检索结果-内蒙古大学图书馆

Electronics and Renewable Systems (ICEARS), International Conference on

作者： Dharanaesh M V. Pushpalatha Yughendaran P Janarthanan S Dinesh A ECE Dept CEG Chennai India

Fatigue or drowsiness is a significant factor that contributes to the occurrence of terrible road accidents. Every day, the number of fatal injuries increases day by day. The paper introduces a novel experimental model that aims to reduce the frequency of accidents by detecting driver drowsiness while also recommending songs based on facial emotions. The existing models use more hardware than necessary, leading to more cost and also do not provide as much accuracy. The proposed system aims to enhance the overall experience by reducing both the computational time required to obtain results and the overall cost of the system. For that, this study has developed a real-time information processing system that captures the video from the car dash camera. Then, an object detection algorithm will be employed to extract multiple facial parts from each frame using a pre-trained deep learning model from image processing libraries like OpenCV. Then, there is MobileNetV2, a lightweight convolutional neural network model that performs transfer learning by freezing feature extraction layers and creating custom dense layers for facial emotion classification, with output labels of happiness, sadness, fear, anger, surprise, disgust, sleepiness, and neutral. The driver's face will be identified using directional analysis from multiple facial parts in a single frame to carry out drowsiness detection to avoid accidents. Then, according to the emotion predicted by multiple users, the application will fetch a playlist of songs from Spotify through a Spotify wrapper and recommend the songs by displaying them on the car's dash screen. Finally, the model will be optimized using various optimization techniques to run on low-latency embedded devices.

关键词： deep learning Computational modeling Face recognition Transfer learning Feature extraction Low latency communication Recommender systems

来源：评论

学校读者我要写书评

暂无评论

deep learning-Based Smart Parking Management System and Business Model 1

引用

5th International Conference on Computer Vision and image processing, CVIP 2020

作者： Kher, Yatharth Saxena, Aditya Tamizharasan, P.S. Joshi, Amit D. Pilani Dubai Campus Dubai United Arab Emirates Department of Computer Engineering and IT College of Engineering Pune Pune MS India

ISBN: (数字)9789811611032

ISBN: (纸本)9789811611025

In this fast-developing world, the increase in the number of vehicles demands a smart parking system in smart cities. The issue of spending a lot of time finding parking slots needs to be addressed. The increase of smartphones provides the space to develop smart applications enabled with AI and deep learning. This paper proposes an AI-based smart parking management system and a business model to provide a solution for both user and the owner of the parking space. Owners of the parking slots can opt for fixed or variable timeslots to make use of their parking spaces. Registered users can check the availability of the parking spaces at the destination in real-time and details of the users such as the time and vehicle details can be detected and updated automatically. Billing for the parking space usage will also be done automatically as per the regulated guidelines. Raspberry Pi and deep learning tools are used for the implementation. The proposed system is cost-effective and reduces time and energy. © 2021, Springer Nature Singapore Pte Ltd.

关键词： Android (operating system)

来源：评论

学校读者我要写书评

暂无评论

Tool Wear Segmentation in Blanking Processes with Fully Convolutional Networks based Digital image processing

arXiv

引用

arXiv 2023年

作者： Schlegel, Clemens Molitor, Dirk Alexander Kubik, Christian Martin, Daniel Michael Groche, Peter Institute for Production Engineering and Forming Machines Technical University of Darmstadt Germany

