# Automatic Pavement Crack Detection Based on Structured Prediction with the Convolutional Neural Network

arXiv: https://arxiv.org/abs/1802.02208 · HTML: https://mlsanity.com/paper/1802.02208/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Feb 2018.

## Built on by (most cited first, 21 of 21)

- [NHA12D: A New Pavement Crack Dataset and a Comparison Study Of Crack Detection Algorithms](https://mlsanity.com/paper/2205.01198/) (2022)
- [Fast and Accurate Road Crack Detection Based on Adaptive Cost-Sensitive Loss Function](https://mlsanity.com/paper/2106.15510/) (2021)
- [Weakly-Supervised Surface Crack Segmentation by Generating Pseudo-Labels using Localization with a Classifier and Thresholding](https://mlsanity.com/paper/2109.00456/) (2021)
- [What's Cracking? A Review and Analysis of Deep Learning Methods for Structural Crack Segmentation, Detection and Quantification](https://mlsanity.com/paper/2202.03714/) (2022)
- [B-BACN: Bayesian Boundary-Aware Convolutional Network for Crack Characterization](https://mlsanity.com/paper/2302.06827/) (2023)
- [Deep Learning Approaches in Pavement Distress Identification: A Review](https://mlsanity.com/paper/2308.00828/) (2023)
- [Pushing the Envelope of Thin Crack Detection](https://mlsanity.com/paper/2101.03326/) (2021)
- [Weakly Supervised Patch Label Inference Networks for Efficient Pavement Distress Detection and Recognition in the Wild](https://mlsanity.com/paper/2203.16782/) (2022)
- [Weakly-Supervised Crack Detection](https://mlsanity.com/paper/2206.06743/) (2022)
- [RHA-Net: An Encoder-Decoder Network with Residual Blocks and Hybrid Attention Mechanisms for Pavement Crack Segmentation](https://mlsanity.com/paper/2207.14166/) (2022)
- [CrackCLF: Automatic Pavement Crack Detection based on Closed-Loop Feedback](https://mlsanity.com/paper/2311.11815/) (2023)
- [RDD4D: 4D Attention-Guided Road Damage Detection And Classification](https://mlsanity.com/paper/2501.02822/) (2025)
- [Deep Learning for Pavement Condition Evaluation Using Satellite Imagery](https://mlsanity.com/paper/2508.01206/) (2025)
- [Image-based Detection of Surface Defects in Concrete during Construction](https://mlsanity.com/paper/2208.02313/) (2022)
- [PicT: A Slim Weakly Supervised Vision Transformer for Pavement Distress Classification](https://mlsanity.com/paper/2209.10074/) (2022)
- [Neuro-DynaStress: Predicting Dynamic Stress Distributions in Structural Components](https://mlsanity.com/paper/2301.02580/) (2022)
- [A Computer Vision Enabled damage detection model with improved YOLOv5 based on Transformer Prediction Head](https://mlsanity.com/paper/2303.04275/) (2023)
- [Deep Learning for Segmentation of Cracks in High-Resolution Images of Steel Bridges](https://mlsanity.com/paper/2403.17725/) (2024)
- [Pavement Fatigue Crack Detection and Severity Classification Based on Convolutional Neural Network](https://mlsanity.com/paper/2407.16021/) (2024)
- [Pixel-Level Pavement Distress Assessment Using Instance Segmentation](https://mlsanity.com/paper/2605.26095/) (2026)
- [YOLO-AMC: An Improved YOLO Architecture with Attention Mechanisms for Building Crack Detection](https://mlsanity.com/paper/2606.12958/) (2026)

## Abstract (arXiv)

Automated pavement crack detection is a challenging task that has been researched for decades due to the complicated pavement conditions in real world. In this paper, a supervised method based on deep learning is proposed, which has the capability of dealing with different pavement conditions. Specifically, a convolutional neural network (CNN) is used to learn the structure of the cracks from raw images, without any preprocessing. Small patches are extracted from crack images as inputs to generate a large training database, a CNN is trained and crack detection is modeled as a multi-label classification problem. Typically, crack pixels are much fewer than non-crack pixels. To deal with the problem with severely imbalanced data, a strategy with modifying the ratio of positive to negative samples is proposed. The method is tested on two public databases and compared with five existing methods. Experimental results show that it outperforms the other methods.