检索结果-内蒙古大学图书馆

IEEE Transactions on Intelligent Vehicles 2024年第5期9卷 1-15页

作者： Fei, Ben Zhang, Rui Yang, Weidong Li, Zhijun Chen, Wen-Ming School of Computer Science Fudan University Shanghai China School of Mechanical Engineering Tongji University Shanghai China Academy for Engineering and Technology Fudan University Shanghai China

Point cloud completion concentrates on completing geometric and topological shapes from incomplete 3D shapes. Nevertheless, the unordered nature of point clouds will hamper the generation of high-quality point clouds without predicting structured and topological information of the complete shapes and introducing noisy points. To effectively address the challenges posed by missing topology and noisy points, we introduce SPOFormer, a novel topology-aware model that utilizes surface-projection optimization in a progressive growth manner. SPOFormer consists of three distinct steps for completing the missing topology: (1) Missing Keypoints Prediction. A topology-aware transformer auto-encoder is integrated for missing keypoint prediction. (2) Skeleton Generation. The skeleton generation module produces a new type of representation named skeletons with the aid of keypoints predicted by topology-aware transformer auto-encoder and the partial input. (3) Progressively Growth. We design a progressive growth module to predict final output under Multi-scale Supervision and Surface-projection Optimization. Surface-projection Optimization is firstly devised for point cloud completion, aiming to enforce the generated points to align with the underlying object surface. Experimentally, SPOFormer model achieves an impressive Chamfer Distance-$\ell _{1}$ (CD) score of 8.11 on PCN dataset. Furthermore, it attains average CD-$\ell _{2}$ scores of 1.13, 1.14, and 1.70 on ShapeNet-55, ShapeNet-34, and ShapeNet-Unseen21 datasets, respectively. Additionally, the model achieves a Maximum Mean Discrepancy (MMD) of 0.523 on the real-world KITTI dataset. These outstanding qualitative and quantitative performances surpass previous approaches by a significant margin, firmly establishing new state-of-the-art performance across various benchmark datasets. Our code is available at https://***/kiddoray/SPOFormer IEEE

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

A Data-driven Pattern Design Method for Watertight Plush Toy Models

引用

computer-Aided Design and Applications 2023年第1期20卷 28-43页

作者： Jin, Yinghan Wang, Jin Li, Jituo Zhang, Dongliang College of Computer Science Zhejiang University China School of Mechanical Engineering Zhejiang University China

This paper presents an automatic pattern design method for plush toys using a data-driven approach. For a watertight toy model, we use a divide-andconquer strategy to convert it into a component-based model, then automatically design patterns of each component. Firstly, a segmentation and componentization method based on deep learning is proposed to decompose a 3D watertight model into several meaningful components. Then, the most similar component template is retrieved and used to generate seam lines of the model. Finally, 2D patterns of the toy model are obtained by flattening 3D surface patches. The segmentation and componentization method can also be used to easily edit the shape and structure of a watertight model. Experimental results show that reasonable patterns can be designed automatically and efficiently using the proposed method. © 2023 CAD Solutions, LLC.

关键词： Toys

来源：评论

学校读者我要写书评

暂无评论

Thermal intelligence: exploring AI’s role in optimizing thermal systems – a review

引用

Interactions 2024年第1期245卷 1-19页

作者： Devasenan, Madhesh Madhavan, Saritha Department of Mechanical Engineering AMET University Tamil Nadu Chennai603112 India Department of Computer Science and Engineering Sri Sivasubramaniya Nadar College of Engineering Tamil Nadu Kalavakkam Chennai603110 India

Artificial Intelligence (AI) is profoundly transforming multiple industries by enhancing efficiency, decision-making, and problem-solving capabilities. This impact is particularly notable in sectors such as thermal engineering, industrial processing, and solar thermal systems. In thermal engineering, AI has addressed challenges in modeling, prediction, control, and optimization, leading to improvements in energy efficiency, system reliability, and environmental sustainability. Innovations in AI-driven models, such as gradient-boosted regression trees and deep reinforcement learning, have advanced the design and management of thermal systems, including energy storage devices, power plants, and HVAC systems. AI is also revolutionizing industrial processing and solar thermal systems. In industrial settings, AI applications such as quality control, process optimization, and predictive maintenance are enhancing operational efficiency and supporting the transition towards Industry 5.0. Despite these advancements, challenges remain in aligning AI solutions with real-world industrial requirements. Similarly, in solar thermal systems, AI is optimizing performance through advanced tracking systems, predictive maintenance, and energy storage strategies. The integration of AI with technologies like the Internet of Things (IoT) and advanced materials is proving effective in boosting system efficiency. Overall, AI is providing innovative solutions that improve efficiency, reliability, and sustainability across various applications. However, ongoing research is essential to address limitations in AI models, enhance their generalizability, and integrate emerging technologies. Continued development in AI methodologies will be crucial for optimizing energy efficiency, system performance, and environmental impact in these sectors. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Predictive maintenance

