检索结果-内蒙古大学图书馆

Multiple-Object Tracking Using Histogram Stamp Extraction in CCTV Environments

computers, Materials & Continua 2023年第12期77卷 3619-3635页

作者： Ye-Yeon Kang Geon Park Hyun Yoo Kyungyong Chung Division of AI Computer Science and Engineering Kyonggi UniversitySuwon16227Korea Contents Convergence Software Research Institute Kyonggi UniversitySuwon16227Korea

Object tracking,an important technology in the field of image processing and computer vision,is used to continuously track a specific object or person in an *** technology may be effective in identifying the same person within one image,but it has limitations in handling multiple images owing to the difficulty in identifying whether the object appearing in other images is the *** tracking the same object using two or more images,there must be a way to determine that objects existing in different images are the same ***,this paper attempts to determine the same object present in different images using color information among the unique information of the ***,this study proposes a multiple-object-tracking method using histogram stamp extraction in closed-circuit television *** proposed method determines the presence or absence of a target object in an image by comparing the similarity between the image containing the target object and other *** this end,a unique color value of the target object is extracted based on its color distribution in the image using three methods:mean,mode,and interquartile *** Top-N accuracy method is used to analyze the accuracy of each method,and the results show that the mean method had an accuracy of 93.5%(Top-2).Furthermore,the positive prediction value experimental results show that the accuracy of the mean method was 65.7%.As a result of the analysis,it is possible to detect and track the same object present in different images using the unique color of the *** the results,it is possible to track the same object that can minimize manpower without using personal information when detecting objects in different *** the last response speed experiment,it was shown that when the mean was used,the color extraction of the object was possible in real time with 0.016954 *** this,it is possible to detect and track the same object in real time when using the proposed

关键词： Data mining deep learning object detection object tracking real-time object detection multiple object image processing

来源：评论

学校读者我要写书评

暂无评论

Defect Detection Model Using Time Series Data Augmentation and Transformation

引用

computers, Materials & Continua 2024年第2期78卷 1713-1730页

作者： Gyu-Il Kim Hyun Yoo Han-Jin Cho Kyungyong Chung Department of Computer Science Kyonggi UniversitySuwon-si16227Korea Contents Convergence Software Research Institue Kyonggi UniversitySuwon-si16227Korea Department of Energy IT Engineering Far East UniversityEumsung-gun27601Korea Division of AI Computer Science and Engineering Kyonggi UniversitySuwon-si16227Korea

Time-series data provide important information in many fields,and their processing and analysis have been the focus of much ***,detecting anomalies is very difficult due to data imbalance,temporal dependence,and ***,methodologies for data augmentation and conversion of time series data into images for analysis have been *** paper proposes a fault detection model that uses time series data augmentation and transformation to address the problems of data imbalance,temporal dependence,and robustness to *** method of data augmentation is set as the addition of *** involves adding Gaussian noise,with the noise level set to 0.002,to maximize the generalization performance of the *** addition,we use the Markov Transition Field(MTF)method to effectively visualize the dynamic transitions of the data while converting the time series data into *** enables the identification of patterns in time series data and assists in capturing the sequential dependencies of the *** anomaly detection,the PatchCore model is applied to show excellent performance,and the detected anomaly areas are represented as heat *** allows for the detection of anomalies,and by applying an anomaly map to the original image,it is possible to capture the areas where anomalies *** performance evaluation shows that both F1-score and Accuracy are high when time series data is converted to ***,when processed as images rather than as time series data,there was a significant reduction in both the size of the data and the training *** proposed method can provide an important springboard for research in the field of anomaly detection using time series ***,it helps solve problems such as analyzing complex patterns in data lightweight.

