检索结果-内蒙古大学图书馆

16th International Conference on Computer and Automation Engineering (ICCAE)

作者： Kumar, Vijeth Murugesan, Malathi Veneri, Giacomo Baker Hughes Bangalore India Baker Hughes Florence Italy

ISBN: (纸本)9798350370058;9798350370164

We propose a computer vision architecture based on Hyperbolic networks, contrastive learning and knowledge distillation to detect unsafe behavior in energy production and oil & gas plants. Data scarcity poses a significant challenge to develop machine learning applications in industry. Indeed, the data may be incomplete, inconsistent, or biased, making it difficult to develop accurate and reliable models. Insufficient data during training phase has direct impact on the models' representation learning capabilities;with the aid of vision Transformers (ViTs), we are able to solve data crunch situations by learning efficient representations of the existing data. We harnessed the power of ViTs, as it incorporates more global information, leading to quantitatively stronger intermediate feature representations. Further, we approached the task with contrastive learning and obtained pairs of samples which are similar, to tackle the limited data availability in our industrial use case. The proposed approach by applying hyperbolic embeddings helps in extracting complex relationships in the data. Furthermore, the size of the model makes it suitable for devices with low computational capabilities such as unmanned robots.

关键词： Hyperbolic networks Contrastive Learning and Knowledge Distillation image processing image Understanding vision Transformers (ViTs) Data Scarcity

来源：评论

学校读者我要写书评

暂无评论

machine learning models and methods aspects of processing unstructured data 1

Machine learning models and methods aspects of processing un...

引用

1st International Workshop on Bioinformatics and Applied Information Technologies, BAIT 2024

作者： Bryk, Oleksandr Mudryk, Ivan Holubovskyi, Mykhailo Stoianov, Yurii Ternopil Ivan Puluj National Technical University 56 Ruska str. Ternopil46001 Ukraine

The ever-increasing amount of unstructured data, including text, images, audio, and video, poses a serious challenge to traditional data mining techniques. machine learning (ML) offers powerful tools and techniques to unlock the valuable insights hidden in this vast amount of information. This article explores the role of machine learning models and methods in processing unstructured data. We delve into key aspects of unstructured data processing, including data cleaning, feature development, and model selection. We explore specific ML techniques developed for different types of data, such as natural language processing (NLP) for text analysis and computer vision for image recognition. The paper also discusses the challenges and considerations involved in building and deploying ML models to handle unstructured data[1]. By understanding the capabilities of ML on unstructured data, organizations can gain a competitive advantage by deriving valuable insights for various applications. This information can range from understanding customer sentiment in social media posts to detecting anomalies in sensor data for predictive maintenance. © 2024 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

Instance-Dependent Noisy Label Learning via Graphical Modelling 23

Instance-Dependent Noisy Label Learning via Graphical Modell...

引用

23rd IEEE/CVF Winter Conference on applications of Computer vision (WACV)

作者： Garg, Arpit Cuong Nguyen Felix, Rafael Thanh-Toan Do Carneiro, Gustavo Univ Adelaide Australian Inst Machine Learning Adelaide SA Australia Monash Univ Dept Data Sci & AI Clayton Vic Australia Univ Surrey Ctr Vis Speech & Signal Proc Guildford Surrey England

ISBN: (纸本)9781665493468

Noisy labels are unavoidable yet troublesome in the ecosystem of deep learning because models can easily overfit them. There are many types of label noise, such as symmetric, asymmetric and instance-dependent noise (IDN), with IDN being the only type that depends on image information. Such dependence on image information makes IDN a critical type of label noise to study, given that labelling mistakes are caused in large part by insufficient or ambiguous information about the visual classes present in images. Aiming to provide an effective technique to address IDN, we present a new graphical modelling approach called InstanceGM, that combines discriminative and generative models. The main contributions of InstanceGM are: i) the use of the continuous Bernoulli distribution to train the generative model, offering significant training advantages, and ii) the exploration of a state-of-the-art noisy-label discriminative classifier to generate clean labels from instance-dependent noisy-label samples. InstanceGM is competitive with current noisy-label learning approaches, particularly in IDN benchmarks using synthetic and real-world datasets, where our method shows better accuracy than the competitors in most experiments(1).

