Professor Orchard's main research interest is computational neuroscience. He uses unsupervised learning methods to train neural network models of the mammalian visual system. The goal is to uncover the algorithm that the brain uses to adjust its connections and encode information. He is an affiliate of the Centre for Theoretical Neuroscience.
He has also done research in the area of image processing and medical imaging. Most recently, he has developed a method to register (align) several images simultaneously; this is a departure from the conventional two-at-a-time registration. He has also published a number of other papers on image registration, image reconstruction for MRI and CT, denoising, image mosaicking, and forensic image processing.
Degrees and Awards
BMath (Waterloo), MSc (British Columbia), PhD (Simon Fraser)
J. Orchard, R. Mann. Registering a Multi-Sensor Ensemble of Images. IEEE Transactions on Image Processing, 19(5):1236-1247, 2010.
Y. Wang, J. Orchard. Fast Discrete Orthonormal Stockwell Transform. SIAM Journal on Scientific Computing, 31(5):4000-4012, 2009.
J. Orchard and C. Kaplan. Cut-Out Image Mosaics. Proceedings of the 6th Symposium on Non-Photorealistic Animation and Rendering (NPAR), pp. 79-87, 2008.
J. Orchard. Multimodal Image Registration using Floating Regressors in the Joint Intensity Scatter Plot. Medical Image Analysis, 12(4):385-396, 2008.
J. Orchard, J.T.W. Yeow. Toward a Flexible and Portable CT Scanner. Medical Image Computing and Computer Assisted Intervention (MICCAI), pp. 188-195, 2008.
J. Orchard. Efficient Least-Squares Multimodal Registration with a Globally Exhaustive Alignment Search. IEEE Transactions on Imaging Processing, 16(10):2526-2534, 2007.
J. Orchard, C. Greif, G.H. Golub, B. Bjornson, and M.S. Atkins. Simultaneous Registration and Activation Detection for fMRI. IEEE Transactions on Medical Imaging special issue on Medical Image Registration, 22(11):1427-1435, 2003.