A Human Detection Approach for Intrusion in Hazardous Areas Using 4D-BIM-Based Spatial-Temporal Analysis and Computer Vision
Abstract
:1. Introduction
- Considering geometric information of the hazardous areas from the BIM model, which is defined through a rule-based model from the plan, then extracting and transforming it into objects for supporting the vision monitoring system.
- The IDC4D also considers the construction schedule through dividing phases during the construction process. Accordingly, the hazardous area in the field of view of a virtual camera can be transformed into a suitable object before construction.
- The prototype system was developed to validate the IDC4D, and the results show the efficiency of IDC4D supported the planner in selecting a plan and detecting the event of on-site persons entering hazardous areas below the scaffolding work.
2. Literature Review
2.1. Preventing Entry to Hazardous Areas on Construction Jobsites
2.2. Intrusion Detection in Hazardous Areas Based on Computer Vision
2.3. BIM for Construction Safety
2.4. Need for BIM-Based Information Extraction to Support Computer Vision System
- Monitoring on-site persons entering hazardous areas: The proposed approach must detect on-site persons in the field of view of the camera and the correlation between them and the hazardous areas.
- Multiple workspace shapes: The approach must detect different hazardous areas at construction jobsites as stipulated in the laws and regulations.
- Inefficient monitoring by the site safety manager: Manually monitoring every construction on-site person is challenging for site managers. Therefore, the suggested monitoring approach must utilize computer vision and BIM to ensure on-site person safety in the danger zone.
- Inconvenience: During the construction phase, the hazardous area in the same workspace may change due to shifting to other activities as stipulated in the schedule. Accordingly, the laborers have to reinstall the sensor tags for different purposes. Hence, the proposed system should predefine the hazardous area before construction to monitor intrusion.
3. Development of the Proposed Approach
3.1. A 4D BIM-Based Safety Planning Module (4BSP)
3.2. Hazardous Areas Registration Module (HAR)
3.3. Hazardous Area Intrusion Detection Module (HAID)
3.3.1. Hazardous-Area Extraction
Algorithm 1. Algorithm of color-based extraction |
Input: Register virtual image (image) Output: The contour zone around the hazardous zone and the intersection point (intersection Color) Library: OpenCV 1: Read image 2: Set “lower” and “upper” RGB range to obtain the hazardous by color 3: Obtain the zone by masking and bitwise the mask 4: Define the kernel for dilation task and perform dilate —kernel size (5 × 5) —dilate (image, kernel, iteration = 1) 5: Find the biggest contour of the white pixel; —maxContour = findMax(findContours(image), key = contourArea) 6: Simplify the biggest contour shape —epsilon = thresholdValue * arcLength(maxContour) —finalContour = approxPolyDP(maxContour, epsilon) 7: Find the intersection coordinate of maxContour —for i in range(finalContour.shape [0]—1): drawline((finalContour[i][0]), (finalContour[i + 1][0])) intersectionCoor.append(tuple(approx[i][0])) |
3.3.2. Computer Vision-Based On-Site Persons Detection
3.3.3. Object Correlation Analysis
3.4. Development of Prototype System Based on Proposed Approach
Dataset Preparation
4. Prototype System Application
4.1. Case Application
4.2. Results
5. Discussions
- Construction sites often contain hazardous areas, and on-site people need to be aware of these areas and take the necessary safety precautions. Common hazardous areas include areas where cranes, forklifts, supply vehicles (especially when driven in the winter), heavy equipment, and hazardous materials are present. Thus, the vision intelligence system needs to have the capability of monitoring multiple types of hazardous-area shapes as stipulated in the relevant laws and regulations. However, previous studies have focused on specific hazardous areas, such as areas around trucks and rectangular openings. In our study, the 3D information on hazardous areas, defined through a rule-based model from the plan, was extracted and transformed to objects for supporting the vision monitoring system. The color-based extraction process was then applied to extract the orange area from registered virtual images. Afterward, the orange-area boundary was extracted as polygons. Accordingly, the hazardous area around the fixed and mobile scaffolds was determined using the intrusion detection algorithm.
