Work on Machine Vision
Within machine vision, Roy has worked on several main topics:
- image processing
- noise elimination
- feature detection
- shape analysis
- Hough transforms
- robust pattern matching
- artificial neural networks.
He has specialised in algorithm design for automated visual inspection, particularly for the food industry.
He also has interests in vision in more unconstrained environments:
- surveillance
- crime detection and prevention
- vehicle driver assistance
- laparoscopic surgery.
He has spent much of his career on real-time implementations of vision, initially on fast hardware,
and latterly on efficient software employing sampling and other techniques.
For details of Roy's research, see the Publications on Machine Vision, which
is accompanied by a breakdown of his research into the following five main headings, each with an analysis of achievements:
- Image filtering
- Feature detection
- Intermediate level analysis
- Real-time operation
- Automated visual inspection.
