Non-Technical Overview of Research Topics

Broadly speaking, my past research topics have been in image processing, image segmentation, data fusion, image synthesis, sea ice monitoring and polarimetric SAR processing, with some medical imaging and MRI during my work terms at the Hospital for Sick Children (Sick Kids). The goal of my image processing work has been how to find ways to get a computer to automatically extract useful information from images, while my work on MRI at Sick Kids focused on improving the accuracy of MRI scans.

Image segmentation is one way a computer can get useful information out of an image: it divides an image up into regions that the computer thinks corresponds to a real-world object. For example, in sea ice monitoring, the segmentation tries to divide a satellite image of sea ice into multiple regions that are individually comprised of only one type of sea ice.

Data fusion is another area I worked on; this refers to combining data from different sources in a quantitative way that makes best use of the different types of information provided by the different data sources. For example, if there are two satellites and each one picks up information in a different frequency of light, then each satellite provides different information about the surface that they are imaging. If both satellites image the same area of the Earth at roughly the same time, then we can combine the different information from them and be able to better understand the properties of that area.

Image synthesis is the creation of synthetic images that look like real images. While this is what we do in 3D / CG Artwork, my research work was a bit more specific: create SAR images of sea ice that have the same properties as real SAR images. Once these images are created, they are used to test image processing algorithms (such as an image segmentation algorithm). Synthetic images are useful because their properties can be controlled so that the testing process can be simplified. Polarimetric SAR processing is simply processing of a type of SAR image that has slightly more information than non-polarimetric SAR. There are some properties that are different so that the mathematics of processing change a bit.

About Peter Yu I am a research and development professional with expertise in the areas of image processing, remote sensing and computer vision. I received BASc and MASc degrees in Systems Design Engineering at the University of Waterloo. My working experience covers industries ranging from district energy to medical imaging to cinematic visual effects. I like to dabble in 3D artwork, I enjoy cycling recreationally and I am interested in sustainable technology. More about me...

Feel free to contact me with any questions about this site at [user]@[host] where [user]=web and [host]

Copyright © 1997 - 2021 Peter Yu