- Surveillance camera systems are installed at construction jobsites to monitor potential hazards at specific workspaces. Notably, the hazardous area in the same workspace may change due to shifting to different activities as per the schedule. This study investigated a 4D BIM model that divided the schedule into phases by extracting the hazardous area of each site layout. Accordingly, the hazardous area in the field of view of a virtual camera can be transformed to a suitable object before construction. In the case study, the hazardous areas were defined at phases 1 and 2, which covered the potential hazard area around the scaffolding system, as depicted in Figure 13. In addition, the application of 4D BIM information as a foundation for extracting hazardous-area data assures that the technique may be implemented across projects with diverse scales and levels of intricacy, which may be customized to incorporate the distinctive attributes of the project. The scalability of the project enables the inclusion of developing hazard zones at different phases of development, facilitating their dynamic incorporation. Moreover, a notable characteristic of this technique lies in its capacity to adapt to various sorts of hazardous locations and site layouts. Transferring these hazardous areas as objects into actual images contributes to identifying potential intrusion detection in hazardous areas.
- The IDC4D approach effectively facilitates hazardous-area entry monitoring by establishing a correlation between on-site personnel and designated hazardous zones. This innovative approach leverages 4D Building Information Modeling (BIM) data to extract hazardous area information from virtual environments, subsequently transferring these areas as objects into actual images. Concurrently, on-site person-detection algorithms were developed to identify individuals within the camera’s field of view. Notably, when the identified on-site person objects coincided with the hazardous area objects, an alert was generated and transmitted to safety managers. For instance, Figure 14 illustrated the monitoring results when on-site persons were or were not working near the scaffolding area. In addition, this adaptable approach demonstrates its potential applicability across diverse projects, underlining its versatility in ensuring safety compliance and hazard prevention. Leveraging 4D BIM and camera-based person detection can seamlessly integrate with different project scales, hazardous area types, and site layouts. This versatility positions the approach as a valuable tool for enhancing safety monitoring across various construction endeavors.
6. Conclusions
Author Contributions
Funding
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Zhang, M.; Shi, R.; Yang, Z. A critical review of vision-based occupational health and safety monitoring of construction site workers. Saf. Sci. 2020, 126, 104658. [Google Scholar] [CrossRef]
- Fang, W.; Ding, L.; Love, P.E.D.; Luo, H.; Li, H.; Peña-Mora, F.; Zhong, B.; Zhou, C. Computer vision applications in construction safety assurance. Automat. Constr. 2020, 110, 103013. [Google Scholar] [CrossRef]
- OSHA Fatality Report. Available online: https://www.osha.gov/stop-falls (accessed on 6 October 2022).
- Construction Work|Statistics Korea. Available online: http://kostat.go.kr/portal/eng/pressReleases/4/5/index.board (accessed on 6 October 2022).
- Accidents at Work Statistics—Statistics Explained. Available online: https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Accidents_at_work_statistics#Number_of_accidents (accessed on 2 February 2023).
- Tran, S.V.T.; Khan, N.; Lee, D.; Park, C. A Hazard Identification Approach of Integrating 4D BIM and Accident Case Analysis of Spatial–Temporal Exposure. Sustainability 2021, 13, 2211. [Google Scholar] [CrossRef]
- Luo, X.; Li, H.; Wang, H.; Wu, Z.; Dai, F.; Cao, D. Vision-based detection and visualization of dynamic workspaces. Automat. Constr. 2019, 104, 1–13. [Google Scholar] [CrossRef]
