检索结果-内蒙古大学图书馆

Parallel Architecture for Low Latency UAV Detection and Tracking Using Robotic Telescopes

IEEE TRANSACTIONS ON AEROSPACE AND ELECTRONIC SYSTEMS 2024年第4期60卷 5515-5524页

作者： Ojdanic, Denis Naverschnigg, Christopher Sinn, Andreas Zelinskyi, Daniil Schitter, Georg TU Wien Automat & Control Inst ACIN A-1040 Vienna Austria

This article presents the implementation of a multithreaded parallel architecture, which enables telescope-based optical unmanned aerial vehicle (UAV) detection and tracking in real time. For efficient image processing an accurate deep learning object detector is complemented in parallel by a fast object tracker. A transition strategy between detector and tracker is introduced based on the tracker reliability, which improves the object localization accuracy of the system. The deep learning algorithm initializes the tracker and in the subsequent frames the reliability of the tracker is compared to the confidence value of each newly detected object to determine whether a reinitialization is necessary. The implemented architecture successfully demonstrates the parallel combination of an FRCNN detector and a MEDIANFLOW tracker to achieve visual UAV detection and tracking at 100 fps. The proposed reliability-based strategy outperforms a purely detector and tracker-based strategy by 6% and 14%, respectively, in terms of intersection over union at a threshold of 0.5, in scenarios, when the target UAV is flying in front of a complex background. In addition, the implemented parallel architecture increases the probability for a flight path estimation, which requires at least two localizations, by 49%, when compared to a nonparallel architecture. Field tests are conducted with the proposed architecture using a telescope system demonstrating UAV detection and tracking at 100 fps in distances up to 4000 m in front of a clear background.

关键词： Detectors Autonomous aerial vehicles Reliability deep learning Radar tracking Telescopes Cameras detection parallel tracking unmanned aerial vehicle (UAV)

来源：评论

学校读者我要写书评

暂无评论

Efficient Multimodal Fusion for Hand Pose Estimation With Hourglass Network

引用

IEEE ACCESS 2024年 12卷 113810-113825页

作者： Hoang, Dinh-Cuong Xuan Tan, Phan Pham, Duc-Long Pham, Hai-Nam Bui, Son-Anh Nguyen, Chi-Minh Phi, An-Binh Tran, Khanh-Duong Trinh, Viet-Anh Tran, van-Duc Tran, Duc-Thanh Duong, van-Hiep Phan, Khanh-Toan Nguyen, van-Thiep Vu, van-Duc Nguyen, Thu-Uyen FPT Univ Greenwich Vietnam Hanoi 10000 Vietnam Shibaura Inst Technol Coll Engn Tokyo 1358548 Japan FPT Univ Dept Informat Technol Hanoi 10000 Vietnam

Hand pose estimation is vital for various applications, including virtual reality (VR), augmented reality (AR), gesture recognition, human-computer interaction (HCI), and robotics. Achieving accurate and real-time hand pose estimation is challenging due to factors such as the high degree of articulation in the human hand and the variability in hand shapes and sizes. While multimodal data offers advantages, developing a fast and resource-efficient hand pose estimation system remains challenging. Current state-of-the-art methods often require powerful graphics processing units (GPUs) for high performance, limiting deployment on edge platforms with limited computational resources. There is a critical need for higher efficiency without compromising accuracy, especially in real-world applications like mobile devices and embedded systems. Additionally, real-time performance is essential for practical applications, where systems must respond immediately to user interactions. Unfortunately, most current methods struggle to achieve real-time speeds, even on powerful GPUs, let alone on resource-constrained devices. To address these challenges, we propose an efficient hand pose estimation system that leverages both red-green-blue (RGB) and depth (RGBD) data through a unified fusion strategy. Our method combines appearance and geometric data early in the processing pipeline, significantly reducing computational complexity while maintaining real-time performance on resource-constrained devices. Experimental results show that the proposed model runs at over 110 fps on GPU, and 30 fps on the edge platform of NVidia Jetson NX Xavier, which is 4 to 5 times faster than existing methods, while achieving competitive accuracy.