The extend of tool wear significantly affects blanking processes and has a decisive impact on product quality and productivity. For this reason, numerous scientists have addressed their research to wear monitoring systems in order to identify or even predict critical wear at an early stage. Existing approaches are mainly based on indirect monitoring using time series, which are used to detect critical wear states via thresholds or machine learning models. Nevertheless, differentiation between types of wear phenomena affecting the tool during blanking as well as quantification of worn surfaces is still limited in practice. While time series data provides partial insights into wear occurrence and evolution, direct monitoring techniques utilizing image data offer a more comprehensive perspective and increased robustness when dealing with varying process parameters. However, acquiring and processing this data in real-time is challenging. In particular, high dynamics combined with increasing strokes rates as well as the high dimensionality of image data have so far prevented the development of direct image-based monitoring systems. For this reason, this paper demonstrates how high-resolution images of tools at 600 spm can be captured and subsequently processed using semantic segmentation deep learning algorithms, more precisely Fully Convolutional Networks (FCN). 125,000 images of the tool are taken from successive strokes, and microscope images are captured to investigate the worn surfaces. Based on findings from the microscope images, selected images are labeled pixel by pixel according to their wear condition and used to train a FCN (U-Net). Copyright © 2023, The Authors. All rights reserved.

关键词： Wear of materials

来源：评论

学校读者我要写书评

暂无评论

级联混合模型引导的实时柑橘采摘点定位方法

引用

中国图象图形学报 2024年第10期29卷 3130-3143页

作者：梁云刘云帆林毅申姜伟鹏黄梓帆华南农业大学数学与信息学院广州510642

目的柑橘是我国最常见的水果之一,目前多以人工采摘为主,成本高、效率低等问题严重制约规模化生产,因此柑橘自动采摘成为近年的研究热点。但是,柑橘生长环境复杂、枝条形态各异、枝叶和果实互遮挡严重,如何精准实时地定位采摘点成为自... 详细信息

目的柑橘是我国最常见的水果之一,目前多以人工采摘为主,成本高、效率低等问题严重制约规模化生产,因此柑橘自动采摘成为近年的研究热点。但是,柑橘生长环境复杂、枝条形态各异、枝叶和果实互遮挡严重,如何精准实时地定位采摘点成为自动采摘的关键。通过构建级联混合网络模型,提出了一种通用且高效的柑橘采摘点自动精准定位方法。方法构建团簇框生成模型和枝条稀疏实例分割模型,对两者进行级联混合实现实时柑橘采摘点定位。首先,构建柑橘果实检测网络,提出团簇框生成模型,该模型通过特征提取、果实检测框生成和DBSCAN(density-based spatial clustering of applications with noise)果实密度聚类,实时地生成图像内果实数目最多的团簇框坐标;然后,提出融合亮度先验的枝条稀疏分割模型,该模型以团簇框内的图像作为输入,有效降低背景枝条的干扰,通过融合亮度先验的稀疏实例激活图,实时地分割出与果实相连接枝条实例;最后基于分割结果搜索果实采摘点定位坐标。结果经过长时间户外采集制作了柑橘果实检测数据集CFDD(citrus fruit detection dataset)和柑橘枝条分割数据集CBSD(citrus branch segmentation dataset)。这两个数据集由成熟果实、未成熟果实组成,包含晴天、阴天、顺光和逆光等挑战,总共37000幅图像。在该数据集上本文方法的采摘点定位精准度达到了95.77%,帧率(frames per second,FPS)达到了28.21帧/s。结论本文方法在果实采摘点定位方面取得较好进展,能够快速且准确地获取柑橘采摘点,并且提供配套的机械臂采摘设备可供该采摘点定位算法的落地使用,为柑橘产业发展提供有力支持。

关键词：采摘机器人采摘点定位方法团簇框生成器亮度先验枝条稀疏分割模型

来源：评论

学校读者我要写书评

暂无评论

FPGA-Based Inter-layer Pipelined Accelerators for Filter-Wise Weight-Balanced Sparse Fully Convolutional Networks with Overlapped Tiling

引用

JOURNAL OF SIGNAL processing SYSTEMS FOR SIGNAL image AND VIDEO TECHNOLOGY 2021年第5期93卷 499-512页

作者： Shimoda, Masayuki Sada, Youki Nakahara, Hiroki Tokyo Inst Technol Tokyo Japan

