Community Research and Development Information Service - CORDIS


This paper examines the problem of automatically grouping image curves in computer vision. The majority of techniques developed for grouping are based on dot patterns or straight lines. Studying the grouping of sensory phenomena has identified a number of classes of grouping: proximity, similarity, closure, good continuation and symmetry. This paper concentrates on good continuation (curvilinearity), parallelism and proximity.

Additional information

Authors: ROSIN P L, JRC Ispra (IT)
Bibliographic Reference: Paper presented: British Machine Vision Conference, York (GB), September 13-16, 1994
Availability: Available from (1) as Paper EN 38599 ORA
Record Number: 199411298 / Last updated on: 1994-12-06
Original language: en
Available languages: en