Article
Engineering, Electrical & Electronic
Jianchao Fan, Chuan Liu
Summary: This article designs a multitask generative adversarial networks (MTGANs) oil spill detection model to distinguish oil spills and look-alike oil spills and segment oil spill areas in one framework. The experimental results demonstrate that the proposed MTGANs oil spill detection framework outperforms other models in oil spill classification and semantic segmentation.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Engineering, Electrical & Electronic
Fatemeh Mahmoudi, Shahriar Baradaran Shokouhi, Gholamreza Akbarizadeh
Summary: This article explores the use of SAR imaging and deep neural networks for oil spill detection, finding that the U-NET network is the most accurate in identifying oil spills in SAR images. The authors increased the number of input images and trained two convolutional neural networks to achieve their results.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Lizwe Wandile Mdakane, Waldo Kleynhans
Summary: The study aims to develop a monitoring system for automatically detecting oil spill events caused by vessels in African Oceans and identify critical features for distinguishing oil spills from look-alikes. Investigation of common feature selection methods and classifiers reveals consistent significance of certain features across all methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Engineering, Electrical & Electronic
Cornelius Quigley, A. Malin Johansson, Cathleen E. E. Jones
Summary: The damping ratio is used as an indicator of oil thickness within oil slicks observed in SAR imagery. However, there is a lack of well-defined and evaluated method for calculating the damping ratio. This study reviews prior work on the damping ratio and tests alternative methods on multifrequency datasets. The proposed histogram method provides consistent results even under challenging conditions.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Cornelius Quigley, Camilla Brekke, Torbjorn Eltoft
Summary: In this study, the retrieval results for the dielectric properties of verified oil slick using airborne multifrequency synthetic aperture radar were compared. Two different inversion methods were used and the results showed consistency, especially for low dielectric values.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Fang Chen, Aihua Zhang, Heiko Balzter, Peng Ren, Huiyu Zhou
Summary: This work aims to develop an effective method for marine oil spill segmentation in SAR images by investigating the distribution representation of SAR images. The proposed method utilizes the probability distribution representation of oil spill SAR images to guide the segmentation process. Experimental evaluations demonstrate the effectiveness of the proposed method for different types of marine oil spill SAR image segmentation.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Liang Li, Xiaoling Zhang, Bokun Tian, Chen Wang, Liming Pu, Jun Shi, Shunjun Wei
Summary: An adaptive threshold ROI extraction algorithm is proposed to enhance the anti-interference ability of existing image segmentation methods, significantly improving the effectiveness of target extraction.
Article
Chemistry, Analytical
Mohamed Shaban, Reem Salim, Hadil Abu Khalifeh, Adel Khelifi, Ahmed Shalaby, Shady El-Mashad, Ali Mahmoud, Mohammed Ghazal, Ayman El-Baz
Summary: This paper introduces a two-stage deep-learning framework for the identification of oil spill occurrences from SAR images. The framework shows improved precision and Dice score compared to related work, while addressing the issue of reduced oil spill representation in patches through semantic segmentation.
Article
Geochemistry & Geophysics
Leonardo De Laurentiis, Cathleen E. Jones, Benjamin Holt, Giovanni Schiavon, Fabio Del Frate
Summary: This study uses uninhabited aerial vehicle SAR data to investigate oil slick classification within a deep learning framework, evaluating the capabilities of deep architectures to provide a reliable and accurate three-state classifier for separating mineral oil films from biogenic slicks and clean sea. By exploiting parameters sensitive to dielectric constant and ocean wave damping properties, as well as leveraging CNNs' capability for learning nonlinear features and patterns, significant classification accuracy is achieved, with values up to 0.91, 0.94, 0.98, and 0.99 under real-world spill acquisition conditions.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Engineering, Aerospace
Michael Inggs
Summary: This document summarizes the achievements in synthetic aperture radar (SAR) technology during the 50-year existence of the Aerospace and Electronic Systems Society. Advances in radar technology, driven by the digital revolution, have led to the widespread application of SAR in various fields. The development of coherent radar during World War II enabled the formation of large synthetic apertures, resulting in microwave images with high resolution that are unaffected by time and weather. The article traces the history of SAR technology from airborne platforms to satellites and discusses its achievements. SAR technology has now entered the phase of commercial exploitation, with a significant increase in available systems.
