Object recognition using random image-lines

View-based recognition is a conceptually simple, relatively robust method f or object recognition. Current techniques use small, simplistic object data bases requiring, in many cases, large processing training and/or recognitio n times. In this paper we propose an extension to the view-based object rec ognition paradigm based on the classification of lines subtending an object of 2D image views that achieves high generalisation recognition rates with reduced computational times. (C) 2000 Elsevier Science B.V. All rights res erved.