Article
Environmental Sciences
Zhenyang Hui, Zhuoxuan Li, Penggen Cheng, Yao Yevenyo Ziggah, JunLin Fan
Summary: This study proposed a building extraction method from airborne LiDAR data based on multi-constraints graph segmentation, which converted point-based building extraction into object-based building extraction, and introduced a multi-scale progressive growth optimization method, achieving superior results.
Article
Computer Science, Interdisciplinary Applications
Anett Fekete, Mate Cserep
Summary: With the increasing affordability and accessibility of LiDAR technology, the analysis of point clouds constructed by laser scanning is becoming more prominent. Airborne LiDAR is particularly valuable for land object analysis and classification, while our research aims to define an automated methodology for vegetation segmentation and change detection in urban areas. The algorithm proposed in our study proved to be effective for qualified and quantified change detection of trees, including height and volume changes, providing a robust approach that can scale dynamically to large areas.
COMPUTERS & GEOSCIENCES
(2021)
Article
Robotics
Yeong-Hyeon Kim, Ukcheol Shin, Jinsun Park, In So Kweon
Summary: This study introduces a multi-spectral unsupervised domain adaptation method for thermal image semantic segmentation, aiming to enhance segmentation performance by utilizing RGB image data and segmentation knowledge, addressing data scarcity and achieving high performance through domain adaptation.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2021)
Article
Remote Sensing
Xiaona Wang, Le Wang, Jinyan Tian, Chen Shi
Summary: This study aimed to improve upon the limitations of the Ppf-CM method by incorporating geometric, texture, and contextual information. A new object-based phenological feature composite method (OPpf-CM) was developed, which involved stacking two images acquired during distinctive phenological periods of Spartina alterniflora, deriving spectrally homogeneous objects, and using various features from these objects as input for a support vector machine classifier to generate a map of S. alterniflora with higher accuracy than the pixel-based method.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Environmental Sciences
Haijun Liu, Fenglei Chen, Zhihong Zeng, Xiaoheng Tan
Summary: This paper proposes a simple yet effective add-multiply fusion (AMFuse) module for fusing RGB and thermal information. The module extracts cross-modal complementary features and common features through addition and multiplication operations, respectively. Attention and ASPP modules are incorporated to enhance multi-scale context information. Experimental results demonstrate the effectiveness of the proposed AMFuse module in multi-spectral semantic segmentation and salient object detection.
Article
Multidisciplinary Sciences
Simon Madec, Kamran Irfan, Kaaviya Velumani, Frederic Baret, Etienne David, Gaetan Daubige, Lucas Bernigaud Samatan, Mario Serouart, Daniel Smith, Chrisbin James, Fernando Camacho, Wei Guo, Benoit De Solan, Scott C. Chapman, Marie Weiss
Summary: Applying deep learning to images of cropping systems offers new knowledge and insights in research and commercial applications. Semantic segmentation or pixel-wise classification of ground level RGB images into vegetation and background is crucial for estimating canopy traits. Current CNN-based methodologies trained on controlled or indoor datasets cannot generalize to real-world images, requiring fine-tuning with new labeled datasets. The creation of the VegAnn dataset, consisting of 3775 multicrop RGB images acquired under diverse illumination conditions, aims to improve segmentation algorithm performance and facilitate benchmarking in large-scale crop vegetation segmentation research.
Article
Geography, Physical
Guillaume Lassalle, Matheus Pinheiro Ferreira, Laura Elena Cue La Rosa, Rebecca Del'Papa Moreira Scafutto, Carlos Roberto de Souza Filho
Summary: This study presents the advances in mapping mangrove species using multispectral and hyperspectral imagery. A new framework based on convolutional neural network is proposed for accurate classification at pixel and object level. The study highlights the importance of spatial and spectral resolutions, especially the short-wave infrared bands.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Zahra Ghorbani, Amir H. Behzadan
Summary: The research utilizes AI technologies such as convolutional neural networks to monitor and classify oil spills, enhancing the accuracy of oil spill response and resource allocation. The findings have the potential to improve current practices of oil pollution cleanup and preventive maintenance, leading to healthier and more resilient coastal communities.