IEEE AEROSPACE AND ELECTRONIC SYSTEMS MAGAZINE
(2023)
Article
Geochemistry & Geophysics
Hongyang An, Junjie Wu, Kah Chan Teh, Zhichao Sun, Zhongyu Li, Jianyu Yang
Summary: This article proposes an efficient video formation method for video SAR systems with reduced data, modeling the observed scene as a sum of low-rank and sparse tensors and using a tensor alternating direction method of multiplier. Compared to traditional imaging methods, the proposed approach greatly reduces the amount of data samples.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Tao Xie, Ruihang Ouyang, Will Perrie, Li Zhao, Xiaoyun Zhang
Summary: This paper focuses on using polarization ratio to discriminate between oil spills and seawater. A new method called relative polarization ratio (PRr) is proposed, which enhances the contrast between oil spills and seawater, reduces the difference between low wind areas and ordinary seawater, and provides better image details. The threshold method based on Euclidean distance achieves the highest overall classification accuracy within the allowable error range, and the proposed methods largely avoid misjudging low wind areas as oil spills. With an overall classification accuracy of more than 95%, this method can effectively complement existing oil spill detection methods.
Article
Engineering, Electrical & Electronic
Xin Zhang, Chunlei Huo, Nuo Xu, Hangzhi Jiang, Yong Cao, Lei Ni, Chunhong Pan
Summary: This study introduces a multitask learning-based object detector (MTL-Det) to improve ship detection performance in SAR images by modeling the problem as three cooperative tasks and utilizing auxiliary subtasks to enhance feature learning. The approach effectively addresses the challenges posed by speckle noise in SAR images and outperforms traditional single-task-based object detectors.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Review
Computer Science, Artificial Intelligence
Shuyuan Yang, Guangying Xu, Huixiao Meng, Min Wang
Summary: This study proposes a DCNP algorithm for PolSAR image classification, utilizing information theoretic divergence and divergence-Chebyshev distance to measure and reveal the similarity and affinity between pixels. Inspired by human learning characteristic, the algorithm can progressively determine the labels of unknown pixels.
APPLIED SOFT COMPUTING
(2021)
Article
Engineering, Electrical & Electronic
Yuri Alvarez Lopez, Jaime Laviada, Ana Arboleya, Fernando Las-Heras
Summary: Synthetic aperture radar (SAR)-based microwave imaging systems are widely used in various applications. The scanning speed is a crucial factor in SAR imaging systems, and widening the distance between measurements can increase it. However, this causes the presence of grating lobes that degrade the quality of microwave SAR images. To address this issue, a novel methodology is proposed in this study, which incorporates the amplitude and phase of the field radiated by the transmitting and receiving antennas in the backpropagation imaging algorithm. This method exploits the directive pattern of the antennas to reduce the level of grating lobes in SAR images.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Geochemistry & Geophysics
Yuteng Ma, Junmin Meng, Lina Sun, Peng Ren
Summary: In this research, a two-stage oceanic internal wave (IW) signature segmentation algorithm for synthetic aperture radar (SAR) images is proposed. The algorithm consists of an IW images classification stage and a stripe segmentation stage. Experimental results demonstrate that the proposed algorithm can successfully extract the signature of oceanic IWs from SAR images.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Geochemistry & Geophysics
Fang Chen, Heiko Balzter, Feixiang Zhou, Peng Ren, Huiyu Zhou
Summary: In this article, we propose an effective segmentation framework called DGNet for oil spill segmentation in SAR images. Our framework incorporates the intrinsic distribution of backscatter values in SAR images and utilizes two deep neural modules for inference and generation. Experimental evaluations show that DGNet achieves accurate oil spill segmentation.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geography, Physical
Hao Wang, Shixin Sun, Peng Ren
Summary: Underwater camera images often suffer from visual degradation issues, and existing studies tend to address these issues separately, resulting in inconsistent improvements in underwater image visibility. To overcome this limitation, we propose a smart protocol for underwater image enhancement, which utilizes reinforcement learning to optimize the parameter values for a comprehensive cascade of seven enhancement techniques. Our methodology, referred to as meta underwater camera (MUC), outperforms state-of-the-art methods in improving underwater image visibility.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Yuteng Ma, Junmin Meng, Baodi Liu, Lina Sun, Hao Zhang, Peng Ren
Summary: In this paper, a novel dictionary learning algorithm is proposed for few-shot remote sensing scene classification. By using natural image datasets for pre-training, a feature extractor suitable for remote sensing data is obtained. A kernel space classifier is designed to map the features to a high-dimensional space and embed the label information into the dictionary learning process to improve feature discrimination for classification. Extensive experiments on four popular remote sensing scene classification datasets demonstrate the effectiveness of the proposed dictionary learning method.
