Concrete Crack Detection Based on Well-Known Feature Extractor Model and the YOLO_v2 Network

Abstract

This paper compares the crack detection performance (in terms of precision and computational cost) of the YOLO_v2 using 11 feature extractors, which provides a base for realizing fast and accurate crack detection on concrete structures. Cracks on concrete structures are an important indicator for assessing their durability and safety, and real-time crack detection is an essential task in structural maintenance. The object detection algorithm, especially the YOLO series network, has significant potential in crack detection, while the feature extractor is the most important component of the YOLO_v2. Hence, this paper employs 11 well-known CNN models as the feature extractor of the YOLO_v2 for crack detection. The results confirm that a different feature extractor model of the YOLO_v2 network leads to a different detection result, among which the AP value is 0.89, 0, and 0 for ‘resnet18’, ‘alexnet’, and ‘vgg16’, respectively meanwhile, the ‘googlenet’ (AP = 0.84) and ‘mobilenetv2’ (AP = 0.87) also demonstrate comparable AP values. In terms of computing speed, the ‘alexnet’ takes the least computational time, the ‘squeezenet’ and ‘resnet18’ are ranked second and third respectively; therefore, the ‘resnet18’ is the best feature extractor model in terms of precision and computational cost. Additionally, through the parametric study (influence on detection results of the training epoch, feature extraction layer, and testing image size), the associated parameters indeed have an impact on the detection results. It is demonstrated that: excellent crack detection results can be achieved by the YOLO_v2 detector, in which an appropriate feature extractor model, training epoch, feature extraction layer, and testing image size play an important role.

Comparison of classic object-detection techniques for automated sewer defect detection

“…As for the feature extraction layer, the last CONV of the pretrained CNN model was selected based on empirical analysis. The design of network layer was primarily based on the official manuals of the MATLAB and our previous research results (Teng et al 2021;Lin et al 2022). The previous work has explained the functions of all layers in detail and confirmed that these layers have excellent performance for defect image detection.…”

“…It is difficult to evaluate the detection effects based on P and R individually as they are lack of overall information. Instead, AP is a comprehensive indicator of P and R and is thus commonly used to evaluate the network performance (Maeda et al 2018;Deng et al 2020;Zhang et al 2020;Teng et al 2021). The value range of AP is 0-1, the closer the AP value to 1, the more excellent is the detection effect.…”

Damage detection using in-domain and cross-domain transfer learning

“…Visual inspection datasets are small scale and are expensive to curate. Therefore, several studies adopt standard architectures-such as VGG16, Inception-v3, LeNet, YOLO, or ResNet50 often pre-trained with IMAGENET representations-for the detection of cracks [31,56,61,75], potholes [39], spalls [68], and multiple other damages including corrosion, seapage, and exposed bars [17,28,72,78]. In Table 1, we provide a non-exhaustive list of recent literature studies that have used transfer learning for concrete damage detection tasks.…”

Improved YOLO Based Detection Algorithm for Floating Debris in Waterway

“…Zheng et al [ 24 ] proposed a convolution neural network based on YOLOv3 to detect bearing-cover defects. Teng et al [ 25 ] used a well-known feature extractor model and the YOLOv2 network to detect a concrete crack. Li et al [ 26 ] proposed an enhanced YOLOv3 tiny network for real-time ship detection.…”