- Bao, L.; Tran, S.V.T.; Nguyen, T.L.; Pham, H.C.; Lee, D.; Park, C. Cross-platform virtual reality for real-time construction safety training using immersive web and industry foundation classes. Automat. Constr. 2022, 143, 104565. [Google Scholar] [CrossRef]
- Chen, H.; Li, H.; Goh, Y.M. A review of construction safety climate: Definitions, factors, relationship with safety behavior and research agenda. Saf. Sci. 2021, 142, 105391. [Google Scholar] [CrossRef]
- Jin, R.; Zhang, H.; Liu, D.; Yan, X. IoT-based detecting, locating and alarming of unauthorized intrusion on construction sites. Automat. Constr. 2020, 118, 103278. [Google Scholar] [CrossRef]
- Park, J.; Asce, S.M.; Marks, E.; Cho, Y.K.; Suryanto, W. Performance Test of Wireless Technologies for Personnel and Equipment Proximity Sensing in Work Zones. J. Constr. Eng. Manag. 2015, 142, 04015049. [Google Scholar] [CrossRef]
- Fang, W.; Zhong, B.; Zhao, N.; Love, P.E.D.; Luo, H.; Xue, J.; Xu, S. A deep learning-based approach for mitigating falls from height with computer vision: Convolutional neural network. Adv. Eng. Inform. 2019, 39, 170–177. [Google Scholar] [CrossRef]
- Khan, M.; Khalid, R.; Anjum, S.; Tran, S.V.-T.; Park, C. Fall Prevention from Scaffolding Using Computer Vision and IoT-Based Monitoring. J. Constr. Eng. Manag. 2022, 148, 04022051. [Google Scholar] [CrossRef]
- Yang, J.; Lee, D.; Baek, C.; Park, C.; Lan, B.Q.; Lee, D. Leveraging Blockchain for Scaffolding Work Management in Construction. IEEE Access 2022, 10, 39220–39238. [Google Scholar] [CrossRef]
- Kim, H.; Lee, H.S.; Park, M.; Chung, B.Y.; Hwang, S. Automated hazardous area identification using laborers’ actual and optimal routes. Automat. Constr. 2016, 65, 21–32. [Google Scholar] [CrossRef]
- Mei, X.; Zhou, X.; Xu, F.; Zhang, Z. Human Intrusion Detection in Static Hazardous Areas at Construction Sites: Deep Learning–Based Method. J. Constr. Eng. Manag. 2023, 149, 04022142. [Google Scholar] [CrossRef]
- Akram, R.; Thaheem, M.J.; Nasir, A.R.; Ali, T.H.; Khan, S. Exploring the role of building information modeling in construction safety through science mapping. Saf. Sci. 2019, 120, 456–470. [Google Scholar] [CrossRef]
- Alizadehsalehi, S.; Yitmen, I.; Celik, T.; Arditi, D. The effectiveness of an integrated BIM/UAV model in managing safety on construction sites. Int. J. Occup. Saf. Ergon. 2018, 26, 829–844. [Google Scholar] [CrossRef] [PubMed]
- Khan, M.; Nnaji, C.; Khan, M.S.; Ibrahim, A.; Lee, D.; Park, C. Risk factors and emerging technologies for preventing falls from heights at construction sites. Autom. Constr. 2023, 153, 104955. [Google Scholar] [CrossRef]
- Huang, M.Q.; Ninić, J.; Zhang, Q.B. BIM, machine learning and computer vision techniques in underground construction: Current status and future perspectives. Tunn. Undergr. Space Technol. 2021, 108, 103677. [Google Scholar] [CrossRef]
- Xu, W.; Wang, T.K. Dynamic safety prewarning mechanism of human–machine–environment using computer vision. Eng. Constr. Archit. Manag. 2020, 27, 1813–1833. [Google Scholar] [CrossRef]
- Yang, B.; Zhang, B.; Zhang, Q.; Wang, Z.; Dong, M.; Fang, T. Automatic detection of falling hazard from surveillance videos based on computer vision and building information modeling. Struct. Infrastruct. Eng. 2022, 18, 1049–1063. [Google Scholar] [CrossRef]
- Tang, S.; Shelden, D.R.; Eastman, C.M.; Pishdad-Bozorgi, P.; Gao, X. A review of building information modeling (BIM) and the internet of things (IoT) devices integration: Present status and future trends. Automat. Constr. 2019, 101, 127–139. [Google Scholar] [CrossRef]
- Tran, S.V.; Bao, L.Q.; Nguyen, L.T.; Pedro, A. Development of Computer Vision and BIM-cloud based Automated Status Updating for Construction Safety Monitoring. In Proceedings of the 22nd International Conference on Construction Applications of Virtual Reality (CONVR2022), Seoul, Republic of Korea, 16–19 November 2022. [Google Scholar]
- Winsemius, W. Some ergonomic aspects of safety. Ergonomics 2007, 8, 151–162. [Google Scholar] [CrossRef]
- CPWR|Practice Trench Safety. It Saves Lives. Available online: https://www.cpwr.com/practice-trench-safety-it-saves-lives/ (accessed on 8 December 2022).