关键词： Training Deep learning Visualization Computer vision Biological system modeling Ecosystems Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Future Iris Imaging with Advanced Fuzzified Histogram Equalization

引用

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2024年第2期15卷 877-884页

作者： Mashudi, Nurul Amirah Ahmad, Norulhusna Dziyauddin, Rudzidatul Akmam Noor, Norliza Mohd Univ Teknol Malaysia Razak Fac Technol & Informat Kuala Lumpur Malaysia Univ Teknol Malaysia Wireless Commun Ctr Kuala Lumpur Malaysia

images captured under low lighting frequently exhibit low brightness, low contrast, and a small grayscale. These features can affect the individual's view and severely limit the performance of machine vision systems, particularly when data annotation is involved. Hence, the issues motivate this study to examine the effectiveness of advanced fuzzified histogram equalization for image enhancement. A comparative study was conducted based on the low lighting condition of iris images to evaluate three image enhancement methods: Advanced Fuzzified Histogram Equalization (AFHE), Contrast Limited Adaptive Histogram Equalization (CLAHE), and Fuzzy Contrast Enhancement (FCE) using the MIREIS dataset. The Gaussian membership functions (GMF) were modified accordingly to satisfy the suitable pixel intensity of the input iris images. The results were compared using the peak signal-to-noise ratio (PSNR) value, including the central processing unit (CPU) times. As a result, the AFHE showed a better PSNR value at 76.02db with faster CPU times at 4.04s compared to CLAHE and FCE. Although the PSNR value of HE is slightly lower than CLAHE (0.3%) and FCE (0.7%), AFHE improved the image's quality and brightness, which can help other researchers with the data annotation process. The performance of the proposed methods was validated by comparing them with state-of-the-art methods. The results demonstrated that AFHE, CLAHE, and FCE exceeded other HE, AHE, CLAHE, and hybrid HE using fuzzy approaches that employed PSNR metrics.

关键词： image enhancement fuzzy logic histogram equalization CLAHE iris recognition

来源：评论

学校读者我要写书评

暂无评论

Integration of image segmentation and fuzzy theory to improve the accuracy of damage detection areas in traffic accidents

引用

JOURNAL OF BIG DATA 2021年第1期8卷 1-17页

作者： Amirfakhrian, Majid Parhizkar, Mahboub Univ Calgary Calgary AB Canada Islamic Azad Univ Cent Tehran Branch Dept Math Tehran Iran

In the next decade, machine vision technology will have an enormous impact on industrial works because of the latest technological advances in this field. These advances are so significant that the use of this technology is now essential. machine vision is the process of using a wide range of technologies and methods in providing automated inspections in an industrial setting based on imaging, process control, and robot guidance. One of the applications of machine vision is to diagnose traffic accidents. Moreover, car vision is utilized for detecting the amount of damage to vehicles during traffic accidents. In this article, using image processing and machine learning techniques, a new method is presented to improve the accuracy of detecting damaged areas in traffic accidents. Evaluating the proposed method and comparing it with previous works showed that the proposed method is more accurate in identifying damaged areas and it has a shorter execution time.

关键词： machine vision image processing image segmentation Dimensional reduction Crash damage detection

来源：评论

学校读者我要写书评

暂无评论

Empirical Comparison of Four Stereoscopic Depth Sensing Cameras for Robotics applications

引用

IEEE ACCESS 2025年 13卷 67564-67577页

作者： Rustler, Lukas Volprecht, Vojtech Hoffmann, Matej Czech Tech Univ Fac Elect Engn Dept Cybernet Prague 12135 Czech Republic

Depth sensing is an essential technology in robotics and many other fields. Many depth sensing (or RGB-D) cameras are available on the market and selecting the best one for your application can be challenging. In this work, we tested four stereoscopic RGB-D cameras that sense the distance by using two images from slightly different views. We empirically compared four cameras (Intel RealSense D435, Intel RealSense D455, StereoLabs ZED 2, and Luxonis OAK-D Pro) in three scenarios: (i) planar surface perception, (ii) plastic doll perception, (iii) household object perception (YCB dataset). We recorded and evaluated more than 3,000 RGB-D frames for each camera. For table-top robotics scenarios with distance to objects up to one meter, the best performance is provided by the D435 camera that is able to perceive with an error under 1 cm in all of the tested scenarios. For longer distances, the other three models perform better, making them more suitable for some mobile robotics applications. OAK-D Pro additionally offers integrated AI modules (e.g., object and human keypoint detection). ZED 2 is overall the best camera which is able to keep the error under 3 cm even at 4 meters. However, it is not a standalone device and requires a computer with a GPU for depth data acquisition. All data (more than 12,000 RGB-D frames) are publicly available at https://***/rgbd-comparison

