From Local to Global: Class Feature Fused Fully Convolutional Network for Hyperspectral Image Classification

Current mainstream networks for hyperspectral image (HSI) classification employ image patches as inputs for feature extraction. Spatial information extraction is limited by the size of inputs, which makes networks unable to perform effective learning and reasoning from the global perspective. As a common component for capturing long-range dependencies, non-local networks with pixel-by-pixel information interaction bring unaffordable computational costs and information redundancy. To address the above issues, we propose a class feature fused fully convolutional network (CFF-FCN) with a local feature extraction block (LFEB) and a class feature fusion block (CFFB) to jointly utilize local and global information. LFEB based on dilated convolutions and reverse loop mechanism can acquire the local spectral–spatial features at multiple levels and deliver shallower layer features for coarse classification. CFFB calculates global class representation to enhance pixel features. Robust global information is propagated to every pixel with low computational cost. CFF-FCN considers a fully global class context and obtains more discriminative representation by concatenating high-level local features and re-integrated global features. Experimental results conducted on three real HSI data sets demonstrate that the proposed fully convolutional network is superior to multiple state-of-the-art deep learning-based approaches, especially in the case of a small number of training samples.

MDPrePost-Net: A Spatial-Spectral-Temporal Fully Convolutional Network for Mapping of Mangrove Degradation Affected by Hurricane Irma 2017 Using Sentinel-2 Data

Mangroves are grown in intertidal zones along tropical and subtropical climate areas, which have many benefits for humans and ecosystems. The knowledge of mangrove conditions is essential to know the statuses of mangroves. Recently, satellite imagery has been widely used to generate mangrove and degradation mapping. Sentinel-2 is a volume of free satellite image data that has a temporal resolution of 5 days. When Hurricane Irma hit the southwest Florida coastal zone in 2017, it caused mangrove degradation. The relationship of satellite images between pre and post-hurricane events can provide a deeper understanding of the degraded mangrove areas that were affected by Hurricane Irma. This study proposed an MDPrePost-Net that considers images before and after hurricanes to classify non-mangrove, intact/healthy mangroves, and degraded mangroves classes affected by Hurricane Irma in southwest Florida using Sentinel-2 data. MDPrePost-Net is an end-to-end fully convolutional network (FCN) that consists of two main sub-models. The first sub-model is a pre-post deep feature extractor used to extract the spatial–spectral–temporal relationship between the pre, post, and mangrove conditions after the hurricane from the satellite images and the second sub-model is an FCN classifier as the classification part from extracted spatial–spectral–temporal deep features. Experimental results show that the accuracy and Intersection over Union (IoU) score by the proposed MDPrePost-Net for degraded mangrove are 98.25% and 96.82%, respectively. Based on the experimental results, MDPrePost-Net outperforms the state-of-the-art FCN models (e.g., U-Net, LinkNet, FPN, and FC-DenseNet) in terms of accuracy metrics. In addition, this study found that 26.64% (41,008.66 Ha) of the mangrove area was degraded due to Hurricane Irma along the southwest Florida coastal zone and the other 73.36% (112,924.70 Ha) mangrove area remained intact.

Lithological Mapping Based on Fully Convolutional Network and Multi-Source Geological Data

Deep learning algorithms have found numerous applications in the field of geological mapping to assist in mineral exploration and benefit from capabilities such as high-dimensional feature learning and processing through multi-layer networks. However, there are two challenges associated with identifying geological features using deep learning methods. On the one hand, a single type of data resource cannot diagnose the characteristics of all geological units; on the other hand, deep learning models are commonly designed to output a certain class for the whole input rather than segmenting it into several parts, which is necessary for geological mapping tasks. To address such concerns, a framework that comprises a multi-source data fusion technology and a fully convolutional network (FCN) model is proposed in this study, aiming to improve the classification accuracy for geological mapping. Furthermore, multi-source data fusion technology is first applied to integrate geochemical, geophysical, and remote sensing data for comprehensive analysis. A semantic segmentation-based FCN model is then constructed to determine the lithological units per pixel by exploring the relationships among multi-source data. The FCN is trained end-to-end and performs dense pixel-wise prediction with an arbitrary input size, which is ideal for targeting geological features such as lithological units. The framework is finally proven by a comparative study in discriminating seven lithological units in the Cuonadong dome, Tibet, China. A total classification accuracy of 0.96 and a high mean intersection over union value of 0.9 were achieved, indicating that the proposed model would be an innovative alternative to traditional machine learning algorithms for geological feature mapping.