Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Jielei Wang et al.
Summary: This paper proposes a lightweight network model for ship target detection in synthetic aperture radar (SAR) imagery. The network structure optimization algorithm based on the multi-objective firefly algorithm (NOFA) is designed to encode the filters of a well-performing ship target detection network into a list of probabilities. The multi-objective firefly optimization algorithm (MFA) further optimizes the probability list to output a set of lightweight network encodings.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Chemistry, Analytical
Xianjia Yu et al.
Summary: Over the last decade, robotic perception algorithms have been greatly improved by the advancements in deep learning. This study explored the potential of using general-purpose deep learning perception algorithms for processing image-like outputs of advanced lidar sensors. By preprocessing the data, it was shown that deep learning models can effectively process these images, allowing for their usage in environments where vision sensors have limitations.
Article
Geochemistry & Geophysics
Zheng Zhou et al.
Summary: In this article, a hybrid representation learning enhanced SAR target detection algorithm called HRLE-SARDet is proposed from a lightweight perspective. The algorithm achieves excellent performance on four SAR image datasets through the design of a lightweight and scattering feature extraction backbone, a multiscale feature fusion neck, a lightweight hybrid representation learning enhancement module, and a flexible loss function.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2023)
Article
Environmental Sciences
Kexue Zhou et al.
Summary: In this paper, a new methodology called MSSDNet is proposed for better detection of multi-scale ship objects in SAR images. The method utilizes YOLOv5 with a small model size and incorporates the CSPMRes2 module for improved feature representation and the FC-FPN module for adaptive feature map fusion. The experimental results demonstrate that MSSDNet outperforms state-of-the-art methods in multi-scale ship detection.
Article
Environmental Sciences
Takero Yoshida et al.
Summary: This study proposes an automatic method for detecting cruising ships in spotlight synthetic aperture radar (SAR) images using the You Only Look Once (YOLO) v5 model. The results showed high precision and recall rates in the automatic detection of moving ships using this deep learning method.
Article
Environmental Sciences
Simin Wang et al.
Summary: As an outstanding method for ocean monitoring, synthetic aperture radar (SAR) has received much attention from scholars in recent years. Significant progress has been made in ship detection in SAR images, however, detecting small ships in SAR images remains challenging. To address this issue, we propose an efficient ship detection model named YOLO-SD, which improves accuracy by fusing feature information at different scales and optimizing semantic information. Experimental results show that YOLO-SD outperforms baseline YOLOX and other object detection models in terms of overall performance.
Article
Environmental Sciences
Shou Feng et al.
Summary: In this paper, a multispectral image change detection method based on multi-scale adaptive kernel network and multimodal conditional random field (MSAK-Net-MCRF) is proposed. The method effectively extracts multi-scale features and smooths the detection results through the extension of the encoding path, the design of a selective convolution kernel block, and the embedding of an upsampling module.
Review
Environmental Sciences
Ying Zhang et al.
Summary: This paper reviews SAR image target detection based on Convolutional Neural Network (CNN), analyzes the limitations of traditional detection algorithms, and introduces the principles and basic structure of CNN. The research status of SAR image target detection based on CNN is summarized and compared with traditional algorithms. The challenges of SAR image target detection are discussed and future research directions are proposed.
Article
Computer Science, Artificial Intelligence
Tingting Liang et al.
Summary: In this paper, a novel and flexible backbone framework called CBNet is proposed for constructing high-performance object detectors. CBNet integrates multiple identical backbones and gradually expands the receptive field to improve object detection. The experimental results show that CBNet is a more efficient and effective approach for building high-performance backbone networks compared to simply increasing network depth and width.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2022)
Article
Chemistry, Analytical
Haozhen Zhu et al.
Summary: A SAR ship detector called DB-YOLO is proposed in this paper to address challenges such as real-time performance and multi-scale detection. Experimental results demonstrate that DB-YOLO outperforms other state-of-the-art methods on two SAR ship datasets.
Article
Environmental Sciences
Zhongzhen Sun et al.
Summary: This article proposes a novel YOLO-based arbitrary-oriented SAR ship detector BiFA-YOLO, which effectively detects arbitrary-oriented and densely arranged ships in HR SAR images through bi-directional feature fusion and angular classification. The method achieves good detection performance in ship detection tasks.
Article
Chemistry, Analytical
Yuhang Jiang et al.
Summary: In this paper, a new lightweight anchor-free detector named R-Centernet+ is proposed for ship detection in SAR images, which effectively addresses the shortcomings of traditional models in small ships and complex backgrounds by introducing the CBAM module, FEM module, and a detection head that can output the ship angle map. Experiments on public SAR image datasets demonstrate that the proposed detector achieves higher accuracy in detecting inshore and offshore ships, with an average precision of 95.11% on SSDD and 84.89% on AIR-SARShip, while maintaining a fast detection speed of 33 frames per second.
Review
Environmental Sciences
Tianwen Zhang et al.
Summary: SSDD is the first open dataset widely used for research on ship detection from SAR imagery based on deep learning, with 46.59% of public reports confidently choosing it. However, the initial version's coarse annotations and ambiguous usage standards hinder fair methodological comparisons and effective academic exchanges. To address these challenges, SSDD will be officially released in three versions to cater to different research needs.
Article
Computer Science, Information Systems
Jun Zhang et al.
Summary: This paper proposes a small target detection method based on deep learning, which improves the current deep network architecture and introduces multiple features and multi-scale detection to effectively solve the problem of small target detection and expand the training dataset effectively, solving the difficulties in labeling small target data and overfitting.
Article
Environmental Sciences
Tianwen Zhang et al.
Article
Computer Science, Information Systems
Shunjun Wei et al.
Article
Computer Science, Information Systems
Xuan Nie et al.
Article
Geochemistry & Geophysics
Zongyong Cui et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Computer Science, Artificial Intelligence
Shaoqing Ren et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Environmental Sciences
Miao Kang et al.
Article
Computer Science, Artificial Intelligence
Kaiming He et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2015)
Article
Environmental Sciences
Wang Weixing et al.
JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING
(2015)
Review
Multidisciplinary Sciences
Yann LeCun et al.
Article
Geochemistry & Geophysics
Alberto Moreira et al.
IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE
(2015)
Article
Geochemistry & Geophysics
Stephan Brusch et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2011)
