A Real-Time Detection Method for Concrete Surface Cracks Based on Improved YOLOv4

Abstract

Many structures in civil engineering are symmetrical. Crack detection is a critical task in the monitoring and inspection of civil engineering structures. This study implements a lightweight neural network based on the YOLOv4 algorithm to detect concrete surface cracks. In the extraction of backbone and the design of neck and head, the symmetry concept is adopted. The model modules are improved to reduce the depth and complexity of the overall network structure. Meanwhile, the separable convolution is used to realize spatial convolution, and the SPP and PANet modules are improved to reduce the model parameters. The convolutional layer and batch normalization layer are merged to improve the model inference speed. In addition, using the focal loss function for reference, the loss function of object detection network is improved to balance the proportion of the cracks and the background samples. To comprehensively evaluate the performance of the improved method, 10,000 images (256 × 256 pixels in size) of cracks on concrete surfaces are collected to build the database. The improved YOLOv4 model achieves an mAP of 94.09% with 8.04 M and 0.64 GMacs. The results show that the improved model is satisfactory in mAP, and the model size and calculation amount are greatly reduced. This performs better in terms of real-time detection on concrete surface cracks.

Bibliometric Analysis and Review of Deep Learning-Based Crack Detection Literature Published between 2010 and 2022

“…In terms of processing, YOLOv4 performed better with the fastest detection rate of 10.16 FPS, while in terms of mAP, YOLOv5 achieved the highest mAP of 94.39%. Yao et al (2021) [93] integrated SPP and PANet modules to the YOLov4 architecture to reduce the number of parameters for a concrete surface crack detection model. The model accuracy slightly decreased as compared to the traditional YOLOv4 model; however, the number of parameters and inference time of the model decreased.…”

Lightweight Neural Network for Real-Time Crack Detection on Concrete Surface in Fog

“…The ResBlock_body is the residual block of CSPDarknet53, which can extract the target features of the image and reduce the computational bottleneck and the memory cost. The specific internal Module architecture is shown in Figure 2 (Yao et al, 2021).…”

“…In addition, it has been highlighted in image classification and object detection (Ren et al, 2017). A deep-learning-based method was developed to detect concrete bugholes (Wei et al, 2019;Yao et al, 2019;Wei et al, 2021), concrete cracks (Chen and Jahanshahi, 2018;Dung and Anh, 2019;Sun et al, 2021;Tang et al, 2021a;Yao et al, 2021), pavement cracks (Ji et al, 2020;Mei and Gül, 2020;Guan et al, 2021), and other defects (Lin et al, 2017;Cha et al, 2018;Li et al, 2019;Tang et al, 2021b;Jiang et al, 2021). The existing crack detection methods based on CNN generally have problems, such as complex network structures and excessive training parameters.…”

Special Issue Editorial “Symmetry in Structural Health Monitoring”

“…Yao et al [9] implemented a lightweight neural network based on the YOLOv4 model to detect concrete surface cracks by using the symmetry concept. The model modules were improved to reduce the depth and complexity of the overall network structure, allowing it to performed better in terms of the real-time detection of concrete surface cracks.…”