关键词： Pose estimation robot vision systems intelligent systems deep learning supervised learning machine vision robot vision systems intelligent systems deep learning supervised learning machine vision

来源：评论

学校读者我要写书评

暂无评论

Implementing W-Net deep learning for terahertz image enhancement and segmentation

引用

ENGINEERING RESEARCH EXPRESS 2025年第1期7卷 015246-015246页

作者： Mondal, Shyamal Jampani, Kashyap Raj, R. Akshay Chowdhury, Dibakar Roy Sethi, Abhijit Def Inst Adv Technol Dept Appl Phys Pune 411025 India Anurag Univ Ghatkesar Hyderabad 500088 India

Terahertz images typically suffer from poor image quality and do not allow traditional machine learning methods to detect and identify objects. In this work, an image segmentation algorithm called W-Net is deployed for the first time, to segment concealed objects from the poor-quality Terahertz image samples. This involves developing an image enhancement and segmentation system for terahertz images. The proposed system comprises of two stages: in the first stage, the resolution of low resolution THz images has been enhanced by training a U-Net deep learning model. In the latter stage, another U-Net model is trained to segment the enhanced images using corresponding masks to identify regions of interest. The proposed system is evaluated using a dataset of low-resolution terahertz images with its respective high resolution images and masks. The system performed well with an accuracy, precision, receiver operating characteristics and F1 scores of 99.1%, 0.9143, 0.9977 and 0.9146, respectively for enhancement. Moreover, values of the same evaluation metrics deduced as 99.9%, 0.9851, 0.9999 and 0.9923, respectively for segmentation. Therefore, the demonstrated scheme is expected to identify and classify objects hidden in real-time terahertz images and improve the quality of the same to enable accurate segmentation for applications like security screening, biomedical applications, quality and health monitoring in food industry, etc.

关键词： terahertz image processing segmentation enhancement machine learning

来源：评论

学校读者我要写书评

暂无评论

Outdoor Navigation Assistive System Based on Robust and real-time Visual-Auditory Substitution Approach

引用

SENSORS 2024年第1期24卷 166-166页

作者： Scalvini, Florian Bordeau, Camille Ambard, Maxime Migniot, Cyrille Dubois, Julien Univ Bourgogne Lab ImViA EA 7535 F-21078 Dijon France Univ Bourgogne LEAD CNRS UMR 5022 F-21078 Dijon France

Blindness affects millions of people worldwide, leading to difficulties in daily travel and a loss of independence due to a lack of spatial information. This article proposes a new navigation aid to help people with severe blindness reach their destination. Blind people are guided by a short 3D spatialised sound that indicates the target point to follow. This sound is combined with other sonified information on potential obstacles in the vicinity. The proposed system is based on inertial sensors, GPS data, and the cartographic knowledge of pedestrian paths to define the trajectory. In addition, visual clues are used to refine the trajectory with ground floor information and obstacle information using a camera to provide 3D spatial information. The proposed method is based on a deep learning approach. The different neural networks used in this approach are evaluated on datasets that regroup navigations from pedestrians' point-of-view. This method achieves low latency and real-time processing without relying on remote connections, instead using a low-power embedded GPU target and a multithreaded approach for video processing, sound generation, and acquisition. This system could significantly improve the quality of life and autonomy of blind people, allowing them to reliably and efficiently navigate in their environment.

关键词： navigation aid wearable assistive device obstacle avoidance sensory substitution visual impairment deep learning

来源：评论

学校读者我要写书评

暂无评论

Multi-focus image fusion with visual state space model and dual adversarial learning

引用

COMPUTERS & ELECTRICAL ENGINEERING 2025年 123卷

作者： Xie, Xinzhe Guo, Buyu Li, Peiliang He, Shuangyan Zhou, Sangjun Zhejiang Univ Ocean Coll Zhoushan 316021 Peoples R China Zhejiang Univ Hainan Inst Sanya 572025 Peoples R China Donghai Lab Zhoushan 316021 Peoples R China

In recent years, the two-stage multi-focus image fusion (MFF) method, which utilizes neural networks to first generate decision maps and then calculate the fused image, has witnessed significant advancements. However, after supervised training, many networks become overly reliant on semantic information, making it challenging to discern whether homogeneous regions and flat regions are in focus or not, as these regions lack distinct blur cues. To alleviate this issue, this paper proposes a multi-focus image fusion network named BridgeMFF by applying a visual state space model and developing a general fine-tuning technique named BridgeTune, which bridges the semantic and texture gap via dual adversarial learning. By fine-tuning the entire network in an adversarial manner, decision maps are generated to synthesize clear and blurred images with probability density distributions closely approximating real ones, thereby implicitly learning local spatial patterns and statistical properties of pixel values. Extensive experiments demonstrate that the proposed BridgeMFF achieves superior fusion quality, especially in challenging homogeneous regions. Moreover, BridgeMFF has the smallest model size (0.05M) and fastest processing speed (0.09s per image pair), enabling real-time fusion applications. The codes are available at https://***/Xinzhe99/BridgeMFF.

