SOURCE: AFI
A groundbreaking algorithm developed by Rishabh Bhattacharya, a third-year engineering student at the International Institute of Information Technology, Hyderabad (IIITH), has caught the attention of the Indian Navy. This innovative solution, aimed at detecting and tracking flying objects such as helicopters, aeroplanes, drones, UAVs, and birds, could soon become an integral part of naval operations.
Rishabh’s algorithm was showcased at a nationwide competition hosted by the Indian Navy under the banner of “Swavalamban 2024,” an event dedicated to fostering innovation and self-reliance in defense technology. His work not only won him the first prize but also a cash award of Rs 3 lakh.
Rishabh developed an ‘optical flow tracking algorithm’ that offers real-time tracking with sub-pixel accuracy, crucial for precise motion estimation. This approach uses “optical flow,” a technique in computer vision that tracks the movement of objects by analyzing pixel changes across consecutive frames. This makes it particularly effective in environments with challenging conditions like varying lighting, high-speed object movements, and complex textures.
Speaking to The Times of India (TOI) on Thursday, Rishabh detailed how his algorithm meets the Navy’s stringent requirements. “One of the criteria laid out was for the solution to demonstrate resilience to varying lighting conditions, rapid movements, and complex textures while maintaining efficiency on platforms like drones or embedded systems,” he explained. “My algorithm meets the requirement by ensuring robustness and scalability.”
The Navy, recognizing the potential of this technology, has shown interest in further development and integration into their operational framework. Rishabh mentioned that he received a call from Navy officials requesting him to continue his research at the Indian Institute of Technology, Hyderabad. This collaboration could lead to significant enhancements in how the Navy detects and tracks aerial threats and navigates through aerial environments.
One of the major challenges in developing such an algorithm is dealing with the lack of comprehensive datasets for training. To address this, Rishabh combined datasets from Sekilab, which included images of planes, helicopters, and birds, with a UAV dataset from Kaggle. This amalgamation allowed for a more robust training model, enhancing the algorithm’s performance across different scenarios.
Rishabh’s work not only showcases the potential of young Indian talent in contributing to national defense but also underscores the importance of academic institutions in research and development. His success story could inspire more students to delve into practical, impactful research that directly serves national interests.