Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection

Billah, Umme Hafsa; La, Hung Manh; Tavakkoli, Alireza

doi:10.3390/s20164403

Cited by 21 publications

(7 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, UAV surveillance has been upgraded with many more features of self-controlling, analysis and data processing by integrating UAVs with artificial intelligence (AI) [3] . Using these technologies, UAVs can be trained to perform particular tasks by processing large amounts of images and videos and simultaneously identifying presence region of interests (RoIs) in frames such as rebar detection bridge deck inspection [4] , perform monitoring tasks for early warning of natural disasters [5] or crack detection [6] . Indeed, the integration between UAV with AI technology could help UAV in performing complex tasks rather than surveillance.…”

Section: Methods Detailsmentioning

confidence: 99%

Video Surveillance Processing Algorithms utilizing Artificial Intelligent (AI) for Unmanned Autonomous Vehicles (UAVs)

Nguyen

Truong

2021

MethodsX

View full text Add to dashboard Cite

With the advancement of science and technology, the combination of the unmanned aerial vehicle (UAV) and camera surveillance systems (CSS) is currently a promising solution for practical applications related to security and surveillance operations. However, one of the biggest risks and challenges for the UAV-CSS is analysis, process, and transmission data, especially, the limitations of computational capacity, storage and overloading the transmission bandwidth. Regard to conventional methods, almost the data collected from UAVs is processed and transmitted that cost huge energy. A certain amount of data is redundant and not necessary to be processed or transmitted. This paper proposes an efficient algorithm to optimize the transmission and reception of data in UAV-CSS systems, based on the platforms of artificial intelligence (AI) for data processing. The algorithm creates an initial background frame and update to the complete background which is sent to server. It splits the region of interest (moving objects) in the scene and then sends only the changes. This supports the CSS to reduce significantly either data storage or data transmission. In addition, the complexity of the systems could be significantly reduced. The main contributions of the algorithm can be listed as follows; The developed solution can reduce data transmission significantly. The solution can empower smart manufacturing via camera surveillance. Simulation results have validated practical viability of this approach. The experimental method results show that reducing up to 80% of storage capacity and transmission data.

show abstract

Section: Methods Detailsmentioning

confidence: 99%

Video Surveillance Processing Algorithms utilizing Artificial Intelligent (AI) for Unmanned Autonomous Vehicles (UAVs)

Nguyen

Truong

2021

MethodsX

View full text Add to dashboard Cite

show abstract

“…More researches tend to cascade two or more kinds of deep networks. Billah and Jiang can quickly detect cracks in high-resolution bridge images by cascading target detection and semantic segmentation networks [35,36]. Chen Fu et al [37] propose naïve bayesconvolutional neural network (NB-CNN), which integrates the CNN network with Naive Bayes, significantly improving nuclear reactors' crack detection accuracy in nuclear power plants.…”

Section: Related Workmentioning

confidence: 99%

“…Comparison of IOU and speed on the dataset in this paper using different semantic segmentation networks models based on depth learning (Deepcrack [26], FPHBN [29], APLCNet [32], ANet-FSM [35], HDCB-Net [36]), the performance and output of the different methods are shown in Figures 14 and 15.…”

Section: Figure 13mentioning

confidence: 99%

A deep learning‐based method for pixel‐level crack detection on concrete bridges

Zhang

et al. 2022

IET Image Processing

View full text Add to dashboard Cite

Crack detection of the concrete bridge is an essential index for the safety assessment of bridge structure. It is more important to check the whole structure than to check the accuracy in the damage assessment. However, the traditional deep learning model method cannot completely detect the crack structure, which challenges image-based crack detection. For this reason, we propose deep bridge crack classification (DBCC)-Net as a classification-based deep learning network. By pruning the Yolox, the regression problem of the target detection is converted to the binary classification problem to avoid the network performance degradation caused by the translation invariance of the convolutional neural network (CNN). In addition, the network post-processing and a two-stage crack detection strategy are proposed to enable the network to detect cracks and extract crack morphology in high-resolution images quickly. In the first stage, DBCC-Net realizes the coarse extraction of crack position based on image slice classification. In the second stage, the complete crack morphology is extracted from the location suggested by the semantic segmentation network. Experimental results show that the proposed two-stage method has 19 frames per second (FPS) and 0.79 Miou (mean intersection over union) at the actual bridge images with 2560×2560 pixels. Although FPS is reduced, the Miou value is 7.8% higher than other methods, proving this paper's practical value.

show abstract

“…Billah et al proposed a specific detection framework for crack segmentation using a deep learning network to assist the automatic detection process. This method integrates the feature silencing module and improves the network detection performance [40]. Fang et al proposed a novel hybrid method that combines deep learning models and Bayesian probability analysis to achieve reliable crack detection.…”

Section: Introductionmentioning

confidence: 99%

Bridge Crack Semantic Segmentation Based on Improved Deeplabv3+

Meng

et al. 2021

JMSE

View full text Add to dashboard Cite

Cracks are the main goal of bridge maintenance and accurate detection of cracks will help ensure their safe use. Aiming at the problem that traditional image processing methods are difficult to accurately detect cracks, deep learning technology was introduced and a crack detection method based on an improved DeepLabv3+ semantic segmentation algorithm was proposed. In the network structure, the densely connected atrous spatial pyramid pooling module was introduced into the DeepLabv3+ network, which enabled the network to obtain denser pixel sampling, thus enhancing the ability of the network to extract detail features. While obtaining a larger receptive field, the number of network parameters was consistent with the original algorithm. The images of bridge cracks under different environmental conditions were collected, and then a concrete bridge crack segmentation data set was established, and the segmentation model was obtained through end-to-end training of the network. The experimental results showed that the improved DeepLabv3+ algorithm had higher crack segmentation accuracy than the original DeepLabv3+ algorithm, with an average intersection ratio reaching 82.37%, and the segmentation of crack details was more accurate, which proved the effectiveness of the proposed algorithm.

show abstract

Deep Learning-Based Feature Silencing for Accurate Concrete Crack Detection

Cited by 21 publications

References 46 publications

Video Surveillance Processing Algorithms utilizing Artificial Intelligent (AI) for Unmanned Autonomous Vehicles (UAVs)

Video Surveillance Processing Algorithms utilizing Artificial Intelligent (AI) for Unmanned Autonomous Vehicles (UAVs)

A deep learning‐based method for pixel‐level crack detection on concrete bridges

Bridge Crack Semantic Segmentation Based on Improved Deeplabv3+

Contact Info

Product

Resources

About