Year: 2025
Type: article
Abstract: Intrusion detection in construction scenes can effectively reduce the occurrence of hazardous incidents. Current detection methods, while effective, are often too complex. This paper proposes a lightweight monitoring model based on convolutional neural networks (CNNs). First, the model is trained using a dataset to achieve high accuracy. Then, the model is lightweighted using CNNs. Simulation results show that the model can maintain accuracy while occupying a smaller volume.
DOI: 10.54097/ejhnrv33
Author: Haoyang Zhao
Institution: East China Jiaotong University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W4409870074