关键词： deep learning Multi-focus image fusion Adversarial learning Fine-tuning technology Visual state space model

来源：评论

学校读者我要写书评

暂无评论

Residual Attention Augmented U-Shaped Network for One-Bit SAR image Restoration

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷 1-22页

作者： Guo, Li-Bo Dong, Yang-Yang Dong, Chunxi Xidian Univ Sch Elect Engn Xian 710071 Shaanxi Peoples R China

The application of one-bit sampling technology in synthetic aperture radar (SAR) systems has great potential due to its attractive advantages such as fast sampling speed, low data rate, high real-time performance, cheap hardware cost, and low energy consumption. However, one-bit sampling produces ghost targets in SAR imaging results and causes a significant reduction in the resolution and sharpness of SAR images, which is a challenge for one-bit SAR imaging. We develop a novel residual attention augmented U-shaped network (RAAUNet) with an encoder-and-decoder architecture, capable of learning the nonlinear mapping from one-bit SAR images to high-precision SAR images through end-to-end training. To enhance the efficiency of intermodule information communication at each level, our RAAUNet adopts three types of helpful skip connections that serve distinct roles in improving learning efficiency and convergence for the entire network, reducing information loss, and preserving spatial details during encoding processing, as well as transmitting multiresolution residual features. Furthermore, several specifically designed components are integrated into our network to improve its feature learning and perception abilities, where the attentive residual convolution module with the attention mechanism is employed in both encoders and decoders to endow them with the discriminative learning ability and enhance the nonlinear representation capacity, and the multiresolution fusion recovery module enriches contextual and spatial details by fusing multiresolution residual results, thereby improving the quality of the reconstructed SAR image. Numerical experiments on three synthetic one-bit SAR image datasets demonstrate that the RAAUNet achieves favorable performance against the state-of-the-art methods for one-bit SAR image restoration.

关键词： Attention mechanism deep learning (DL) multiresolution features one-bit sampling one-bit synthetic aperture radar (SAR) image restoration skip connection

来源：评论

学校读者我要写书评

暂无评论

Crack Detection and 3D Scene Creation of Road Bridge Using deep learning and image processing

Crack Detection and 3D Scene Creation of Road Bridge Using D...

引用

International Symposium on Soft Computing and Intelligent Systems (SCIS)

作者： Kotaro Ishida Yasutoshi Nomura Graduate school of science and engineering Ritsumeikan University Kusatsu Japan Civil and environmental engineering Ritsumeikan University Kusatsu Japan

ISBN: (数字)9798350373332

ISBN: (纸本)9798350373349

In Japan, the number of social infrastructure facilities, such as bridges and tunnels, which were constructed during a period of high economic growth and have been in existence for 50 years, has been increasing, and inspection is important for managing them. In 2014, it became obligatory to inspect structures once every 5 years. From the above decision, the inspection of structures by engineers using close-up visual observation and visible light images has been carried out periodically. However, damage inspection by visual inspection has become a problem from the viewpoint of cost such as human cost, the viewpoint of labor, and time cost. The periodical inspection is determined to utilize close-up visual observation by engineers or the equivalent diagnosis, and digitization and sharing of the inspection results are important in the management of infrastructure. Therefore, the construction of a system that carries out the diagnosis equivalent to close-up visual observation and digitizes the result is tried.

关键词： YOLO Visualization Three-dimensional displays Costs Accuracy Structural panels Transfer learning Training data Inspection Rendering (computer graphics)

来源：评论

学校读者我要写书评

暂无评论

A novel lightweight real-time traffic sign detection method based on an embedded device and YOLOv8

引用

JOURNAL OF real-time image processing 2024年第2期21卷 1-10页

作者： Luo, Yuechen Ci, Yusheng Jiang, Shixin Wei, Xiaoli Harbin Inst Technol Sch Transportat Sci & Engn Harbin Peoples R China Harbin Inst Technol Fac Comp Harbin Peoples R China