ENVIRONMENTAL POLLUTION
(2021)
Article
Biology
Manisha Saini, Seba Susan
Summary: Screening and diagnosis of diabetic retinopathy disease is a significant problem in the biomedical domain. The use of medical imagery from a patient's eye for computer-aided diagnosis has greatly advanced with the success of deep learning. However, challenges with imbalanced datasets, inconsistent annotations, limited samples, and inappropriate evaluation metrics have impacted the performance of deep learning models.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Forestry
Bruno Rodrigues de Oliveira, Arlindo Ananias Pereira da Silva, Larissa Pereira Ribeiro Teodoro, Gileno Brito de Azevedo, Glauce Tais de Oliveira Sousa Azevedo, Fabio Henrique Rojo Baio, Renato Lustosa Sobrinho, Carlos Antonio da Silva Junior, Paulo Eduardo Teodoro
Summary: The study evaluated the use of ML techniques to classify the growth of five species of eucalyptus and Corymbria citriodora, recognizing the species based on their growth using vegetation indices and spectral bands.
FOREST ECOLOGY AND MANAGEMENT
(2021)
Article
Agronomy
Alessia Cogato, Shaikh Yassir Yousouf Jewan, Lihua Wu, Francesco Marinello, Franco Meggio, Paolo Sivilotti, Marco Sozzi, Vinay Pagay
Summary: The study suggests that physiological and spectral responses of grapevines to water stress in hot environments may differ from expectations, providing insights for winegrowers in managing abiotic stress.
Article
Computer Science, Information Systems
Hao Li, Ziwen Sun, Chong Ling, Chao Xu
Summary: Tracking multiple states of the target simultaneously is a research hotspot in object tracking. Siamese contour segmentation network (SiamCS) is proposed to address this issue by formulating object tracking as region classification and contour regression. Experimental results on multiple datasets show that SiamCS outperforms other state-of-the-art trackers.
Article
Engineering, Multidisciplinary
Dheerendra Pratap Singh, Manohar Yadav
Summary: This paper proposes a method called 3D-MFDNN for vegetation segmentation based on ALS data. The method generates feature descriptors, trains the 3D-MFDNN model, and performs testing and performance comparison using ALS datasets. It achieves accurate segmentation of tree points in complex cases and outperforms several state-of-the-art methods, with F1-score and accuracy of 83.94% and 92.13%, respectively, on six datasets with different levels of scene complexity.
Article
Multidisciplinary Sciences
Jianguang Wen, Xiaodan Wu, Qing Xiao, Qinhuo Liu, Mingguo Ma, Xingming Zheng, Yonghua Qu, Rui Jin, DongQin You, Yong Tang, Xingwen Lin, Wenpin Yu, Baochang Gong, Jian Yang, Yuan Han
Summary: This study presents a unique library of field spectra, providing full-band, multi-angle, multi-scale spectral measurements of the main surface elements of China. The library covers a large spatial extent over a 10-year period, and serves as a vital link between ground measurements and satellite observations.
Article
Environmental Sciences
F. Javier Cardama, Dora B. Heras, Francisco Arguello
Summary: In this study, the change detection problem in very-high-spatial-resolution remote sensing images was investigated, focusing on the vegetation corresponding to crops and natural ecosystems. A consensus multi-scale binary change detection technique based on object-based features extraction was developed to address the challenge of similar spectral signatures of vegetation elements. Different detectors based on various segmentation algorithms were utilized at different scales to capture changes at different granularity levels. The proposed approach, including the use of CVA-SAM at the segment level, demonstrated effectiveness in identifying changes over land cover vegetation images with different types, spatial and spectral resolutions.
