Detection of dimension sets in engineering drawings

Abstract

This paper presents a system f o r detecting dimension sets in engineering drawings that are drawn t o ANSI drafting standards.A new rule-based text/graphics separation algorithm and a model-based procedure for detecting arrowheads in any ofientation have been developed. Arrowhead tracking and search methods are used to extract leaders, tails, and witness lines from segmented images containing only graphics. Text blocks and feature control frames extracted from the segmented images are then associated with their corresponding leaders to obtain complete dimension sets. Object lines are then separated from centerlines and hatching lines. Experimental results are presented.

Genericity in graphics recognition algorithms

“…Leader (arrowhead) detection has been investigated by Dori et al [15], Lai and Kasturi [16], and Das and Langrana [17]. Dori et al [15] search areas at wire endpoints for potential arrowheads by tracing lines orthogonal to the wire.…”

“…Joseph and Pridmore [9] dealt with finding dashed lines in engineering drawings by looking for chains of short lines within the ANON system. Lai and Kasturi [10] have done work on detecting dashed lines in drawings and maps. They attempted to recognize dashed lines by linking short isolated bars under certain conditions in three passes.…”

“…The dashed lines are not necessarily straight, as is the case in maps. Chen et al [11] use the same method as in [10] to detect dashed lines of several patterns in the refinement of vectorized mechanical drawings. Agam et al [12] and Dori et al [13] have recently investigated the detection of dashed and dash-dotted lines with straight and curved shapes.…”

“…Vector-based arc segmentation has also been investigated [6]. Line style detection has been studied by several groups [7][8][9][10][11][12][13]. Pao et al [7] use a HT-based method to detect dashed circles and dashed straight lines in several steps.…”

“…Further test requires that there be a tail and a pointing reference. Lai and Kasturi [16] use the characteristics of arrowhead skeletonization, which are a long line and three short segments at the end, as a clue for arrowhead presence. Further test requires significant thickness change and enough black pixels in the arrowhead area.…”