关键词： Defect detection time series deep learning data augmentation data transformation

来源：评论

学校读者我要写书评

暂无评论

Search and Recommendation Systems with Metadata Extensions 26

Search and Recommendation Systems with Metadata Extensions

引用

26th International Conference on Advanced Communications Technology, ICACT 2024

作者： Kim, Woo-Hyeon Kim, Joo-Chang Kyonggi University Division of Ai Computer Science and Computer Engineering Korea Republic of Contents Convergence Software Research Institute Kyonggi University Korea Republic of

ISBN: (纸本)9791188428120

This paper proposes an AI-based video metadata extension model to overcome the limitations of video search and recommendation systems in the multimedia industry. Current video searches and recommendations utilize pre-added metadata. Metadata includes filenames, keywords, tags, genres, etc. This makes it impossible to make direct predictions about the content of a video without pre-added metadata. These platforms also analyze your previous search history, viewing history, etc. to understand your interests in order to serve you personalized videos. This may not reflect the actual content and may raise privacy concerns. In addition, recommendation systems suffer from a cold start problem, which is the lack of an initial target, as well as a bubble effect. Therefore, this study proposes a search and recommendation system by expanding metadata in videos using techniques such as shot boundary detection, speech recognition, and text mining. The proposed method selects the main objects required by the recommendation system based on the object frequency and extracts the corresponding objects from the video frame by frame. In addition, we extract the speech from the video separately, convert the speech to text to extract the script and apply text mining techniques to the extracted script to quantify it. Then, we synchronize the object frequency and the transcript to create a single contextual data. After that, we group videos and clips based on the contextual data and index them. Finally, we utilize Shot Boundary Detection to segment videos based on their content. To ensure that the generated contextual data is appropriate for the video, the proposed model compares the extracted script with the video's subtitle data to check and calibrate its accuracy. The model can then be fine-tuned by tuning and cross-validating the hyperparameter to improve its performance. These models can be incorporated into a variety of content discovery and recommendation platforms. By using expanded

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

DrDiff: Drug Response Prediction Through Controllable Diffusion-GE and Graph Attention Network 27

DrDiff: Drug Response Prediction Through Controllable Diffus...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Choi, Seungyeon Seo, Samgmin Choi, Jonghwan Park, Chihyun Park, Sanghyun Department of Computer Science Yonsei University Seoul Korea Republic of Division of Software Hallym University Chuncheon Korea Republic of Department of Computer Science and Engineering Kangwon National University Chuncheon Korea Republic of

ISBN: (纸本)9781643685489

The accurate prediction of drug responses based on the genomic profile of a patient is essential to progress in the field of precision *** advent of various deep-learning algorithms based on publicly available large-scale omics datasets is the driving force behind research in this *** characteristics of biological datasets, characterized by high dimensions and low sample sizes, pose challenges of overfitting and limited generalization in prediction ***, constructing prediction models using biological data such as gene expression is further complicated by the need to account for the complex relationships among genes, which exacerbates the aforementioned *** address these challenges, we propose a drug response prediction framework (DrDiff) that integrates a denoising diffusion probabilistic model (DDPM) based data augmentation module with a graph attention network based drug response prediction *** proposed model showed a 10% higher AUC than the state-of-the-art models for drug response prediction for the six drugs considered in the study, suggesting the superior generalization performance of DrDiff over other baseline ***, we demonstrated the feasibility of generative models, which form one of the modules of the proposed framework, in overcoming the fundamental limitations of omics *** experiments bear out the feasibility of generative models, which form one of the modules of the proposed framework, in augmenting gene expression data. © 2024 The Authors.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Comparison of Missing Data Imputation Methods in Time Series Forecasting

引用

computers, Materials & Continua 2022年第1期70卷 767-779页

作者： Hyun Ahn Kyunghee Sun Kwanghoon Pio Kim Division of Software Convergence Hanshin UniversityGyeonggi18101Korea Contents Convergence Software Research Institute Kyonggi UniversityGyeonggi16227Korea Division of AI Computer Science and Engineering Kyonggi UniversityGyeonggi16227Korea

Time series forecasting has become an important aspect of data analysis and has many real-world ***,undesirable missing values are often encountered,which may adversely affect many forecasting *** this study,we evaluate and compare the effects of imputationmethods for estimating missing values in a time *** approach does not include a simulation to generate pseudo-missing data,but instead perform imputation on actual missing data and measure the performance of the forecasting model created *** an experiment,therefore,several time series forecasting models are trained using different training datasets prepared using each imputation ***,the performance of the imputation methods is evaluated by comparing the accuracy of the forecasting *** results obtained from a total of four experimental cases show that the k-nearest neighbor technique is the most effective in reconstructing missing data and contributes positively to time series forecasting compared with other imputation methods.

