End-to-end tire defect detection model based on transfer learning techniques DOI Creative Commons
Radhwan A. A. Saleh, Mehmet Zeki Konyar, Kaplan Kaplan

et al.

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(20), P. 12483 - 12503

Published: April 22, 2024

Abstract Visual inspection of defective tires post-production is vital for human safety, as faulty can lead to explosions, accidents, and loss life. With the advancement technology, transfer learning (TL) plays an influential role in many computer vision applications, including tire defect detection problem. However, automatic difficult two reasons. The first presence complex anisotropic multi-textured rubber layers. Second, there no standard X-ray image dataset use detection. In this study, a TL-based model proposed using new from global company. First, we collected labeled consisting 3366 images 20,000 qualified tires. Although covers 15 types defects arising different design patterns, our primary focus on binary classification detect or absence defects. This challenging was split into 70, 15, 15% training, validation, testing, respectively. Then, nine common pre-trained models were fine-tuned, trained, tested dataset. These are Xception, InceptionV3, VGG16, VGG19, ResNet50, ResNet152V2, DenseNet121, InceptionResNetV2, MobileNetV2. results show that fine-tuned DenseNet21 InceptionNet achieve compatible with literature. Moreover, Xception outperformed compared TL literature methods terms recall, precision, accuracy, F1 score. it achieved testing 73.7, 88, 80.2, 94.75% score, respectively, validation 73.3, 90.24, 80.9, 95%

Language: Английский

DenseSPH-YOLOv5: An automated damage detection model based on DenseNet and Swin-Transformer prediction head-enabled YOLOv5 with attention mechanism DOI

Arunabha M. Roy,

Jayabrata Bhaduri

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 56, P. 102007 - 102007

Published: April 1, 2023

Language: Английский

Citations

167
WilDect-YOLO: An efficient and robust computer vision-based accurate object localization model for automated endangered wildlife detection DOI

Arunabha M. Roy,

Jayabrata Bhaduri,

Teerath Kumar

et al.

Ecological Informatics, Journal Year: 2022, Volume and Issue: 75, P. 101919 - 101919

Published: Nov. 18, 2022

Language: Английский

Citations

156
Densely Knowledge-Aware Network for Multivariate Time Series Classification DOI
Zhiwen Xiao, Huanlai Xing, Rong Qu

et al.

IEEE Transactions on Systems Man and Cybernetics Systems, Journal Year: 2024, Volume and Issue: 54(4), P. 2192 - 2204

Published: Jan. 9, 2024

Multivariate time series classification (MTSC) based on deep learning (DL) has attracted increasingly more research attention. The performance of a DL-based MTSC algorithm is heavily dependent the quality learned representations providing semantic information for downstream tasks, e.g., classification. Hence, model's representation ability critical enhancing its performance. This article proposes densely knowledge-aware network (DKN) MTSC. DKN's feature extractor consists residual multihead convolutional (ResMulti) and transformer-based (Trans), called ResMulti-Trans. ResMulti five blocks capturing local patterns data while Trans three transformer extracting global data. Besides, to enable dense mutual supervision between lower-and higher-level information, this adapts dual self-distillation (DDSD) mining rich regularizations relationships hidden in Experimental results show that compared with 5 state-of-the-art variants, proposed DDSD obtains 13/4/13 terms "win"/"tie"/"lose" gains lowest-AVG_rank score. In particular, pure ResMulti-Trans, DKN 20/1/9 regarding win/tie/lose. Last but not least, overweighs 18 existing algorithms 10 UEA2018 datasets achieves

Language: Английский

Citations

97
Deep Learning-Based Cost-Effective and Responsive Robot for Autism Treatment DOI Creative Commons
Aditya Singh, Kislay Raj,

Teerath Kumar

et al.

