检索结果-内蒙古大学图书馆

Generative Adversarial Networks Based Framework for Music Genre Classification

SN computer science 2024年第8期5卷 1-16页

作者： Dwivedi, Pulkit Islam, Benazir School of Computer Science and Engineering IILM University Greater Noida India New Jersey Institute of Technology Newark United States

Music genre classification plays a crucial role in organizing and exploring large music collections, enabling personalized music recommendations, and enhancing music-related services. This paper presents a novel approach to music genre classification using Generative Adversarial Networks (GANs), Fourier Transform, and Wavelet Transform. The main objective is to leverage the power of GANs to extract discriminative features from audio data and accurately classify music into different genres. The proposed methodology involves two key components: the generator and the discriminator. The generator generates synthetic audio samples that resemble real music, while the discriminator learns to distinguish between real and synthetic audio samples. By training the GAN on a diverse dataset of music samples from various genres, the discriminator becomes proficient in recognizing genre-specific features. To enhance classification accuracy, Fourier Transform and Wavelet Transform are applied to extract both frequency and time-domain features from the audio data. Additionally, classifiers such as support vector machines and neural networks are employed to effectively distinguish between different music genres. The experimental results demonstrate the effectiveness of the proposed approach across multiple datasets. The method achieves 98.97% accuracy on the GTZAN dataset, 92.47% accuracy on the FMA-Small dataset, and 92.98% accuracy on the ISMIR Genre dataset, significantly outperforming traditional classification methods These results highlight the power of GANs, Fourier Transform, and Wavelet Transform in enhancing the accuracy and robustness of music genre classification. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Classification models Deep learning Feature extraction Generative adversarial networks (GANs) Music genre classification

来源：评论

学校读者我要写书评

暂无评论

GSD-YOLO: a Lightweight Framework for Traffic Sign Detection

IAENG International Journal of Computer Science

引用

IAENG International Journal of computer science 2025年第6期52卷 1815-1823页

作者： Jianbang, Ding Tianwei, Shi Zhehao, Zhang Ruiqi, Wang School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan114051 China School of Computer Science and Software Engineering University of Science and Technology Liaoning Anshan China

To address the challenges of detecting small traffic signs, low detection accuracy, complex detection network models, and large parameter counts, this paper proposes a lightweight traffic sign detection framework called GSD-YOLO. This framework integrates the GCSconv module, SENetV2 module, and DySample module. Specifically, the Slim-net architecture, formed by combining the GCSconv and VOVGSCP modules, is employed to replace the Conv and C2f modules in the original YOLOv8 neck network. This substitution not only reduces the number of parameters and computational load during training but also achieves a lightweight design. The lightweight SENetV2 module is incorporated into the backbone of the detection model, further lightening the backbone. In the neck network, the DySample module replaces the original upsampling module, enhancing the model’s upsampling capability while reducing the parameter count. Compared to the original YOLOv8 detection network, Grad-CAM visualization shows that traffic signs in the detected regions appear darker and more concentrated. On the CCTSDB dataset, the proposed framework achieves an mAP@0.5 of 95.5%, which is 2.3% higher than the original YOLOv8 (93.3%). The model size is reduced to 4.2MB, which is 32.2% smaller than the original YOLOv8. The GFLOPs are reduced by 35%, and the overall parameter count is decreased by 40%. The lightweight detection framework proposed in this paper effectively reduces model size, simplifies model complexity, decreases parameter count, and improves detection accuracy, achieving significant improvements in the field of traffic sign detection. © (2025), (International Association of Engineers). All rights reserved.

关键词： Road and street markings

来源：评论

学校读者我要写书评

暂无评论

Influencing Factors Analysis of Occupational Values of Undergraduates in IT Major and Countermeasures 23

Influencing Factors Analysis of Occupational Values of Under...