Article
Environmental Sciences
Ruonan Chen, Liangyun Liu, Xinjie Liu, Zhunqiao Liu, Lianhong Gu, Uwe Rascher
Summary: This study presents methods to accurately estimate sub-daily GPP from SIF in evergreen needleleaf forests and demonstrates that the interactions among light, canopy structure, and leaf physiology regulate the SIF-GPP relationship at the canopy scale.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Daniel L. Goldberg, Madankui Tao, Gaige Hunter Kerr, Siqi Ma, Daniel Q. Tong, Arlene M. Fiore, Angela F. Dickens, Zachariah E. Adelman, Susan C. Anenberg
Summary: A novel method is applied in this study to directly use satellite data to evaluate the spatial patterns of urban NOx emissions inventories. The results show that the 108 spatial surrogates used by NEMO are generally appropriate, but there may be underestimation in areas with dense intermodal facilities and overestimation in wealthy communities.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Zhuoyue Hu, Xiaoyan Li, Liyuan Li, Xiaofeng Su, Lin Yang, Yong Zhang, Xingjian Hu, Chun Lin, Yujun Tang, Jian Hao, Xiaojin Sun, Fansheng Chen
Summary: This paper proposes a whisk-broom imaging method using a long-linear-array detector and high-precision scanning mirror to achieve high-resolution and wide-swath thermal infrared data. The method has been implemented in the SDGs satellite and has shown promising test results.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Dandan Wang, Leiqiu Hu, James A. Voogt, Yunhao Chen, Ji Zhou, Gaijing Chang, Jinling Quan, Wenfeng Zhan, Zhizhong Kang
Summary: This study evaluates different schemes for determining model coefficients to quantify and correct the anisotropic impact from remote sensing LST for urban applications. The schemes have consistent results and accurately estimate parameter values, facilitating the broadening of parametric models.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Jamie Tolan, Hung - Yang, Benjamin Nosarzewski, Guillaume Couairon, Huy V. Vo, John Brandt, Justine Spore, Sayantan Majumdar, Daniel Haziza, Janaki Vamaraju, Theo Moutakanni, Piotr Bojanowski, Tracy Johns, Brian White, Tobias Tiecke, Camille Couprie
Summary: Vegetation structure mapping is crucial for understanding the global carbon cycle and monitoring nature-based approaches to climate adaptation and mitigation. This study presents the first high-resolution canopy height maps for California and Sao Paulo, achieved through the use of very high resolution satellite imagery and aerial lidar data. The maps provide valuable tools for forest structure assessment and land use monitoring.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Regina Eckert, Steffen Mauceri, David R. Thompson, Jay E. Fahlen, Philip G. Brodrick
Summary: In this paper, a mathematical framework is proposed to improve the retrieval of surface reflectance and atmospheric parameters by leveraging the expected spatial smoothness of the atmosphere. Experimental results show that this framework can reduce the surface reflectance retrieval error and surface-related biases.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Chongya Jiang, Kaiyu Guan, Yizhi Huang, Maxwell Jong
Summary: This study presents the Field Rover method, which uses vehicle-mounted cameras to collect ground truth data on crop harvesting status. The machine learning approach and remote sensing technology are employed to upscale the results to a regional scale. The accuracy of the remote sensing method in predicting crop harvesting dates is validated through comparison with satellite data.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Oksana V. Lunina, Anton A. Gladkov, Alexey V. Bochalgin
Summary: In this study, an unmanned aerial vehicle (UAV) was used to detect and map surface discontinuities with displacements of a few centimeters, indicating the presence of initial geological deformations. The study found that sediments of alluvial fans are susceptible to various tectonic and exogenous deformational processes, and the interpretation of ultra-high resolution UAV images can help recognize low-amplitude brittle deformations at an early stage. UAV surveys are critical for discerning neotectonic activity and its related hazards over short observation periods.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Feng Zhao, Weiwei Ma, Jun Zhao, Yiqing Guo, Mateen Tariq, Juan Li
Summary: This study presents a data-driven approach to reconstruct the terrestrial SIF spectrum using measurements from the TROPOMI instrument on Sentinel-5 precursor mission. The reconstructed SIF spectrum shows improved spatiotemporal distributions and demonstrates consistency with other datasets, indicating its potential for better understanding of the ecosystem function.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Stephen Stehman, John E. Wagner
Summary: This article investigates optimal sample allocation in stratified random sampling for estimation of accuracy and proportion of area in applications where the target class is rare. The study finds that precision of estimated accuracy has a stronger impact on sample allocation than estimation of proportion of area, and the trade-offs among these estimates become more pronounced as the target class becomes rarer. The results provide quantitative evidence to guide sample allocation decisions in specific applications.
REMOTE SENSING OF ENVIRONMENT
(2024)
Article
Environmental Sciences
Jingyao Zheng, Tianjie Zhao, Haishen Lu, Defu Zou, Nemesio Rodriguez-Fernandez, Arnaud Mialon, Philippe Richaume, Jianshe Xiao, Jun Ma, Lei Fan, Peilin Song, Yonghua Zhu, Rui Li, Panpan Yao, Qingqing Yang, Shaojie Du, Zhen Wang, Zhiqing Peng, Yuyang Xiong, Zanpin Xing, Lin Zhao, Yann Kerr, Jiancheng Shi
Summary: Soil moisture and freeze/thaw (F/T) play a crucial role in water and heat exchanges at the land-atmosphere interface. This study reports the establishment of a wireless sensor network for soil moisture and temperature over the permafrost region of Tibetan Plateau. Satellite-based surface soil moisture (SSM) and F/T products were evaluated using ground-based measurements. The results show the reliability of L-band passive microwave SSM and F/T products, while existing F/T products display earlier freezing and later thawing, leading to unsatisfactory accuracy.
REMOTE SENSING OF ENVIRONMENT
(2024)