Related Experiment Video
Updated: Jan 17, 2026

Author Spotlight: AI-Driven Trypanosome Species Detection from Microscopic Images
Published on: October 27, 2023
YOLO-based intelligent recognition system for hidden dangers at construction sites
Hang Li1, Peijian Jin1,2, Long Zhan3
1School of Emergency Science and Engineering, Jilin Jianzhu University, Changchun, Jilin, China.
This study introduces an improved YOLOv5n hazard recognition system for construction sites, enhancing safety through lightweight design and high accuracy in detecting unsafe behaviors and conditions.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Construction Safety
Background:
- Construction industry faces high accident rates.
- Existing hazard recognition systems may lack efficiency or accuracy.
Purpose of the Study:
- To develop an optimized, lightweight hazard recognition system for construction sites.
- To improve the detection of unsafe human behaviors and unsafe object conditions.
Main Methods:
- Incorporated ECA attention, ghost module, SIoU loss, and EIoU-NMS into YOLOv5n.
- Trained two ultra-small models (approx. 2.5 MBs) on a custom dataset.
- Deployed the system on a Jetson Nano B01 edge platform.
Main Results:
- Achieved high mean Average Precision (mAP@0.5) scores: 93.6% for unsafe human behaviors and 99.5% for unsafe object conditions.
- Demonstrated lightweight acceleration and improved accuracy.
- Validated efficient onsite hazard detection capabilities.
Conclusions:
- The proposed YOLOv5n-based system effectively enhances construction site safety.
- The optimized model offers a practical solution for real-time, edge-based hazard detection.
Related Concept Videos
Design Example: Identifying the Locations of Monuments in the Field Using Global Positioning System Device
Survey Safety
Detection of Black Holes
Their closest cousins are neutron stars, which are composed almost entirely of neutrons packed against each other, making them extremely dense. A neutron star has the same mass as the Sun but its diameter is only a few kilometers. Therefore, the escape velocity from their surface is close to the speed of light.
Not until the 1960s, when the first neutron...
Masking and Demasking Agents
There are many masking agents, such as cyanide, fluoride, triethanolamine, thiourea, and 2,3-bis(sulfanyl)propan-1-ol (formerly 2,3-dimercapto-1-propanol), with the masking agent chosen based on...
Force Classification
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Hazard Rate
