检索结果-内蒙古大学图书馆

Nighttime object detection system with lightweight deep network for internet of vehicles

JOURNAL OF real-time image processing 2021年第4期18卷 1141-1155页

作者： Jhong, Sin-Ye Chen, Yung-Yao Hsia, Chih-Hsien Lin, Shih-Chang Hsu, Kuo-Hua Lai, Chin-Feng Natl Cheng Kung Univ Dept Engn Sci Tainan Taiwan Natl Taiwan Univ Sci & Technol Dept Elect & Comp Engn Taipei Taiwan Natl Taipei Univ Technol Grad Inst Automat Technol Taipei Taiwan

Autonomous driving systems in internet of vehicles (IoV) applications usually adopt a cloud computing mode. In these systems, information got at the edge of the cloud computing center for data analysis and situation response. However, the conventional IoV face enormous challenges to meet the requirements in terms of storage, communication, and computing problems because of the considerable amount of information on the traffic environment. The environment perception during the nighttime is poorer than that during the daytime that this problem also requires addressing. To solve these problems, we propose a nighttime object detection scheme based on a lightweight deep learning model in the edge computing mode. First, the pedestrian detection and the vehicle detection algorithm that using the thermal images based on the YOLO architecture. We can implement the model on edge devices that can achieve real-time detection through the designed lightweight strategy. Next, a spatial prior information and temporal prior information into the detection algorithm and divide the frames into key and non-key frames to increase the performance and speed of the system simultaneously. Finally, we implemented the detection network for performance and feasibility verification on the Jetson TX2 edge device. The experimental results show that the proposed system can achieve real-time and high-accuracy object detection on edge devices.

关键词： Internet of vehicles Thermal infrared images Object detection deep neural network Prior information

来源：评论

学校读者我要写书评

暂无评论

An efficient model for real-time wildfire detection in complex scenarios based on multi-head attention mechanism

引用

JOURNAL OF real-time image processing 2023年第4期20卷 66-66页

作者： Wang, Xiaotian Pan, Zhongjie Gao, Hang He, Ningxin Gao, Tiegang Nankai Univ Coll Software Tianjin 300071 Peoples R China Tsinghua Univ Dept Engn Phys Beijing 100084 Peoples R China

Wildfire is a common natural disaster that destroys the ecological environment and endangers the lives and property of people. The average precision and speed of wildfire detection algorithms are critical in complex wildfire detection scenarios. However, the complex wildfire scenarios and multi-scale flaming objects decrease the detection accuracy of the model. How to design an accurate and fast wildfire detection model is a challenging study. To improve the average detection precision of the model while reducing its computational cost as much as possible, we propose an efficient model for real-time wildfire detection in complex scenarios called FireDetn. First, we use four different detection heads to help FireDetn detect different size flame objects. Then, we integrate Transformer Encoder blocks with multi-head attention in FireDetn to enhance its ability to capture global feature information and contextual information, thus improving its average precision in complex scenarios. Finally, we fuse the spatial pyramid pooling fast structure, which benefits in detecting multi-scale flame objects at a lower computational cost. Our experimental results show that the FireDetn can detect about 112 frames per second, meeting the requirement of real-time detection. We also compare FireDetn with other algorithms, demonstrating that FireDetn obtains satisfactory results in average precision and speed.

关键词： Wildfire detection Object detection deep learning YOLO Transformer

来源：评论

学校读者我要写书评

暂无评论

Multi-dimensional data modelling of video image action recognition and motion capture in deep learning framework

引用

IET image processing 2020年第7期14卷 1257-1264页

作者： Gao, Peijun Zhao, Dan Chen, Xuanang Jiujiang Univ Sch Phys Culture Jiujiang 332005 Peoples R China Jiujiang Maternal & Child Hlth Hosp Jiujiang 322000 Peoples R China

In order to improve the accuracy of small-range human motion recognition in video and the computational efficiency of large-scale data sets, a multi-dimensional data model of motion recognition and motion capture in video image based on deep-learning framework was proposed. First, the moving foreground of the target is extracted by the Gauss mixture model, and the human body is recognised by the gradient histogram. At the second level, the dense trajectory feature and the deep learning feature are fused, according to the integration of global encoding algorithm and convolutional neural network. In the deep learning feature, the fusion of the deep video feature and the video RGB tricolour feature is taken as the feature of deep learning. Finally, the classification is based on the deep learning network model. The simulation experiments based on large-scale real data sets and small-scale gesture data sets show that the algorithm has high recognition accuracy for large-scale data sets and small-scale gesture actions. In addition, Imperial Computer Vision & learning Lab human behaviour data set is used to classify the experimental data. The average classification accuracy is 85.79%. The algorithm can run at a speed of about 20 frames per second.