引用

7th International Conference on Digital technology in Education, ICDTE 2023

作者： Chen, Haowen Zhang, Yuanbin School of Computer Science and Engineering South China University of Technology Guangzhou China

ISBN: (纸本)9798400708527

Establishing morality and educating people is the fundamental task of higher education in China. Cultivating undergraduate' correct occupational values is an important part of colleges and universities. In order to better understand the occupational values of IT undergraduates and their influencing factors, by using Ning Weiwei's initially revised Chinese version of the Occupational Values Scale to conduct a questionnaire survey on IT undergraduates in a university in Guangzhou, we draw the following conclusions. First, in terms of the gender, there are significant differences between IT majors' occupational values in terms of intellectual development and colleague relationship. Second, in terms of the place of the origin, IT college students have significant differences in their occupational values in the dimensions of altruism, reputation, supervision and creativity due to the convenience of society, family, interpersonal relationship, etc. Third, in terms of grades, IT majors' professional values have significant differences in the dimension of colleague relationship. Fourth, there is a significant difference in the understanding of the IT industry development in the two dimensions of occupational values management and creativity. Fifth, there are significant differences in the clarity of self-planning in the three dimensions of the occupational values, namely, economic remuneration, management and creativity. In view of these influences, we made targeted analysis and proposed corresponding countermeasures to better help undergraduates in IT major establish correct occupational values. © 2023 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Differential Evolution Algorithm Based Hyper-Parameters Selection of Convolutional Neural Network for Speech Command Recognition 15

Differential Evolution Algorithm Based Hyper-Parameters Sele...

引用

15th International Joint Conference on Computational Intelligence, IJCCI 2023

作者： Dhar, Sandipan Sen, Anuvab Bandyopadhyay, Aritra Jana, Nanda Dulal Ghosh, Arjun Sarayloo, Zahra Computer Science and Engineering National Institute of Technology West Bengal Durgapur India Electronics and Telecommunication Indian Institute of Engineering Science and Technology Shibpur Howrah India Computer Science and Technology Indian Institute of Engineering Science and Technology Shibpur Howrah India School of Computer Science University of Waterloo ON Canada

ISBN: (纸本)9789897586743

Speech Command Recognition (SCR), which deals with identification of short uttered speech commands, is crucial for various applications, including IoT devices and assistive technology. Despite the promise shown by Convolutional Neural Networks (CNNs) in SCR tasks, their efficacy relies heavily on hyperparameter selection, which is typically laborious and time-consuming when done manually. This paper introduces a hyperparameter selection method for CNNs based on the Differential Evolution (DE) algorithm, aiming to enhance performance in SCR tasks. Training and testing with the Google Speech Command (GSC) dataset, the proposed approach showed effectiveness in classifying speech commands. Moreover, a comparative analysis with Genetic Algorithm-based selections and other deep CNN (DCNN) models highlighted the efficiency of the proposed DE algorithm in hyperparameter selection for CNNs in SCR tasks. © 2023 by SCITEPRESS – science and technology Publications, Lda.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Pre-Welding Defect Detection in Hull Construction Based on Dual-Branch Attention Feature Enhancement 9

Pre-Welding Defect Detection in Hull Construction Based on D...

引用

9th IEEE International Conference on Progress in Informatics and Computing, PIC 2024

作者： Kong, Yue Dou, Quansheng Ma, Haoxu Wang, Bo School of Information and Electronic Engineering Shandong Technology and Business University Yantai China School of Computer Science and Technology Shandong Technology and Business University Yantai China

ISBN: (纸本)9798350363203

In hull construction, manually identifying and assessing pre-welding defects is a challenging and time-consuming task, with potential defects posing significant safety risks. Fortunately, with the advancement of deep learning, convolutional neural network (CNN)-based recognition models have been successfully applied to computer vision tasks. However, due to the diversity of interfering factors in pre-welding defect images, current models struggle to capture critical defect features. When using attention mechanisms to address this issue, attention shifts can occur, leading to lower model recognition accuracy. To address the issue of attention shift, this paper proposes a Dual-Branch Attention Enhancement module (DBAE). This module extends traditional attention mechanisms by introducing an additional branch focused on enhancing defect features, guiding the attention towards these critical features, and improving feature extraction. Experimental results demonstrate that the DBAE module has superior feature capture capabilities compared to other classic attention modules, with the most significant improvement reaching 7.25 %, consistently outperforming other attention mechanisms. Furthermore, the results confirm that DBAE is effective across various classic classification networks, exhibiting good generalizability. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

CFD simulation of pumping ventilation in a three-story isolated building with internal partitioning:Effects of partition widths,heights and locations

