Japanese / English

Detail of Publication

Text Language English
Authors Seiichi Uchida, Masakazu Iwamura, Shinichiro Omachi, and Koichi Kise
Title Design and recognition of human-readable and machine-readable patterns
Journal 1st Joint Workshop on Machine Perception and Robotics (MPR2005)
Number of Pages 8 pages
Location Beijin, China
Reviewed or not Reviewed
Month & Year September 2005
Abstract In this paper, design and recognition of human-readable and machine-readable patterns are investigated. Specifically speaking, we design character images printed with a horizontal stripe pattern, called a cross ratio pattern. The cross ratio derived from the cross ratio pattern represents the class information of the character. Since the cross ratio is invariant to projective distortion, the class information is extracted correctly regardless of camera angle. The character image itself is human-readable and therefore the character image with the cross ratio pattern is not only humanreadable and but also machine-readable and can be used as a medium for human-machine communication.
Back to list