关键词： feature extraction computer vision image classification image motion analysis learning (artificial intelligence) gesture recognition video signal processing image colour analysis mixture models image capture convolutional neural nets motion capture deep-learning framework Gauss mixture model human body dense trajectory feature deep learning feature deep video feature video RGB tricolour feature deep learning network model small-scale gesture data sets high recognition accuracy large-scale data sets small-scale gesture actions Imperial Computer Vision & learning Lab human behaviour data set multidimensional data modelling video image action recognition deep learning framework small-range human motion recognition gradient histogram global encoding algorithm convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Scale Efficient Training for Large Datasets

arXiv

引用

arXiv 2025年

作者： Zhou, Qing Gao, Junyu Wang, Qi Northwestern Polytechnical University China

The rapid growth of dataset scales has been a key driver in advancing deep learning research. However, as dataset scale increases, the training process becomes increasingly inefficient due to the presence of low-value samples, including excessive redundant samples, overly challenging samples, and inefficient easy samples that contribute little to model improvement. To address this challenge, we propose Scale Efficient Training (SeTa) for large datasets, a dynamic sample pruning approach that losslessly reduces training time. To remove low-value samples, SeTa first performs random pruning to eliminate redundant samples, then clusters the remaining samples according to their learning difficulty measured by loss. Building upon this clustering, a sliding window strategy is employed to progressively remove both overly challenging and inefficient easy clusters following an easy-to-hard curriculum. We conduct extensive experiments on large-scale synthetic datasets, including ToCa, SS1M, and ST+MJ, each containing over 3 million samples. SeTa reduces training costs by up to 50% while maintaining or improving performance, with minimal degradation even at 70% cost reduction. Furthermore, experiments on various scale real datasets across various backbones (CNNs, Transformers, and Mambas) and diverse tasks (instruction tuning, multi-view stereo, geo-localization, composed image retrieval, referring image segmentation) demonstrate the powerful effectiveness and universality of our approach. Code is available at https://***/mrazhou/SeTa. © 2025, CC BY-NC-SA.

关键词： Stereo image processing

来源：评论

学校读者我要写书评

暂无评论

Hybrid deep learning and machine learning approach for passive image forensic

引用

IET image processing 2020年第10期14卷 1952-1959页

作者： Thakur, Abhishek Jindal, Neeru Thapar Inst Engn & Technol Elect & Commun Engn Dept Patiala Punjab India

image forgery detection using traditional algorithms takes much time to find forgeries. The new emerging methods for the detection of image forgery use a deep neural network algorithm. A hybrid deep learning (DL) and machine learning-based approach is used in this study for passive image forgery detection. A DL algorithm classifies images into the forged and not forged categories, whereas colour illumination localises forgery. The simulated results are compared to other algorithms on public datasets. The simulated results achieved 99% accuracy for CASIA1.0, 98% accuracy for CASIA2.0, 98% accuracy for BSDS300, 97% accuracy for DVMM, and 99% accuracy for CMFD image manipulation dataset.

