Article
Environmental Sciences
Chuanlong Ye, Shanwei Liu, Mingming Xu, Bo Du, Jianhua Wan, Hui Sheng
Summary: A multiscale resampling endmember bundle extraction (MSREBE) method is proposed to address spectral variability in hyperspectral unmixing. The experiments show that the endmembers extracted by this method are superior to those extracted by compared methods, maintaining optimal results in abundance estimation.
Article
Environmental Sciences
Jie Yu, Bin Wang, Yi Lin, Fengting Li, Jianqing Cai
Summary: Spectral unmixing has attracted attention in data processing. The novel inequality-constrained weighted linear mixture model (IWLMM) proposed in this study outperforms other linear and nonlinear unmixing models by addressing endmember variability and improving decomposing and reconstructing abilities.
REMOTE SENSING OF ENVIRONMENT
(2021)
Article
Environmental Sciences
Ying Cheng, Liaoying Zhao, Shuhan Chen, Xiaorun Li
Summary: Spectral unmixing is an important topic in hyperspectral image analysis as it deals with the presence of multiple sources in images and the variability in spectral signatures caused by environmental conditions. Various spectral mixing models have been proposed, but their interpretation is often insufficient and the corresponding unmixing algorithms are classical techniques. This paper introduces a spectral unmixing network based on a scaled and perturbed linear mixing model, incorporating deep learning techniques for determining abundances, scales, and perturbations. The proposed approach outperforms other competitors in both synthetic and real data sets.
Article
Environmental Sciences
Linke Ouyang, Caiyan Wu, Junxiang Li, Yuhan Liu, Meng Wang, Ji Han, Conghe Song, Qian Yu, Dagmar Haase
Summary: Impervious surface area (ISA) is a crucial indicator of urbanization. Spectral mixture analysis (SMA), commonly used to estimate ISA from remotely sensed data, faces challenges due to endmember spectral variability and plant phenology. This study developed a novel approach, PF-LSMA, which incorporates phenology with Fisher transformation, and demonstrated its effectiveness in accurately extracting ISA.
Article
Geochemistry & Geophysics
Ricardo Augusto Borsoi, Tales Imbiriba, Jose Carlos Moreira Bermudez, Cedric Richard
Summary: This study proposes a library augmentation strategy to improve the diversity of existing spectral libraries by utilizing deep generative models to learn the statistical distribution of endmembers, resulting in an enhanced quality of the spectral unmixing process.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Remote Sensing
Zhenfeng Shao, Yuan Zhang, Cheng Zhang, Xiao Huang, Tao Cheng
Summary: Impervious surface mapping is crucial for urban environmental studies. This study proposes a method that integrates a hierarchical strategy and spatially varied endmember spectra to map impervious surface abundance. The research finds that this method outperforms other methods in terms of estimation accuracy and shows significant performance improvement in less developed areas. Furthermore, spatially varied endmember spectra facilitate the reduction of heterogeneous surface material variations.
GEO-SPATIAL INFORMATION SCIENCE
(2022)
Article
Environmental Sciences
Jianbin Tao, Xinyue Zhang, Yiqing Liu, Qiyue Jiang, Yang Zhou
Summary: Agricultural cropping intensity plays a crucial role in evaluating food security and sustainable agriculture. Existing indicators lack precision and new methods are needed. This study introduces a new temporal mixture analysis method using time series remote sensing data to accurately map cropping intensity.
Article
Agriculture, Multidisciplinary
Nestor E. Caicedo Solano, Guisselle A. Garcia Llinas, Jairo R. Montoya-Torres
Summary: This paper presents a Mixed Integer Nonlinear Programming Problem model to address the planning challenges in agricultural production systems, aiming to reduce production costs and waste. The model is validated using data from banana farmers in the Caribbean region of Colombia, offering a new perspective for operational planning in agriculture production.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Environmental Sciences
Linghong Meng, Danfeng Liu, Liguo Wang, Jon Atli Benediktsson, Xiaohan Yue, Yuetao Pan
Summary: Spectral unmixing (SU) is an important preprocessing task for handling hyperspectral images (HSI), but its process is affected by nonlinearity and spectral variability (SV). Currently, SV is considered within the framework of linear mixing models (LMM), which ignores the nonlinear effects in the scene. To address that issue, we investigate the effects of SV on SU and propose an augmented generalized bilinear model to address spectral variability (AGBM-SV).