关键词： Cameras Robot vision systems Robot sensing systems Sensors Robots Graphics processing units Accuracy Stereo image processing Three-dimensional printing Plastics Depth sensing Intel RealSense Luxonis OAK-D Pro ZED 2 object detection RGB-D segmentation

来源：评论

学校读者我要写书评

暂无评论

Survey of automated crack detection methods for asphalt and concrete structures

引用

INNOVATIVE INFRASTRUCTURE SOLUTIONS 2024年第11期9卷 1-36页

作者： Khlifati, Oumaima Baba, Khadija Tayeh, Bassam A. Mohammed V Univ Mohammadia Engn Sch Civil Engn & Environm Lab LGCE Rabat Morocco Islamic Univ Gaza Fac Engn Civil Engn Dept POB 108 Gaza Strip Palestine

Crack detection in civil infrastructure, including roads, bridges, and buildings, is crucial for maintaining structural safety and functionality. Traditional manual inspection methods are time-consuming and prone to errors, highlighting the need for automated solutions. This study evaluates state-of-the-art computer vision techniques for automatically detecting cracks in both asphalt and concrete surfaces from 2013 to 2024. The study assesses the effectiveness and limitations of image processing, traditional machine learning, and deep learning methods for crack detection. A comparative analysis of commonly used models is presented, utilizing public datasets: SDNET2018, CCIC, and BCD for concrete images, and AigleRN, CFD, CRACK500, and GAPs for asphalt images. Based on the comparison results, advanced deep learning models such as YOLOv5 and U-Net have demonstrated superior performance in crack detection for both asphalt and concrete structures, significantly outperforming traditional methods. For concrete crack detection, YOLOv5l exhibited exceptional performance on the SDNET2018 dataset, achieving a precision of 97.7%, recall of 96.7%, and a mAP@.5 of 99.3%, with a rapid inference time of 1.1 ms, making it highly suitable for real-time applications. For asphalt crack detection, U-Net achieved outstanding results, particularly on the GAPs dataset, with a near-perfect precision of 99.53%, and on the CFD dataset, with a precision of 92.54% and an F1-score of 89.90%. The study also highlights public concrete and asphalt datasets, providing details on methodology, including the number of images, image sizes, and noted noise factors. Additionally, it discusses the impact of data source variability on crack detection methods, showcasing the applications, strengths, and limitations of multi-sensor fusion techniques. Finally, unresolved challenges such as imbalanced datasets, high inference times, and complex network architectures are identified, with suggestions for future

关键词： Crack detection Asphalt Concrete image processing machine learning Deep learning Public datasets Data source variability Multi-sensor fusion

来源：评论

学校读者我要写书评

暂无评论

machine learning applications for early detection of esophageal cancer: a systematic review

引用

BMC MEDICAL INFORMATICS AND DECISION MAKING 2023年第1期23卷 1-17页

作者： Hosseini, Farhang Asadi, Farkhondeh Emami, Hassan Ebnali, Mahdi Shahid Beheshti Univ Med Sci Sch Allied Med Sci Dept Hlth Informat Technol & Management Tehran Iran Harvard Med Sch Dept Emergency Med Boston MA USA

IntroductionEsophageal cancer (EC) is a significant global health problem, with an estimated 7th highest incidence and 6th highest mortality rate. Timely diagnosis and treatment are critical for improving patients' outcomes, as over 40% of patients with EC are diagnosed after metastasis. Recent advances in machine learning (ML) techniques, particularly in computer vision, have demonstrated promising applications in medical image processing, assisting clinicians in making more accurate and faster diagnostic decisions. Given the significance of early detection of EC, this systematic review aims to summarize and discuss the current state of research on ML-based methods for the early detection of *** conducted a comprehensive systematic search of five databases (PubMed, Scopus, Web of Science, Wiley, and IEEE) using search terms such as "ML", "Deep Learning (DL (", "Neural Networks (NN)", "Esophagus", "EC" and "Early Detection". After applying inclusion and exclusion criteria, 31 articles were retained for full *** results of this review highlight the potential of ML-based methods in the early detection of EC. The average accuracy of the reviewed methods in the analysis of endoscopic and computed tomography (CT (images of the esophagus was over 89%, indicating a high impact on early detection of EC. Additionally, the highest percentage of clinical images used in the early detection of EC with the use of ML was related to white light imaging (WLI) images. Among all ML techniques, methods based on convolutional neural networks (CNN) achieved higher accuracy and sensitivity in the early detection of EC compared to other *** findings suggest that ML methods may improve accuracy in the early detection of EC, potentially supporting radiologists, endoscopists, and pathologists in diagnosis and treatment planning. However, the current literature is limited, and more studies are needed to investigate the clinical applications of these met