- Hinze, J.; Godfrey, R. An evaluation of safety performance measures for construction projects. J. Constr. Res. 2003, 04, 5–15. [Google Scholar] [CrossRef]
- Xu, Q.; Chong, H.Y.; Liao, P.C. Collaborative information integration for construction safety monitoring. Automat. Constr. 2019, 102, 120–134. [Google Scholar] [CrossRef]
- Kanan, R.; Elhassan, O.; Bensalem, R. An IoT-based autonomous system for workers’ safety in construction sites with real-time alarming, monitoring, and positioning strategies. Automat. Constr. 2018, 88, 73–86. [Google Scholar] [CrossRef]
- Shuang, D.; Heng, L.; Skitmore, M.; Qin, Y. An experimental study of intrusion behaviors on construction sites: The role of age and gender. Saf. Sci. 2019, 115, 425–434. [Google Scholar] [CrossRef]
- Huang, H.; Hu, H.; Xu, F.; Zhang, Z.; Tao, Y. Skeleton-based automatic assessment and prediction of intrusion risk in construction hazardous areas. Saf. Sci. 2023, 164, 106150. [Google Scholar] [CrossRef]
- Wan, H.-P.; Zhang, W.-J.; Ge, H.-B.; Luo, Y.; Todd, M.D. Improved Vision-Based Method for Detection of Unauthorized Intrusion by Construction Sites Workers. J. Constr. Eng. Manag. 2023, 149, 04023040. [Google Scholar] [CrossRef]
- Kim, D.; Liu, M.; Lee, S.H.; Kamat, V.R. Remote proximity monitoring between mobile construction resources using camera-mounted UAVs. Automat. Constr. 2019, 99, 168–182. [Google Scholar] [CrossRef]
- Hua, Y.; He, J.; Gong, J.; Zhao, J. Hazardous Area Risk-Based Evacuation Simulation and Analysis of Building Construction Sites. J. Constr. Eng. Manag. 2020, 146, 5. [Google Scholar] [CrossRef]
- Tran, S.; Ali, A.K.; Khan, N.; Lee, D.; Park, C. A Framework for Camera Planning in Construction Site using 4D BIM and VPL. In Proceedings of the 37th International Symposium on Automation and Robotics in Construction (ISARC), Kitakyushu, Japan, 27–28 October 2020. [Google Scholar]
- Feng, C.-W.; Lu, S.-W. Using BIM to Automate Scaffolding Planning for Risk Analysis at Construction Sites. In Proceedings of the 34th International Symposium on Automation and Robotics in Construction, ISARC 2017, Taipei, Taiwan, 28 June–July 2017; International Association for Automation and Robotics in Construction: Edinburgh, UK, 2017; pp. 610–617. [Google Scholar]
- Getuli, V.; Capone, P.; Bruttini, A.; Isaac, S. BIM-based immersive Virtual Reality for construction workspace planning: A safety-oriented approach. Automat. Constr. 2020, 114, 103160. [Google Scholar] [CrossRef]
- Tran, S.V.T.; Nguyen, T.L.; Chi, H.L.; Lee, D.; Park, C. Generative planning for construction safety surveillance camera installation in 4D BIM environment. Automat. Constr. 2022, 134, 104103. [Google Scholar] [CrossRef]
- Arslan, M.; Cruz, C.; Ginhac, D. Visualizing intrusions in dynamic building environments for worker safety. Saf. Sci. 2019, 120, 428–446. [Google Scholar] [CrossRef]
- Valinejadshoubi, M.; Moselhi, O.; Bagchi, A.; Salem, A. Development of an IoT and BIM-based automated alert system for thermal comfort monitoring in buildings. Sustain. Cities Soc. 2021, 66, 102602. [Google Scholar] [CrossRef]
- Zhou, X.; Li, H.; Wang, J.; Zhao, J.; Xie, Q.; Li, L.; Liu, J.; Yu, J. CloudFAS: Cloud-based building fire alarm system using Building Information Modelling. J. Build. Eng. 2022, 53, 104571. [Google Scholar] [CrossRef]
- Tran, S.V.T.; Nguyen, T.L.; Park, C. A BIM Integrated Hazardous Zone Registration Using Image Stitching. In Proceedings of the 38th International Symposium on Automation and Robotics in Construction, Dubai, United Arab Emirates, 2–4 November 2021; pp. 176–181. [Google Scholar] [CrossRef]
