Arrowhead recognition during automated data capture

Authors:
G. Priestnall;R. E. Marston;D. G. Elliman
Affiliations:
Department of Geography, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom;Department of Computer Science, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom;Department of Computer Science, University of Nottingham, University Park, Nottingham, NG7 2RD, United Kingdom
Venue:
Pattern Recognition Letters
Year:
1996

Citing 4
Cited 2

The grammar of dimensions in machine drawings

Computer Vision, Graphics, and Image Processing
A syntactic geometric approach to recognition of dimensions in engineering machine drawings

Computer Vision, Graphics, and Image Processing
Preprocessing of engineering drawings for 3D reconstruction

ISCI '90 Proceedings of the first international conference on systems integration on Systems integration '90
Knowledge-Directed Interpretation of Mechanical Engineering Drawings

IEEE Transactions on Pattern Analysis and Machine Intelligence

Recognition of Arrows in Line Drawings based on the Aggregation of Geometric Criteria using the Choquet Integral

ICDAR '03 Proceedings of the Seventh International Conference on Document Analysis and Recognition - Volume 1
A New Way to Detect Arrows in Line Drawings

IEEE Transactions on Pattern Analysis and Machine Intelligence

Quantified Score

Hi-index	0.10

Visualization

Abstract

This paper presents a method of recognising arrowheads on engineering drawings during the automatic conversion of a paper drawing to a computer readable vector representation. The method takes advantage of a new data structure formed from the vector outlines extracted from the scanned bi-level image. Opposite outline vectors are paired where possible, creating a structure of linked outlines representing linear features, and unpaired areas at junctions, corners and endpoints. This data structure is descriptive and information preserving, and allows rule-based feature extraction modules to be developed. This paired outline data structure, and the method of extracting one type of standard arrowhead are described in detail. Results are presented and the potential for exploiting the paired outline data structure for recognition of different types of arrowhead and for features in other domains is discussed.