A Fast Detection Method via Region‐Based Fully Convolutional Neural Networks for Shield Tunnel Lining Defects

219

Deep learning has ushered in many breakthroughs in vision‐based detection via convolutional neural networks (CNNs), but the vibration‐based structural damage detection by CNN remains being refined. Thus, this study proposes a simple one‐dimensional CNN that detects tiny local structural stiffness and mass changes, and validates the proposed CNN on actual structures. Three independent acceleration databases are established based on a T‐shaped steel beam, a short steel girder bridge (in test field), and a long steel girder bridge (in service). The raw acceleration data are not pre‐processed and are directly used as the training and validation data. The well‐trained CNN almost perfectly identifies the locations of small local changes in the structural mass and stiffness, demonstrating the high sensitivity of the proposed simple CNN to tiny structural state changes in actual structures. The convolutional kernels and outputs of the convolutional and max pooling layers are visualized and discussed as well.

Section: Introductionmentioning

confidence: 99%

Vibration‐based structural state identification by a 1‐dimensional convolutional neural network

Zhang

Miyamori

Mikami

et al. 2019

219

“…ML and data science has shown great potential for predicting, designing, and discovering materials (Ley & Bordas, ). In civil engineering and construction, ML has been extensively used in a variety of applications such as structural heal monitoring (Gao & Mosalam, ; Rafiei & Adeli, , ; Xue & Li, ), reliability analysis (Dai & Cao, ; Grande, Castillo, Mora, & Lo, ; Nabian & Meidani, ), transportation (Dharia & Adeli, ; García‐Ródenas, López‐García, & Sánchez‐Rico, ; Yu, Wang, Shan, & Yao, ; Zhang & Ge, ), and prediction and estimation (Adeli & Wu, ; Chou & Pham, ; Rafiei, Khushefati, Demirboga, & Adeli, ; Zhao & Ren, ). In concrete‐related studies, DeRousseau, Kasprzyk, and Srubar () recently reviewed the application of ML to optimize mixture design of concrete.…”

Section: Introductionmentioning

confidence: 99%

Deep neural network with high‐order neuron for the prediction of foamed concrete strength

Nguyen

Kashani

Ngo

et al. 2018

214

The article presents a deep neural network model for the prediction of the compressive strength of foamed concrete. A new, high‐order neuron was developed for the deep neural network model to improve the performance of the model. Moreover, the cross‐entropy cost function and rectified linear unit activation function were employed to enhance the performance of the model. The present model was then applied to predict the compressive strength of foamed concrete through a given data set, and the obtained results were compared with other machine learning methods including conventional artificial neural network (C‐ANN) and second‐order artificial neural network (SO‐ANN). To further validate the proposed model, a new data set from the laboratory and a given data set of high‐performance concrete were used to obtain a higher degree of confidence in the prediction. It is shown that the proposed model obtained a better prediction, compared to other methods. In contrast to C‐ANN and SO‐ANN, the proposed model can genuinely improve its performance when training a deep neural network model with multiple hidden layers. A sensitivity analysis was conducted to investigate the effects of the input variables on the compressive strength. The results indicated that the compressive strength of foamed concrete is greatly affected by density, followed by the water‐to‐cement and sand‐to‐cement ratios. By providing a reliable prediction tool, the proposed model can aid researchers and engineers in mixture design optimization of foamed concrete.

“…Undeniably, these methods achieve excellent performance in realistic situations when only one damage type is detected. However, the CNN method has to adopt a sliding window technique to localize the detected damage (Cha et al., ; Xue & Li, ). Subsequently, Cha et al.…”

Section: Introductionmentioning

confidence: 99%

Automatic pixel‐level multiple damage detection of concrete structure using fully convolutional network

Zhao

Zhou

2019

369

158

Deep learning‐based structural damage detection methods overcome the limitation of inferior adaptability caused by extensively varying real‐world situations (e.g., lighting and shadow changes). However, most deep learning‐based methods detect structural damage at the image level and grid‐cell level. To provide pixel‐level detection of multiple damages, a Fully Convolutional Network (FCN)‐based multiple damages detection method for concrete structure is proposed. To realize this method, a database of 2,750 images (with 504 × 376 pixels) including crack, spalling, efflorescence, and hole images in concrete structure is built, and the four damages included in those images are labeled manually. Then, the architecture of the FCN is modified, trained, validated, and tested using this database. A strategy of model‐based transfer learning is used to initialize the parameters of the FCN during the training process. The results show 98.61% pixel accuracy (PA), 91.59% mean pixel accuracy (MPA), 84.53% mean intersection over union (MIoU), and 97.34% frequency weighted intersection over union (FWIoU). Subsequently, the robustness and adaptability of the trained FCN model is tested and the damage is extracted, where damage areas are provided according to a calibrated relation between the ratio (the pixel area and true area of the detected object) and the distance from the smartphone to the concrete surface using a laser range finder. A comparative study is conducted to examine the performance of the proposed FCN‐based approach using a SegNet‐based method. The results show that the proposed method substantiates quite better performance and can indeed detect multiple concrete damages at the pixel level in realistic situations.