Physics Guided Remote Sensing Image Synthesis Network for Ship Detection

Zhang, Weichang; Zhang, Rui; Wang, Guoqing; Li, Wei; Liu, Xun; Yang, Yang; Hu, Die

doi:10.1109/tgrs.2023.3248106

Cited by 11 publications

(5 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This innovative approach has the potential to significantly improve the accuracy and efficiency of SAR image analysis. Zhang et al [22] proposed a novel sensor-related image synthesis framework, dubbed remote sensing-image synthesis pipeline (RS-ISP), which was developed in response to the scarcity of on-orbit remote sensing imagery. Zhang et al [23] used the feature of DCT blocks to extract horizon information for efficient ship detection.…”

Section: Deep-learning-based Ship Monitoring Methodsmentioning

confidence: 99%

LFLD-CLbased NET: A Curriculum-Learning-Based Deep Learning Network with Leap-Forward-Learning-Decay for Ship Detection

Sun

et al. 2023

JMSE

View full text Add to dashboard Cite

Ship detection in the maritime domain awareness field has seen a significant shift towards deep-learning-based techniques as the mainstream approach. However, most existing deep-learning-based ship detection models adopt a random sampling strategy for training data, neglecting the complexity differences among samples and the learning progress of the model, which hinders training efficiency, robustness, and generalization ability. To address this issue, we propose a ship detection model called the Leap-Forward-Learning-Decay and Curriculum Learning-based Network (LFLD-CLbased NET). This model incorporates innovative strategies as Leap-Forward-Learning-Decay and curriculum learning to enhance its ship detection capabilities. The LFLD-CLbased NET is composed of ResNet as the feature extraction unit, combined with a difficulty generator and a difficulty scheduler. The difficulty generator in LFLD-CLbased NET effectively expands data samples based on real ocean scenarios, and the difficulty scheduler constructs corresponding curriculum training data, enabling the model to be trained in an orderly manner from easy to difficult. The Leap-Forward-Learning-Decay strategy, which allows for flexible adjustment of the learning rate during curriculum training, is proposed for enhancing training efficiency. Our experimental findings demonstrate that our model achieved a detection accuracy of 86.635%, approximately 10% higher than other deep-learning-based ship detection models. In addition, we conducted extensive supplementary experiments to evaluate the effectiveness of the learning rate adjustment strategy and curriculum training in ship detection tasks. Furthermore, we conducted exploratory experiments on different modules to compare performance differences under varying parameter configurations.

show abstract

Section: Deep-learning-based Ship Monitoring Methodsmentioning

confidence: 99%

LFLD-CLbased NET: A Curriculum-Learning-Based Deep Learning Network with Leap-Forward-Learning-Decay for Ship Detection

Sun

et al. 2023

JMSE

View full text Add to dashboard Cite

show abstract

“…Discussion and limitations. The reconstruction effect of GAN-based methods has more realistic texture details and is widely used in remote sensing image processing [14,113], but the large number of network parameters in this approach makes network training unstable and slows down inference. Section 3.2.2 highlights the super-resolution methods based on degradation modeling using generative adversarial networks.…”

Section: Realsr(2020)mentioning

confidence: 99%

Network architecture for single image super‐resolution: A comprehensive review and comparison

Zhang,

Han,

Zhu

et al. 2024

IET Image Processing

View full text Add to dashboard Cite

Single image super‐resolution (SISR) is a promising research direction in computer vision and image processing for improving the visual perception of low‐quality images. In recent years, deep learning algorithms have driven tremendous development in SR, and SR methods based on various network architectures have significantly improved the quality of reconstructed images. Although there has been a large amount of reviews focusing on SISR, few studies have focused specifically on network architectures for SISR. This paper aims to provide a systematic overview of the design ideas of SISR using multiple architectures, including Convolutional Neural Networks (CNN), Generative Adversarial Networks (GAN), Transformer, and Diffusion model. In addition, an experimental analysis and comparison of state‐of‐the‐art SR algorithms have been performed on publicly available quantitative and qualitative datasets. Finally, some future directions are discussed that may help other community researchers.

show abstract

“…Zhou et al [16] proposed a data augmentation technique and training set partitioning method to speed up the training process and alleviate model overfitting to improve the overall performance of deep neural network based conversational recommendation algorithms. Based on the idea of GRU4Rec, researcher in [17,18] proposed a multi-layer recurrent neural network model [19], which separately considered the user's dependencies and interest changes between different sessions to provide more reliable recommendation results [20,21].…”

Section: Recommendation Based On Neural Networkmentioning

confidence: 99%

Deep Attention Network with Clustering Design for Recommendation System

Zhou,

2023

J. Comb. Math. Comb. Comput.

View full text Add to dashboard Cite

In order to provide users with better recommendations, it is particularly important to analyze the behavior of users tagging different resources. In this paper, an attention mechanism based on deep learning is designed to effectively capture the features related to the user’s long-term interests and current interests in the session simultaneously, and alleviate the impact of the user’s interest drift that is difficult to deal with by the current session recommendation algorithm on the recommendation accuracy. The main community discovery algorithms are applied to the clustering analysis of the label system, and their performances on different data sets are compared. Besides, we design a personalized recommendation algorithm for the label system. The experimental results show that the proposed algorithm can find the interests of different users and improve the quality of the recommendation system.

show abstract

Physics Guided Remote Sensing Image Synthesis Network for Ship Detection

Cited by 11 publications

References 45 publications

LFLD-CLbased NET: A Curriculum-Learning-Based Deep Learning Network with Leap-Forward-Learning-Decay for Ship Detection

LFLD-CLbased NET: A Curriculum-Learning-Based Deep Learning Network with Leap-Forward-Learning-Decay for Ship Detection

Network architecture for single image super‐resolution: A comprehensive review and comparison

Deep Attention Network with Clustering Design for Recommendation System

Contact Info

Product

Resources

About