Visibility degradation caused by fog, haze, smoke, dust, and turbid water remains a major challenge for transportation, autonomous systems, defense, marine operations, industrial inspection, and search and rescue. Existing solutions are typically designed for a single environment or depend on expensive thermal, infrared, LiDAR, or specialized imaging hardware, limiting their large scale deployment.

Our innovation is a Universal Visibility Restoration Platform that enhances visual perception across both atmospheric and underwater scattering environments using standard RGB cameras and real time computational imaging. Instead of developing separate solutions for individual scenarios, the platform is built on a generalized framework that adapts to the optical characteristics of different scattering media, restoring scene visibility while preserving natural appearance.

The technology has been demonstrated in two distinct domains. The first addresses atmospheric visibility degradation, where the system restores visibility in foggy environments to improve situational awareness for railway operations, autonomous vehicles, surveillance, and outdoor monitoring. The second addresses underwater visibility degradation caused by water turbidity, enabling clearer imaging for remotely operated vehicles, autonomous underwater vehicles, seabed inspection, navigation, marine research, and underwater search operations. Demonstrating the same core technology across two fundamentally different environments validates the versatility and scalability of the platform.

The system operates in real time at up to 30 frames per second and has been developed for deployment on practical edge computing hardware. The platform has already been evaluated through working demonstrations and prototype implementations, illustrating its ability to improve scene visibility under challenging environmental conditions while maintaining low deployment cost by using conventional visible spectrum cameras.

Unlike conventional approaches that require different algorithms or specialized sensors for different operating conditions, this platform provides a unified computational imaging solution that can be integrated into existing vision systems with minimal hardware modification. This significantly reduces implementation cost while extending the operational capability of cameras used in transportation, defense, robotics, marine systems, industrial automation, and critical infrastructure monitoring.

Potential applications include intelligent railway safety systems, autonomous ground vehicles, unmanned aerial vehicles, maritime navigation, autonomous underwater vehicles, border surveillance, disaster response, industrial inspection, mining, smart cities, and environmental monitoring. The technology is readily adaptable to future multimodal perception systems through integration with radar, LiDAR, thermal imaging, and artificial intelligence based object detection.

By providing a generalized, real time, and cost effective visibility restoration platform for multiple scattering environments, this innovation establishes a new approach to computational imaging that improves safety, operational efficiency, and autonomous perception across a wide range of industries. Its broad applicability, demonstrated feasibility, and commercialization potential position it as an enabling technology for next generation intelligent vision systems.