Related Experiment Video
Updated: Jul 10, 2025
03:31
Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
Published on: December 15, 2023
559
Deep Learning Based Fire Risk Detection on Construction Sites
1Vibration Engineering Section, Faculty of Environment, Science, and Economics, University of Exeter, Exeter EX4 4QF, UK.
Sensors (Basel, Switzerland)
|November 25, 2023
Summary
Computer vision technology can proactively detect construction site fire risks by identifying ignition sources and combustible materials. Yolov5 deep learning model demonstrated superior performance and learning efficiency for this crucial fire prevention task.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Fire Safety Engineering
Background:
- Large-scale fires on South Korean construction sites necessitate advanced fire risk detection.
- Current methods lack proactive capabilities to prevent fire incidents.
Purpose of the Study:
- To develop a proactive fire risk detection system using computer vision.
- To identify the coexistence of ignition sources (sparks) and combustible materials (urethane foam, Styrofoam).
Main Methods:
- Object detection applied to surveillance camera images.
- Statistical analysis of South Korean construction site fire data.
- Comparison of deep learning models Yolov5 and EfficientDet.
Main Results:
- Yolov5 models achieved mean Average Precision (mAP) from 87% to 90%.
- EfficientDet models achieved mAP from 82% to 87%.
- Yolov5 demonstrated advantages in learning speed and ease.
Conclusions:
- Computer vision, specifically Yolov5, offers a viable solution for proactive fire risk detection on construction sites.
- The system's effectiveness can be enhanced through improved labeling and long-distance object detection.
Related Concept Videos
01:19Survival Tree
87
Survival trees are a non-parametric method used in survival analysis to model the relationship between a set of covariates and the time until an event of interest occurs, often referred to as the "time-to-event" or "survival time." This method is particularly useful when dealing with censored data, where the event has not occurred for some individuals by the end of the study period, or when the exact time of the event is unknown.
Building a Survival Tree
Constructing a...
Building a Survival Tree
Constructing a...
87
01:28Survey Safety
30
Surveying near highways, rough terrain, or power lines involves significant risks. Working along highways is particularly dangerous and requires the use of warning signs and flagmen. It is safest to avoid working directly on roads and use offsets whenever possible. When highway work is unavoidable, it must follow all safety guidelines. Surveyors should wear bright clothing, such as orange reflective vests, to ensure visibility to motorists, coworkers, and hunters. In construction zones, wearing...
30