Next Article in Journal
Numerical Study of the Coupling of Sub-Terahertz Radiation to n-Channel Strained-Silicon MODFETs
Next Article in Special Issue
Identification of Crop Type in Crowdsourced Road View Photos with Deep Convolutional Neural Network
Previous Article in Journal
Semantic Segmentation Leveraging Simultaneous Depth Estimation
Previous Article in Special Issue
Automated Sensing System for Real-Time Recognition of Trucks in River Dredging Areas Using Computer Vision and Convolutional Deep Learning
Open AccessArticle

A CNN-Based Length-Aware Cascade Road Damage Detection Approach

by 1,2, 2,3,* and 4,5
1
Chengdu Institute of Computer Applications, Chinese Academy of Sciences, Chengdu 610041, China
2
University of Chinese Academy of Sciences, Beijing 100049, China
3
Harbin Institute of Technology, Shenzhen 518055, China
4
Guangzhou Electronic Technology Co., Ltd., Chinese Academy of Sciences, Guangzhou 510070, China
5
Guangzhou Institute of Electronic Technology, Chinese Academy of Sciences, Guangzhou 510070, China
*
Author to whom correspondence should be addressed.
Sensors 2021, 21(3), 689; https://doi.org/10.3390/s21030689
Received: 23 December 2020 / Revised: 14 January 2021 / Accepted: 15 January 2021 / Published: 20 January 2021
(This article belongs to the Special Issue Image Sensing and Processing with Convolutional Neural Networks)
Accurate and robust detection of road damage is essential for public transportation safety. Currently, deep convolutional neural networks (CNNs)-based road damage detection algorithms to localize and classify damage with a bounding box have achieved remarkable progress. However, research in this field fails to take into account two key characteristics of road damage: weak semantic information and abnormal geometric properties, resulting in inappropriate feature representation and suboptimal detection results. To boost the performance, we propose a CNN-based cascaded damage detection network, called CrdNet. The proposed model has three parts: (1) We introduce a novel backbone network, named LrNet, that reuses low-level features and mixes suitable range dependency features to learn high-to-low level feature fusions for road damage weak semantic information representation. (2) We apply multi-scale and multiple aspect ratios anchor mechanism to generate high-quality positive samples regarding the damage with abnormal geometric properties for network training. (3) We designed an adaptive proposal assignment strategy and performed cascade predictions on corresponding branches that can establish different range dependencies. The experiments show that the proposed method achieves mean average precision (mAP) of 90.92% on a collected road damage dataset, demonstrating the good performance and robustness of the model. View Full-Text
Keywords: deep convolutional neural network (CNN); road damage detection; abnormal geometric properties; length-aware; multi-scale attention deep convolutional neural network (CNN); road damage detection; abnormal geometric properties; length-aware; multi-scale attention
Show Figures

Figure 1

MDPI and ACS Style

Xu, H.; Chen, B.; Qin, J. A CNN-Based Length-Aware Cascade Road Damage Detection Approach. Sensors 2021, 21, 689. https://doi.org/10.3390/s21030689

AMA Style

Xu H, Chen B, Qin J. A CNN-Based Length-Aware Cascade Road Damage Detection Approach. Sensors. 2021; 21(3):689. https://doi.org/10.3390/s21030689

Chicago/Turabian Style

Xu, Huiqing; Chen, Bin; Qin, Jian. 2021. "A CNN-Based Length-Aware Cascade Road Damage Detection Approach" Sensors 21, no. 3: 689. https://doi.org/10.3390/s21030689

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Search more from Scilit
 
Search
Back to TopTop