Onsite video mining for construction hazards identification with visual relationships

Xiong, Ruoxin; Song, Yong; Li, Heng; Wang, Yu-Xuan

doi:10.1016/j.aei.2019.100966

Cited by 37 publications

(14 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Risk identification technology based on the semantic network is a method to provide semantic representation to reuse domain knowledge, but it is accompanied by a large amount of manual rule extraction. For example, Xiong et al (2019) used computer vision techniques to extract semantic representations from videos. The automatic hazard identification system had been proved to have good performance in the detection of workers not wearing safety helmets and walking beneath the crane, besides aiming at the problem of knowledge management confusion caused by the complexity of construction safety information.…”

Section: Intelligent Safety Risk Identificationmentioning

confidence: 99%

See 1 more Smart Citation

Integrating knowledge management and BIM for safety risk identification of deep foundation pit construction

Chen

Zhou

Duan

et al. 2022

ECAM

View full text Add to dashboard Cite

PurposeThe outbreak of COVID-19 pandemic has posed severe challenges to infrastructure construction in China. Particularly, the complex technology and high process uncertainty of deep foundation pit construction make its safety risk identification a challenging issue of general concern. To address these challenges, Building Information Modeling (BIM) can be used as an important tool to enhance communication and decision-making among stakeholders during the pandemic. The purpose of this study is to propose a knowledge management and BIM-integrated safety risk identification method for deep foundation pit construction to improve the management efficiency of project participants.Design/methodology/approachThis paper proposes a risk identification method that integrates BIM and knowledge management for deep foundation pit construction. In the framework of knowledge management, the topological relationships between objects in BIM are extracted and visualized in the form of knowledge mapping. After that, formal expressions of codes are established to realize the structured processing of specification provisions and special construction requirements. A comprehensive plug-in for deep foundation pit construction is designed based on the BIM software.FindingsThe proposed method was verified by taking a sub-project in deep foundation pit project construction as an example. The result showed the new method can make full use of the existing specification and special engineering requirements knowledge. In addition, the developed visual BIM plug-in proves the feasibility and applicability of the proposed method, which can help to increase the risk identification efficiency and refinement.Originality/valueThe deep foundation pit safety risk identification is challenged by the confusion of deep foundation pit construction safety knowledge and the complexity of the BIM model. By establishing the standardized expression of normative knowledge and special construction requirements, the efficiency and refinement of risk identification are improved while ensuring the comprehensiveness of results. Moreover, the topology-based risk identification method focuses on the project objects and their relations in the way of network, eliminating the problem of low efficiency from the direct BIM-based risk identification method due to massive data.

show abstract

Section: Intelligent Safety Risk Identificationmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Integrating knowledge management and BIM for safety risk identification of deep foundation pit construction

Chen

Zhou

Duan

et al. 2022

ECAM

View full text Add to dashboard Cite

show abstract

“…Such ontologies have been applied for safety hazard identification in other studies. Xiong et al (2019) and Fang et al (2020d) used a safety-specific ontology when understanding visual relationships and identifying hazardous factors from jobsite images, and Zhang et al (2020c) evaluated spatial risks between construction workers and equipment by feeding only the object detection results into a safety ontology. Liu et al (2020) also leveraged an ontology to understand and describe construction scene images in natural language.…”

Section: Object-to-object Interaction Analysismentioning

confidence: 99%

Visual Analytics for Operation-Level Construction Monitoring and Documentation: State-of-the-Art Technologies, Research Challenges, and Future Directions

Kim

2020

Front. Built Environ.

View full text Add to dashboard Cite

Operation-level vision-based monitoring and documentation has drawn significant attention from construction practitioners and researchers. To automate the operation-level monitoring of construction and built environments, there have been much effort to develop computer vision technologies. Despite their encouraging findings, it remains a major challenge to exploit technologies in real construction projects, implying that there are knowledge gaps in practice and theory. To fill such knowledge gaps, this study thoroughly reviews 119 papers on operation-level vision-based construction monitoring, published in mainstream construction informatics journals. Existing research papers can be categorized into three sequential technologies: (1) camera placement for operation-level construction monitoring, (2) single-camera-based construction monitoring and documentation, and (3) multi-camera-based onsite information integration and construction monitoring. For each technology, state-of-the-art algorithms, open challenges, and future directions are discussed.

show abstract

“…Such a triplet is often called a subject-predicate-object (SPO) triplet [23]. Due to the expressiveness and flexibility of SPO triplets, they have been widely used to capture regulatory knowledge in the construction industry [24,25]. Accordingly, the triplet representation can also be used to represent various types of constraints expressed in building codes.…”

Section: Semantic Types Of Named Entities and Relationsmentioning

confidence: 99%

Joint Extraction of Multiple Relations and Entities from Building Code Clauses

Song

Shan

2020

Applied Sciences

Self Cite

View full text Add to dashboard Cite

The extraction of regulatory information is a prerequisite for automated code compliance checking. Although a number of machine learning models have been explored for extracting computer-understandable engineering constraints from code clauses written in natural language, most are inadequate to address the complexity of the semantic relations between named entities. In particular, the existence of two or more overlapping relations involving the same entity greatly exacerbates the difficulty of information extraction. In this paper, a joint extraction model is proposed to extract the relations among entities in the form of triplets. In the proposed model, a hybrid deep learning algorithm combined with a decomposition strategy is applied. First, all candidate subject entities are identified, and then, the associated object entities and predicate relations are simultaneously detected. In this way, multiple relations, especially overlapping relations, can be extracted. Furthermore, nonrelated pairs are excluded through the judicious recognition of subject entities. Moreover, a collection of domain-specific entity and relation types is investigated for model implementation. The experimental results indicate that the proposed model is promising for extracting multiple relations and entities from building codes.

show abstract

Onsite video mining for construction hazards identification with visual relationships

Cited by 37 publications

References 36 publications

Integrating knowledge management and BIM for safety risk identification of deep foundation pit construction

Integrating knowledge management and BIM for safety risk identification of deep foundation pit construction

Visual Analytics for Operation-Level Construction Monitoring and Documentation: State-of-the-Art Technologies, Research Challenges, and Future Directions

Joint Extraction of Multiple Relations and Entities from Building Code Clauses

Contact Info

Product

Resources

About