引用

Building Simulation 2024年第2期17卷 267-284页

作者： Huai-Yu Zhong Jie Sun Chao Lin Song-Heng Wu Jin Shang Hideki Kikumoto Fu-Ping Qian Carlos Jimenez-Bescos Fu-Yun Zhao School of Energy and Environment Anhui University of TechnologyMa-An-ShanAnhui Province243002China State Key Laboratory of Green Building in Western China Xi’an University of Architecture&TechnologyXi’anShaanxi ProvinceChina School of Power and Mechanical Engineering Wuhan UniversityWuhanHubei Province430072China Institute of Industrial Science The University of Tokyo4-6-1 KomabaMeguro-kuTokyo153-8505Japan School of Built and Natural Environment University of DerbyDerbyUK

Pumping ventilation(PV),a special single-sided ventilation(SSV),has been certified as an effective strategy to improve the air exchange rate of ***,most studies targeted on the single space,and few studies have been focused on the effect of internal partitioning on *** paper aims to evaluate the ventilation performance of PV influenced by different configurations of internal *** fluid dynamics(CFD)simulation was used to predict the flow fields and ventilation *** width(w/H),height(h/H)and location(d/H)are the three main internal partition parameters considered in this *** simulation results showed that the total,mean and fluctuating ventilation rates all decrease with wider internal *** normalized total ventilation rate decreases by 7.6%when w/H is increased from 50%to 75%.However,the reduction rate is only 0.23%between w/H=0 and 25%,and only 0.61%between w/H=25%and 50%.The ventilation rate is hardly reduced by increasing the partition width when w/H<50%,whereas greatly reduced by wider partition for w/H>50%.Increasing the partition height will reduce the mean ventilation rate but promote the fluctuating and total ventilation rate in some *** increase of total ventilation rate by 1.4%is observed from h/H=50%to 75%.The ventilation rate is larger when the internal partition is attached to the leeward or windward *** total,mean and fluctuating ventilation rates for d/H=50%are relatively higher than d/H=0 by 1.5%,3.1%and 0.8%,*** the internal partition should be mounted attached to the windward wall so as to obtain the greatest pumping ventilation *** periodicity of pumping flow oscillation and pumping frequency are independent of the partition *** peak power of pumping flow is the lowest for the widest internal partition and is negatively affected by the partition height,but it generally has a positive correlation with the distance between the partition and leeward wa

关键词： pumping ventilation single-sided ventilation internal partitioning periodic oscillation ventilation rate

来源：评论

学校读者我要写书评

暂无评论

Distributed Ensemble Hyper-Heuristic Algorithm for Aircraft Arrival Sequencing and Scheduling Problem

Distributed Ensemble Hyper-Heuristic Algorithm for Aircraft ...

引用

2024 Genetic and Evolutionary Computation Conference Companion, GECCO 2024 Companion

作者： Wang, Xian-Ping Xu, Xin-Xin Jiang, Yi Bi, Sheng Liang, Yan-Chun Zhan, Zhi-Hui School of Computer Science and Engineering South China University of Technology Guangzhou510006 China School of Computer Science and Technology Ocean University of China Qingdao266100 China Zhuhai College of Science and Technology Zhuhai519041 China College of Artificial Intelligence Nankai University Tianjin300350 China

ISBN: (纸本)9798400704956

Aircraft arrival sequencing and scheduling (ASS) is a challenging NP-hard problem, which aims to arrange the order of aircraft landing at an airport to optimize the use of runway capacity and reduce delays. As air traffic continues to grow, airports face the constant challenge of efficiently managing the flow of incoming flights, emphasizing the need for a fast and efficient algorithm. However, existing algorithms are often extremely time-consuming to construct a near-optimal solution for ASS. To overcome the deficiency in existing algorithms, this paper proposes a fast and efficient distributed ensemble hyper-heuristic algorithm (DEHHA). To fast construct an optimal solution for ASS, DEHHA proposes an ensemble hyper-heuristic-based solution construct (EHHSC) strategy and a distributed cooperative evolution (DCE) strategy. The EHHSC strategy ensembles multiple effective hyper-heuristic optimizers to find diverse solutions and enables the communication of diverse optimizers to effectively find global optimal solution. The DCE strategy integrates the strengths of distributed computing, which enables multiple hyper-heuristic optimizers to cooperatively evolve in parallel to reduce running time. Extensive experiments results indicate DEHHA is not only capable of achieving superior performance in terms of solution quality but also requires a mere 42.9% of the time needed by the state-of-the-art algorithms. © 2024 held by the owner/author(s).