关键词： machine learning Deep learning Esophagus Esophageal Cancer Early detection

来源：评论

学校读者我要写书评

暂无评论

A machine vision Algorithm Approach for Angle Detection in Industrial applications 12th

A Machine Vision Algorithm Approach for Angle Detection in I...

引用

12th International Symposium on Intelligent Manufacturing and Service Systems, IMSS 2023

作者： Kayğusuz, Mehmet Öz, Barış Çelik, Ayberk Akgül, Yunus Emre Şimşek, Gözde Sarıgüzel, Ebru Gezgin Ar-Ge Merkezi Mamur Teknoloji Sistemleri Istanbul Turkey Graduate School of Science and Engineering Yildiz Technical University Istanbul Turkey Graduate School of Science and Engineering Kocaeli University Kocaeli İzmir Turkey

ISBN: (纸本)9789819960613

In automatic feeding systems, feeding of characteristic workpieces by mechanical tools causes accuracy and cost difficulties. For this reason, in systems where special workpieces are fed, image processing applications are necessary to obtain characteristic features of a product. In this study, a novel image processing algorithm is proposed for feeding a workpiece which has characteristic geometrical structures. The proposed algorithm is based on obtaining geometrical and rotational properties of the product and the gradient-based analysis as follows. The first step is to extract features from the shape of the workpiece, this step includes noise reduction, filtering, and edge detection operations. The gradient values of the edge information are used to create the angle-length vector pair in the second step. The workpiece rotation information is derived from length values indexed with angle information. The last step involves determination of the workpiece position in the 2D coordinate system. The coordinate information is used to determine the position of the gripper holder. The coordinates and angle are transmitted to the feed control. The proposed algorithm is applied on the 800 images that are collected from manufactured products. Rotation angle of the workpiece is determined by a tolerance of 1.5°. It is seen that results have sufficient accuracy for industrial applications. © 2024, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Design and simulation of autonomous military vehicle control system based on machine vision and ensemble movement approach

引用

JOURNAL OF SUPERCOMPUTING 2022年第15期78卷 17309-17347页

作者： Ahmadi, Kourosh Dadashtabar Rashidi, Ali Jabar Moghri, Ali Massomi Malek Ashtar Univ Technol Dept Elect & Comp Engn Tehran Iran

On the battlefield, early detection of armored vehicles can have a positive effect. Because according to this issue, timely and appropriate reactions can be done. The purpose of this study is to achieve the required algorithm in the vehicle control system by considering the car sensor vision, which is necessary to identify and determine the equipment needed to control the military drone based on car sensor vision. Today, the use of wireless networks, especially inter-vehicle wireless networks, in military applications is inevitable. Therefore, in the first step of this research, a new method has been proposed to control and steer unmanned vehicles based on car vision. In the proposed method, two 180-degree panoramic cameras with horizontal vision are used from the recorded images. The simulation results of the proposed method show increased accuracy and reduced implementation cost compared to using LIDAR and RADAR technologies. In the second step, a new approach is introduced to identify four common classes of armored vehicles (tanks, personnel carriers, firing tanks, and military vehicles) that are more likely to be present on battlefields. For this purpose, the latest image processing methods, which is deep learning, have been used. The results of the simulation of the proposed approach show the high accuracy of the proposed approach in detecting armored vehicles in a short time. In the third step of this research, a new method has been proposed to increase the connection of wireless networks. In the proposed method, queue theory is used and the results of the simulation of the proposed method show the high efficiency of the method. As a result, accurate and fast detection with unique features makes the users of the system superior.

关键词： Inter-vehicular wireless network Military applications machine vision Detection Connectivity

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：