关键词： Missing data imputation method time series forecasting LSTM

来源：评论

学校读者我要写书评

暂无评论

Future Intelligent Autonomous Robots, Ethical by Design. Lessons Learned from Autonomous Cars Ethics 14

Future Intelligent Autonomous Robots, Ethical by Design. Les...

引用

14th International Conference on Society and Information Technologies, ICSIT 2023

作者： Dodig-Crnkovic, Gordana Holstein, Tobias Pelliccione, Patrizio Thavarasa, Jathoosh Division of Computer Science and Software Engineering Mälardalen University Västerås Sweden Department of Computer Science and Engineering Chalmers University of Technology Gothenburg Sweden Department of Information Engineering Computer Science and Mathematics University of l'Aquila L'Aquila Italy Software Engineering Efrei Paris Villejuif France

ISBN: (纸本)9781950492701

The "ethical by design" approach involves examining all stages of a lifecycle of technology to ensure that they are ethically justifiable and socially sustainable. Building on our work on the ethics of autonomous intelligent robocars, and studies of the literature on the ethics of robotics, we propose for robot applications a set of values and ethical principles including safety, security, privacy, transparency, and explainability, accountability, fairness, human control, well-being, autonomy and freedom, and sustainability. This may help stakeholders in the field of intelligent autonomous robotics to connect ethical principles with their applications. Most ethical considerations we identified in our work on autonomous cars are relevant to all AI-powered robots, but robots require additional examination depending on their application domain, such as social robots (care robots, personal companions, robots used in education, health care, elderly care, education, entertainment, chat-bots), industrial robots, etcetera. Thus, existing ethical frameworks need to be applied in a context-sensitive way, by assessments in interdisciplinary, multi-competent teams through multi-criteria analysis. Furthermore, we argue for the need for continuous development of ethical principles, guidelines, and regulations, informed by the progress of technologies and involving relevant stakeholders. This implies designing the socio-technical system as an intelligent learning ecology. © 2023 Proceedings ICSIT, International Conference on Society and Information Technologies. All rights reserved.

关键词： Intelligent robots

来源：评论

学校读者我要写书评

暂无评论

Optimizing auditory brainstem response detection through NSGA-II guided feature selection

引用

Neural Computing and Applications 2025年第15期37卷 8925-8943页

作者： Majidpour, Jafar Hassanzadeh, Hiwa Khezri, Edris Arabi, Hossein Department of Software Engineering University of Raparin Kurdistan Region Rania Iraq Department of Computer Science University of Raparin Kurdistan Region Rania Iraq Department of Computer Engineering Islamic Azad University Boukan Branch Boukan Iran Division of Nuclear Medicine and Molecular Imaging Department of Medical Imaging Geneva University Hospital Geneva1211 Switzerland

Hearing-impaired people undergo auditory brainstem response (ABR) testing to assess their peripheral auditory nerve system. Audiologists apply diagnostic labels to ABR data using reference-based indicators such as peak latency, waveform shape, amplitude, and others. ABR test scoring errors may invalidate auditory nerve system integrity results. Machine learning, especially deep learning, may reduce human error in ABR analysis. This work offers a complete methodology to handle ABR analysis problems. First, extract ABR test images from PDF reports to create a dataset. The next step uses Elastic Distortion approach to create high-quality ABR images while conserving data. ResNet-50 deep learning extracts ABR image characteristics in the third phase. The fourth step finds the most relevant features using a trainable multi-objective non-dominated sorting genetic algorithm II optimizer. Finally, ABR images are classified using machine learning algorithms. The proposed model achieved significant performance metrics: an accuracy of 99.46%, precision of 99.33%, specificity of 99.57%, sensitivity of 99.57%, F1-score of 99.43%, and a mean squared error of 0.0054 when using a support vector machine classifier with both augmented and original images. These findings demonstrate the effectiveness of trainable feature selection and high-quality ABR images in improving ABR analysis accuracy. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

Optimization Performance Analysis of Different Initialization Methods in Single-Objective and Multi-Objective Chaotic Evolution 13

Optimization Performance Analysis of Different Initializatio...

