Vision Enhancement Solution for Loco Pilot During Fog, Rain, Curves

Prashant Sagar

Abstract: This project presents a concept for improving locomotive driving visibility and collision avoidance by integrating a smart digital display with layered protective glazing and a thin electrochromic glass layer, supported by an onboard processing unit for real-time vision and navigation functions. The proposed setup combines a transparent display, an infrared thermal camera feed, and embedded image-processing methods for track and obstacle detection under low-visibility conditions such as fog and heavy rain. A compact GPU-CPU class processing board runs the operating system, camera pipelines, and AI-based detection models- including CNN and YOLO-based frameworks- with attention to installation steps and field integration constraints. In parallel, an intra-railway navigation system is proposed using real-time GPS/GNSS tracking to display nearby locomotive positions, speeds, and route geometry on the cab screen, aiming to reduce rail-to-rail accidents, derailment risk, operational delays, and infrastructure losses. This approach complements existing Indian Railway safety initiatives such as the Kavach Automatic Train Protection system by adding a vision-layer and a pilot-accessible navigational display. A supporting business case is discussed using reported accident-loss figures and the growing relevance of service reliability for both public and private rail operations.

Keywords: electrochromic glass, thermal infrared camera, obstacle detection, onboard image processing, railway GPS navigation, CNN, YOLO, loco pilot vision, intra-railway navigation, Kavach

How to Cite?: Prashant Sagar, "Vision Enhancement Solution for Loco Pilot During Fog, Rain, Curves", Volume 15 Issue 2, February 2026, International Journal of Science and Research (IJSR), Pages: 1669-1675, https://www.ijsr.net/getabstract.php?paperid=SR26222111913, DOI: https://dx.dx.doi.org/10.21275/SR26222111913

Download Citation: APA | MLA | BibTeX | EndNote | RefMan