[go: up one dir, main page]
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
5.3
2.5
Journals
Applied Sciences
Special Issues
Generative Artificial Intelligence Technologies and Applications for Road Environment Understanding
applsci-logo
Submit to Special Issue Submit Abstract to Special Issue Review for Applied Sciences Propose a Special Issue

Journal Menu

► Journal Menu

Journal Browser

► Journal Browser
share announcement

Generative Artificial Intelligence Technologies and Applications for Road Environment Understanding

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: 20 May 2025 | Viewed by 2663

Special Issue Editors

Prof. Dr. Rung-Ching Chen

E-Mail Website
Guest Editor
Department of Information Management, Chaoyang University of Technology, Taichung 413310, Taiwan
Interests: network technology; pattern recognition and machine learning; IoT and data analysis; applications of artificial intelligence; deep learning; generative adversarial networks
Special Issues, Collections and Topics in MDPI journals
Prof. Dr. Long-Sheng Chen

E-Mail Website
Guest Editor
Department of Information Management, Chaoyang University of Technology, Taichung 413310, Taiwan
Interests: data mining; social networking; quality management; customer relationship management
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

This Special Issue explores the latest advancements and applications of generative AI technologies to enhance understanding road environments. Generative AI models have demonstrated remarkable capabilities in synthesizing realistic data, enabling novel solutions for tasks such as scene generation, object detection, and semantic segmentation in road environments. The topics will cover theoretical foundations, methodological developments, and the practical applications of generative AI for road scene analysis, 3D environment modeling, traffic flow prediction, and other related areas. The Special Issue will provide a platform for disseminating innovative ideas, fostering interdisciplinary collaborations, and driving progress in this rapidly evolving field. The topics include but are not limited to the following:

  1. Generative adversarial networks for road scene synthesis and augmentation;
  2. Variational autoencoders for road environment modeling and reconstruction;
  3. Diffusion models for road scene generation and editing;
  4. Generative models for 3D road environment reconstruction;
  5. Conditional generative models for road object detection and segmentation;
  6. Generative models for traffic flow prediction and simulation;
  7. Adversarial training techniques for road environment understanding;
  8. Interpretability and explainability of generative models in road scene analysis;
  9. Multimodal generative models for road environments (e.g., combining vision and LiDAR data);
  10. Generative models for data augmentation and domain adaptation in road scene understanding;
  11. Generative models for road sign and lane marking synthesis;
  12. Applications of generative AI in autonomous driving, intelligent transportation systems, and road safety.

Prof. Dr. Rung-Ching Chen
Prof. Dr. Long-Sheng Chen
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • deep learning
  • synthetic data generation
  • road environments
  • scene understanding
  • intelligent transportation systems
  • autonomous driving
  • object detection
  • semantic segmentation
  • generative adversarial networks (GANs)
  • variational autoencoders (VAEs)

Benefits of Publishing in a Special Issue

  • Ease of navigation: Grouping papers by topic helps scholars navigate broad scope journals more efficiently.
  • Greater discoverability: Special Issues support the reach and impact of scientific research. Articles in Special Issues are more discoverable and cited more frequently.
  • Expansion of research network: Special Issues facilitate connections among authors, fostering scientific collaborations.
  • External promotion: Articles in Special Issues are often promoted through the journal's social media, increasing their visibility.
  • e-Book format: Special Issues with more than 10 articles can be published as dedicated e-books, ensuring wide and rapid dissemination.

Further information on MDPI's Special Issue policies can be found here.

Published Papers (2 papers)

Download All Papers
Order results
Result details
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:

Research

17 pages, 4838 KiB  
Article
Improved Detection of Multi-Class Bad Traffic Signs Using Ensemble and Test Time Augmentation Based on Yolov5 Models
by Ibrahim Yahaya Garta, Shao-Kuo Tai and Rung-Ching Chen
Appl. Sci. 2024, 14(18), 8200; https://doi.org/10.3390/app14188200 - 12 Sep 2024
Viewed by 972
Abstract
Various factors such as natural disasters, vandalism, weather, and environmental conditions can affect the physical state of traffic signs. The proposed model aims to improve detection of traffic signs affected by partial occlusion as a result of overgrown vegetation, displaced signs (those knocked [...] Read more.
Various factors such as natural disasters, vandalism, weather, and environmental conditions can affect the physical state of traffic signs. The proposed model aims to improve detection of traffic signs affected by partial occlusion as a result of overgrown vegetation, displaced signs (those knocked down, bent), perforated signs (those damaged with holes), faded signs (color degradation), rusted signs (corroded surface), and de-faced signs (placing graffiti, etc., by vandals). This research aims to improve the detection of bad traffic signs using three approaches. In the first approach, Spiral Pooling Pyramid-Fast (SPPF) and C3TR modules are introduced to the architecture of Yolov5 models. SPPF helps provide a multi-scale representation of the input feature map by pooling at different scales, which is useful in improving the quality of feature maps and detecting bad traffic signs of various sizes and perspectives. The C3TR module uses convolutional layers to enhance local feature extraction and transformers to boost understanding of the global context. Secondly, we use predictions of Yolov5 as base models to implement a mean ensemble to improve performance. Thirdly, test time augmentation (TTA) is applied at test time by using scaling and flipping to improve accuracy. Some signs are generated using stable diffusion techniques to augment certain classes. We test the proposed models on the CCTSDB2021, TT100K, GTSDB, and GTSRD datasets to ensure generalization and use k-fold cross-validation to further evaluate the performance of the models. The proposed models outperform other state-of-the-art models in comparison. Full article
(This article belongs to the Special Issue Generative Artificial Intelligence Technologies and Applications for Road Environment Understanding)
►▼ Show Figures

