Article
Multidisciplinary Sciences
Yazhou Yao, Tao Chen, Hanbo Bi, Xinhao Cai, Gensheng Pei, Guoye Yang, Zhiyuan Yan, Xian Sun, Xing Xu, Hai Zhang
Summary: This paper presents the background and results of the Automated Object Recognition in Optical Remote Sensing Imagery track in the 2022 International Algorithm Case Competition, and provides a summary of the challenges, champion solutions, and future directions.
NATIONAL SCIENCE REVIEW
(2023)
Article
Geochemistry & Geophysics
Xueqing Chen, Li Ma, Qian Du
Summary: The study proposes a one-stage anchor-free network based on searching four corner points of an object to fit objects with different shapes and orientations. By combining two strategies to detect corners, the results of the experiment demonstrate that the proposed method achieves superior performance to both anchor-based and anchor-free methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Jiaqing Zhang, Jie Lei, Weiying Xie, Zhenman Fang, Yunsong Li, Qian Du
Summary: In this article, the authors propose SuperYOLO, an accurate and fast object detection method for remote sensing images. By fusing multimodal data and utilizing assisted super resolution learning, SuperYOLO achieves high-resolution object detection on multiscale objects while considering the computation cost. Experimental results show that SuperYOLO outperforms state-of-the-art models in terms of accuracy and computational efficiency.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Remote Sensing
Zhipeng Dong, Yanxiong Liu, Yikai Feng, Yanli Wang, Wenxue Xu, Yilan Chen, Qiuhua Tang
Summary: This paper proposes a CNN-based method for object detection in high spatial resolution remote-sensing images. The method obtains optimal object anchor scales through adaptive learning and designs a detection framework based on these scales. Experimental results demonstrate its superior performance compared to other state-of-the-art algorithms.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Tong Zhang, Yin Zhuang, He Chen, Guanqun Wang, Lihui Ge, Liang Chen, Hao Dong, Lianlin Li
Summary: In this article, a novel one-stage anchor-free detector called PIIDet is proposed for Arbitrary-oriented object detection (AOOD). The detector utilizes an object-aware posterior guidance structure and a hierarchical feature fusion module for better learning specific categories and parametric information of objects. It also solves the negative optimization of angle regression using a binary classification embedded angle regression space.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Geochemistry & Geophysics
Sunan Shi, Yanfei Zhong, Ji Zhao, Pengyuan Lv, Yinhe Liu, Liangpei Zhang
Summary: In this article, a class-prior object-oriented conditional random field (COCRF) framework is proposed for high spatial resolution (HSR) remote sensing image change detection. The proposed framework utilizes class-prior knowledge to improve the construction of unary potential and provides constraint between binary change detection and multiclass change detection. Experimental results show that the proposed COCRF framework outperforms other state-of-the-art CD methods on two HSR remote sensing image datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xian Sun, Yingfei Liu, Zhiyuan Yan, Peijin Wang, Wenhui Diao, Kun Fu
Summary: The paper proposes a robust shape robust anchor-free network (SRAF-Net) for garbage dump detection, utilizing a context-based deformable (CBD) module and multitask detection strategy to enhance detection accuracy and efficiency, while successfully creating a garbage dump dataset (GDD) to validate the effectiveness of the method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Yin Zhuang, Yuqun Liu, Tong Zhang, He Chen
Summary: Under the multiscale distribution, a concise and effective one-stage anchor-free contour modeling detector called CMDet is proposed for accurate arbitrary-oriented ship detection. Different from currently existed methods, CMDet resolves the oriented bounding box (OBB) modeling by jointly regressing the contour information. The proposed CMDet achieves competitive results compared to state-of-the-art detectors in extensive experiments on public OBB ship detection datasets.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Environmental Sciences
Ruchan Dong, Licheng Jiao, Yan Zhang, Jin Zhao, Weiyan Shen
Summary: In this study, a multi-scale spatial attention region proposal network (MSA-RPN) is proposed for high-resolution optical remote sensing imagery, which focuses on improving the recall rate for small targets in object detection and achieves higher accuracy.
Article
Computer Science, Artificial Intelligence
Zhuo Zheng, Yanfei Zhong, Junjue Wang, Ailong Ma, Liangpei Zhang
Summary: In this paper, a foreground-aware relation network (FarSeg++) is proposed to address the issues of scale variation, large intra-class variance of background, and foreground-background imbalance in high spatial resolution remote sensing imagery. The network improves the discrimination of foreground features, achieves balanced optimization, and enhances objectness representation. Experimental results demonstrate that FarSeg++ outperforms state-of-the-art semantic segmentation methods and achieves a better trade-off between speed and accuracy.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Environmental Sciences
Jianxiang Li, Yan Tian, Yiping Xu, Zili Zhang
Summary: This paper proposes a novel anchor-free two-stage oriented object detector, which successfully addresses the issues of imbalanced positive and negative samples and appearance ambiguity faced by existing methods, achieving high detection performance.