关键词： Heuristic algorithms

来源：评论

学校读者我要写书评

暂无评论

Broad Learning and Convolutional Neural Network aided Principal Component Analysis for Detecting Faults in High-speed Trains

Broad Learning and Convolutional Neural Network aided Princi...

引用

2023 International Conference on Automation Control, Algorithm, and Intelligent Bionics, ACAIB 2023

作者： Zhou, Mingyue Zhao, Hongyang Cheng, Chao School of Computer Science and Engineering Changchun University of Technology Changchun130012 China

ISBN: (纸本)9781510667662

Traction systems provide the traction power of high-speed trains. Because the complex operation mechanism of train under actual working conditions and the measured data are nonlinear and non-Gaussian, and the sampling frequency of the sensor is high in actual working conditions. Directly using a neural network or the multivariate statistical method is challenging to obtain the ideal fault detection (FD) result. Therefore, this paper proposes a data-driven method based on broad learning system (BLS) and convolutional neural network (CNN) assisted principal component analysis (PCA). Two neural networks are used to enhance the robustness of the algorithm, so that the proposed method has better fault detection ability in nonlinear and non-Gaussian systems. The advantage of this method is that it does not require the establishment of a complex high-speed train data model. Instead, by processing the collected data, the proposed algorithm can ensure good fault detection capabilities. Finally, the effectiveness and feasibility of the proposed method are verified on the simulation platform of traction drive control system (TDCS). © 2023 SPIE.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Face Forgery Detection Based on Fine-Grained Clues and Noise Inconsistency

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第1期6卷 144-158页

作者： Zhang, Dengyong He, Ruiyi Liao, Xin Li, Feng Chen, Jiaxin Yang, Gaobo Changsha University of Science and Technology Hunan Provincial Key Laboratory of Intelligent Processing of Big Data on Transportation the School of Computer and Communication Engineering Changsha410114 China Hunan University Changsha410082 China

Deepfake detection has gained increasing research attention in media forensics, and a variety of works have been produced. However, subtle artifacts might be eliminated by compression, and the convolutional neural networks (CNNs)-based detectors are invalidated for fake face images with compression. In this work, we propose a two-stream network for deepfake detection. We observed that high-frequency noise features and spatial features are inherently complementary to each other. Thus, both spatial features and high-frequency noise features are exploited for face forgery detection. Specifically, we design a double-frequency transformer module (DFTM) to guide the learning of spatial features from local artifact regions. To effectively fuse spatial features and high-frequency noise features, a dual-domain attention fusion module (DDAFM) is designed. We also introduce a local relationship constraint loss, which requires only image-level labels, for model training. We evaluate the proposed approach on five large-scale benchmark datasets, and extensive experimental results demonstrate the proposed approach outperforms most SOTA works. © 2024 IEEE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

Accessible e-News Summary Browser Extension for Visually Challenged Users 3

Accessible e-News Summary Browser Extension for Visually Cha...

引用

3rd International Conference on Innovative Sustainable Computational Technologies, CISCT 2023

作者： Rajyalakshmi, Tamma Kuppusamy, K.S. School of Engineering and Technology Pondicherry University Department of Computer Science Puducherry India

ISBN: (纸本)9798350303360

The news media is a vital global information source, evolving from physical newspapers to digital platforms. technology has revolutionized dissemination, ensuring broader accessibility. The main goal of news is to provide significant information and raise public awareness about a wide range of issues, notably jobs, healthcare, governmental initiatives, and alerts about natural disasters. This study delves into the challenges faced by individuals with visual impairments when accessing e-news online. The paramount goal is to ensure unbridled access to news for everyone, despite barriers. In this paper, We undertake a web accessibility evaluation of the homepages of 12 Indian news websites under three conformance levels of the WCAG 2.0 recommendation using AChecker and the WAVE Tool. Our evaluation is grounded in quantifying accessibility barriers. In particular, We address issues related to web accessibility and highlight the adverse effects of pop-up advertisements and automatic video play on news websites. We propose a browser extension for ad-free summaries, enhancing the news experience for the visually impaired. Evaluation insights guide ad format decisions, optimizing characteristics and placements on websites. © 2023 IEEE.

关键词： Websites

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：