Jeyashree, A and Subash Kumar, C S and Mokshitha, Jayachandran and Niveditha, A V S and Santhosh, M (2025) Optidepth:Edge-AI Optimized Monocular Depth Estimation for Real-Time Heavy Machinery Safety. 2025 Second International Conference on Intelligent Technologies for Sustainable Electric and Communications Systems (iTech SECOM). pp. 1-6.
Full text not available from this repository.Abstract
Optidepth is an AI-powered safety system that enables real-time collision prevention for heavy machinery at a fraction of the cost of traditional solutions. By leveraging optimized monocular depth estimation with quantized MiDaS neural networks on edge devices such as Jetson Nano and Raspberry Pi 4, the system transforms standard cameras into intelligent monitoring tools. Optidepth achieves 15-20 FPS, a detection range of 0.5−5 meters, and sub- 100 ms alert latency with an accuracy of 92%, requiring no additional hardware modifications for deployment. Delivering LiDAR-grade perception capabilities purely through software, the system addresses critical safety challenges in construction, including worker protection, prevention of utility strikes, and reduction of equipment damage. Its scalable, retrofit-ready architecture with support for over-the-air (OTA) updates makes Optidepth a practical and cost-effective solution for enhancing industrial safety in a market valued at over $ 50 billion.
| Item Type: | Article |
|---|---|
| Subjects: | Artificial Intelligence and Data Science > Deep Learning Computer Science and Engineering > Embedded and Real-Time Systems Computer Science and Engineering > Wireless Network |
| Divisions: | Electrical and Electronics Engineering |
| Depositing User: | Dr Krishnamurthy V |
| Date Deposited: | 22 Apr 2026 09:02 |
| Last Modified: | 22 Apr 2026 09:02 |
| URI: | https://ir.psgitech.ac.in/id/eprint/1810 |
Dimensions
Dimensions