learning [16,17] methods capable of adaptively learning [18,19] these features have gradually become a focal point in ship recognition research.…”