Bamidele Ogunjumelo

Bamidele's research lies at the intersection of computer vision, optical imaging, artificial intelligence, and combustion measurement, with a focus on applying deep learning to light field (LF) imaging for flame diagnostics. Bamidele develops data-driven methods that enable depth-aware refocusing and 3-D analysis of burner flames from a single optical capture, supporting real-time, non-intrusive measurement of combustion phenomena.

A key contribution of Bamidele's work is the development of a deep learning–based alternative to traditional LF refocusing algorithms, such as shift-and-sum, which are computationally expensive and impractical for real-time applications. This approach significantly reduces computational cost while maintaining refocusing accuracy, enabling near real-time performance.

Overall, Bamidele's research demonstrates how AI-driven computer vision techniques can enhance optical measurement systems, enabling efficient 3-D flame imaging and refocusing without complex optical setups. These outcomes contribute to improved combustion diagnostics, energy efficiency, and emissions reduction, with broader relevance to real-time optical imaging and intelligent measurement systems.  

• Deep Learning
• Computer Vision
• Light Field Imaging and Embedded Systems

Supervised by Dr Moinul Hossain and Dr Gang Lu.