Figure 1

Figure 1
<p>Sample images representing all the classes in the dataset: (<b>a</b>) occluded; (<b>b</b>) displaced; (<b>c</b>) faded; (<b>d</b>) perforated; (<b>e</b>) good; (<b>f</b>) rusted; (<b>g</b>) defaced.</p> Full article ">Figure 2
<p>Structure of the Yolov5 model.</p> Full article ">Figure 3
<p>Structures of C3 and C3TR modules.</p> Full article ">Figure 4
<p>Flowchart of the proposed ensemble model.</p> Full article ">Figure 5
<p>Flowchart of the proposed test time augmentation.</p> Full article ">Figure 6
<p>Comparison of accuracy of all classes for base and improved models.</p> Full article ">Figure 7
<p>Precision–recall curve of the mean ensemble.</p> Full article ">Figure 8
<p>F1 score of the TTA model.</p> Full article ">Figure 9
<p>Graph showing mAP@50 of the proposed models.</p> Full article ">Figure 10
<p>Detection results on some public datasets: (<b>a</b>) Detection results on the GTSRD by TTA; (<b>b</b>) detection results on the TT100K test image by mean ensemble; (<b>c</b>) detection result showing misclassification and detection of good traffic signs on GTSRD image by improved Yolov5m; (<b>d</b>) detection result by improved Yolov5s on CCTSDB2021; (<b>e</b>) misdetection by Yolov5s on CCTSDB2021 as rusted traffic sign and correctly detect good traffic sign; (<b>f</b>) detection result by Yolov5m on GTSDB dataset.</p> Full article ">
14 pages, 2101 KiB  
Article
Temperature Behavior in Headlights: A Comparative Analysis between Battery Electric Vehicles and Internal Combustion Engine Vehicles
by Tabea Schlürscheid, Tran Quoc Khanh, Alexander Buck and Stefan Weber
Appl. Sci. 2024, 14(15), 6654; https://doi.org/10.3390/app14156654 - 30 Jul 2024
Cited by 1 | Viewed by 1071
Abstract
In the context of a global shift towards renewable energies and climate change mitigation, the market for electric vehicles has experienced a remarkable upswing, with battery electric vehicles (BEVs) leading this transformative wave. The appeal of BEVs lies in their ability to significantly [...] Read more.
In the context of a global shift towards renewable energies and climate change mitigation, the market for electric vehicles has experienced a remarkable upswing, with battery electric vehicles (BEVs) leading this transformative wave. The appeal of BEVs lies in their ability to significantly curtail CO2 emissions by supplanting the traditional internal combustion engine (ICE) with an electric motor. This pivotal change in vehicular technology extends its influence to various subsystems, including automotive lighting. Headlights are particularly sensitive to the thermal environment they operate in, which can profoundly affect their functionality and durability. The removal of an ICE in BEVs typically results in a reduction in heat exposure to headlight components, prompting a potential reevaluation of their design. This article presents a comprehensive analysis of temperature distributions within headlight units, comparing BEVs and ICE vehicles. The study encompasses a robust dataset of nearly 30,000 vehicles from around the globe, taking into account the impact of ambient temperature on headlight operation. The investigation delineates the distinct thermal behaviors of the two vehicle categories and offers strategic recommendations for conceptual modifications of headlights in BEVs. These adjustments are aimed at enhancing headlight efficacy, prolonging lifespan, and furthering the sustainability objectives of electric mobility. Full article
(This article belongs to the Special Issue Generative Artificial Intelligence Technologies and Applications for Road Environment Understanding)
►▼ Show Figures

Figure 1

Figure 1
<p>Histogram of the mileage in km for all vehicles for the BEV and ICE engine types.</p> Full article ">Figure 2
<p>Number of BEV (light blue) and ICE (dark blue) vehicles analyzed in the different ambient temperature classes.</p> Full article ">Figure 3
<p>Distribution of headlamp temperature histogram across all vehicles, all ambient temperature classes, and all light functions for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 4
<p>Distribution of the headlamp temperature histogram across all vehicles and all lighting functions for ambient temperature class 1 of &lt;15 °C annual mean ambient temperature for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 5
<p>Distribution of the headlamp temperature histogram across all vehicles and all lighting functions for ambient temperature class 2 of 15–20 °C annual mean ambient temperature for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 6
<p>Distribution of the headlamp temperature histogram across all vehicles and all lighting functions for ambient temperature class 3 of 20–25 °C annual mean ambient temperature for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 7
<p>Distribution of the headlamp temperature histogram across all vehicles and all lighting functions for ambient temperature class 4 of 25–30 °C annual mean ambient temperature for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 8
<p>Distribution of the headlamp temperature histogram across all vehicles and all lighting functions for ambient temperature class 5 of &gt;30 °C annual mean ambient temperature for BEVs (light blue) and ICEs (dark blue).</p> Full article ">Figure 9
<p>The calculated temperature center of gravity in the headlamps of the respective ambient temperature classes for ICEs and BEVs along with their linear fit for BEVs (light blue) and ICEs (dark blue).</p> Full article ">
Show export options expand_more Show export options expand_less
Select all
Export citation of selected articles as:
 
Displaying articles 1-2
Back to TopTop