Article
Remote Sensing
Ariolfo Camacho, Edwin Vargas, Henry Arguello
Summary: This paper proposes a hyperspectral-multispectral image fusion algorithm based on spectral unmixing, which improves the accuracy and effectiveness of fusion methods by considering spectral variability. The proposed method is validated on realistic data and crop images, and outperforms state-of-the-art fusion methods in different metrics.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Ge Zhang, Shaohui Mei, Yan Feng, Qian Du
Summary: This study introduces a spectral-spatial constrained unmixing method based on nonnegative matrix factorization (NMF), which improves the performance of unmixing by imposing spatial and spectral constraints. Experimental results on synthetic and real-world datasets demonstrate the effectiveness of the proposed method in comparison to state-of-the-art NMF-based unmixing algorithms.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Shuaikai Shi, Lijun Zhang, Yoann Altmann, Jie Chen
Summary: In this article, a variational autoencoder-based deep generative model is proposed for spatial-spectral unmixing in complex hyperspectral images. The model can handle both spatially correlated materials distributions and spectral variability by linking generated endmembers to the probability distributions of endmember bundles extracted from the images. Experimental results demonstrate that the proposed model generates more accurate endmembers compared to other state-of-the-art methods.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Longfei Ren, Danfeng Hong, Lianru Gao, Xu Sun, Min Huang, Jocelyn Chanussot
Summary: Hyperspectral unmixing is important for estimating pure spectral signatures in each pixel, but spectral variability can cause significant errors. To address this, a new method called orthogonal subspace unmixing (OSU) was developed, utilizing orthogonal subspace projection. By jointly performing subspace learning and unmixing, the proposed OSU method removes complex spectral variability. Experiments showed the effectiveness and superiority of this framework in mitigating spectral variability compared to classical linear unmixing methods or variability accounting approaches.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Remote Sensing
Yingying Xu
Summary: In this paper, a blind hyperspectral unmixing model driven by local similarity is proposed to improve the unmixing performance by constructing a coefficient matrix and utilizing pixel-wise weight of spatial constraints. Experimental results show that the proposed method is less time consuming and outperforms or is comparable to the state-of-the-art methods.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2021)
Article
Chemistry, Multidisciplinary
Baohua Jin, Yunfei Zhu, Wei Huang, Qiqiang Chen, Sijia Li
Summary: The purpose of hyperspectral unmixing is to obtain the spectral features and proportions of materials in a hyperspectral image. However, spectral variabilities make it difficult to accurately extract these features. To address this issue, this study proposes an efficient attention-based convolutional neural network and a convolution block attention module. Experimental results demonstrate that this method outperforms other unmixing methods.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
Dengkai Chi, Koenraad Van Meerbeek, Kang Yu, Jeroen Degerickx, Ben Somers
Summary: The study investigated the physiological and phenological leaf plasticity of Tiliaxeuchlora trees in response to the urban heat island effect and soil sealing. Results showed that leaf water content and specific leaf area were higher in cooler zones, while chlorophyll and carotenoid contents were lower. Spatiotemporal variations in leaf functional traits were successfully tracked using spectral indices calculated from leaf reflectance measurements, demonstrating the adaptability of Tiliaxeuchlora trees to changes in the urban environment and the potential for using leaf optical traits as a proxy for studying urban tree responses to environmental factors.
