Christopher James Henry

My academic career started with research into how the similarity of objects can be assessed within computing applications. Inspired by the human ability to quantify similarity based on the comparison of object features, I developed formal and systematic processes for quantifying the similarity of objects with a focus on practical image analysis problems and applications. This work was computationally complex and naturally led me to general-purpose computing using GPUs (GPGPU) in the late aughts. Building on this experience I co-established the Applied Parallel Computing and Collaborative Research Laboratory. My background in quantifying similarity (which is closely related to pattern classification) and GPGPU meant I was also uniquely positioned to explore machine learning as the deep learning revolution unfolded in the early to mid-2010s. Since then I have been largely involved in the theory and application of machine learning with a focus on applications in digital agriculture and remote sensing. This led me to co-create and co-lead the TerraByte research group whose focus is on automated methods, such as embedded and robotic systems, to create labelled datasets and the development of machine learning models to solve agricultural problems. Today my research interests include machine learning, deep learning, domain adaptation, generative image generation, digital agriculture, remote sensing, GPGPU and image analysis.