Related references
Note: Only part of the references are listed.
Article
Environmental Sciences
Xiangrong Zhang et al.
Summary: The aim of remote sensing image captioning (RSIC) is to describe a given RSI using coherent sentences. In this paper, a multi-source interactive stair attention mechanism is proposed to model the semantics of preceding sentences and visual regions of interest separately. Furthermore, a CIDEr-based reward reinforcement learning is used to enhance the quality of the generated sentences.
Article
Environmental Sciences
Jie Chen et al.
Summary: A high-quality built environment is crucial for human well-being. This paper proposes an intelligent evaluation method based on high-resolution remote sensing images to assess the urban built environment. The results demonstrate that the deep-learning-driven method is effective for representing high-resolution remote-sensing image scenes and assessing large-scale urban built areas.
Article
Computer Science, Information Systems
Yunhao Zhou et al.
Summary: This paper proposes a re-ranking network that utilizes contextual information to optimize person or vehicle re-identification (re-ID) ranking list, improving retrieval performance. The network predicts correlations between probe and top-ranked neighbor samples, and expands and enhances the feature embeddings of query and gallery images using a linear combination of their neighbors. The effectiveness of the proposed method is demonstrated through experiments on widely-used person and vehicle re-ID benchmarks.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Geochemistry & Geophysics
Yuanjia Chen et al.
Summary: Aircraft detection in synthetic aperture radar (SAR) automatic target recognition (ATR) is valuable yet challenging. To address this, we propose a novel SAR aircraft detector called deformable scattering feature correlation network (DSFCN), which leverages the scattering characteristics of multiscale SAR aircraft. DSFCN utilizes a scalable Swin Transformer backbone (SSTB) to extract hierarchical scattering features and a deformable region correlation module (DRCM) to flexibly correlate strong scattering regions (SSRs) carrying aircraft salient features. Experiments on a real-measured Gaofen-3 SAR aircraft dataset demonstrate the superiority and reliability of DSFCN over other representative CNN-based methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Geochemistry & Geophysics
Xu Tang et al.
Summary: In this article, an unsupervised change detection method called GMCD is proposed based on graph convolutional network (GCN) and metric learning. Through the use of Siamese fully convolution network, multiscale dynamic GCN, and a pseudolabel generation mechanism, GMCD outperforms existing state-of-the-art methods on four HRRS image change detection datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Environmental Sciences
Zihao Ren et al.
Summary: Remote sensing image captioning is challenging due to the variability in scale, distribution, and number of objects. This research proposes a mask-guided Transformer network with a topic token to improve accuracy and diversity in caption generation. Experimental results demonstrate that the method can generate captions with high accuracy and diversity.
Article
Engineering, Civil
Lei Tong et al.
Summary: UAV-based automatic railway inspection has the potential to reform railway inspection, and real-time railway scene parsing is essential in this area. This paper proposes a novel algorithm called deep fully decoupled residual convolutional network to address the challenge of balancing real-time demands and limited computation resources. The algorithm achieves this by using fully decoupled residual blocks and a customized auxiliary line loss function. Experimental results demonstrate the superior performance of the approach in terms of parameters, computation cost, and accuracy.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Environmental Sciences
Haonan Zhou et al.
Summary: In this study, we propose a self-learning method named SFRC for few-shot remote sensing image captioning. SFRC improves the performance in few-shot scenarios by ameliorating the way and efficiency of learning on limited data. The experiments show that SFRC outperforms recent methods in terms of performance evaluation metric scores.
Article
Geochemistry & Geophysics
Chenyang Liu et al.
Summary: This article introduces a new task of remote sensing image change captioning, proposes the RSICCformer model, and builds a large-scale dataset LEVIR-CC. Experimental results show that the model outperforms existing synthetic image change captioning methods on the LEVIR-CC dataset.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xiuwei Zhang et al.
Summary: In this paper, we propose an attentive differential high-resolution change detection network for remote sensing images. Our network effectively extracts multiscale substantive changed features and accurately detects small changed objects by utilizing a high-resolution backbone with a differential pyramid module and a multiscale spatial feature attention module. The experimental results demonstrate that our method outperforms the state-of-the-art methods in terms of both quantitative metrics and visual details.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Genc Hoxha et al.
Summary: This article introduces a system for describing changes in bitemporal images through change sentences to provide user-friendly interpretation. The system utilizes convolutional neural networks to extract features from bitemporal images and uses recurrent neural networks or support vector machines to generate coherent change descriptions, achieving promising experimental results.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Gong Cheng et al.
Summary: This paper proposes ISNet to improve the separability for remote sensing image change detection through feature refinement and two strategies. Experimental results demonstrate that ISNet achieves state-of-the-art performance on multiple datasets.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Yong Wang et al.
Summary: This research proposes a multiscale multiinteraction remote sensing image captioning model to address the differences between remote sensing images and natural images. By refining the CNN backbone on remote sensing images and using a multiinteraction feature representation module, the model achieves improved performance in generating captions for remote sensing images.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Xiaofeng Ma et al.
Summary: The optical remote-sensing image-captioning task has become a research hotspot, with proposed multiscale methods addressing incomplete image representation and outperforming traditional approaches.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Geochemistry & Geophysics
Wei Huang et al.
Summary: By utilizing the denoising-based multi-scale feature fusion mechanism, the quality of generating captions for remote sensing images has been improved, with experimental results showing the effectiveness of the proposed method on two public datasets.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Computer Science, Artificial Intelligence
Jianjun Gao et al.
Summary: Interpersonal relations play a crucial role in understanding how people interact with each other. This paper proposes a new architecture to reason interpersonal relations based on higher-order graph networks and multi-scale features, achieving over 5% improvement in recognizing social relations in images compared to state-of-the-art work.
Article
Geochemistry & Geophysics
Xuelong Li et al.
Summary: This article investigates the overfitting phenomenon in Remote Sensing Image Captioning caused by CE loss and proposes a new TCE loss method to alleviate the problem. Extensive experiments demonstrate the positive impact of TCE loss on RSIC.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Gencer Sumbul et al.
Summary: In this article, a novel summarization-driven approach is proposed for remote sensing image captioning, which enhances the quality and accuracy of captions by summarizing ground-truth captions and eliminating redundancy.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Geochemistry & Geophysics
Xueli Peng et al.
Summary: The study introduces a new end-to-end change detection network called DDCNN, which improves the accuracy and effectiveness of extracting change features in bitemporal optical remote sensing images by combining dense attention and DE unit. It demonstrates superior performance on five challenging data sets and particularly excels in seasonal change detection data set.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Yue Qiu et al.
Summary: The study introduces a multi-change captioning method that addresses the issues through simulation-based dataset, benchmarking existing single change captioning methods, and proposing Multi-Change Captioning transformers (MCCFormers). The proposed method achieved high scores in multi-change captioning evaluation and showed high effectiveness in change localization.
2021 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2021)
(2021)
Article
Environmental Sciences
Yangyang Li et al.
Article
Robotics
Yue Qiu et al.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2020)
Article
Geochemistry & Geophysics
Tengfei Bao et al.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2020)
Article
Engineering, Electrical & Electronic
Genc Hoxha et al.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2020)
Article
Geochemistry & Geophysics
Sudipan Saha et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Article
Geochemistry & Geophysics
Xiaoqiang Lu et al.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2018)