来源：评论

学校读者我要写书评

暂无评论

A Framework for Industrial Inspection System using Deep Learning

引用

Annals of Data science 2024年第2期11卷 445-478页

作者： Hridoy, Monowar Wadud Rahman, Mohammad Mizanur Sakib, Saadman Department of Mechatronics & amp Industrial Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh Department of Mechanical Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh Department of Computer Science & amp Engineering Chittagong University of Engineering & ampTechnology Chattogram Bangladesh

Industrial Inspection systems are an essential part of Industry 4.0. An automated inspection system can significantly improve product quality and reduce human labor while making their life easier. However, a deep learning-based camera inspection system requires a large amount of data to classify the defective products accurately. In this paper, a framework is proposed for an industrial inspection system with the help of deep learning. Additionally, A new dataset of hex-nut products is proposed containing 4000 images, i.e., 2000 defective and 2000 non-defective. Moreover, different CNN architectures, i.e., Custom CNN, Inception ResNet v2, Xception, ResNet 101 v2, ResNet 152 v2, are experimented with the concept of transfer learning on the new hex-nut dataset. Fine-tuning the CNN architectures is performed by freezing the last 14 layers, which provided the optimal architecture, i.e., Xception (last 14 layers trainable, excluding the fully connected layer). The proposed framework can efficiently separate the defective products from the non-defective products with 100% accuracy on the hex nut dataset. Furthermore, the proposed optimal Xception architecture has experimented on a publicly available casting material dataset which produced 99.72% accuracy, outperforming existing methods. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2022. Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s);author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

关键词： Industry 4.0

来源：评论

学校读者我要写书评

暂无评论

Classifying chest x-rays for COVID-19 through transfer learning: a systematic review

引用

Multimedia Tools and Applications 2025年第2期84卷 689-748页

作者： Mallick, Devanshi Singh, Arshdeep Ng, Eddie Yin-Kwee Arora, Vinay Thapar Institute of Engineering & Technology Punjab Patiala India Wipro Limited Gurugram India School of Mechanical and Aerospace Engineering Nanyang Technological University Singapore Singapore Computer Science Engineering Department Thapar Institute of Engineering & Technology Punjab Patiala India

This study makes a comprehensive assessment of the predominant Transfer Learning (TL) techniques employed for the classification of COVID-19 cases in Chest X-rays (CXR) images. The methodologies have been selected on the basis of their merits and demerits, suitability, and possible impact on the development of the region being studied. The study examines the various methods of TL employed in the classification of COVID-19 cases with the objective to gain a deeper understanding about all the aspects of these methodologies. It can be of great significance for the researchers and medical professionals in making well-informed decisions about the implementation of these techniques to improve the precision and effectiveness of COVID-19 diagnosis. The practical consequences of these techniques help in early identification of such cases for having a suitable intervention. As many as 48 studies conducted during the period 2020–2023 have been included in the current research work for having an assessment about the problem under investigation. The study has specifically focused on transfer learning-based models utilized for the identification of COVID-19 through CXR pictures. It highlights the challenges posed by dataset dynamics, methodological variations, and performance metrics of different models. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

Enhancement of mechanical properties in AA5083 aluminum metal matrix composites through alumina and titanium carbide reinforcement

引用

Interactions 2024年第1期245卷 1-12页

作者： Vadivel, M. Seenivasan, S. Satishkumar, P. Gomkale, Shruti Saminathan, Rajasekaran Department of Civil Engineering Nehru Institute of Technology Tamilnadu Coimbatore India Department of Mechanical Engineering Rathinam Technical Campus Tamilnadu Coimbatore India Applied Chemistry Yeshwantrao Chavan College of Engineering Maharashtra Nagpur India Department of Mechanical Engineering College of Engineering and Computer Science Jazan University Jazan Saudi Arabia

