TD-RD: A Top-Down Benchmark with Real-Time Framework for Road Damage Detection

2025-01-24

AI Summary

To address the scarcity of top-down-view data and insufficient real-time performance in road damage detection, this paper introduces TDRD—the first dedicated benchmark dataset comprising 7,088 high-resolution images and 12,882 annotated instances across three damage types: cracks, potholes, and patches. We further propose TD-YOLOv10, a lightweight, real-time detection framework featuring multi-scale feature enhancement, dynamic label assignment, a streamlined neck architecture, and a high-precision bounding-box regression module. Evaluated on TDRD, TD-YOLOv10 achieves 68.3% mAP@0.5 at 42 FPS on a Tesla V100 GPU, outperforming state-of-the-art models including YOLOv8, YOLOv10, and RT-DETR. This work bridges two critical gaps in intelligent infrastructure inspection: the lack of high-quality, open-source top-down road damage data and efficient, real-time detection models tailored for practical deployment.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Computer Vision: Diffusion Models for Vision
• Intelligent Robots: State Estimation

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Object detection has witnessed remarkable advancements over the past decade, largely driven by breakthroughs in deep learning and the proliferation of large scale datasets. However, the domain of road damage detection remains relatively under explored, despite its critical significance for applications such as infrastructure maintenance and road safety. This paper addresses this gap by introducing a novel top down benchmark that offers a complementary perspective to existing datasets, specifically tailored for road damage detection. Our proposed Top Down Road Damage Detection Dataset (TDRD) includes three primary categories of road damage cracks, potholes, and patches captured from a top down viewpoint. The dataset consists of 7,088 high resolution images, encompassing 12,882 annotated instances of road damage. Additionally, we present a novel real time object detection framework, TDYOLOV10, designed to handle the unique challenges posed by the TDRD dataset. Comparative studies with state of the art models demonstrate competitive baseline results. By releasing TDRD, we aim to accelerate research in this crucial area. A sample of the dataset will be made publicly available upon the paper's acceptance.

Problem

• Road Damage Detection
• Accuracy
• Real-time Performance

Innovation

• TD-RD Standard
• TDYOLOV10 System
• TDRD Dataset