AI-Based Road Safety Audit Automated Detection and Deterioration Assessment of Highway Safety Elements


Authors : Pitabasa Mohapatra R; Renukuntla Kranthi Kiran R; Deepika Sanga R; Bharani Kumar Depuru

Volume/Issue : Volume 10 - 2025, Issue 3 - March

Google Scholar : https://tinyurl.com/59wve3re

Scribd : https://tinyurl.com/567jw7pe

DOI : https://doi.org/10.38124/ijisrt/25mar1276

PlumX Metrics

Semantic Scholar

ResearchGate

IJISRT Repository

Google Scholar

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 15 to 20 days to display the article.

Abstract : Ensuring road safety requires continuous inspection and maintenance of critical infrastructure such as lane markings, signboards, and barriers. Traditional manual inspections are time-consuming, expensive, and prone to inconsistencies, leading to delays in identifying deteriorated safety products and increasing accident risks. This study presents an AI-powered solution that automates road safety audits using computer vision[1]. An object detection model identifies road safety elements, and a segmentation model evaluates their deterioration levels by classifying defects such as rust, fading, or structural damage. The deterioration percentage determines the classification: Good (≤30%) – No immediate action required; Moderate (31–70%) – Requires maintenance within a reasonable timeframe; Bad (>70%) – Requires urgent replacement or repair. The implemented system achieves a minimum accuracy rate of 87.5% in detecting and classifying road safety elements, contributing to a 40% reduction in inspection costs and enabling proactive maintenance scheduling. By automating road safety audits, this system enhances detection accuracy, reduces manual inspection costs, and enables scalable, real-time monitoring of highways[11].

Keywords : Road Safety Audit, Highway Safety, Object Detection, Image Segmentation, YOLOv8, Deterioration Assessment, Automated Maintenance, Streamlit , Deep Learning Models

References :

  1. Zhang, Z., Liu, Q., & Wang, Y. (2018). Automated road crack detection using deep convolutional neural networks. IEEE Transactions on Intelligent Transportation Systems.​
    https://www.researchgate.net/publication/330622399_Automated_Road_Crack_Detection_Using_Deep_Convolutional_Neural_Networks​ResearchGate
  2. Ren, S., He, K., Girshick, R., & Sun, J. (2015). Deep learning-based traffic sign detection and recognition for autonomous vehicles. IEEE Conference on Computer Vision and Pattern Recognition(CVPR). https://www.sciencedirect.com/science/article/pii/S0386111219301566​ScienceDirect
  3. Wu, J., Zhang, X., & Li, J. (2019). Automatic road marking detection and classification using machine learning algorithms. IEEE Transactions on Intelligent Transportation Systems.​
     https://dl.acm.org/doi/abs/10.1007/s00138-022-01302-0​ACM Digital Library
  4. Li, H., Li, Y., & Wang, Y. (2019). Real-time traffic sign recognition based on YOLOv3 model. IEEE International Conference on Image Processing (ICIP).​ https://www.mdpi.com/2412-3811/8/2/20​MDPI
  5. Yang, F., Zhang, H., & Li, S. (2019). Automated pavement distress detection using deep learning approaches. IEEE Transactions on Intelligent Transportation Systems.​ https://journals.sagepub.co m/doi/10.1177/03611981211004974​SAGE Journals
  6. Maeda, Y., Sekimoto, T., & Kato, S. (2018). Deep learning-based automated pavement distress detection and quantification using unmanned aerial vehicles. Automation in Construction.​
    https://www.sciencedirect.com/science/article/pii/S2046043023001028​ScienceDirect
  7. Lee, J., Kim, K., & Kim, S. (2019). A deep learning approach for road surface condition classification using acoustic signals. Sensors.​ https://www.m dpi.com/2076-3417/10/19/6997​MDPI
  8. Guan, Y., Li, J., & Wang, C. (2015). Automated detection of road markings from mobile LiDAR data using a novel point cloud segmentation approach. IEEE Transactions on Intelligent Transportation Systems.https://journals.sagepub.com/doi/10.1177/03611981211004974​SAGE Journals
  9. Fan, Z., Wu, Y., & Zhang, L. (2018). Deep learning-based automated pavement crack detection and measurement using UAV imagery. Computer-Aided Civil and Infrastructure Engineering. https://www.sci encedirect.com/science/article/pii/S2046043023001028
  10. Mario Soilán a, Diego González-Aguilera a, Ana del-Campo-Sánchez a, David Hernández-López b, Susana Del Pozo a . Road marking degradation analysis using 3D point cloud data acquired with a low-cost Mobile Mapping System .​ https://www.sciencedirect.com/sci ence/article/pii/S0926580522003193
  11. George J. Giummarra, Tim Martin, Zahidul Hoque, and Ron Roper Establishing Deterioration Models for Local Roads in Australia https://sci-hub.se/https://journals.sagepub.com/doi/abs/10.3141/1989-73
  12. Huang, Y., Xu, B., & Yu, L. (2018). Automated detection of pavement distresses using image-based deep learning. Construction and Building Materials.​https://www.sciencedirect.com/science/article/pii/S2046043023001028

 

Ensuring road safety requires continuous inspection and maintenance of critical infrastructure such as lane markings, signboards, and barriers. Traditional manual inspections are time-consuming, expensive, and prone to inconsistencies, leading to delays in identifying deteriorated safety products and increasing accident risks. This study presents an AI-powered solution that automates road safety audits using computer vision[1]. An object detection model identifies road safety elements, and a segmentation model evaluates their deterioration levels by classifying defects such as rust, fading, or structural damage. The deterioration percentage determines the classification: Good (≤30%) – No immediate action required; Moderate (31–70%) – Requires maintenance within a reasonable timeframe; Bad (>70%) – Requires urgent replacement or repair. The implemented system achieves a minimum accuracy rate of 87.5% in detecting and classifying road safety elements, contributing to a 40% reduction in inspection costs and enabling proactive maintenance scheduling. By automating road safety audits, this system enhances detection accuracy, reduces manual inspection costs, and enables scalable, real-time monitoring of highways[11].

Keywords : Road Safety Audit, Highway Safety, Object Detection, Image Segmentation, YOLOv8, Deterioration Assessment, Automated Maintenance, Streamlit , Deep Learning Models

Video Explanation for Published paper

Never miss an update from Papermashup

Get notified about the latest tutorials and downloads.

Subscribe by Email

Get alerts directly into your inbox after each post and stay updated.
Subscribe
OR

Subscribe by RSS

Add our RSS to your feedreader to get regular updates from us.
Subscribe