In order to build inversion model of dust-fall weight by hyperspectral data, 30 samples were collected in Beijing. Through electronic balance and Analytical Spectral Devices FieldSpec Pro (ASD) analysis, the "dust leaves" and the "clean leaves" weight and spectral reflectance were determined respectively, which also obtained information of dust weight and spectral features. Then, based on tradition and partial least squares (PLS) model's analysis, the relationship between dust weight and spectral reflectance was explored. The results showed that 350-700, 780-1 300 and 1 900-2 500 nm bands had apparently variations when they response to the different dust weights. In general, there was a negative relationship between dust weight and spectral reflectance, the maximum negative value -0.8 occurred at 737 band which belonged to near-infrared bands. In the analysis of dust weight with multi-band, it was indicated that NDVI index which was formed by 948 and 945 bands had a significant correlation (r = 0.76) to dust. Finally, through accuracy assessment of regression model, the PLS could obtain a more accurate result than the traditional model.
Download full-text PDF |
Source |
|---|
Publication Analysis
Top Keywords
Similar Publications
Superpixel guided spectral-spatial feature extraction and weighted feature fusion for hyperspectral image classification with limited training samples.
Sci Rep
January 2025
School of Information Engineering, Tianjin University of Commerce, Tianjin, China.
Deep learning is a double-edged sword. The powerful feature learning ability of deep models can effectively improve classification accuracy. Still, when the training samples for each class are limited, it will not only face the problem of overfitting but also significantly affect the classification result.
View Article and Find Full Text PDFEnhancing spectral imaging with multi-condition image fusion.
Sci Rep
January 2025
Centre for Applied Photonics, INESC TEC, Rua do Campo Alegre 687, 4169-007, Porto, Portugal.
Spectral Imaging techniques such as Laser-induced Breakdown Spectroscopy (LIBS) and Raman Spectroscopy (RS) enable the localized acquisition of spectral data, providing insights into the presence, quantity, and spatial distribution of chemical elements or molecules within a sample. This significantly expands the accessible information compared to conventional imaging approaches such as machine vision. However, despite its potential, spectral imaging also faces specific challenges depending on the limitations of the spectroscopy technique used, such as signal saturation, matrix interferences, fluorescence, or background emission.
View Article and Find Full Text PDFMapping nutrient pollution in inland water bodies using multi-platform hyperspectral imagery and deep regression network.
J Hazard Mater
January 2025
Shanghai Municipal Institute of Surveying and Mapping, Shanghai, 200063, China.
Inland waters face multiple threats from human activities and natural factors, leading to frequent water quality issues, particularly the significant challenge of eutrophication. Hyperspectral remote sensing provides rich spectral information, enabling timely and accurate assessment of water quality status and trends. To address the challenge of inaccurate water quality mapping, we propose a novel deep learning framework for multi-parameter estimation from hyperspectral imagery.
View Article and Find Full Text PDFAssessment of antimicrobial efficacy of leather coating using chitosan modified TiO-CuO nanocomposites.
RSC Adv
January 2025
Nanoscience Research Laboratory, Department of Chemistry, Shivaji University Kolhapur 416 004 Maharashtra India
This research investigates the microbial inactivation potential of ternary TiO-CuO-chitosan nanocomposites (TCC NCs) applied as surface coatings on cowhide leather. Initially, bare TiO nanoparticles (NPs) and binary TiO-CuO (TC) NCs, with varying CuO NPs content, were prepared using an sol-gel method. These binary TC NCs were then modified with chitosan at varying weight percentages (2%, 4%, 6%, and 8%).
View Article and Find Full Text PDFHybrid generative adversarial network based on frequency and spatial domain for histopathological image synthesis.
BMC Bioinformatics
January 2025
Centre for Big Data Research in Health, University of New South Wales, Sydney, Australia.
Background: Due to the complexity and cost of preparing histopathological slides, deep learning-based methods have been developed to generate high-quality histological images. However, existing approaches primarily focus on spatial domain information, neglecting the periodic information in the frequency domain and the complementary relationship between the two domains. In this paper, we proposed a generative adversarial network that employs a cross-attention mechanism to extract and fuse features across spatial and frequency domains.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!