Traffic sign recognition, as one of the key steps of intelligent driving technologies, effectively avoids most traffic accidents by detecting the location and type of traffic signs in real time and providing the information to drivers or autonomous vehicles promptly. In addition, edge devices close to users has become an inevitable requirement for the development of IoT technology, real-time computing and the realization of network edge intelligence. Nowadays, the YOLO algorithm in object detection, developed to YOLOv8, is accompanied by various lightweight networks and lightweight methods to win by "fast", so this paper will propose an algorithm, fusing Ghost module and Efficient Multi-Scale Attention Module into YOLOv8, so that the model can improve the computing speed while maintaining the original characteristics. Furthermore, we choose Raspberry Pi as the object detection device due to its many characteristics such as lightweight, low power consumption. Through experiments, the model is trained on CCTSDB dataset, and the improved algorithm is tested on Raspberry Pi 4B. The results show that for three types of traffic signs, namely prohibited, indication and warning, the recognition accuracy mAP reaches 93.5% on the poor weather test set and 82.9% on the original test set, and the inference delay of Raspberry Pi reaches 0.79 s, which is effective in actual road test experiments. The improved model's accuracy has increased by 6.4% and 3.8% on two separate test sets compared to the original model, while the detection time has been reduced by 0.12 s. This research is of great significance to the maturation of assisted driving and autonomous driving technologies.

关键词： Traffic sign recognition GhostNet Efficient multi-scale attention Yolov8 Raspberry Pi deep learning

来源：评论

学校读者我要写书评

暂无评论

Ensemble-based deep learning model for welding defect detection and classification

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年 136卷

作者： Vasan, Vinod Sridharan, Naveen Venkatesh Balasundaram, Rebecca Jeyavadhanam Vaithiyanathan, Sugumaran Vellore Inst Technol Sch Mech Engn SMEC Chennai 600127 India York St John Univ Dept Comp Sci York YO31 7EX N Yorkshire England Lulea Univ Technol Div Operat & Maintenance Engn Lulea Sweden

This study introduces an ensemble-based deep learning approach for monitoring and detecting submerged arc weld defects in weld beads and adjacent zones during Non-Destructive Testing (NDT). The methodology utilizes an open-source image dataset to categorize submerged arc weld conditions into three defect types: cracks, lack of penetration and porosity, and ideal welds (no defects). To enhance image analysis, we employ preprocessing and feature extraction in both spatial and frequency domains using segmentation techniques like grey level difference method (GLDM), grey level co-occurrence matrix (GLCM), discrete wavelet transform (DWT), Fast Fourier Transform (FFT), and texture analysis. Following image preparation, the preprocessed data undergoes training and testing in our proposed ensemble-based deep learning model. The model 's performance is thoroughly assessed using key metrics such as precision, recall, F1 score, receiver operating characteristics (ROC) curve, and a confusion matrix, resulting in an impressive accuracy rate of 93.12% for detecting and classifying weld faults. In comparing our model to state-of-the-art techniques in existing literature, our ensemble-based deep learning approach consistently demonstrates a high fault detection and classification performance in the face of a simplistic MLP-based core architecture, ratifying the robustness of our feature ensemble approach. The study concludes that this model has significant potential for integration into current inspection systems, offering an efficient and robust solution for real-time condition monitoring of welds along weldment joints.

关键词： Feature ensemble deep neural network image segmentation Welding defects Automated defect identification Radiographic images Non-destructive testing

来源：评论

学校读者我要写书评

暂无评论

Model Inference Using TVM on AMD GPU: image Classification and LLM Case Studies 11

Model Inference Using TVM on AMD GPU: Image Classification a...

引用

11th International Conference on Electrical, Electronic and Computing Engineering, IcETRAN 2024

作者： Malbaša, Andrija Petrović, Nenad Arsić, Branko Syrmia LLC University of Niš Faculty of Electornic Engineering Serbia Syrmia LLC University of Kragujevac Faculty of Science Serbia

ISBN: (纸本)9798350386998

deep learning models have revolutionized the field of artificial intelligence by exhibiting exceptional capabilities in understanding complex patterns and making accurate predictions. However, the growing complexity of these models presents a challenge in achieving real-time inference, which is crucial for practical applications. As deep learning models become more popular, there is an increasing demand for faster inference times to enable real-time decision-making in critical scenarios, such as fire detection and classification. In recent years, the advancement in parallel computing hardware, such as graphics processing units (GPUs), specialized tensor processing units (TPUs) and application-specific integrated circuits (ASICs), has provided a significant boost to deep learning inference. One of the challenges in accelerating deep learning inference on diverse hardware platforms is the lack of optimized implementations of mathematical operations specifically tailored for each platform. This limitation can significantly hinder the performance of deep learning models on these platforms. In this paper, we focus on Advanced Micro Devices (AMD) GPU graphics cards and explore the utilization of the Apache Tensor Virtual Machine (TVM) compiler framework to accelerate deep learning inference for Generative Pre-trained Transformer 2 (GPT-2) large language model (LLM) and ResNet-18 image classification model, represented in PyTorch and Open Neural Network Exchange (ONNX) formats, respectively. © 2024 IEEE.

关键词： Application specific integrated circuits

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

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：