检索结果-内蒙古大学图书馆

7th International Symposium on Autonomous systems, ISAS 2024

作者： Wang, Tianqi Yang, Liying Chang, Yanchun Huang, Zhaoxiong Jiang, Han Zheng, Yongjie Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang Institute of Automation Shenyang110016 China Chinese Academy of Sciences Institutes For Robotics and Intelligent Manufacturing Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China

ISBN: (纸本)9798350363173

Unmanned aerial vehicles (UAVs) have seen widespread applications, bringing new changes to many fields, such as forest fire monitoring, factory warehouse etc. Consequently, ensuring the safety of UAVs during flight missions is of paramount importance. Enhancing the capability of UAVs to avoid obstacles is important in ensuring their flight safety. While algorithms for avoiding static obstacles have been extensively studied, there is still room for improvement in the ability to avoid dynamic obstacles. This paper provides a retrospective review of research papers from recent years that focus on UAVs avoiding dynamic obstacles. It outlines the fundamental framework for UAVs to navigate around dynamic obstacles, categorizing it into three main components: environmental perception, trajectory prediction, and trajectory planning. The study summarizes various algorithms for UAVs to avoid dynamic obstacles and compares the differences of these algorithms. Finally, the aspects of dynamic obstacle avoidance that should continue to be improved are discussed. © 2024 IEEE.

关键词： Collision avoidance

来源：评论

学校读者我要写书评

暂无评论

Study on gesture recognition based on improved YOLOV8

Study on gesture recognition based on improved YOLOV8

引用

2024 International Conference on Mechatronic Engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Wang, Shiyi Zhang, Hengrui School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

Based on YOLOV8 hand gesture signal recognition in today's hand gesture signal recognition field has a great development and application prospects, which improve the detection accuracy of the hand gesture signal is the most important thing. Hand gesture signal recognition and detection and other areas of target detection is slightly different, hand gesture in the whole picture often belongs to a smaller target, and hand gesture signal recognition of complex application scenarios, In an effort to enhance the detection precision of hand gesture signal, this paper in initial YOLOV8 algorithm based on YOLOV8 algorithm: in YOLOV8 backbone network tail to join the Multi-Scale Convolutional Attention (MSCA) module is added to the tail of YOLOV8 backbone network;Large Selective Kernel Network (LSKNet) module is included into the front of YOLOV8 detection head;and the loss function of YOLOV8 is exchanged for Wise-IoU. The improved YOLOV8 is evaluated against initial YOLOV8 in control experiments, and the experimental results indicate that all the exactness measurements of YOLOV8 are enhanced in the detection and identification of hand gesture signal, and it can be proved that this paper's algorithm can enhance the exactness of YOLOV8. The experimental conclusions indicate that the improved YOLOV8 has enhanced all the accuracy measures in the recognition and detection of hand gesture signal, and it can be demonstrated that the improved YOLOV8 suggested in this paper has certain effect in the recognition and detection of hand gesture signal. © 2025 SPIE.

关键词： Palmprint recognition

来源：评论

学校读者我要写书评

暂无评论

Simulation and implementation of deep learning based expression recognition

Simulation and implementation of deep learning based express...

引用

2024 International Conference on Mechatronic Engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Wang, Shiyi Chen, Jiayu School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

Facial expressions are a challenging area in image recognition, and ensuring the universality of the model in the presence of a large sample size is a difficult point in such research. This article conducts repeatability experiments on two extreme scenarios: high-quality small samples and low-quality large samples, to explore solutions. Through transfer learning, data augmentation, and substantive evaluation of confusion matrices, the article demonstrates the accurate classification of seven facial expressions and the design of a 4-class model with high universality, which is presented in the form of an APP. © 2025 SPIE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Real-Time Detection of Surface Floating Garbage Based on Improved YOLOv7 16th

Real-Time Detection of Surface Floating Garbage Based on Imp...

