Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{199127,
        author = {TUSHAK R and Dr.M.S.Bennet Praba},
        title = {Enhancing Worker Safety at Heights using Deep Learning and Transformer-based Object Detection},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14257-14263},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199127},
        abstract = {Accidents caused by falls from heights remain one of the leading factors of fatalities and serious injuries in construction and industrial environments. Traditional safety monitoring methods rely heavily on manual supervision, which is inefficient, error prone, and difficult to scale. This paper proposes an intelligent vision based safety monitoring framework leveraging deep learning and transformer based object detection models to enhance worker safety at heights. The system automatically detects workers, personal protective equipment (PPE), and hazardous situations such as missing harnesses, unsafe proximity to edges, and unguarded elevated platforms. By integrating convolutional neural networks (CNNs) with transformer architectures for global contextual understanding, the proposed approach achieves robust detection accuracy even in complex, cluttered, and dynamic construction scenes. Experimental evaluation on real world and benchmark datasets demonstrates that transformer based detectors outperform traditional CNN only approaches in both accuracy and robustness, highlighting their potential for deployment in real time safety surveillance systems.},
        keywords = {},
        month = {April},
        }