关键词： Gabor filters image classification iris recognition feature extraction learning (artificial intelligence) neural nets image forensics image segmentation image coding machine learning-based approach passive image forgery detection DL algorithm forged forged categories not forged categories CMFD image manipulation dataset machine learning approach passive image forensic forgeries emerging methods deep neural network algorithm

来源：评论

学校读者我要写书评

暂无评论

WB LUTS: CONTRASTIVE learning FOR WHITE BALANCING LOOKUP TABLES

arXiv

引用

arXiv 2024年

作者： Manne, Sai Kumar Reddy Wan, Michael Roux Institute Northeastern University United States Institute for Experiential AI Northeastern University United States

Automatic white balancing (AWB), one of the first steps in an integrated signal processing (ISP) pipeline, aims to correct the color cast induced by the scene illuminant. An incorrect white balance (WB) setting or AWB failure can lead to an undesired blue or red tint in the rendered sRGB image. To address this, recent methods pose the post-capture WB correction problem as an image-to-image translation task and train deep neural networks to learn the necessary color adjustments at a lower resolution. These low resolution outputs are post-processed to generate high resolution WB corrected images, forming a bottleneck in the end-to-end run time. In this paper we present a 3D Lookup Table (LUT) based WB correction model called WB LUTs that can generate high resolution outputs in real time. We introduce a contrastive learning framework with a novel hard sample mining strategy, which improves the WB correction quality of baseline 3D LUTs by 25.5%. Experimental results demonstrate that the proposed WB LUTs perform competitively against state-of-the-art models on two benchmark datasets while being 300× faster using 12.7× less memory. Our model and code are available at https://***/skrmanne/3DLUT_sRGB_WB1 © 2024, CC BY.

关键词： Table lookup

来源：评论

学校读者我要写书评

暂无评论

deep learning for real-time Leaf Disease Detection: Revolutionizing Apple Orchard Health

Deep Learning for Real-Time Leaf Disease Detection: Revoluti...

引用

Big Data Analytics and Practices (IBDAP), International Conference on

作者： Maherun Nessa Isty Nishat Vasker Fahmida Anjum Warda Ruhin Parsub Amit Roy Akibul Haque Mahamudul Hasan Maheen Islam Department of Computer Science and Engineering East West University Dhaka Bangladesh

deep learning approach, particularly integrated neural architecture for real-time Single Shot multibox detector (INAR-SSD), is the most effective approach in the context of fast detection and classification of apple leaf diseases in apple orchards. Advanced data augmentation and complex image annotation techniques are using in the development of apple leaf diseases detection dataset (ALDD), which is the main purpose of this study. Apple leaf diseases detection dataset plays a vital role in the field of identifying apple leaf disease which is the key challenge of this study. The novel INAR-SSD model, created by the combination of SSD architecture with an integrated neural architecture, is proposed in this research. Utilizing the various functionalities of GoogLeNet, Inception Structure and Rainbow concentration the innovative INAR-SSD model is trained to identify various types of apple leaf diseases that effect on apple leaves. To categorize the diseases precisely, the INAR-SSD model follows digital image processing such as noise removing, overfitting reducing, feature extraction etc. Another purpose of study is to report the findings from the experiments that evaluate the performance of the proposed model. The INAR-SSD model successfully detects the illnesses in apple leaves with an accuracy of 97%, evidenced by its mean Average Precision(mAP), and the processing speed of 23.13 Frames Per Second (FPS).

关键词：

来源：评论

学校读者我要写书评

暂无评论

A TinyML based Residual Binarized Neural Network for real-time image Classification

A TinyML based Residual Binarized Neural Network for real-ti...

引用

International conference of Electronics, Communication and Aerospace Technology (ICECA)

作者： C Srinivasan P Sridhar V Hari Priya S Swathi Department of ECE Saveetha Engineering College Chennai Tamilnadu India Department of ECE Sri Ramakrishna Engineering College Coimbatore Tamilnadu India

ISBN: (纸本)9781665482721

image processing is an important requirement in IoT applications such as robotics, augmented reality, computer vision, Industry 4.0 etc. The capabilities of IoT devices for image processing are limited to sensing the environment, processing and communicating the results. Tiny machine learning (TinyML) is a new paradigm that takes advantage of the IoT to deploy deep learning models to perform complex tasks in resource constrained embedded devices. image classification is an important task in IoT to interpret images of a particular scene or class. Currently, this task is performed in embedded devices using Binarized Neural Networks (BNNs), which can be converted to a set of weights using a one-hot encoding process. These networks integrated with hardware accelerators can be trained to perform image processing tasks in real-time. This paper proposes a BNN for image classification based on residual learning paradigm, called Tiny-BNN which exploits the skip connections to reduce information loss, and improve the training time and accuracy. Experimental results show that the model achieves a classification accuracy of 90.1 % and 91.6% on the on CIFAR-10 and MNIST datasets respectively.