引用

16th International Conference on Intelligent robotics and Applications, ICIRA 2023

作者： Li, Liang Li, Yiping Jiang, Zhibin Wang, Hailin State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China Key Laboratory of Marine Robotics Liaoning Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China

ISBN: (纸本)9789819964796

Cleaning up surface floating garbage is of great significance to improve the ecological environment of the waters, of which the detection of surface floating garbage is a crucial step. It is difficult to distinguish the background and garbage and solve the problem of small targets of the existing target detection methods. Therefore, an improved YOLOv7 network is proposed to achieve efficient and high-precision detection of surface floating garbage. The newly proposed DS method is applied to original YOLOv7 network for multi-scale feature fusion. Meanwhile, CBAM attention mechanism module is applied to network backbone to improve detection precision. The public data set FLOW is used to validate the method in this paper. The simulation test results show that the proposed method significantly improves the detection precision without reducing too much detection speed, which proves the effectiveness of the proposed method. Finally, the proposed method is deployed to an NVIDIA Jetson device and applied to real-time surface floating garbage detection, which proves its practicability. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

MRI Super-Resolution via Hybrid Information Enhancement Network based on Multi-Attention and Adaptive Convolution

MRI Super-Resolution via Hybrid Information Enhancement Netw...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Ma, Jixin Yu, Hongjian Du, Zhijiang Hua, Xin Li, Zibo Zhao, Hui Harbin Institute of Technology State Key Laboratory of Robotics and System Harbin China The 6th Medical Center Chinese Pla General Hospital Senior Department of Otolaryngology Head and Neck Surgery Beijing China

ISBN: (纸本)9798350386226

Deep learning-based super-resolution (SR) reconstruction is a critical approach which is used to generate high-resolution images from corresponding low-resolution images. However, the CNN-based methods are ineffective in capturing global information, whereas the Transformer-based methods have limited ability to model long-range dependencies caused by window self-attention. Besides, it is a challenging task to recover lost high-frequency information from downsampled images. In this paper, a Hybrid Information Enhanced Network (HIEN) is proposed for MRI super-resolution task. Specifically, we propose a Spatial-Channel Hybrid Attention (SCHA) to enhance specific semantics representation by combining spatial and channel self-attention together. To recover more high-frequency components, we propose a Dynamic High-Frequency Pass Filter (DHPF) to preserve high-frequency information adaptively in pixel-wise. The results of our extensive experiments indicate that HIEN outperforms other state-of-the-art methods. © 2024 IEEE.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

HSV-enhanced canny-hough algorithm for robust lane line recognition

HSV-enhanced canny-hough algorithm for robust lane line reco...

引用

2024 International Conference on Mechatronic Engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Wang, Shiyi Xu, Xiong School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

Lane line recognition plays an important role in road safety and intelligent transport system. This paper focuses on the key technology of lane line recognition, which is divided into three parts: firstly, the original image is preprocessed, including de-distortion and grey scale, to reduce noise and interference. Then, lane line information extraction is explored in terms of colour features and edge detection. By analysing the RGB and HSV colour modes and their conversion, the colour features are used to improve the accuracy and robustness of recognition. Meanwhile, the lane line edges are extracted by combining the improved Canny algorithm and commonly used edge detection operators, and the methods of threshold segmentation and Hough transform are applied to further describe the geometric features of lane lines. The direction of vehicle motion is determined by calculating the polar coordinate equation and vanishing point coordinates. Finally, this paper verifies the lane line recognition algorithm based on HSV-improved Canny-Hough through simulation experiments, and proves that it is more effective and accurate in lane line recognition compared with traditional methods. © 2025 SPIE.

关键词： Edge detection

来源：评论

学校读者我要写书评

暂无评论

Self-Regulating Pre-Tensioning Parameter Strategy for Control of Cable-Driven Knee Exoskeleton

Self-Regulating Pre-Tensioning Parameter Strategy for Contro...

引用

Chinese Automation Congress (CAC)

作者： Jiaqi Han Weihao Yin Zhibo Jing Jianda Han Juanjuan Zhang Tianjin Key Laboratory of Intelligent Robotics Institute of Robotics and Automatic Information System Nankai University Tianjin China

ISBN: (数字)9798350368604

ISBN: (纸本)9798350368611

Knee exoskeletons can offer additional strength and stability for individuals with mobility impairments by providing proper assistance. However, the precise timing of assistance torque for the knee joint poses a challenge to the control of cable-driven exoskeleton systems. Conventional torque-position switching control with a pre-tensioning strategy often fails to meet the precision requirements. In this paper, we developed a self-regulating pre-tensioning parameter strategy to enhance the control performance of cable-driven knee exoskeletons. A specific knee exoskeleton capable of providing both extension and flexion assistance was selected for study. Through geometric analysis of the human-exoskeleton interaction system, a more accurate kinematic relationship has been established, and a state variable named “tightness” has been introduced to describe the cable state. During exoskeleton-assisted walking, the pre-tensioning parameter is iteratively regulated based on tightness, aiming to bring the cable to a critically-tensioned state at the moment of heel strike. Preliminary experimental results demonstrate the strategy significantly improves control precision, the root-mean-square-error of torque tracking with this strategy are reduced by 88.7% and 77.4% for extension and flexion assistance, respectively, compared to scenarios without pre-tensioning. This enhanced control strategy shows promise for further advancing the capabilities of exoskeletons in facilitating human locomotion.

