Unveiling the potential of diffusion model-based framework with transformer for hyperspectral image classification.

Hyperspectral imaging has gained popularity for analysing remotely sensed images in various fields such as agriculture and medical. However, existing models face challenges in dealing with the complex relationships and characteristics of spectral-spatial data due to the multi-band nature and data redundancy of hyperspectral data. To address this limitation, we propose a novel approach called DiffSpectralNet, which combines diffusion and transformer techniques. The diffusion method is able extract diverse and meaningful spectral-spatial features, leading to improvement in HSI classification. Our approach involves training an unsupervised learning framework based on the diffusion model to extract high-level and low-level spectral-spatial features, followed by the extraction of intermediate hierarchical features from different timestamps for classification using a pre-trained denoising U-Net. Finally, we employ a supervised transformer-based classifier to perform the HSI classification. We conduct comprehensive experiments on three publicly available datasets to validate our approach. The results demonstrate that our framework significantly outperforms existing approaches, achieving state-of-the-art performance. The stability and reliability of our approach are demonstrated across various classes in all datasets.

Contributions to the biodiversity of Vietnam - Results of VIETBIO inventory work and field training in Cuc Phuong National Park.

VIETBIO [Innovative approaches to biodiversity discovery and characterisation in Vietnam] is a bilateral German-Vietnamese research and capacity building project focusing on the development and transfer of new methods and technology towards an integrated biodiversity discovery and monitoring system for Vietnam. Dedicated field training and testing of innovative methodologies were undertaken in Cuc Phuong National Park as part and with support of the project, which led to the new biodiversity data and records made available in this article collection. VIETBIO is a collaboration between the Museum für Naturkunde Berlin - Leibniz Institute for Evolution and Biodiversity Science (MfN), the Botanic Garden and Botanical Museum, Freie Universität Berlin (BGBM) and the Vietnam National Museum of Nature (VNMN), the Institute of Ecology and Biological Resources (IEBR), the Southern Institute of Ecology (SIE), as well as the Institute of Tropical Biology (ITB); all Vietnamese institutions belong to the Vietnam Academy of Science and Technology (VAST). The article collection "VIETBIO" (https://doi.org/10.3897/bdj.coll.63) reports original results of recent biodiversity recording and survey work undertaken in Cuc Phuong National Park, northern Vietnam, under the framework of the VIETBIO project. The collection consist of this "main" cover paper - characterising the study area, the general project approaches and activities, while also giving an extensive overview on previous studies from this area - followed by individual papers for higher taxa as studied during the project. The main purpose is to make primary biodiversity records openly available, including several new and interesting findings for this biodiversity-rich conservation area. All individual data papers with their respective primary records are expected to provide useful baselines for further taxonomic, phylogenetic, ecological and conservation-related studies on the respective taxa and, thus, will be maintained as separate datasets, including separate GUIDs also for further updating.