关键词： Performance evaluation deep learning Computer vision Computational modeling Neural networks Robot sensing systems real-time systems

来源：评论

学校读者我要写书评

暂无评论

Enhancing IoT-Edge Computation with Data Forwarding based Decentralized deep Neural Networks

Enhancing IoT-Edge Computation with Data Forwarding based De...

引用

International Conference on image Information processing

作者： Bmg Prasad Rachappa Jopate Pankaj Savita A Basi Reddy B Prabu Shankar M S Arunkumar Department of CSE Ravindra College of Engineering for Women(RCEW Autonomous) Kurnool Andhra Pradesh India Department of Information Technology University of Technology And Applied Sciences - Al Mussanah Al Muladha Muscat Sultanate of Oman Department of CSE Sagar Institute of Science & Technology Bhopal Organization-Sagar Institute of Science & Technology Bhopal Madhya Pradesh India Department of Computer Science and Engineering School of Computing Mohan Babu University Tirupati Andhra Pradesh India Department of Computer Science and Engineering Alliance College of Engineering and Design Alliance University Banglore Karnataka India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Chennai Tamilnadu India

ISBN: (数字)9798350371406

ISBN: (纸本)9798350371413

The incorporation of edge computing has become essential in the big data and Internet of Things age to tackle issues related to latency, real-time processing, and data scale. This paper delves into the imperative need for edge computing within the context of big data environments. Distributed deep learning methodologies are explored, encompassing a diverse array of deep learning models. Particularly, a novel paradigm named Data Forwarding based Decentralized deep Neural Network (DF-DDNN) is introduced to achieve low latency IoT-Edge computation. In order to reduce latency in Io T ecosystems, the DF-DDNN model makes use of edge devices’ processing capabilities. By strategically distributing data processing and computation at the edge, this model seeks to overcome inherent limitations of conventional Io T architectures, with a keen focus on network, processing, and overall latencies. The research provides a thorough performance evaluation that compares the DF-DDNN model’s performance versus conventional IoT models in a number of areas. The outcomes demonstrate how much better the suggested DF-DDNN model is than traditional IoT systems. Notably, improvements are evident in terms of reduced network latency, accelerated processing speeds, and overall latency mitigation. This research underscores the critical importance of embracing edge computing in the context of big data and deep learning applications. The DF-DDNN model emerges as a promising avenue for advancing IoT-Edge computation, effectively addressing latency concerns and augmenting the performance of IoT-enabled systems.

关键词： deep learning Performance evaluation Computational modeling Biological system modeling Artificial neural networks Big Data Data models

来源：评论

学校读者我要写书评

暂无评论

Active Noise Control of Airborne Road Noise Based on Artificial Intelligent Computed Road Classification and Adaptive Digital Filter 25th

Active Noise Control of Airborne Road Noise Based on Artific...

引用

25th International Conference on Human-Computer Interaction, HCII 2023

作者： Lee, Sang-Kwon Okcu, Orhun An, Kanghyun Acoustics and Vibration Signal Processing Laboratory Inha University Incheon22212 Korea Republic of

ISBN: (纸本)9783031360039

The study presents a new method for real time noise control based on road surface classification technology using deep learning and active noise cancellation technology using digital adaptive filter. Recently active noise control (ANC) method for structure vibration induced road noise is developed and applied to real car for road noise control. This method is effective for ANC of road noise due to structural vibration of car body but is not effective for ANC of road noise due to air borne noise source such as interaction noise between tire and road surface. Acceleration signals measured on the car body have been used for ANC of structure borne noise. However, a new reference signal is required for ANC of air borne noise. Interaction noise between tire and road surface is severe on the concrete road than the other road such as asphalt. The reference signal should be correlated to interaction noise between tire and road surface. In the paper, the type of road surface is classified in real time throughout deep learning of the measured interaction noise based on convolutional neural network (CNN). The reference signal of interaction noise on concrete road was obtained by calculating instantaneous frequency of interaction noise and was successfully applied to ANC of airborne induced road noise. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Convolution

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：