An integration–competition network for bridge crack segmentation under complex scenes

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2023, Volume and Issue: 39(4), P. 617 - 634

Published: Oct. 16, 2023

Abstract The segmentation accuracy of bridge crack images is influenced by high‐frequency light, complex scenes, and tiny cracks. Therefore, an integration–competition network (complex [CCSNet]) proposed to address these problems. First, a grayscale‐oriented adjustment algorithm solve the light problem. Second, mechanism detach backgrounds grayscale features Finally, attention extract shallow CCSNet outperforms seven state‐of‐the‐art methods in both generalization comparison experiments on self‐built dataset four public datasets. It also achieved excellent performance practical tests. effective auxiliary method for lowering cost safety detection.

Language: Английский

---

A generative adversarial network approach for removing motion blur in the automatic detection of pavement cracks

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2024, Volume and Issue: 39(22), P. 3412 - 3434

Published: May 20, 2024

Abstract Advancements in infrastructure management have significantly benefited from automatic pavement crack detection systems, relying on image processing enhanced by high‐resolution imaging and machine learning. However, motion blur substantially challenge the accuracy of analysis. Nevertheless, research mitigating remains sparse. This study introduces an effective system adept at deblurring segmentation, employing a generative adversarial network (GAN) with UNet as generator Wasserstein GAN Gradient Penalty (WGAN‐gp) loss function. approach performs exceptionally images improves segmentation accuracy. Models were trained sharp artificially blurred images, WGAN‐gp surpassing other functions effectiveness. innovatively suggests assessing quality through addition to peak signal‐to‐noise ratio (PSNR) structural similarity (SSIM), revealing that PSNR SSIM may not fully capture effectiveness for images. An extensive evaluation various generators, including UNet, lightweight TransUNet, DeblurGAN, DeblurGAN‐v2, MIMO‐UNet, identifies superior performance simulated blur. Validation actual motion‐blurred confirms proposed model. These findings demonstrate GAN‐based models great potential overcoming challenges marking notable advancement field.

Language: Английский

---

Semi-supervised semantic image segmentation by deep diffusion models and generative adversarial networks

International Journal of Neural Systems, Journal Year: 2024, Volume and Issue: 34(11)

Published: July 5, 2024

Typically, deep learning models for image segmentation tasks are trained using large datasets of images annotated at the pixel level, which can be expensive and highly time-consuming. A way to reduce amount required training is adopt a semi-supervised approach. In this regard, generative models, concretely Generative Adversarial Networks (GANs), have been adapted tasks. This work proposes MaskGDM, architecture combining some ideas from EditGAN, GAN that jointly their segmentations, together with diffusion model. With careful integration, we find model improve EditGAN performance results in multiple datasets, both multi-class binary labels. According quantitative obtained, proposed improves when compared DatasetGAN respectively, by [Formula: see text] text]. Moreover, ISIC dataset, our proposal other up

Language: Английский

---

A response‐compatible ground motion generation method using physics‐guided neural networks

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2024, Volume and Issue: 39(15), P. 2350 - 2366

Published: March 31, 2024

Abstract Selecting or generating ground motions (GMs) that elicit seismic responses matching specific standards expected benchmarks for nonlinear time‐history analysis (NLTHA) is crucial ensuring the rationality of structural design and analysis. Typical GM inputs NLTHA, either natural artificial, are normally spectrum‐compatible, which may produce significant variations in results, even using multiple GMs. This paper introduces a response‐compatible motion generation (RCGMG) method GMs tailored to be response‐compatible. NLTHA results only few these artificial can closely approximate mean from large set spectrum‐compatible target responses. The RCGMG adopts response diagram time domain (RDTD) characterize nonstationary features their impacts on dynamic A physics‐guided conditional generative adversarial network developed RDTDs with These generated then mapped into through feedforward neural network. To verify effectiveness RCGMG, different structure models under various site conditions spectra conducted. Seismic RCGMG‐generated compared demonstrate closer responses, fewer robust generalization performance.

Language: Английский

---

Intelligent evaluation of pavement friction at high speeds with artificial intelligence powered three-dimensional laser imaging technology

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 150, P. 110580 - 110580

Published: March 23, 2025

Language: Английский

---

A survey of generative models for image-based structural health monitoring in civil infrastructure

Journal of Infrastructure Intelligence and Resilience, Journal Year: 2025, Volume and Issue: unknown, P. 100138 - 100138

Published: Jan. 1, 2025

Language: Английский

---

Network models for temporal data reconstruction for dam health monitoring

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 12, 2025

Abstract The reconstruction of monitoring data is an important step in the process structural health monitoring. Monitoring involves generating values that are close to true or expected values, and then using generated replace anomalous fill missing data. Deep learning models can be used reconstruct dam data, but current suffer from inabilities when dataset significantly incomplete, accuracy speed have needs for improvement. To this end, paper proposes a temporal nets (DTRN) based on generative adversarial nets, which accurately cases incomplete datasets. improve embeds gated recurrent unit network sequence‐to‐sequence model into DTRN extract features In addition, given random matrices with different distributions lead results, maximum probability multiple filling adopted. Finally, several experiments show (1) not only applicable various types (e.g., displacement seepage pressure seam gauge etc.) also applied other relatively smooth time series (2) average root mean square error (0.0618) indicates its 92.3%, 57.5%, 71.99% higher than imputation (GAIN), timing GAIN (TGAIN), (DMDRN), respectively. (3) elapsed (522.6 s) 68.45% 48.10% shorter TGAIN DMDRN,

Language: Английский

---

Underwater bridge pier morphology measurement method via refraction correction and multi‐camera calibration

Computer-Aided Civil and Infrastructure Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 21, 2025

Abstract Underwater structural inspection is essential for ensuring the safety and longevity of bridges. To improve efficiency accuracy these inspections, this paper presents a method measuring morphology bridge piers through refraction correction multi‐camera calibration. Using an underwater visual platform with appropriate lighting, measurement equipment mitigates low visibility challenges. A coplanar camera parameter calibration based on encoded markers proposed to reduce effects refraction, along development multi‐refraction model. Additionally, novel extrinsic introduced stitch point clouds. comparative analysis two methods, conducted both in air underwater, has been performed validate approach. Finally, circular cross‐section shape pier was successfully measured, results defect localization were effectively presented.

Language: Английский

---

Image Super-Resolution Reconstruction Algorithm Based on SRGAN and Swin Transformer

Symmetry, Journal Year: 2025, Volume and Issue: 17(3), P. 337 - 337

Published: Feb. 24, 2025

Existing methods have problems such as loss of details and insufficient reconstruction effect when processing complex images. To improve the quality efficiency image super-resolution reconstruction, this study proposes an improved algorithm based on generative adversarial network Swin Transformer. Firstly, ground traditional network, combined with global feature extraction capability Transformer, model’s capacity to capture multi-scale features restore is enhanced. Subsequently, by utilizing perceptual further optimize training process, image’s visual improved. The results show that optimization had high PSNR structural similarity index values in multiple benchmark test datasets, highest reaching 43.81 0.94, respectively, which are significantly better than comparison algorithm. In practical applications, demonstrated higher accuracy reconstructing images textures rich edge details. could reach 98.03%, time was low 0.2 s or less. summary, model can greatly details, reduce detail loss, provide efficient reliable solution for tasks.

Language: Английский

---

Scaling from macro to micro: A novel approach to bridging gaps in multiple pavement texture scales using generative neural networks

Transportation Research Part C Emerging Technologies, Journal Year: 2025, Volume and Issue: 174, P. 105108 - 105108

Published: March 25, 2025

Language: Английский

---