SCIENCE OF THE TOTAL ENVIRONMENT
(2022)
Article
Agronomy
Willem W. Verstraeten, Rostislav Kouznetsov, Lucie Hoebeke, Nicolas Bruffaerts, Mikhail Sofiev, Andy W. Delcloo
Summary: Airborne birch pollen can cause allergies and have a negative impact on public health. By using a modeling approach, researchers have reconstructed and predicted the spatial distribution of birch pollen levels in Belgium from 1982 to 2019. The study shows that the amount of birch pollen in the air has been increasing, with climate change and variations in pollen emission sources playing a significant role.
AGRICULTURAL AND FOREST METEOROLOGY
(2022)
Article
Ecology
Camille Van Eupen, Dirk Maes, Marc Herremans, Kristijn R. R. Swinnen, Ben Somers, Stijn Luca
Summary: This study collected traits of birds, butterflies, and dragonflies and linked them to the effects of data quality filtering on species distribution model performance. The results showed that both taxonomic group and relative species traits play a role in defining the impact of data quality filtering on model performance.
ECOLOGICAL MODELLING
(2022)
Article
Environmental Sciences
Dengkai Chi, Raf Aerts, An Van Nieuwenhuyse, Mariska Bauwelinck, Claire Demoury, Michelle Plusquin, Tim S. Nawrot, Lidia Casas, Ben Somers
Summary: This study quantified the number of trees and relevant 3D structural traits in Brussels and investigated their associations with sales of medication commonly prescribed for mood disorders and cardiovascular disease. The results showed that an increase in crown volume was associated with lower medication sales, while an increase in stem density was associated with higher medication sales. This suggests that conserving large trees in urban environments can have positive effects on human health.
ENVIRONMENTAL HEALTH PERSPECTIVES
(2022)
Article
Environmental Sciences
Jingli Yan, Stijn van der Linden, Yunyu Tian, Jo Van Valckenborgh, Veerle Strosse, Ben Somers
Summary: Domestic gardens provide residents with immediate access to landscape amenities and ecological provisions. This study developed a method to map garden parcels in a medieval city using land use layers and satellite imagery. The results showed that multi-sourced satellite imagery improved the accuracy of identifying vegetation types, and the phenology and 3D structure of plants were significant factors in garden mapping. The study also revealed variations in greenspace landscapes between urban and rural gardens. More attention should be given to gardens and interdisciplinary studies conducted to maximize the benefits to residents in the face of environmental changes and health risks.
Article
Biochemistry & Molecular Biology
Yinthe Dockx, Esmee Bijnens, Nelly Saenen, Raf Aerts, Jean-Marie Aerts, Lidia Casas, Andy Delcloo, Nicolas Dendoncker, Catherine Linard, Michelle Plusquin, Michiel Stas, an Van Nieuwenhuyse, Jos Van Orshoven, Ben Somers, Tim Nawrot
Summary: Numerous studies have shown that green space has a significant impact on adult cognition and childhood neurodevelopment, and these effects may be partly driven by epigenetic modifications. This study investigated the influence of green space on epigenetic processes in placental tissue during foetal development. The results indicate that increased maternal green space exposure is associated with increased methylation of HTR2A in placental tissue. These findings are important for understanding the relationship between epigenetic changes, placental functional modifications, and foetal development.
Article
Biodiversity Conservation
Johanna Van Passel, Wanda de Keersmaecker, Paulo N. Bernardino, Xin Jing, Nikolaus Umlauf, Koenraad Van Meerbeek, Ben Somers
Summary: Extreme precipitation and drought events are expected to increase in frequency and intensity in the Amazon rainforest. This study found that the forest has resistance and resilience to drought, but previous climatic events have lasting effects on the forest's response to drought.
GLOBAL CHANGE BIOLOGY
(2022)
Article
Ecology
Jeroen Krols, Raf Aerts, Naomi Vanlessen, Valerie Dewaelheyns, Sebastien Dujardin, Ben Somers
Summary: The study found associations between residential green space quality, green space related activities, socioeconomic background variables of gardeners, and self-reported health, with nature relatedness playing a key role in linking gardens to improved health.