关键词： Knee Legged locomotion Torque Accuracy Exoskeletons Process control Switches Control systems Stability analysis Timing

来源：评论

学校读者我要写书评

暂无评论

Fruit classification and quality detection system based on improved YOLOv8 model

Fruit classification and quality detection system based on i...

引用

2024 International Conference on Mechatronic Engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Tang, Xiaoyu Huang, Yukun School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

With the expansion of fruit planting scale, the inefficiency of manual labor has gradually become a bottleneck limiting the development of the industry. In recent years, with the advancement of technology, research into modern intelligent agricultural technologies has also become a research hotspot. This paper is based on the YOLOv8 model for fruit classification and quality detection. First, image data enhancement is performed using operations such as power law transformation, and random operations like translation, flipping, and scaling are applied to augment the image dataset. A dataset is then established. Next, EfficientViT is introduced to replace the backbone network, and the SA attention mechanism is incorporated into the detection head, making the model more accurate in fruit classification and quality detection. Finally, a GUI (Graphical User Interface) visualization design is implemented to achieve a user-friendly interaction and is applied in the overall system design. © 2025 SPIE.

关键词： Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

Query Attack by Multi-Identity Surrogates

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第2期5卷 684-697页

作者： Chen, Sizhe Huang, Zhehao Tao, Qinghua Huang, Xiaolin Department of Automation and Institute of Medical Robotics Shanghai Jiao Tong University Shanghai200240 China MOE Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Shanghai200240 China ESAT-STADIUS Katholieke Universiteit Leuven Leuven3000 Belgium

Deep neural networks (DNNs) are acknowledged as vulnerable to adversarial attacks while the existing black-box attacks require extensive queries on the victim DNN to achieve high success rates. For query-efficiency, surrogate models of the victim are used to generate transferable adversarial examples (AEs) because of their gradient similarity (GS), i.e., surrogates' attack gradients are similar to the victim's ones. However, it is generally neglected to exploit their similarity on outputs, namely the prediction similarity (PS), to filter out inefficient queries by surrogates without querying the victim. To jointly utilize and also optimize surrogates' GS and PS, we develop QueryNet, a unified attack framework that can significantly reduce queries. QueryNet creatively attacks by multi-identity surrogates, i.e., crafts several AEs for one sample by different surrogates and also uses surrogates to decide on the most promising AE for the query. After that, the victim's query feedback is accumulated to optimize not only surrogates' parameters but also their architectures, enhancing both the GS and the PS. Although QueryNet has no access to pretrained surrogates' prior, it reduces queries by averagely about an order of magnitude compared to alternatives within an acceptable time, according to our comprehensive experiments: 11 victims (including two commercial models) on MNIST/CIFAR10/ImageNet, allowing only 8-b image queries, and no access to the victim's training data. The code is available at https://***/Sizhe-Chen/QueryNet. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on face tracking method based on improved CAMSHIFT and Kalman filter

Research on face tracking method based on improved CAMSHIFT ...

引用

2024 International Conference on Mechatronic Engineering and Artificial Intelligence, MEAI 2024

作者： Qin, Lijuan Tang, Xiaoyu Gao, Rongjun School of Information Science and Engineering Shenyang Ligong University Shenyang China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Science Shenyang China

ISBN: (纸本)9781510689176

With the rapid development of information technology, biometric recognition technologies have become a research hotspot in the field of pattern recognition due to their security, reliability, and effectiveness. Face detection and tracking technologies have become an active exploration topic for researchers both domestically and internationally, as they combine intelligence with everyday life. This paper employs a skin color model-based method for face detection, while for face tracking, an improved combination of the Kalman filter algorithm and the Camshift algorithm is used. The aim is to construct an efficient model system for face tracking. The movement or rapid changes in the face do not interrupt the tracking process, which enhances the real-time performance and robustness of face tracking. © 2025 SPIE.

关键词： Biometrics

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：