As a result of its remarkable resistance to corrosion, strong wear resistance, high strength, and low weight, aluminum metal matrix composites (AMMC’s) have garnered tremendous widespread recognition for structural and aeronautical applications. Several mechanical properties of aluminum alloy (AA5083) are examined in this experimental study, which focuses on the impact of reinforcing particles (alumina and titanium carbide). The stir casting method was used to manufacture AMMC. Fabrication was carried out using the less difficult ex-situ approach. Reinforcement particle weight percentages ranging from 4 to 12% were used in the MMC preparation process. Machine operations were performed on the ingots following manufacturing in order to prepare test samples. To investigate the mechanical and tribological properties and effects of different reinforcement weight percentages, tests were conducted including tensile, compression, impact, hardness and wear tests. Adding alumina and titanium carbide resulted in a gradual rise of the tensile and compression strengths. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Aluminum corrosion

来源：评论

学校读者我要写书评

暂无评论

A semi-supervised deep neuro-fuzzy iterative learning system for automatic segmentation of hippocampus brain MRI

引用

Mathematical Biosciences and engineering 2024年第12期21卷 7830-7853页

作者： Nisha, M. Kannan, T. Sivasankari, K. Department of Computer Science and Engineering Akshaya College of Engineering and Technology Tamil Nadu Coimbatore India Department of Mechanical Engineering Amrita College of Engineering and Technology Tamil Nadu Nagercoil India Department of Electronics and Engineering Akshaya College of Engineering and Technology Tamil Nadu Coimbatore India

The hippocampus is a small, yet intricate seahorse-shaped tiny structure located deep within the brain’s medial temporal lobe. It is a crucial component of the limbic system, which is responsible for regulating emotions, memory, and spatial navigation. This research focuses on automatic hippocampus segmentation from Magnetic Resonance (MR) images of a human head with high accuracy and fewer false positive and false negative rates. This segmentation technique is significantly faster than the manual segmentation methods used in clinics. Unlike the existing approaches such as UNet and Convolutional Neural Networks (CNN), the proposed algorithm generates an image that is similar to a real image by learning the distribution much more quickly by the semi-supervised iterative learning algorithm of the Deep Neuro-Fuzzy (DNF) technique. To assess its effectiveness, the proposed segmentation technique was evaluated on a large dataset of 18,900 images from Kaggle, and the results were compared with those of existing methods. Based on the analysis of results reported in the experimental section, the proposed scheme in the Semi-Supervised Deep Neuro-Fuzzy Iterative Learning System (SS-DNFIL) achieved a 0.97 Dice coefficient, a 0.93 Jaccard coefficient, a 0.95 sensitivity (true positive rate), a 0.97 specificity (true negative rate), a false positive value of 0.09 and a 0.08 false negative value when compared to existing approaches. Thus, the proposed segmentation techniques outperform the existing techniques and produce the desired result so that an accurate diagnosis is made at the earliest stage to save human lives and to increase their life span. © 2024 the Author(s),

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A systematic literature review of visual feature learning: deep learning techniques, applications, challenges and future directions

引用

Multimedia Tools and Applications 2024年第19期84卷 20439-20496页

作者： Abdullahi, Mohammed Oyelade, Olaide Nathaniel Kana, Armand Florentin Donfack Bagiwa, Mustapha Aminu Abdullahi, Fatimah Binta Junaidu, Sahalu Balarabe Iliyasu, Ibrahim Ore-ofe, Ajayi Chiroma, Haruna Department of Computer Science Ahmadu Bello University Zaria Zaria Nigeria Department of Mechanical Engineering Ahmadu Bello University Zaria Nigeria Department of Computer Engineering Ahmadu Bello University Zaria Nigeria College of Computer Science and Engineering University of Hafr Al Batin Hafar Al-Batin31991 Saudi Arabia School of Electronics Electrical Engineering and Computer Science Queen’s University Belfast Belfast United Kingdom