- Lin, T.Y.; Maire, M.; Belongie, S.; Hays, J.; Perona, P.; Ramanan, D.; Dollár, P.; Zitnick, C.L. Microsoft COCO: Common objects in context. In Lecture Notes in Computer Science; Springer: Berlin/Heidelberg, Germany, 2014; pp. 740–755. [Google Scholar] [CrossRef]
- Arya, K.V.; Bhadoria, R.S. The Biometric Computing: Recognition and Registration; CRC Press: Boca Raton, FL, USA, 2019; ISBN 1351013424. [Google Scholar]
Reference | Solution Approach | Determining On-Site Persons | Type of Hazardous Area | Method of Hazardous Area Annotation |
---|---|---|---|---|
Xu [28] | Sensor | Sensor embedded | Opening | signal range of sensors |
Jin [10] | Sensor | Sensor embedded | Opening | signal range of sensors |
Huang [31] | Computer vision | Skeleton identification | Opening, Lifting operation area | manual annotation |
Mei [16] | Computer vision | Object detection | Foundation pitch, opening, limb, scaffold | manual annotation |
Kim [15] | Sensor | Sensor embedded | Extracted geometric information from BIM | The 3D information on hazardous areas was extracted |
Wan [32] | Computer vision | Helmet color detection | Foundation pitch, opening, limb, scaffold, machinery | manual annotation |
Kim [33] | Computer vision | Object detection | Area around excavator | excavator object annotation |
Park [11] | Sensor | Sensor embedded | Construction equipment | signal range of sensors |
This study | Computer vision + 4D BIM | Object detection | Extracted geometric information from 4D BIM | The 3D information on hazardous areas was extracted and transformed into objects for supporting the vision monitoring system |
Reference | Functionality | Benefits for Safety Management |
---|---|---|
Tran [6] | 4D BIM | Identifying hazards of spatial–temporal exposure cases |
Arslan [39] | 3D visualization | Visualization of intrusions in dynamic construction settings with the objective of enhancing worker safety |
Feng [36] | Rule-based modeling | Using BIM to automate planning for scaffolding on construction sites |
Tran [38] | 4D BIM | Developing a surveillance camera installation plan for safety purposes |
Zhou [41] | Cloud-based BIM | Developing building fire alarm system |
Hazardous Entity | Rule-Based Analysis | |
---|---|---|
Condition | Rule Modeling | |
Foundation ditch | Depth > 2 m | Edge extends outwards by 1 m |
- | Edge extends outwards/inwards by 1 m | |
Opening | 0.5 m < dimension< 1 m | Edge extends outwards by 0.5 m |
dimension > 1 m | Edge extends outwards by 1 m | |
Limb | Height > 2 m | Edge extends outwards by 1 m |
Scaffold | Height > 2 m | Above the operation area |
- | Vertical projection extends outwards by 2 m |
Type | Actual Intrusion | Predicted Intrusion |
---|---|---|
TP | Yes | Yes |
FN | Yes | No |
FP | No | Yes |
TN | No | No |
Performance Matrices | Precision | Recall | F1-Score | Accuracy | |
---|---|---|---|---|---|
Phase | |||||
1 | 95.3% | 83.7% | 89.1% | 92.5% | |
2 | 95.1% | 87.9% | 91.4% | 94.1% |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Tran, S.V.-T.; Lee, D.; Bao, Q.L.; Yoo, T.; Khan, M.; Jo, J.; Park, C. A Human Detection Approach for Intrusion in Hazardous Areas Using 4D-BIM-Based Spatial-Temporal Analysis and Computer Vision. Buildings 2023, 13, 2313. https://doi.org/10.3390/buildings13092313
Tran SV-T, Lee D, Bao QL, Yoo T, Khan M, Jo J, Park C. A Human Detection Approach for Intrusion in Hazardous Areas Using 4D-BIM-Based Spatial-Temporal Analysis and Computer Vision. Buildings. 2023; 13(9):2313. https://doi.org/10.3390/buildings13092313
Chicago/Turabian StyleTran, Si Van-Tien, Doyeop Lee, Quy Lan Bao, Taehan Yoo, Muhammad Khan, Junhyeon Jo, and Chansik Park. 2023. "A Human Detection Approach for Intrusion in Hazardous Areas Using 4D-BIM-Based Spatial-Temporal Analysis and Computer Vision" Buildings 13, no. 9: 2313. https://doi.org/10.3390/buildings13092313