Article
Geochemistry & Geophysics
Shunzhou Wang, Tianfei Zhou, Yao Lu, Huijun Di
Summary: This paper proposes a lightweight super-resolution network that reduces the network burden by replacing convolution layers while maintaining performance. The authors introduce a lightweight convolution layer called the contextual transformation layer (CTL) for remote-sensing image super-resolution. Experimental results demonstrate the effectiveness of the proposed method in remote-sensing image super-resolution, natural image super-resolution, and denoising tasks.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Jing Zhang, Minhao Shao, Zekang Wan, Yunsong Li
Summary: This study proposes a Multi-Scale Feature Mapping Network (MSFMNet) to adaptively learn the prior information of hyperspectral images (HSIs). By simplifying the network structure and designing multi-scale feature generation and fusion modules, MSFMNet aims to solve the issues of existing HSI super-resolution methods.
Article
Environmental Sciences
Masoud Mahdianpari, Jean Elizabeth Granger, Fariba Mohammadimanesh, Sherry Warren, Thomas Puestow, Bahram Salehi, Brian Brisco
Summary: This study aims to produce the first high-resolution wetland map of the City of St. John's in Canada using advanced machine learning algorithms, very high-resolution satellite imagery, and airborne LiDAR technology. By applying an object-based random forest algorithm to features extracted from WorldView-4, GeoEye-1, and LiDAR data, the study characterizes five wetland classes within an urban area with an overall accuracy of 91.12% and produces wetland surface water flow connectivity using LiDAR data.
JOURNAL OF ENVIRONMENTAL MANAGEMENT
(2021)
Article
Ecology
Miao Liu, Jun Ma, Rui Zhou, Chunlin Li, Dikang Li, Yuanman Hu
Summary: This study established a model to predict the urban floor area in mainland China using remote sensing data, resulting in the generation of the FPFA map. The distribution of floor area was found to be uneven, with higher intensity in the southern urban areas along the coastline compared to the northern regions, and significant disparities among provinces.
LANDSCAPE AND URBAN PLANNING
(2021)
Article
Computer Science, Artificial Intelligence
Hongyuan Chen, Yanting Pei, Hongwei Zhao, Yaping Huang
Summary: This paper proposes a Super-Resolution guided Knowledge Distillation (SRKD) framework to address the challenge of low-resolution image classification. By enhancing the features of low-resolution images and minimizing the difference between the features of high-resolution images and super-resolution images, the proposed method achieves significant improvement in experimental results.
PATTERN RECOGNITION LETTERS
(2022)
Article
Computer Science, Artificial Intelligence
Jun Lyu, Guangyuan Li, Chengyan Wang, Qing Cai, Qi Dou, David Zhang, Jing Qin
Summary: This study develops a novel multicontrast MRI super resolution network, McMRSR(++), which utilizes the Transformer technique to capture long-range dependencies and introduces a multiscale feature matching and aggregation method for high-quality super resolution image reconstruction.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Geochemistry & Geophysics
Yang Zhang, Ruohan Zong, Lanyu Shang, Dong Wang
Summary: This article presents a novel deep convolutional neural network architecture, SCLearn, to address the classification and super-resolution coupling problem in remote urban sensing applications. The evaluation results demonstrate that SCLearn consistently outperforms the state-of-the-art baselines by achieving better classification accuracy and higher reconstructed image quality.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Xiaohong Liu, Kangdi Shi, Zhe Wang, Jun Chen
Summary: Current deep-learning-based Video Super-Resolution methods using videos generated by the camera ISP as inputs are suboptimal due to information loss and inconsistency. This study proposes a new VSR method that utilizes camera sensor data directly, with a carefully built Raw Video Dataset. Through Successive Deep Inference and reconstruction modules, the proposed method achieves superior VSR results compared to the state-of-the-art by leveraging the informativeness of camera raw data and separation of super-resolution and color correction processes.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Remote Sensing
Siqi Du, Hongsheng Huang, Fan He, Heng Luo, Yumeng Yin, Xiaoming Li, Linfu Xie, Renzhong Guo, Shengjun Tang
Summary: In this study, an unsupervised aquaculture ponds extraction method based on hyperspectral imagery super-resolution, feature fusion, and stepwise extraction strategy was proposed. The resolution of the original hyperspectral imagery was enhanced through deep learning-based super-resolution method, and feature fusion method was introduced to enhance feature sensitivity to aquaculture ponds. The aquaculture ponds were extracted via a stepwise extraction strategy. Experimental results showed that the proposed method achieved 97.9% overall accuracy in aquaculture pond extraction and can be generalized for object extraction in multispectral datasets.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2023)
Article
Multidisciplinary Sciences
Luciene Sales Dagher Arce, Lucas Prado Osco, Mauro dos Santos de Arruda, Danielle Elis Garcia Furuya, Ana Paula Marques Ramos, Camila Aoki, Arnildo Pott, Sarah Fatholahi, Jonathan Li, Fabio Fernando de Araujo, Wesley Nunes Goncalves, Jose Marcato Junior
Summary: The use of deep learning approach successfully detected and geolocated the Buriti palm tree, showing improved accuracy compared to other methods and presenting potential applications for mapping individual tree species in dense forest environments.
SCIENTIFIC REPORTS
(2021)
Article
Acoustics
Scott Schoen, Zhigen Zhao, Ashley Alva, Chengwu Huang, Shigao Chen, Costas Arvanitis
Summary: The generation of super-resolution ultrasound images through the localization of individual microbubbles has allowed for improved visualization of microvascular structure and flow. A method based on morphological image reconstruction has been proposed to increase peak detection and spatial resolution, with robustness to noise. This computationally efficient method shows promise for enhancing the capabilities of super-resolution ultrasound imaging.
IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL
(2021)
Article
Computer Science, Artificial Intelligence
Zhi-Song Liu, Wan-Chi Siu, Yui-Lam Chan
Summary: This paper introduces a new approach, Hierarchical CNN based Random Forests (HCRF), for face super-resolution by combining convolutional neural networks and random forests. The proposed method is capable of handling facial images under various conditions without preprocessing. By incorporating the advantages of deep learning with random forests, two novel CNN models are proposed for coarse facial image super-resolution and segmentation, along with new random forests for refining local facial features based on the segmentation results. Extensive benchmark experiments demonstrate that HCRF achieves comparable speed and competitive performance compared to state-of-the-art super-resolution approaches for very low-resolution images.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Cheng Ma, Yongming Rao, Jiwen Lu, Jie Zhou
Summary: Structures play a crucial role in single image super-resolution (SISR), and this paper proposes a structure-preserving super-resolution (SPSR) method to address the issue of structural distortions. By utilizing gradient guidance and a learnable neural structure extractor (NSE), our method achieves superior results in both detail recovery and structure preservation.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Mathematics
Ganbayar Batchuluun, Se Hyun Nam, Chanhum Park, Kang Ryoung Park
Summary: This study proposes a novel plant classification method based on both thermal and visible-light images, which shows higher accuracies than existing methods. It is the first study to perform super-resolution reconstruction using visible-light and thermal plant images, and a method to improve classification performance is proposed using generative adversarial network (GAN)-based super-resolution reconstruction.
Article
Environmental Sciences
Jiechen Tang, Hengjian Tong, Fei Tong, Yun Zhang, Weitao Chen
Summary: Superpixel-based classification using Active Learning (AL) has shown great potential in high spatial resolution remote sensing image classification tasks. However, in existing models, the neighboring superpixels are ignored and only the selected informative superpixel is labeled. This paper proposes a Similar Neighboring Superpixels Search and Labeling (SNSSL) method to fully utilize the expert labeling information and improve the classification accuracy.
Article
Environmental Sciences
Frank Badu Osei, Alfred Stein, Anthony Ofosu
INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH
(2019)
Article
Remote Sensing
Mengmeng Li, Wietske Bijker
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2019)
Article
Environmental Sciences
E. Goumehei, V. Tolpekin, A. Stein, W. Yan
WATER RESOURCES RESEARCH
(2019)
Article
Engineering, Electrical & Electronic
Kiledar S. Tomar, Shashi Kumar, Valentyn A. Tolpekin
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2019)
Article
Environmental Sciences
Khairiya Mudrik Masoud, Claudio Persello, Valentyn A. Tolpekin
Article
Environmental Sciences
Ge Qiu, Yuhai Bao, Xuchao Yang, Chen Wang, Tingting Ye, Alfred Stein, Peng Jia
Article
Environmental Sciences
Shashwat Shukla, Valentyn Tolpekin, Shashi Kumar, Alfred Stein
Article
Chemistry, Analytical
Milad Mahour, Valentyn Tolpekin, Alfred Stein
Article
Environmental Sciences
Mengmeng Li, Alfred Stein
Article
Remote Sensing
Mariana Belgiu, Wietske Bijker, Ovidiu Csillik, Alfred Stein
Summary: Crop type mapping is important for food security applications, and supervised classification methods are commonly used for generating data from satellite images. Various solutions like transfer learning, temporal-spectral signatures, re-utilization of inventories, and crowdsourcing are applied to generate samples for coarser classifications, but rarely for generating crop type samples. This study proposes a method that leverages phenology information to automatically generate crop samples, showing promising results for classes with reduced inter-class similarity. However, the method may not perform as well for crops with high inter-class similarity, particularly in regions with imbalanced crop samples. Despite its shortcomings, the proposed methodology offers a viable option for generating crop samples in regions with limited ground labels.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Geography, Physical
Getachew Workineh Gella, Wietske Bijker, Mariana Belgiu
Summary: This study successfully mapped crop types in fragmented agricultural landscapes using a combination of Synthetic Aperture Radar (SAR) and crop phenological information with different Dynamic Time Warping implementation strategies, demonstrating the potential for crop type mapping in smallholder farming areas.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Environmental Sciences
Anurag Kulshrestha, Ling Chang, Alfred Stein
Summary: The study introduces a Sinkhole Scanner to detect sinkholes in prone areas efficiently, using a mathematical model and numerical approach to search for subsiding regions resembling sinkhole shapes in sinkhole-prone regions.
Article
Remote Sensing
Wisdom Simataa Moola, Wietske Bijker, Mariana Belgiu, Mengmeng Li
Summary: This study developed a fuzzy classifier based on TWDTW distances to map vegetable types from Sentinel-1A SAR image time series. By calculating fuzzy memberships for each pixel, assessing classification uncertainty, and applying thresholds during defuzzification, the classification accuracy of the image was improved.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Environmental Sciences
Hossein Aghababaei, Giampaolo Ferraioli, Alfred Stein, Luis Gomez Deniz
Summary: This paper compares the effects of radar antenna polarization design on the probability of detecting persistent scatterers. It introduces an optimized method based on synthesizing polarimetric responses to improve the performance of detecting multiple scatterers. Experimental results demonstrate that polarization waveform optimization outperforms existing full-polarization-based detection methods in terms of PSs detection, particularly in increasing the density of detected PSs.
Article
Environmental Sciences
Ratna Sari Dewi, Wietske Bijker
EGYPTIAN JOURNAL OF REMOTE SENSING AND SPACE SCIENCES
(2020)
