Multispectral Image Quality Improvement Based on Global Iterative Fusion Constrained by Meteorological Factors

Shi, Yuetian; Fu, Bin; Wang, Nan; Chen, Yaxiong; Fang, Jie

doi:10.1007/s12559-023-10207-7

Cited by 2 publications

(1 citation statement)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When acquiring images of a target area using a UAV-borne hyperspectral imager, the flight altitude is generally located under clouds, and it takes time cycles to capture multiangle information. The illumination changes over time during the cycle, which results in the acquired reflectance information being captured with inconsistent illumination [14]. Directly using such data with illumination bias for BRDF modeling, both the model parameters and reflectance inversion will produce large deviations, and the changes in illumination are generally random, making the random effects on BRDF modeling difficult to eliminate.…”

Section: Introductionmentioning

confidence: 99%

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Wang,

Li,

Wang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

The Bidirectional Reflectance Distribution Function (BRDF) is a critical spatial distribution parameter in the quantitative research of remote sensing and has a wide range of applications in radiometric correction, elemental inversion, and surface feature estimation. As a new means of BRDF modeling, UAV push-broom hyperspectral imaging is limited by the push-broom imaging method, and the multi-angle information is often difficult to obtain. In addition, the random variation of solar illumination during UAV low-altitude flight makes the irradiance between different push-broom hyperspectral rows and different airstrips inconsistent, which significantly affects the radiometric consistency of BRDF modeling and results in the difficulty of accurately portraying the three-dimensional spatial reflectance distribution in the UAV model. These problems largely impede the application of outdoor BRDF. Based on this, this paper proposes a fast multi-angle information acquisition scheme with a high-accuracy BRDF modeling method considering illumination variations, which mainly involves a lightweight system for BRDF acquisition and three improved BRDF models considering illumination corrections. We adopt multi-rectangular nested flight paths for multi-gray level targets, use multi-mode equipment to acquire spatial illumination changes and multi-angle reflectivity information in real-time, and introduce the illumination correction factor K through data coupling to improve the kernel, Hapke, and RPV models, and, overall, the accuracy of the improved model is increased by 20.83%, 11.11%, and 31.48%, respectively. The results show that our proposed method can acquire multi-angle information quickly and accurately using push-broom hyperspectral imaging, and the improved model eliminates the negative effect of illumination on BRDF modeling. This work is vital for expanding the multi-angle information acquisition pathway and high-efficiency and high-precision outdoor BRDF modeling.

show abstract

Section: Introductionmentioning

confidence: 99%

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Wang,

Li,

Wang

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

Constrained Spectral–Spatial Attention Residual Network and New Cross-Scene Dataset for Hyperspectral Classification

Li,

Chen,

Wang

et al. 2024

Electronics

View full text Add to dashboard Cite

Hyperspectral image classification is widely applied in several fields. Since existing datasets focus on a single scene, current deep learning-based methods typically divide patches randomly on the same image as training and testing samples. This can result in similar spatial distributions of samples, which may incline the network to learn specific spatial distributions in pursuit of falsely high accuracy. In addition, the large variation between single-scene datasets has led to research in cross-scene hyperspectral classification, focusing on domain adaptation and domain generalization while neglecting the exploration of the generalizability of models to specific variables. This paper proposes two approaches to address these issues. The first approach is to train the model on the original image and then test it on the rotated dataset to simulate cross-scene evaluation. The second approach is constructing a new cross-scene dataset for spatial distribution variations, named GF14-C17&C16, to avoid the problems arising from the existing single-scene datasets. The image conditions in this dataset are basically the same, and only the land cover distribution is different. In response to the spatial distribution variations, this paper proposes a constrained spectral attention mechanism and a constrained spatial attention mechanism to limit the fitting of the model to specific feature distributions. Based on these, this paper also constructs a constrained spectral–spatial attention residual network (CSSARN). Extensive experimental results on two public hyperspectral datasets and the GF14-C17&C16 dataset have demonstrated that CSSARN is more effective than other methods in extracting cross-scene spectral and spatial features.

show abstract

Multispectral Image Quality Improvement Based on Global Iterative Fusion Constrained by Meteorological Factors

Cited by 2 publications

References 55 publications

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Methodology and Modeling of UAV Push-Broom Hyperspectral BRDF Observation Considering Illumination Correction

Constrained Spectral–Spatial Attention Residual Network and New Cross-Scene Dataset for Hyperspectral Classification

Contact Info

Product

Resources

About