Convolutional neural networks (CNNs) exhibit state-of-the-art performance while performing computer-vision tasks. CNNs require high-speed, low-power, and high-accuracy hardware for various scenarios, such as edge environments. However, the number of weights is so large that embedded systems cannot store them owing to their limited on-chip memory. A different method is used to minimize the input image size, for real-time processing, but it causes a considerable drop in accuracy. Although pruned sparse CNNs and special accelerators are proposed, the requirement of random access incurs a large number of wide multiplexers for a high degree of parallelism, which becomes more complicated and unsuitable for FPGA implementation. To address this problem, we propose filter-wise pruning with distillation and block RAM (BRAM)-based zero-weight skipping accelerator. It eliminates weights such that each filter has the same number of nonzero weights, performing retraining with distillation, while retaining comparable accuracy. Further, filter-wise pruning enables our accelerator to exploit inter-filter parallelism, where a processing block for a layer executes filters concurrently, with a straightforward architecture. We also propose an overlapped tiling algorithm, where tiles are extracted with overlap to prevent both accuracy degradation and high utilization of BRAMs storing high-resolution images. Our evaluation using semantic-segmentation tasks showed a 1.8 times speedup and 18.0 times increase in power efficiency of our FPGA design compared with a desktop GPU. Additionally, compared with the conventional FPGA implementation, the speedup and accuracy improvement were 1.09 times and 6.6 points, respectively. Therefore, our approach is useful for FPGA implementation and exhibits considerable accuracy for applications in embedded systems.

关键词： FPGA deep learning Fully convolutional network Pruning

来源：评论

学校读者我要写书评

暂无评论

Combining CNN deep learning and SVM for Vehicle Classification

Combining CNN Deep Learning and SVM for Vehicle Classificati...

引用

2021 International Conference on Computer, Control, Informatics and Its Applications - learning Experience: Raising and Leveraging the Digital Technologies During the COVID-19 Pandemic, IC3INA

作者： Khusni, Uus Insani, Asep Center for Data and Documentation National Research and Innovation Agency Jakarta Indonesia Instrumentation Development Center National Research and Innovation Agency Bandung Indonesia

ISBN: (纸本)9781450385244

Vehicle classification plays an important role in an intelligent transportation system. The development of research in image processing, pattern recognition, and deep learning has encouraged more research interest in categorizing vehicles using deep learning technology. In everyday life, Automatic vehicle classification can be found in vehicle search, real-time traffic monitoring, automatic toll gates, and others. The use of vehicle classifications as a vehicle search problem is commonly used by a police officer in the case of car robbery crimes. The use of a human expert, in this case, is completely accurate but requires time and money. In this case, automatic categorization of a vehicle using deep learning technology can be one solution. Vehicles can be categorized into four categories: Minivans, Pickup Trucks, Sedans, and SUV. The method used in this research is to use a combination method of Convolution Neural Networks (CNN) as a feature extractor with Support Vector Machine (SVM) as a classifier to increase accuracy. The deep learning architecture used is CNN (AlexNet, VGG16, and Resnet50). The dataset used in this research is the AI Standford Car Dataset which consists of 16,185 images and 196 vehicle classes. From 196 classes, four classes were selected (Minivan, Pickup Truck, Sedan, and SUV). The accuracy obtained is AlexNet + SVM 38.50 %, VGG16 + SVM 82.93 % and ResNet50 + SVM 90.87 %. © 2021 ACM.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Lane departure warning systems and lane line detection methods based on image processing and semantic segmentation:A review

引用

Journal of Traffic and Transportation Engineering(English Edition) 2020年第6期7卷 748-774页

作者： Weiwei Chen Weixing Wang Kevin Wang Zhaoying Li Huan Li Sheng Liu School of Information Engineering Chang’an UniversityXi'an 710064China Xi’an Aeronautical Polytechnic Institute Xi'an 710089China Royal Institute of Technology Stockholm 10044Sweden Audible Inc. NewarkNJ 07102USA School of Computer Science and Engineering Xi’an University of TechnologyXi'an 710048China