引用

Joint 13th International Conference on Soft Computing and Intelligent Systems and 25th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2024

作者： Zhang, Xiaoyi Pei, Yan Jiang, Huiyan Graduate School of Computer Science and Engineering University of Aizu Aizuwakamatsu965-8580 Japan University of Aizu Computer Science Division Aizuwakamatsu965-8580 Japan Northeastern University College of Software Shenyang110819 China

ISBN: (纸本)9798350373332

Chaotic Evolution (CE) is a significantly faster and more robust method for solving single-objective and multi-objective optimization problems. However, there are various factors that can impact the performance of CE, with population initialization being one of them. A well-executed initialization method can contribute to the discovery of better solutions and enhance both convergence rate and accuracy. This paper conducts a comparative experiment on CE algorithms using six different initialization methods. The performance of the CE algorithms utilizing these various initialization methods is analyzed and discussed. To assess convergence rate and accuracy, eight single-objective optimization benchmark functions and five multi-objective optimization benchmark functions are employed in our analysis. The evaluation results reveal significant disparities among the six initialization methods, indicating that different approaches to initialization have a substantial influence on the optimization performance of the CE algorithm. © 2024 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

Interference cancellation in RIS-assisted FD relay transmission 14

Interference cancellation in RIS-assisted FD relay transmiss...

引用

14th International Conference on Ubiquitous and Future Networks, ICUFN 2023

作者： Kim, Donghyeon Choi, Wooyeol Chun, Chanjun Kim, Yohan Kim, Yonggang Chosun University Department of Computer Engineering Korea Republic of Dongseo University Department of Software Korea Republic of Kongju National University Division of Computer Science and Engineering Korea Republic of

ISBN: (纸本)9798350335385

Reconfigurable intelligent surface (RIS) is one of the promising technology for the next-generation wireless networks. The RIS reflects the received signal with phase shift introduced by reflecting elements without analog to digital conversion. In relay transmission scenario, RIS could be exploited to enhance the network coverage and improve transmission performance. In this paper, we consider RIS-assisted full-duplex (FD) relay transmission scenario. In the considered scenario, RIS and FD relay node performs relay transmission. Although the beamformed signal could be generated by phase shift of RIS, the interference from RIS to FD relay node may degrade the throughput performance. The proposed method cancels both of interference from RIS and self-interference at the FD relay node by estimating interference channels in the preparation phase before data transmission. © 2023 IEEE.

关键词： Analog to digital conversion

来源：评论

学校读者我要写书评

暂无评论

SlowFast Based Real-Time Human Motion Recognition with Action Localization

引用

computer Systems science & engineering 2023年第11期47卷 2135-2152页

作者： Gyu-Il Kim Hyun Yoo Kyungyong Chung Department of Computer Science Kyonggi UniversitySuwon16227Korea Contents Convergence Software Research Institute Kyonggi UniversitySuwon16227Korea Division of AI Computer Science and Engineering Kyonggi UniversitySuwon16227Korea

Artificial intelligence is increasingly being applied in the field of video analysis,particularly in the area of public safety where video surveillance equipment such as closed-circuit television(CCTV)is used and automated analysis of video information is ***,various issues such as data size limitations and low processing speeds make real-time extraction of video data *** analysis technology applies object classification,detection,and relationship analysis to continuous 2D frame data,and the various meanings within the video are thus analyzed based on the extracted basic *** recognition is key in this *** recognition is a challenging field that analyzes human body movements,requiring the interpretation of complex movements of human joints and the relationships between various *** deep learning-based human skeleton detection algorithm is a representative motion recognition ***,motion analysis models such as the SlowFast network algorithm,have also been developed with excellent ***,these models do not operate properly in most wide-angle video environments outdoors,displaying low response speed,as expected from motion classification extraction in environments associated with high-resolution *** proposed method achieves high level of extraction and accuracy by improving SlowFast’s input data preprocessing and data structure *** input data are preprocessed through object tracking and background removal using YOLO and DeepSORT.A higher performance than that of a single model is achieved by improving the existing SlowFast’s data structure into a frame unit *** on the confusion matrix,accuracies of 70.16%and 70.74%were obtained for the existing SlowFast and proposed model,respectively,indicating a 0.58%increase in *** detection,based on behavioral classification,the existing SlowFast detected 2,341,164 cases,whereas the proposed model detected 3,11

关键词： Artificial intelligence convolutional neural network video analysis human action recognition skeleton extraction

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：