Article
Engineering, Civil
Hao Wang, Shixin Sun, Xiao Bai, Jian Wang, Peng Ren
Summary: This article investigates the problem of enhancing underwater visual observations for accurate underwater object detection. Existing algorithms tend to follow human vision preference and may not be effective for object detection. The study proposes a reinforcement learning paradigm to configure visual enhancement for object detection in underwater scenes. Experimental results validate the effectiveness of the proposed method in improving detection results.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2023)
Article
Automation & Control Systems
Huajun Song, Laibin Chang, Hao Wang, Peng Ren
Summary: Considering the limitations of single physics-based or deep learning-based methods, this paper proposes an effective Dual-model methodology for underwater image enhancement. The Dual-model consists of a Revised Imaging Network-model (RIN-model) and a Visual Perception Correction-model (VPC-model). The RIN-model estimates ambient-light and direct-transmission parameters to reconstruct a preliminary enhanced image based on a revised underwater image formation model, and the VPC-model further enhances contrast and restores color to correct the visual perception of the image.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Review
Remote Sensing
Yuxi Lu, Wu Wen, Kostromitin Konstantin Igorevich, Peng Ren, Hongxia Zhang, Youxiang Duan, Hailong Zhu, Peiying Zhang
Summary: With the rapid development of 5G and 6G communications, there is increasing interest in space-air-ground integrated networks (SAGINs) to achieve seamless all-area, all-time coverage. Flying ad hoc networks (FANETs) play a key role in SAGINs, particularly in the agriculture and transportation sectors. This study analyzes the unique communication architecture of FANETs in SAGINs, presents existing routing protocols, reviews recent research advances in routing algorithms, and discusses future research trends for FANET routing algorithms in SAGINs.
Article
Geochemistry & Geophysics
Xinrong Lyu, Jun Zhou, Peng Ren, Alejandro C. Frery
Summary: This manuscript presents a model based on self-sampling and semicorrelated co-training to improve the detection accuracy of algal bloom. The self-sampling module enhances the efficiency of extracting useful information from training sets, while the semicorrelated co-training module addresses the issue of limited labeled samples. Experimental results show that the proposed model has contributed to the improvement of detection accuracy.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Xiaoyu Sun, Xi Zhang, Weimin Huang, Zongjun Han, Xinrong Lyu, Peng Ren
Summary: This article investigates sea ice classification on a remote sensing image with just a small number of labeled pixels. An effective method is developed to extract context features from a classification map, and an iterative learning paradigm is established to guide the update of these features. The paradigm is referred to as mutually guided contexts, which comprehensively characterizes the sea ice image representation for training and classification even with limited training data.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Shicai Wei, Yang Luo, Xiaoguang Ma, Peng Ren, Chunbo Luo
Summary: Learning-based multimodal data is of increasing interest in remote sensing due to its strong performance. This work proposes a framework called modality-shared hallucination network (MSH-Net) that reconstructs complete modality-shared features from incomplete inference modalities to address the problem of missing modalities in multimodal remote sensing image processing. Additionally, a novel joint adaptation distillation (JAD) method is developed to guide the hallucination model in learning modality-shared knowledge from the multimodal model. Extensive experiments demonstrate that MSH-Net achieves state-of-the-art performance in addressing the problem of missing modalities. Code is available at: https://github.com/shicaiwei123/MSHNet
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Jingpeng Han, Peng Li, Yimin Tao, Peng Ren
Summary: This article investigates encrypting remote sensing images to counter the Hashing for Localization task. Two cues are characterized for the task: visually similar encrypted patches and different hash codes. Based on an adversarial method, an encrypted patch generator is developed, leading to two encryption frameworks causing nonlocalization and mislocalization to the HfL task.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Engineering, Civil
Yongqing Li, Xinrong Lyu, Peng Ren
Summary: We investigate the problem of oil spill timely backtracking and propose a self-attention temporal convolutional network (SaTCN) to obtain a timely and accurate estimate of historical wind fields for reliable oil spill backtracking. The SaTCN consists of a self-attention network and a temporal convolutional network, which capture spatial correlations and temporal characteristics from wind field sequences. Extensive experiments validate the effectiveness of the proposed method in addressing oil spill accidents.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2023)
Article
Geochemistry & Geophysics
Peng Ren, Yimin Tao, Jingpeng Han, Peng Li
Summary: In this article, a method of fast geolocation of query ground images using geo-tagged aerial images is proposed. By converting aerial images into ground-view aerial images with the same angle of view as the ground image, and generating hash codes for images using a feature extraction model and a hash encoder, the geographic location of the ground image is efficiently retrieved by searching for matched geo-tagged aerial images based on the hash codes.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Yuxin Fang, Peng Li, Jie Zhang, Peng Ren
Summary: This study investigates the efficient localization of sketch depicted scenes in a remote sensing image dataset. Hashing techniques are explored to achieve efficient retrieval and linear mapping models are used to generate hash codes for sketches and remote sensing images separately. The proposed method enables efficient remote sensing image retrieval with sketch queries and has been validated to be effective and efficient.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Huajun Song, Laibin Chang, Ziwei Chen, Peng Ren
Summary: This paper presents a comprehensive underwater visual reconstruction paradigm that includes three procedures: the E-procedure, the R-procedure, and the H-procedure. By enhancing original images, registering multiple enhanced images, and homogenizing the registered images, the paradigm achieves restored colors, sharpened edges, and improved visibility in wide-field underwater images.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)