Recently,the development and application of lane line departure warning systems have been in the *** any of the systems,the key part of lane line tracking,lane line identification,or lane line departure warning is whether it can accurately and quickly detect lane *** 1990 s,they have been studied and implemented for the situations defined by the good viewing conditions and the clear lane markings on *** then,the accuracy for particular situations,the robustness for a wide range of scenarios,time efficiency and integration into higher-order tasks define visual lane line detection and tracking as a continuing research *** present,these kinds of lane marking line detection methods based on machine vision and image processing can be divided into two categories:the traditional image processing and semantic segmentation(includes deep learning)*** former mainly involves feature-based and model-based steps,and which can be classified into similarity-and discontinuity-based ones;and the model-based step includes different parametric straight line,curve or pattern *** semantic segmentation includes different machine learning,neural network and deep learning methods,which is the new trend for the research and application of lane line departure warning *** paper describes and analyzes the lane line departure warning systems,image processing algorithms and semantic segmentation methods for lane line detection.

关键词： Traffic engineering Lane departure warning Lane line detection image processing image analysis Semantic segmentation

来源：评论

学校读者我要写书评

暂无评论

Extraction of laser stripe centerlines from translucent optical components using a multi-scale attention deep neural network

引用

MEASUREMENT SCIENCE AND TECHNOLOGY 2024年第8期35卷 085404-085404页

作者： Jiang, Hao Fu, Weijie Zhang, Xinming Zhang, Chaoxia Foshan Univ Sch Mechatron Engn & Automat Foshan 528001 Peoples R China Changchun Univ Technol Sch Mech & Elect Engn Changchun 130022 Peoples R China Minist Educ Key Lab Cross Scale Micro & NanoMfg Changchun 130022 Peoples R China

The precise extraction of laser stripe centerlines is critical for line-laser 3D scanning systems. However, conventional methods relying on threshold segmentation and morphological operations face significant challenges when confronted with pervasive optical phenomena, including specular reflection, scattering, and bleeding, which are commonly observed in translucent optical components. These methods typically require complex preprocessing procedures and often yield poor precision in centerline extraction. In this paper, we introduce a novel learning-based approach, complemented by a meticulously curated dataset, explicitly designed to address these challenges. Our proposed method leverages a multi-scale attention U-Net-like architecture, initially tasked with the segmentation of laser stripes from the complex background environment. Subsequently, it employs the Steger algorithm for the precise extraction of laser stripe centerlines. The experimental results, obtained by comprehensively evaluating real-world captured images, clearly demonstrate the effectiveness of our deep neural network combined with the Steger algorithm. This combined approach exhibits exceptional accuracy even when challenged by the interferences from specular reflection, scattering, and bleeding artifacts. Specifically, our method achieves a mean intersection over union (mIoU) of 84.71% for the laser stripe detection task, accompanied by a mean square error (MSE) of 10.371 pixels. Also, the average execution time for the centerline extraction task is notably efficient at 0.125 s.

关键词： optical metrology centerline extraction structured light image processing

来源：评论

学校读者我要写书评

暂无评论

deep learning and IoT Based Robotics to Monitor the Traffic 2nd

Deep Learning and IoT Based Robotics to Monitor the Traffic

引用

2nd International Conference on Cyber Warfare, Security and Space Computing, SpacSec 2024

作者： Priya, V. Vishwa Pande, Soumitra S. Ilyas, Md Jayasudha, R. Ramesh, Janjhyam Venkata Naga Suganthi, D. Department of Computer Science Vels Institute of Science Technology and Advanced Studies Chennai India Ericsson Global India Private Limited Bangalore India Prestige Institute of Engineering Management and Research Indore India Dr. N. G. P. Institute of Technology Coimbatore India Koneru Lakshmaiah Education Foundation Andhra Pradesh Guntur Vaddeswaram India Saveetha College of Liberal Arts and Sciences SIMATS Thandalam Chennai India