Drones, Journal Year: 2023, Volume and Issue: 7(2), P. 81 - 81

Published: Jan. 23, 2023

Recent studies state that, for a person with autism spectrum disorder, learning and improvement is often seen in environments where technological tools are involved. A robot an excellent tool to be used therapy teaching. It can transform teaching methods, not just the classrooms but also in-house clinical practices. With rapid advancement deep techniques, robots became more capable of handling human behaviour. In this paper, we present cost-efficient, socially designed called ‘Tinku’, developed assist special needs children. ‘Tinku’ low cost full features has ability produce human-like expressions. Its design inspired by widely accepted animated character ‘WALL-E’. capabilities include offline speech processing computer vision—we light object detection models, such as Yolo v3-tiny single shot detector (SSD)—for obstacle avoidance, non-verbal communication, expressing emotions anthropomorphic way, etc. uses onboard technique localize objects scene information semantic perception. We have several lessons training using these features. sample lesson about brushing discussed show robot’s capabilities. Tinku cute, loaded lots features, management all processes mind-blowing. supervision experts its condition application taken care of. small survey on appearance discussed. More importantly, it tested children acceptance technology compatibility terms voice interaction. helps autistic kids state-of-the-art models. Autism Spectral disorders being increasingly identified today’s world. The that prone interact comfortably than instructor. To fulfil demand, presented cost-effective solution form some common autism-affected child.

Language: Английский

Citations

95
An efficient and robust Phonocardiography (PCG)-based Valvular Heart Diseases (VHD) detection framework using Vision Transformer (ViT) DOI
Sonain Jamil,

Arunabha M. Roy

Computers in Biology and Medicine, Journal Year: 2023, Volume and Issue: 158, P. 106734 - 106734

Published: March 1, 2023

Language: Английский

Citations

69
Deep learning-accelerated computational framework based on Physics Informed Neural Network for the solution of linear elasticity DOI

Arunabha M. Roy,

Rikhi Bose,

Veera Sundararaghavan

et al.

Neural Networks, Journal Year: 2023, Volume and Issue: 162, P. 472 - 489

Published: March 13, 2023

Language: Английский

Citations

59
A data-driven physics-constrained deep learning computational framework for solving von Mises plasticity DOI

Arunabha M. Roy,

Suman Guha

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 122, P. 106049 - 106049

Published: March 7, 2023

Language: Английский

Citations

45
Exploring collaborative decision-making: A quasi-experimental study of human and Generative AI interaction DOI Creative Commons
Xinyue Hao, Emrah Demir, Daniel Eyers

et al.

Technology in Society, Journal Year: 2024, Volume and Issue: 78, P. 102662 - 102662

Published: July 17, 2024

This paper explores the effects of integrating Generative Artificial Intelligence (GAI) into decision-making processes within organizations, employing a quasi-experimental pretest-posttest design. The study examines synergistic interaction between Human (HI) and GAI across four group scenarios three global organizations renowned for their cutting-edge operational techniques. research progresses through several phases: identifying problems, collecting baseline data on decision-making, implementing AI interventions, evaluating outcomes post-intervention to identify shifts in performance. results demonstrate that effectively reduces human cognitive burdens mitigates heuristic biases by offering data-driven support predictive analytics, grounded System 2 reasoning. is particularly valuable complex situations characterized unfamiliarity information overload, where intuitive, 1 thinking less effective. However, also uncovers challenges related integration, such as potential over-reliance technology, intrinsic 'out-of-the-box' without contextual creativity. To address these issues, this proposes an innovative strategic framework HI-GAI collaboration emphasizes transparency, accountability, inclusiveness.

Language: Английский

Citations

26
Deep Learning Models for Diagnosis of Schizophrenia Using EEG Signals: Emerging Trends, Challenges, and Prospects DOI
Rakesh Ranjan, Bikash Chandra Sahana, Ashish Kumar Bhandari

et al.

Archives of Computational Methods in Engineering, Journal Year: 2024, Volume and Issue: 31(4), P. 2345 - 2384

Published: Jan. 6, 2024

Language: Английский

Citations

20
Combining artificial intelligence and computational fluid dynamics for optimal design of laterally perforated finned heat sinks DOI Creative Commons
Seyyed Amirreza Abdollahi, Ali Basem, As’ad Alizadeh

et al.

Results in Engineering, Journal Year: 2024, Volume and Issue: 21, P. 102002 - 102002

Published: March 1, 2024

The efficient design of heat sinks is a severe challenge in thermo-fluid engineering. A creative and innovative way applying lateral perforations to parallel finned sinks. significance achieving an optimal for perforated (PFHSs) has inspired the present authors introduce novel hybrid designing approach that combines computational fluid dynamics (CFD), machine learning (ML), multi-objective optimization (MOO), multi-criteria decision-making (MCDM). variables considered include size (0.25<φ < 0.5) shape (square, circular, hexagonal) perforations, as well airflow Reynolds number (2000

Language: Английский

Citations

19