Related references
Note: Only part of the references are listed.
Article
Geochemistry & Geophysics
Gong Cheng et al.
Summary: The article introduced an effective defense framework PSGAN for RSI scene classification, which trains the classifier by generating examples to combat known and unknown attacks. Experimental results demonstrated the great effectiveness of PSGAN.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2022)
Article
Engineering, Electrical & Electronic
Mingqi Lu et al.
Summary: The study proposed a pose-guided model using keypoint action features to recognize driving behaviors, which showed significant improvements in feature extraction and behavior recognition, achieving superior performance on two datasets.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2022)
Article
Engineering, Civil
Binbin Qin et al.
Summary: The study introduces a new D-HCNN model for distracted driving detection, which uses HOG feature images, L2 weight regularization, dropout, and batch normalization to achieve high accuracy and speed. Experimental evaluations on two public datasets show that the accuracy of the D-HCNN model is higher than many other state-of-the-art methods.
IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
(2022)
Article
Computer Science, Information Systems
Ahmad Ali et al.
Summary: Accurate and timely prediction of citywide traffic flow is crucial for public safety and traffic management in smart cities. While existing methods have leveraged LSTM and CNN to explore temporal and spatial relations separately, the proposed DHSTNet model combines both aspects and outperforms traditional statistical methods. The AAtt-DHSTNet model, which incorporates an attention mechanism, further enhances prediction accuracy compared to the DHSTNet technique.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Yang Wang
Summary: With the advancement of web technology, the analysis of multi-modal data has become a focus, with the fusion of multi-modal feature spaces being a key factor in enhancing performance. Deep neural networks have shown excellent performance in handling multi-modal data, and research from shallow to deep spaces is gradually expanding, with collaboration, adversarial competition, and fusion playing important roles in this field.
ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Ashlesha Kumar et al.
Summary: The study aims to improve driver distraction classification by enhancing driver posture recognition techniques, using an ensemble of six independent deep neural architectures. The approach achieved high accuracy on two datasets, with a quick inference time on a specific machine setup.
ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING (ICAISC 2021), PT II
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Hong Vin Koay et al.
Summary: Research uses transformer to handle computer vision tasks, and hierarchical transformer limits self-attention computation through representation calculated with shifted windows, improving computational efficiency. Results show that shifted-window hierarchical transformer achieves a high classification accuracy in detecting distracted drivers.
2021 IEEE REGION 10 SYMPOSIUM (TENSYMP)
(2021)
Proceedings Paper
Automation & Control Systems
Duy-Linh Nguyen et al.
Summary: Driving is a complex activity that can be easily affected by various distractions. Therefore, developing assistant applications to warn distracted drivers is necessary. This paper proposes a lightweight Convolutional Neural Network for a distracted driver warning system, achieving accuracy rates of 95.36% and 99.95% on different datasets.
IECON 2021 - 47TH ANNUAL CONFERENCE OF THE IEEE INDUSTRIAL ELECTRONICS SOCIETY
(2021)
Article
Computer Science, Information Systems
Faiqa Sajid et al.
Summary: The number of road accidents has been increasing globally, with 45% of crashes attributed to distracted drivers according to the national highway traffic safety administration. This study introduces a new distracted driver detection model using a dataset, with EfficientDet-D3 identified as the best performing model.
Article
Optics
Ankit Pal et al.
Summary: Driver distraction is a significant cause of road accidents, posing a threat to both the driver and others on the road. A system must be developed to monitor and alert drivers of distractions to prevent accidents.
OPTICAL MEMORY AND NEURAL NETWORKS
(2021)
Proceedings Paper
Engineering, Industrial
Bing-Ting Dong et al.
Summary: This paper presents techniques for simultaneously detecting fatigue and distracted driving behaviors using vision and learning based approaches. Facial features and machine learning models are utilized to achieve this. Experimental results show that the proposed methods outperform previous approaches in terms of accuracy and computation time.
2021 22ND IEEE INTERNATIONAL CONFERENCE ON INDUSTRIAL TECHNOLOGY (ICIT)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Mingyan Wu et al.
Summary: A novel multi-feature fusion network based on pose estimation is proposed for detecting distracted driving behaviors in images. The method first detects hands, extracts features using hand and human body posture information, and fuses global, hand, and pose features through weighted combinations and concatenation for state-of-the-art performance.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Francois Nel et al.
Summary: This paper proposes using ResNet with 3D kernels for distracted driver behavior recognition. The study finds a significant positive correlation between model accuracy and network depth, as well as the importance of dataset quality in determining model generalization ability.
2021 SOUTHERN AFRICAN UNIVERSITIES POWER ENGINEERING CONFERENCE/ROBOTICS AND MECHATRONICS/PATTERN RECOGNITION ASSOCIATION OF SOUTH AFRICA (SAUPEC/ROBMECH/PRASA)
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Paola Canas et al.
Summary: This article introduces the newly presented Driver Monitoring Dataset (DMD) and explores the application of action recognition methods to detect driver distraction. The study compares different state-of-the-art models for image and video classification, discussing the feasibility of implementing image-based or video-based models in a real-context driver monitoring system. Preliminary results are presented as a reference for future work on the DMD.
VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 5: VISAPP
(2021)
Article
Engineering, Electrical & Electronic
Xiaoliang Qian et al.
Summary: A new co-saliency detection method TSE-GAN is proposed in this paper, which achieves good performance in handling co-saliency and semantic consistency by using a two-stage training strategy.
IEEE SIGNAL PROCESSING LETTERS
(2021)
Article
Computer Science, Information Systems
Jing Wang et al.
Summary: A data augmentation method based on driving operation areas for distracted driving detection was proposed in this paper, achieving a high classification accuracy rate and helping to identify dangerous driving behaviors to prevent accidents.
Article
Computer Science, Information Systems
Sahil Sharma et al.
MULTIMEDIA TOOLS AND APPLICATIONS
(2020)
Article
Computer Science, Artificial Intelligence
Lin Wu et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2020)
Article
Computer Science, Artificial Intelligence
Chaojie Ou et al.
IEEE TRANSACTIONS ON INTELLIGENT VEHICLES
(2020)
Article
Engineering, Electrical & Electronic
Lin Wu et al.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2020)
Proceedings Paper
Computer Science, Hardware & Architecture
Ahmad Ali et al.
2019 IEEE 25TH INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED SYSTEMS (ICPADS)
(2019)