ISBN: (纸本)9783031734939

In densely populated areas, managing traffic laws has become more difficult due to the rising number of cars on the road. One of the most critical components of effective traffic management and improved mobility is real-time traffic monitoring systems. Consequently, precise and dependable real-time traffic information has always been essential for cars and drivers. Many other approaches have been suggested as potential answers to the issues plaguing traffic conditions recently. A different approach is VCC or vehicle cloud computing. In addition, an advanced model called an Iot-aided robotic (IoRT), which uses cameras and Internet of Things (IoT) detector nodes to gather actual traffic information, has been created. Two deep learning methods, one based on improved LeNet-5 for real-time signs for traffic identification and the other on the Inception-V3 model for detecting and identifying traffic lights, are the critical contributions provided by this research effort. Additionally, to decrease road accidents, the ideal distance between barriers and ultrasonic sensors was determined by analyzing the timing and speed of ultrasonic waves. Data acquired by cameras and sensors is analyzed using a variety of image processing algorithms before being uploaded to the cloud and made accessible to commuters and drivers via a mobile app. According to the test findings, the suggested models are far more accurate. The expanded GTSRB (EGTSRB) dataset achieved 98.89% accuracy while the German Traffic Sign Recognition Benchmark (GTSRB) dataset achieved 98.23% accuracy using the modified LeNet-5. The second model achieved a precision of 99.7 percent using data from the Laboratory for the Intelligent and Safe Automobiles (LISA). The results of this study beat previous research on comparable traffic monitoring systems by 2.13% for traffic light recognition and detection and by 4.89% for traffic sign identification. © The Author(s), under exclusive license to Springer Nature Switzerland AG

关键词： Traffic signs

来源：评论

学校读者我要写书评

暂无评论

Bio-Inspired Optimization Algorithm on Cloud based image Retrieval System using deep Features

Bio-Inspired Optimization Algorithm on Cloud based Image Ret...

引用

2022 International Conference on Augmented Intelligence and Sustainable Systems, ICAISS 2022

作者： Vijayakumar, T. Ramalakshmi, K. Priyadharsini, C. Vasanthakumar, S. Shaina Sharma, Abhishek Sri Eshwar College of Engineering Department of Computer Science and Business Systems Coimbatore India Alliance College of Engineering and Design Department of Computer Science and Engineering Karnataka Benglure India Karunya Institute of Technology and Sciences Department of Computer Science and Engineering Coimbatore India Karpagam Academy of Higher Education Department of Computer Science Coimbatore India GNA University Department of Computer Science and Engineering Punjab Phagwara144401 India Shri Vaishnav Vidyapeeth Vishwavidyalaya Madhya Pradesh Indore India

ISBN: (纸本)9781665489621

With the tremendous improvement of cloud computing (CC) technique, increasing count of users choosing to outsource image data to the clouds. Cloud resource provider decreases the storing problem on local hardware by outsourcing massive image database to cloud server and exploiting the cloud computation ability to image processing application. learning effective feature representation and similarity measures are essential for the performance of content-based image retrieval (CBIR). Despite wide-ranging research efforts for decades, it remains a challenge that significantly hinders the success of real-time CBIR system. Since the trail and error hyperparameter selection is a tedious process, metaheuristic optimization can be used. This paper introduces an Evolutionary Optimization Algorithm for Cloud Based image Retrieval System (EOA-CIRS) technique. The presented EOA-CIRS technique derives a new CBIR model for retrieving highly related images. To obtain this, the EOA-CIRS technique initially employs MobileNetv3 model as a feature extractor, which derives features from the query and database images. Next, the bamboo forest growth optimizer (BFGO) approach was applied as a hyperparameter optimizer to appropriately tune the hyperparameters of the MobileNetv3 approach. Finally, Euclidean distance based similarity measurement is utilized for retrieving the related images. The experimental validation of the EOA-CIRS model on Corel10K dataset demonstrates the effectual efficacy of the EOA-CIRS model over other recent approaches. © 2022 IEEE.

关键词： Search engines

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：