저비용 드론을 이용한 딥러닝 기반 콘크리트 건축물 안전진단 자동화 시스템
Deep Learning-Based Safety Inspection Automation System for Concrete Buildings with Low-Cost Drones
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
건축물의 안전과 기능 유지를 위한 안전 점검은 건물 노후화 및 관리 미비로 인한 대형 사상 사고를 예방하기 위한 효과적인 대비책이다. 하지만, 현장 방문을 통한 건축물 점검을 정기적으로 수행할 검사원의 수가 제한되고, 검사원의 숙련도와 주관적 평가에 의해 점검 결과가 크게 영향을 받을 수 있는 점은 개선이 필요하다. 본 논문에서는 현행 건축물 안전 점검의 한계와 제약을 개선하기 위해 비용 효율적이며 객관적인 정량 계측이 가능한 안전 점검 자동화 기술을 제안한다. 제안하는 기법은 콘크리트 건축물의 실내 균열 탐지를 목표로 하며, 자율비행 드론을 이용한 현장 조사와 딥러닝 비전 기반의 균열탐지 기법으로 구성된다. 본 논문은 안전점검 활동의 비용 절감을 위해 저가 드론 사용을 제안하며, 고도 유지 및 고정 거리 비행 기능을 통해 실내 환경에서 음영지역 없이 자율적으로 현장 조사를 수행하고 건축물 내부 이미지 및 좌표 정보를 수집하기 위한 알고리즘을 제안한다. 안전 점검 기간 최소화를 위해 적은 수의 이미지로 효과적 균열 탐지가 가능한 DRAEM 딥러닝 모델을 사용하며, 균열 탐지 정확도 개선을 위해 임계치 기반의 픽셀 마스킹 기법을 제안한다. 실 환경에서의 테스트를 통해 제안하는 시스템의 자율적 현장 조사 및 균열 탐지 기능의 정확성을 검증했다.
Abstract
The safety inspection for guaranteeing the safety and functionality of buildings is an effective measure to prevent hazardous accidents caused by aging and inadequate management of buildings. However, there is a need for improvement as the number of inspectors who perform on-site inspections regularly is limited, and the inspection results can be significantly influenced by the skill level of inspectors and subjective evaluations. In this paper, we propose a cost-effective and objective safety inspection automation system to overcome the limitations and constraints of current building safety inspections. The proposed technique focuses on detecting indoor cracks in concrete structures and consists of on-site investigation using autonomous drones and crack detection using deep learning vision. In this paper we propose the use of low-power drones to reduce the cost of safety inspection activities and propose algorithms for autonomous on-site investigation in indoor environments without shadow areas through altitude maintenance and fixed-distance flight capabilities and collecting images along with their location information. To minimize the inspection time, the proposed system utilizes the DRAEM deep learning model, which allows effective crack detection with a small number of images, and proposes a threshold-based pixel masking technique to improve crack detection accuracy. The proposed system has been validated through tests in real-world environments.
