Enhanced Intelligent Identification of Concrete Cracks Using Multi-Layered Image Preprocessing-Aided Convolutional Neural Networks

Summary The seismic performance of a building must be evaluated after it has been affected by an earthquake load. In the evaluation process, building codes and standards require that the drift of the structure is determined to assess structural performance. This study provides an innovative method that helps engineers in measuring the deflection of reinforced concrete (RC) beams. An imagery deep learning model, called residual networks (ResNet), is used to classify the deflection based on observation by computer vision. However, determining the optimal values of the hyperparameters of this model is a challenge. Therefore, a hybrid model that integrates the bio‐inspired optimization (i.e., jellyfish search [JS] algorithm) and ResNet is developed. The input data that are used to train the model are images that are collected in RC structural experiments. This experiment involved 29 cantilever beams with various RC designs. These specimen RC beams were tested under simulated seismic loads with lateral displacement control. After each load had been applied to the beam, four single‐lens digital cameras captured images from the east, west, north, and south. Then, the performance of computer vision‐based JS–ResNet was evaluated by comparing its accuracy with that of the original ResNet using default hyperparameters. The results of the analysis show that the proposed JS–ResNet model achieves higher accuracy than conventional ResNet. Therefore, the hybrid model can provide insights in similar visual surveillance tasks.

show abstract

“…Equation 1 provides the softmax function.

S_{j} goodbreak= \frac{e^{italicCj}}{\sum_{j = 1}^{K} e^{Cj}},

where S j is the softmax function, C j is the j th element in set C to be classified, and K is the total number of categories 47 …”

Section: Methodsmentioning

confidence: 99%

“…where S j is the softmax function, C j is the jth element in set C to be classified, and K is the total number of categories. 47 ResNet is a deep 48 Wang et al 49 used a deep ResNet framework for structural health monitoring.…”

Section: Residual Networkmentioning

confidence: 99%

Identifying deflections of reinforced concrete beams under seismic loads by bio‐inspired optimization of deep residual learning

Chou

Karundeng

Truong

et al. 2022

Structural Contr & Hlth

View full text Add to dashboard Cite

show abstract

“…P and R in F1 represent precision and recall, respectively. PA is the ratio of the number of pixels with correct prediction category to the total number of pixels [47]. F1 is a harmonic mean value.…”

Section: Training Process To Save Training Time and Generalizementioning

confidence: 99%

A Modified Fully Convolutional Network for Crack Damage Identification Compared with Conventional Methods

Meng

Zhu

Chen

et al. 2021

Modelling and Simulation in Engineering

View full text Add to dashboard Cite

Large-scale structural health monitoring and damage detection of concealed underwater structures are always the urgent and state-of-art problems to be solved in the field of civil engineering. With the development of artificial intelligence especially the combination of deep learning and computer vision, greater advantages have been brought to the concrete crack detection based on convolutional neural network (CNN) over the traditional methods. However, these machine learning (ML) methods still have some defects, such as it being inaccurate or not strong, having poor generalization ability, or the accuracy still needs to be improved, and the running speed is slow. In this article, a modified fully convolutional network (FCN) with more robustness and more effectiveness is proposed, which makes it convenient and low cost for long-term structural monitoring and inspection compared with other methods. Meanwhile, to improve the accuracy of recognition and prediction, innovations were conducted in this study as follows. Moreover, differed from the common simple deconvolution, it also includes a subpixel convolution layer, which can greatly reduce the sampling time. Then, the proposed method was verified its practicability with the overall recognition accuracy reaching up to 97.92% and 12% efficiency improvement.

show abstract

“…Image inversion, blurring, histogram stretching, and equalization methods were used to pre-process thermal images for person detection using CNN [2]. Homomorphic filtering and OTSU thresholding methods are proposed in [17] for improving image quality for concrete cracks detection using CNN.…”

Section: Related Workmentioning

confidence: 99%

Effects of Pre-processing on the Performance of Transfer Learning Based Person Detection in Thermal Images

Huda

Gade

Moeslund

2021

2021 IEEE 2nd International Conference on Pattern Recognition and Machine Learning (PRML)

View full text Add to dashboard Cite

Thermal images have the property of identifying objects even in low light conditions. However, person detection in thermal is tricky, due to varying person representations depending upon the surrounding temperature. Three major polarities are commonly observed in these representations i.e., 1. person warmer than the background, 2. person colder than the background and 3. person's body temperature is similar to background. In this work, we have studied and analyzed the performance of the detection network by using the data in its original form and by harmonizing the person representation in two ways i.e., dark persons in the light background and light persons in a darker background. The data passed to each testing scenario was first pre-processed using histogram stretching to enhance the contrast. The work also presents the method to separate the three kinds of images from thermal data. The analysis is performed on publicly available outdoor AAU-PD-T and OSU-T datasets. Precision, recall, and F1 score is used to evaluate network performance. The results have shown that network performance is not enhanced by performing the mentioned pre-processing. Best results are obtained by using the data in its original form.

show abstract

Enhanced Intelligent Identification of Concrete Cracks Using Multi-Layered Image Preprocessing-Aided Convolutional Neural Networks

Cited by 17 publications

References 37 publications

Identifying deflections of reinforced concrete beams under seismic loads by bio‐inspired optimization of deep residual learning

Identifying deflections of reinforced concrete beams under seismic loads by bio‐inspired optimization of deep residual learning

A Modified Fully Convolutional Network for Crack Damage Identification Compared with Conventional Methods

Effects of Pre-processing on the Performance of Transfer Learning Based Person Detection in Thermal Images

Contact Info

Product

Resources

About