Article
Geochemistry & Geophysics
Xu He, Shiping Ma, Linyuan He, Le Ru
Summary: Object detection is an important research field in optical remote sensing image analysis. This letter proposes a high-resolution polar network (HRPNet) for target detection in remote sensing images. By converting the detection task into polar coordinates, the HRPNet significantly reduces computational complexities and demonstrates competitive advantages in experiments.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2022)
Article
Geochemistry & Geophysics
Tong Zhang, Yin Zhuang, Guanqun Wang, Shan Dong, He Chen, Lianlin Li
Summary: This article proposes a powerful one-stage detector named MSFC-Net for optical remote sensing object detection, which aims to improve performance through semantic fusion and novel convolution layer design. Experimental results demonstrate that MSFC-Net outperforms other state-of-the-art one-stage detectors on the DOTA and DIOR datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Agronomy
Rongxin Deng, Gao Yang, Ying Li, Zhengran Xu, Xing Zhang, Lu Zhang, Chunjing Li
Summary: This study developed a method for accurately identifying shelterbelt width using high-resolution remote sensing monitoring. The method was applied in a study area in Northeast China, and the results showed high accuracy and reliability. The proposed method can help determine shelterbelt structure and enable remote simulation, contributing to accurate monitoring and management of shelterbelt structure and function.
AGROFORESTRY SYSTEMS
(2022)
Article
Chemistry, Analytical
Xinxin Hu, Changming Zhu
Summary: This article addresses the problem of misalignment in multi-scale features for oriented object detection in aerial images. The authors propose different modules to align the features in spatial, axial, and semantic dimensions, achieving state-of-the-art accuracy and inference speed on challenging aerial benchmarks.
Article
Engineering, Electrical & Electronic
Chen Wang, Jun Shi, Zenan Zhou, Liang Li, Yuanyuan Zhou, Xiaqing Yang
Summary: The paper introduces a training mechanism for concealed object detection network based on normalized accumulation map, which accurately reveals the positions of concealed objects by calculating the average binary mask, thus improving the performance of the object detection network.
IEEE SENSORS JOURNAL
(2021)
Article
Environmental Sciences
Bokun Tian, Xiaoling Zhang, Liang Li, Ling Pu, Liming Pu, Jun Shi, Shunjun Wei
Summary: A fast Bayesian compressed sensing algorithm via relevance vector machine is proposed for linear array synthetic aperture radar (LASAR) 3D sparse imaging in this paper. The method optimizes hyperparameters, extracts target areas, and improves scattering coefficient estimation accuracy using truncated singular value decomposition algorithm. Simulation and experimental results demonstrate improved computational efficiency and imaging quality compared to state-of-the-art CS-based methods.
Article
Engineering, Electrical & Electronic
Liang Li, Xiaoling Zhang, Yuanyuan Zhou, Liming Pu, Jun Shi, Shunjun Wei
Summary: A new algorithm for target extraction in 3-D SAR images is proposed in this paper, which successfully overcomes the influence of background interference and achieves high-precision target extraction using the adaptive morphological reconstruction fuzzy C-means algorithm. Experimental results demonstrate that the algorithm outperforms other algorithms in terms of performance and computational efficiency.
DIGITAL SIGNAL PROCESSING
(2021)
Article
Geochemistry & Geophysics
Chen Wang, Jun Shi, Yuanyuan Zhou, Liang Li, Xiaqing Yang, Tianwen Zhang, Shunjun Wei, Xiaoling Zhang, Chongben Tao
Summary: This article proposes a loss curve-fitting-based method to address the issue of wrong labels in large-scale datasets. By modeling label noise through unsupervised clustering and using augmented samples, the proposed method effectively corrects noisy labels. Experimental results validate the effectiveness of the method.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Liming Pu, Xiaoling Zhang, Zenan Zhou, Liang Li, Liming Zhou, Jun Shi, Shunjun Wei
Summary: This paper introduces a robust least squares phase unwrapping method based on deep learning, which can achieve more accurate and robust results by learning global phase features and the phase gradient between adjacent pixels compared to traditional methods.
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
Environmental Sciences
Liming Pu, Xiaoling Zhang, Liming Zhou, Liang Li, Jun Shi, Shunjun Wei
Summary: This paper proposes a nonlocal InSAR filtering method using deep learning and nonlocal feature selection strategy, which outperforms traditional methods and other deep learning methods in terms of accuracy and speed.
Article
Computer Science, Information Systems
Liming Zhou, Xiaoling Zhang, Yangyang Wang, Liang Li, Liming Pu, Jun Shi, Shunjun Wei
Article
Computer Science, Information Systems
Liang Li, Xiaoling Zhang, Ling Pu, Liming Pu, Bokun Tian, Liming Zhou, Shunjun Wei
Article
Computer Science, Information Systems
Ling Pu, Xiaoling Zhang, Jun Shi, Shunjun Wei, Liang Li, Xinxin Tang
Article
Computer Science, Information Systems
Bokun Tian, Xiaoling Zhang, Shunjun Wei, Jing Ming, Jun Shi, Liang Li, Xinxin Tang
