Traffic sign recognition is a classification problem that poses challenges for computer vision and machine learning algorithms. Although both computer vision and machine learning techniques have constantly been improved to solve this problem, the sudden rise in the number of unlabeled traffic signs has become even more challenging. Large data collation and labeling are tedious and expensive tasks that demand much time, expert knowledge, and fiscal resources to satisfy the hunger of deep neural networks. Aside from that, the problem of having unbalanced data also poses a greater challenge to computer vision and machine learning algorithms to achieve better performance. These problems raise the need to develop algorithms that can fully exploit a large amount of unlabeled data, use a small amount of labeled samples, and be robust to data imbalance to build an efficient and high-quality classifier. In this work, we propose a novel semi-supervised classification technique that is robust to small and unbalanced data. The framework integrates weakly-supervised learning and self-training with self-paced learning to generate attention maps to augment the training set and utilizes a novel pseudo-label generation and selection algorithm to generate and select pseudo-labeled samples. The method improves the performance by: (1) normalizing the class-wise confidence levels to prevent the model from ignoring hard-to-learn samples, thereby solving the imbalanced data problem; (2) jointly learning a model and optimizing pseudo-labels generated on unlabeled data; and (3) enlarging the training set to satisfy the hunger of deep learning models. Extensive evaluations on two public traffic sign recognition datasets demonstrate the effectiveness of the proposed technique and provide a potential solution for practical applications.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC7248915 | PMC |
http://dx.doi.org/10.3390/s20092684 | DOI Listing |
Traffic Inj Prev
January 2025
School of Civil and Hydraulic Engineering, NingXia University, YinChuan, China.
Objective: This study aims to address the issue of driving safety on highways in the desert region of Northwest China during extreme weather conditions such as sandstorms, with the goal of reducing driver risk. It explores driver behavior under extreme conditions of sandstorms and sand accumulation, proposing safety speed recommendations and warning models for different environments to calculate the optimal warning distance in windy and sandy conditions.
Methods: Natural driving simulation experiments were conducted in windy and sandy environments, collecting driving behavior data from 45 drivers under varying visibility and road conditions with or without sand accumulation.
Sensors (Basel)
December 2024
Computer Engineering, Brandenburg University of Technology, Cottbus-Senftenberg, 03046 Cottbus, Germany.
Occasionally, four cars arrive at the four legs of an unsignalized intersection at the same time or almost at the same time. If each lane has a stop sign, all four cars are required to stop. In such instances, gestures are used to communicate approval for one vehicle to leave.
View Article and Find Full Text PDFLeg Med (Tokyo)
February 2025
Department of Forensic Medicine, Medical School, University of Pécs, Pécs, Hungary. Electronic address:
Motor vehicle accidents (MVA) are the leading cause of death in childhood and young adult age. One of the most important factors behind MVA is driving under the influence of alcohol (DUIA) and drugs (DUID). The importance of DUID is rising together with the increasing drug abuse.
View Article and Find Full Text PDFTraffic Inj Prev
December 2024
Intelligent Transportation Systems Research Center, Wuhan University of Technology, Wuhan, Hubei, China.
Objective: Exit ramps are accident-prone areas of freeways. One of the reasons for this is the information overload induced by destination signs, which makes them challenging to recognize and may even result in tension or mistakes. This study examined the cognitive workload that destination signs place on drivers and the compensatory behavior they use to counteract the additional workload.
View Article and Find Full Text PDFJ Abdom Wall Surg
November 2024
Department of Surgery, UD of Medicine of Vall d'Hebron, Universitat Autònoma de Barcelona, Abdominal Wall Surgery Unit, General and Digestive Surgery Department, Hospital Universitari Vall d'Hebrón, Barcelona, Spain.
Aim: To discuss extended retrorectal abscess secondary to blunt abdominal trauma as a cause of abdominal wall (AW) infection and impairment.
Methods: According to the CARE checklist, we describe a rare case of blunt abdominal trauma with late diagnosis of jejunal perforation with an abscess that extensively dissected the retromuscular space.
Results: A 65 years-old female patient experienced multiple traumas after a traffic collision.
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!