참고문헌 (34)
- Korean Law Information Center, "Construction technology promotion act, No. 18933, Enf. Date:10 Jun. 2022", ~ (2022)
- Research on measures to improve construction supervision, "E. Hwang", Korea Institute of Construction Technology, ~ (2013)
- Korea Institute of Construction Technology, "Presentation of responsibility management system implementation results and mid- to long-term development direction", Korea Association of Construction Engineering and Management, ~ (2011)
- "<Responsible Management 20th Anniversary Special> 20 years since introduction of responsible management, site is safe, public is safe", Daehan Economy, 29 Oct. 2023, ~ (2023)
- Korean Law Information Center, "Special act on the safety control and maintenance of establishments, No. 17946, Enf. Date: 17 Sep. 2021", ~ (2021)
- H. Par, "Proposed improvements to the safety inspection system: An analysis of the current status of building disaster accidents and safety inspection systemsm", J. Korean Soc. Hazard Mitig, 19(5), pp.11~21 (2019)
- S. Lee, "Development of robust crack segmentation and thickness measurement model using deep learning", Journal of Korean Institute of Communications and Information Sciences, 48(5), pp.554~566 (2023)
- A. Kim, "Crack detection of concrete structure using deep learning and image processing method in geotechnical engineering", Journal of the Korean Geotechnical Society, 34(12), pp.145~154 (2018)
- G. Kim, "Robust detection deep learning model in the various exterior wall cracks", Proceedings of the Korean Society of Computer Information Conference, 29(2), pp.53~56 (2021)
- G. Kim, "Deep learning-based exterior wall crack detection method for building safety diagnosis", Journal of The Korean Institute of Plant Engineering, 26(4), pp.31~43 (2021)
- X. Wu, "Building crack identification and total quality management method based on deep learning", Pattern Recognition Letters, 145, pp.225~231 (2021)
- M. Gavilan, "Adaptive road crack detection system by pavement classification", Sensors, 11(10), pp.9628~9657 (2011)
- Q. Zou, "CrackTree : automatic crack detection from pavement images", Pattern Recognition Letter, 33(3), pp.227~238 (2012)
- R. Amhaz, "Automatic crack detection on twodimensional pavement images : an algorithm based on minimal path selection", Transactions on Intelligent Transportation Systems, 17(10), pp.2718~2729 (2016)
- Q. Mei, "A cost effective solution for pavement crack inspection using cameras and deep neural networks", Construction and Building Materials, 256, pp.1~15 (2020)
- Y. Shi, "Automatic road crack detection using random structured forests", Transactions on Intelligent Transportation Systems, 17(12), pp.3434~3445 (2016)
- Y. Hu, "Automatic pavement crack detection using texture and shape descriptors", IETE Technical Review, 27(5), pp.398~405 (2014)
- S. Mathavan, "Use of a self-organizing map for crack detection in highly textured pavement images", Journal of Infrastructure Systems, 21(3), pp.04014052~ (2014)
- J. Ha, "A development of road crack detection system using deep learning-based segmentation and object detection", Journal of Society for e-Business Studies, 26(1), pp.93~106 (2021)
- Y. Hamishebahar, "A comprehensive review of deep learningbased crack detection approaches", Applied Sciences, 12(3), pp.1~24 (2022)
- X. Peng, "The feasibility assessment study of bridge crack width recognition in images based on special inspection UAV", Hindawi, 2020, pp.1~17 (2020)
- Y. Li, "A novel approach for UAV image crack detection", Sensors, 22(9), pp.1~15 (2022)
- D. Choi, "UAVDriven Structural Crack Detection and Location Determination Using Convolutional Neural Networks", Sensors, 21(8), pp.1~20 (2021)
- B. Kim, "Dataset augmentation technique for crack detection of wood building", Proceedings of the Korean Society of Computer Information Conference, 29(2), pp.645~647 (2021)
- S. Hong, "Image labeling technology analysis and training set generation model for detecting damage and cracks in road pavement", Journal of Korean Society for Geospatial Information Science, 28(4), pp.119~125 (2020)
- V. Zavrtanik, "DRAEM-A discriminatively trained reconstruction embedding for surface anomaly detection", Proceedings of the IEEE, 2021, 8330~8339
- K. Perlin, "An image synthesizer", ACM Siggraph Computer Graphics, 19(3), pp.287~296 (1985)
- K. Batzner, "Efficientad : Accurate visual anomaly detection at millisecond-level latencies", Proceedings of the IEEE, 2024, 128~138
- Z. Wang, "Image quality assessment : from error visibility to structural similarity", Transactions on Image Processing, 13(4), pp.600~612 (2004)
- O. Ronneberger, "U-net : Convolutional networks for biomedical image segmentation", Proceedings of the Medical Image Computing and Computer-Assisted Intervention (MICCAI), pp.234~241 (2015)
- T. -Y. Lin, "Focal loss for dense object detection", Transactions on Pattern Analysis and Machine Intelligence, 42(2), pp.318~327 (2020)
- M. Cimpoi, "Describing textures in the wild", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.3606~3613 (2014)
- "Geometric Image Transformations,OpenCV"[URL]
- P. Bergmann, "MVTec AD--A comprehensive real-world dataset for unsupervised anomaly detection", Proceedings of the IEEE, 2019, 9592~9600