相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Engineering, Electrical & Electronic
Yunzuo Zhang et al.
Summary: In this paper, we propose the Enhancement Multi-module Network (EMN) to detect leaky cable fixtures in railway tunnels and improve the mean accuracy. The network includes multi-scale module, multi-region module, fusion module, and relation module, which explore intra-class similarities and inter-class differences and enhance feature expression. Experimental results show that our algorithm outperforms other methods based on traditional handcrafted feature algorithm, deep learning algorithm, and few-shot learning algorithm.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2023)
Article
Geochemistry & Geophysics
Dawen Yu et al.
Summary: In this article, a novel object detection framework RSADet is proposed for remote sensing images, considering the spatial distribution, scale, and orientation/shape variations of the objects. The framework utilizes scale-attention boosted CNN heatmaps and deformable convolutions to improve detection performance, and introduces a bounding box confidence prediction branch to eliminate unreliable boxes.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Information Systems
Ziwen Chen et al.
Summary: This paper proposes an adaptive clipping algorithm based on the YOLO object detection algorithm to address the feature loss caused by high-resolution image compression. Experimental results show that the algorithm improves precision, recall, and mAP@0.5 compared to the original algorithm for vehicle detection.
MOBILE INFORMATION SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Ruiqian Zhang et al.
Summary: Object detection in UAV imagery is crucial in various fields, but it faces challenges due to the complex characteristics of the images. To address this, a novel Adaptive Dense Pyramid Network (ADPN) is proposed, which incorporates object distribution information and density prediction to improve detection accuracy.
Article
Computer Science, Artificial Intelligence
Vishnu Chalavadi et al.
Summary: This paper proposes a novel network called mSODANet for multi-scale object detection in aerial images. By using hierarchical dilated convolutions to learn contextual information of different types of objects, the model enhances the detection capability and achieves state-of-the-art performance on challenging datasets.
PATTERN RECOGNITION
(2022)
Article
Computer Science, Artificial Intelligence
Pengfei Zhu et al.
Summary: Drones equipped with cameras have been widely used in various fields, and the automatic understanding of visual data collected from drones has become highly demanding. To promote the development of object detection and tracking algorithms, challenge workshops have been organized, and a large-scale drone captured dataset, VisDrone, has been provided. The dataset enables extensive evaluation and investigation of visual analysis algorithms. The paper analyzes the current state of the field and proposes future directions.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Geochemistry & Geophysics
Yueming Sun et al.
Summary: The study proposes a lightweight and efficient dual contextual parsing network (EDCPNet) to address the challenges of conducting urban traffic information extraction using UAV images. The network surpasses competing methods in car and road extraction from UAV images, demonstrating superior performance and adaptability in complex urban scenes.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Geochemistry & Geophysics
Weihao Bo et al.
Summary: This article proposes a novel approach of utilizing salient object detection for burned area segmentation and introduces an efficient network model (BASNet) to improve the accuracy and speed of high-resolution UAV image segmentation. By utilizing two modules, BASNet significantly outperforms existing methods in both quantitative and qualitative evaluations.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Fang Peng et al.
Summary: The study proposed an automatic detection method of underwater sea cucumber based on deep learning, which improves the multi-scale feature fusion strategy through shortcut connection. The results demonstrate that this method achieves higher detection accuracy in complex underwater environments.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Haijun Zhang et al.
Summary: This paper presents a large-scale benchmark dataset, MOHR, aiming at performing multi-scale object detection in high-resolution UAV images, with a total of 90,014 annotated object instances. Experimental results show promising detection performance, but also demonstrate that the dataset is quite challenging for adopting natural image-based object detection models for UAV images.
Article
Engineering, Civil
Yomna Youssef et al.
Summary: The study examined the effectiveness of drones in collecting data in urban environments, using Cascade R-CNN networks to achieve automatic vehicle counting and tracking. The proposed technique allows for precise detection and classification of vehicles, presenting significant advantages in gathering traffic data.
TRANSPORTATION RESEARCH RECORD
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Weiping Yu et al.
Summary: DSHNet is a solution proposed for the long-tail distribution problem in UAV images, utilizing Class-Biased Samplers and Bilateral Box Heads to handle tail and head classes, significantly improving the performance of tail classes and achieving state-of-the-art results on VisDrone and UAVDT datasets.
2021 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WACV 2021
(2021)
Article
Computer Science, Artificial Intelligence
Sutao Deng et al.
Summary: Object detection has seen significant improvement in performance with the use of deep learning methods, but drone-view object detection remains challenging due to tiny-scale objects and uneven object distribution. This paper proposes a global-local self-adaptive network (GLSAN) to address these challenges, incorporating a global-local fusion strategy and adaptively refining detection through a progressive scale-varying network. The SARSA algorithm dynamically crops crowded regions in input images, while the LSRN enlarges cropped images for finer-scale feature extraction, contributing to data augmentation and enhancing the detector's robustness.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Environmental Sciences
Ruiqian Zhang et al.
Proceedings Paper
Computer Science, Artificial Intelligence
Zhenyu Wu et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Fan Yang et al.
2019 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV 2019)
(2019)
Proceedings Paper
Computer Science, Artificial Intelligence
Ross Girshick
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2015)