Visual Feature Learning (VFL) is a critical area of research in computer vision that involves the automatic extraction of features and patterns from images and videos. The applications of VFL are vast, including object detection and recognition, facial recognition, scene understanding, medical image analysis, and autonomous vehicles. In this paper, we propose to conduct extensive systematic literature review (SLR) on VFL based on deep learning algorithms. The paper conducted an SLR covering deep learning algorithms such as Convolutional Neural Networks (CNNs), Autoencoders, and Generative Adversarial Networks (GANs) including their variants. The review highlights the importance of VFL in computer vision and the limitations of traditional feature extraction techniques. Furthermore, it provides an in-depth analysis of the strengths and weaknesses of various deep learning algorithms for solving problems in VFL. The discussion of the applications of VFL provides an insight into the impact of VFL on various industries and domains. The review also analyzed the challenges faced by VFL, such as data scarcity and quality, overfitting, generalization, interpretability, and explainability. The discussion of future directions for VFL includes hybrid techniques, unsupervised feature learning, continual learning, attention-based models, and explainable AI. These techniques aim to address the challenges faced by VFL and improve the performance of the models. The systematic literature review concludes that VFL is a rapidly evolving field with the potential to transform many industries and domains. The review highlights the need for further research in VFL and emphasizes the importance of responsible use of VFL models in various applications. The review provides valuable insights for researchers and practitioners in the field of computer vision, who can use these insights to enhance their work and ensure the responsible use of VFL models. © The Author(s), under exclusive licence to

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

GRIPPY: A VR Grip Controller for Combating Sarcopenia in Elderly

GRIPPY: A VR Grip Controller for Combating Sarcopenia in Eld...

引用

2025 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops, VRW 2025

作者： Leong, Hoilam Wu, Jiang Lai, Yoke Chin Faulon, Loïc Sun, Xu Lee, Boon Giin University of Nottingham Faculty of Science and Engineering Department of Mechanical Materials and Manufacturing Engineering Ningbo China School of Computer Science China

ISBN: (纸本)9798331514846

As the global population ages, sarcopenia - age-related muscle decline - demands innovative solutions. This paper introduces GRIPPY, a VR grip controller that transforms basic handgrip exercises into immersive, gamified tasks. By integrating sensor-based interaction and VR gameplay, older adults can strengthen grip while maintaining higher motivation. A preliminary test with three participants (ages 68-80) revealed key usability insights, underscoring GRIPPY's potential for enhanced rehabilitation. Future work will refine its design and validate long-term efficacy. © 2025 IEEE.

关键词： Elderly Rehabilitation Grip Strength Immersive Exercise Interactive Training Sarcopenia Virtual Reality

来源：评论

学校读者我要写书评

暂无评论

An experimental and numerical approach for optimization of internal and external corrugations with various geometries in cylindrical thin-walled concertina energy absorbers under quasi-static axial load

引用

Mechanics of Advanced Materials and Structures 2024年第25期31卷 6628-6642页

作者： Rahpeima, Reza Nateghi Boroujeni, Iman Nia, Ali Alavi Graduate institute of biomedical electronics and bioinformatics College of Electrical Engineering and Computer Science National Taiwan University Taipei Taiwan Department of Mechanical Engineering Tarbiat Modares University Tehran Iran Department of Mechanical Engineering Bu-Ali Sina University Hamedan Iran

Thin-walled tubes have been extensively used as energy absorbers in many engineering structures that are under dynamic or quasi-static loads. In this article, the effect of corrugation and its geometry on the collapse characteristics of aluminum concertina tubes under quasi-static axial load is investigated both experimentally and numerically. The numerical simulations were performed with the use of LS-Dyna commercial software. Then, several chosen samples were tested experimentally to verify the results of the simulation. Forty different samples with four different corrugation geometries including singly isosceles triangular corrugation, continual isosceles triangular corrugation, continual scalene triangular corrugation, and continual semi-circular corrugation, and each with 10 different size dimensions were simulated. Among all of these simulations, six of them, each with two replicates were subjected to experimental tests. Experimental examples are aluminum cylindrical thin-walled tubes with the same thickness, height, and diameter that corrugations with different geometries are applied to them by the use of the cold rolling method. Finally, to describe the findings in more theoretical terms, a comparison between our obtained results and earlier theoretical investigations is made. The results show a significant influence of corrugations application on the improvement of the collapse characteristics of the cylindrical tubes so that the collapse style of intact tubes can be controlled by applying corrugations. The results show that by increasing the amplitude and number of corrugations, the absorbed energy, and the initial maximum force decrease. It is also found that among the different corrugation geometries, the corrugations with singly isosceles triangular geometry have the highest energy absorption as well as the lowest initial maximum force. © 2023 Taylor & Francis Group, LLC.

关键词： Geometry

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

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

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：