LANDSCAPE AND URBAN PLANNING
(2022)
Article
Environmental Sciences
Tshilidzi Manyanya, Janne Teerlinck, Ben Somers, Bruno Verbist, Nthaduleni Nethengwe
Summary: The LCZ framework is widely used for studying urban climate. While the standard LCZ typology is specific to Western cities, it is applicable to African cities. Research shows that locally calibrated LCZ classification improves accuracy by 17-30% in an African context.
Article
Engineering, Civil
Ine Rosier, Jan Diels, Ben Somers, Jos Van Orshoven
Summary: Flooding in settlements and agricultural areas in Europe is a growing concern. The use of vegetated landscape elements (vLEs) such as hedges, lines of trees, and grass buffers along parcel boundaries is recognized as a way to mitigate flood risk, but scientific evidence supporting their implementation is lacking. Using the Landlab modelling framework, we studied the importance of vLE presence and characteristics in a 26 ha watershed in Belgium. Our model results showed that vLE density and upstream area control the total runoff volume, peak discharge rate, and lag time in small watersheds.
JOURNAL OF HYDROLOGY
(2023)
Article
Environmental Sciences
Willem W. Verstraeten, Nicolas Bruffaerts, Rostislav Kouznetsov, Letty de Weger, Mikhail Sofiev, Andy W. Delcloo
Summary: Changes in climate and land-use can lead to increased emission of allergenic pollen, causing a rise in respiratory allergies. A research study using the SILAM model found that the increase in birch pollen concentrations is associated with increasing radiation, decreasing precipitation, and decreasing wind speed, while the decrease in grass pollen concentrations is driven by a decreasing trend in grass pollen sources and decreasing precipitation.
ATMOSPHERIC ENVIRONMENT
(2023)
Article
Environmental Sciences
Stephanie Delalieux, Tom Hardy, Michel Ferry, Susi Gomez, Lammert Kooistra, Maria Culman, Laurent Tits
Summary: This study evaluates the potential of a remote sensing approach for the timely and reliable detection of Red Palm Weevil (RPW) infestation on palm canopies. The study found that thermal data can detect RPW infestation before visible canopy symptoms appear, and spectral vegetation index monitoring can detect infested palms before canopy symptoms are visible.
Article
Environmental Sciences
Nick Gutkin, Valens Uwizeyimana, Ben Somers, Bart Muys, Bruno Verbist
Summary: Eastern Rwanda has diverse land cover types including agroforestry, forest patches, and shrubland, all of which have tree cover. The use of automated methods and satellite imagery, such as Google Earth Engine and the random forests algorithm, allows for cost-effective and time-efficient mapping and monitoring of the landscape. This study combined Sentinel-2 satellite imagery with various vegetation indices, texture metrics, principal components, and non-spectral layers to classify land cover types in the study area. The results showed high classification accuracies for forest, shrubland, and agroforestry, with non-spectral layers and texture metrics being important for accurate classification.
Article
Allergy
Joren Buekers, Michiel Stas, Raf Aerts, Nicolas Bruffaerts, Sebastien Dujardin, An Van Nieuwenhuyse, Jos Van Orshoven, Guillaume Chevance, Ben Somers, Jean-Marie Aerts, Judith Garcia-Aymerich
Summary: This study aimed to estimate the associations between daily allergy burden and heart rate characteristics of adults with allergic rhinitis. The results showed that an increase in allergy symptom score was associated with an increase in next-day resting heart rate, and an increase in mood score was associated with an increase in same-day sample entropy. These findings suggest that daily allergy burden has systemic effects beyond the respiratory system.
CLINICAL AND TRANSLATIONAL ALLERGY
(2023)
Article
Agriculture, Multidisciplinary
Maria Culman, Stephanie Delalieux, Bart Beusen, Ben Somers
Summary: This study proposes an automatic labeling method to accelerate the adaptation of on-tree fruit counting solutions based on deep learning. By minimizing human intervention in data labeling and using automatically generated labels, accurate fruit quantity calculation can be achieved in pear orchards.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)