Image classification for sub-surface crack identification in concrete dam based on borehole CCTV images using deep dense hybrid model

IF 3.9 3区环境科学与生态学 Q1 ENGINEERING, CIVIL

Stochastic Environmental Research and Risk Assessment Pub Date : 2024-07-08 DOI:10.1007/s00477-024-02743-x

Qianwei Dai, Muhammad Ishfaque, Saif Ur Rehman Khan, Yu-Long Luo, Yi Lei, Bin Zhang, Wei Zhou

{"title":"Image classification for sub-surface crack identification in concrete dam based on borehole CCTV images using deep dense hybrid model","authors":"Qianwei Dai, Muhammad Ishfaque, Saif Ur Rehman Khan, Yu-Long Luo, Yi Lei, Bin Zhang, Wei Zhou","doi":"10.1007/s00477-024-02743-x","DOIUrl":null,"url":null,"abstract":"<p>The research investigates the significance of identifying structure discontinuities, such as cracks, in concrete dams to ensure dam safety and stability. A novel automatic image classification method is developed, employing Deep Dense Transfer Learning (DDTL) with pre-trained models, including EfficientNetB1, ResNet50, and a hybrid model to identify the detection of cracks in sub-surfaces at pillow dams in Sichuan province, China. The developed model was trained, validated, and tested, with the Hybrid model demonstrating superior performance. The results showed that the DDTL models had high classification accuracies, surpassing Convolutional identification techniques for sub-surface cracks. Consequently, this study suggests that automatic image classification techniques can effectively identify and localize structural defects in concrete dams. This is an innovative approach to predicting normal borehole images and crack recognition using CCTV borehole images.</p>","PeriodicalId":21987,"journal":{"name":"Stochastic Environmental Research and Risk Assessment","volume":"23 1","pages":""},"PeriodicalIF":3.9000,"publicationDate":"2024-07-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Stochastic Environmental Research and Risk Assessment","FirstCategoryId":"93","ListUrlMain":"https://doi.org/10.1007/s00477-024-02743-x","RegionNum":3,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, CIVIL","Score":null,"Total":0}

引用次数: 0

Abstract

The research investigates the significance of identifying structure discontinuities, such as cracks, in concrete dams to ensure dam safety and stability. A novel automatic image classification method is developed, employing Deep Dense Transfer Learning (DDTL) with pre-trained models, including EfficientNetB1, ResNet50, and a hybrid model to identify the detection of cracks in sub-surfaces at pillow dams in Sichuan province, China. The developed model was trained, validated, and tested, with the Hybrid model demonstrating superior performance. The results showed that the DDTL models had high classification accuracies, surpassing Convolutional identification techniques for sub-surface cracks. Consequently, this study suggests that automatic image classification techniques can effectively identify and localize structural defects in concrete dams. This is an innovative approach to predicting normal borehole images and crack recognition using CCTV borehole images.

Abstract Image

查看原文本刊更多论文

基于钻孔 CCTV 图像的图像分类，利用深密混合模型识别混凝土大坝的地下裂缝

该研究探讨了识别混凝土大坝结构不连续性（如裂缝）对确保大坝安全和稳定的重要意义。该研究开发了一种新的自动图像分类方法，利用深度密集迁移学习（DDTL）和预先训练好的模型，包括 EfficientNetB1、ResNet50 和混合模型，来识别检测中国四川省枕木大坝下表面的裂缝。对所开发的模型进行了训练、验证和测试，其中混合模型表现出卓越的性能。结果表明，DDTL 模型具有很高的分类精度，超过了卷积识别技术对次表层裂缝的分类精度。因此，这项研究表明，自动图像分类技术可以有效地识别和定位混凝土大坝的结构缺陷。这是一种利用 CCTV 井眼图像预测正常井眼图像和裂缝识别的创新方法。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Stochastic Environmental Research and Risk Assessment 环境科学-工程：环境

CiteScore

7.10

自引率

9.50%

发文量

189

审稿时长

3.8 months

期刊介绍： Stochastic Environmental Research and Risk Assessment (SERRA) will publish research papers, reviews and technical notes on stochastic and probabilistic approaches to environmental sciences and engineering, including interactions of earth and atmospheric environments with people and ecosystems. The basic idea is to bring together research papers on stochastic modelling in various fields of environmental sciences and to provide an interdisciplinary forum for the exchange of ideas, for communicating on issues that cut across disciplinary barriers, and for the dissemination of stochastic techniques used in different fields to the community of interested researchers. Original contributions will be considered dealing with modelling (theoretical and computational), measurements and instrumentation in one or more of the following topical areas: - Spatiotemporal analysis and mapping of natural processes. - Enviroinformatics. - Environmental risk assessment, reliability analysis and decision making. - Surface and subsurface hydrology and hydraulics. - Multiphase porous media domains and contaminant transport modelling. - Hazardous waste site characterization. - Stochastic turbulence and random hydrodynamic fields. - Chaotic and fractal systems. - Random waves and seafloor morphology. - Stochastic atmospheric and climate processes. - Air pollution and quality assessment research. - Modern geostatistics. - Mechanisms of pollutant formation, emission, exposure and absorption. - Physical, chemical and biological analysis of human exposure from single and multiple media and routes; control and protection. - Bioinformatics. - Probabilistic methods in ecology and population biology. - Epidemiological investigations. - Models using stochastic differential equations stochastic or partial differential equations. - Hazardous waste site characterization.