Amirhossein Ghodrati

Citation

Thesis Title
Sketch Recognition of Digital Ink Diagrams

Read article at Massey Research Online:

Pen-and-paper sketches are used to externalise ideas and concepts. Sketching on computers has become easier following developments in electronic pen-based media. There is consequent interest in automatic recognition of diagrams by computer. This research focuses on the simultaneous grouping and recognition of shapes in digital ink diagrams. Mr Ghodrati’s grouping technique hypothesises multiple shape candidates, among which some are invalid. He presents a novel rejection technique that uses proximity measures to identify valid shape candidates and also presents a novel connector recognition system. The full comparative study results show that Mr Ghodrati’s approach is significantly more accurate in finding shapes and significantly faster on process diagrams. Furthermore, his approach is shown to be more accurate in finding and recognising the shapes in a publicly available dataset.

Supervisors
Professor Hans Guesgen
Professor Stephen Marsland
Dr Rachel Blagojevic