Comparison of CNN-Based Models for Pothole Detection in Real-World Adverse Conditions: Overview and Evaluation

Abstract

Potholes pose a significant problem for road safety and infrastructure. They can cause damage to vehicles and present a risk to pedestrians and cyclists. The ability to detect potholes in real time and with a high level of accuracy, especially under different lighting conditions, is crucial for the safety of road transport participants and the timely repair of these hazards. With the increasing availability of cameras on vehicles and smartphones, there is a growing interest in using computer vision techniques for this task. Convolutional neural networks (CNNs) have shown great potential for object detection tasks, including pothole detection. This study provides an overview of computer vision algorithms used for pothole detection. Experimental results are then used to evaluate the performance of the latest CNN-based models for pothole detection in different real-world road conditions, including rain, sunset, evening, and night, as well as clean conditions. The models evaluated in this study include both conventional and the newest architectures from the region-based CNN (R-CNN) and You Only Look Once (YOLO) families. The YOLO models demonstrated a faster detection response and higher accuracy in detecting potholes under clear, rain, sunset, and evening conditions. R-CNN models, on the other hand, performed better in the worse-visibility conditions at night. This study provides valuable insights into the performance of different CNN models for pothole detection in real road conditions and may assist in the selection of the most appropriate model for a specific application.

Oil and Gas Automatic Infrastructure Mapping: Leveraging High-Resolution Satellite Imagery Through Fine-Tuning of Object Detection Models

“…It was show that YOLO v4 achieved the best trade-off between speed and accuracy with a precision of 90.3%. [18] also concludes that YOLO has better accuracy (increasing with version) via a broad comparison of RCNN and YOLO models and their variants;…”

Exploring the Depths of Visual Understanding: A Comprehensive Review on Real-Time Object of Interest Detection Techniques

“…Overall these studies have investigated the performance of DenseNet for object detection problems, considering diverse datasets. From the review, it was observed while high success was recorded for metrics such as accuracy, recall, mAP, the delay during object detection [87,88] may affect its reliability when deployed as the computer vision model [76] for guidance assistance system for blind navigation. In addition, the maps despite their success need to be improved.…”

Pothole detection for autonomous vehicles using deep learning: a robust and efficient solution

“…A semi-supervised learning technique based on the encodings learned through a combination of a class-conditional variation autoencoder and a Wasserstein generative adversarial network is proposed for classifying and identifying damage into various severity levels. Lastly, in the study (Jakubec et al, 2023), the risk and research evaluation method (SPFPN) YOLO V4 tiny is suggested by combining the two techniques of spatial pyramid pooling and feature pyramid network with Darknet-53. After the data augmentation, the dataset was divided into three sets: training, validation, and testing, with 70% of the data being used for training.…”