Download presentation
Presentation is loading. Please wait.
Published byCarmella Newton Modified over 8 years ago
1
Introduction to Vectorization of Engineering Drawings Song Jiqiang 18/9/2001
2
Concept Vectorization: the conversion from a raster image to its vector-form file. Engineering drawings in paper form Raster images Vector-form (CAD files) ScanVectorization
3
Why do vectorization? A lot of old drawings to be reused, and CAD files are more editable than images. Preprocess of automatic drawing understanding systems (information statistic, 3D reconstruction). Save the storage space.
4
History Begin research on vectorization at late 70’s; Begin that for engineering drawings at late 80’s; Related organizations & publications –IAPR TC-10, IEEE –IEEE T.PAMI, PR, PRL, CVIU, CVGIP, etc. –ICPR, ICDAR, IEEE CVPR, etc.
5
State of arts K. Tombre (LNCS vol.1389, 1998) : “None of these methods works. … Actually, the methods do work, but none of them is perfect.” 《 CADALYST 》 performs the annual evaluation on commercial vectorization systems.
6
Review of existing methods Thinning based CT(Contour Tracking) based RLE(Run Length Encoding) based SPT(Sparse Pixel Tracking) based
7
Thinning-based methods
8
CT-based methods
9
RLE-based methods
10
SPT-based methods P3P3 P0P0 P1P1 P2P2 P0P0 P1P1 P2P2 P3P3 P4P4
11
Existing difficulties Lines with intersections broken into pieces. Texts touch lines misrecognition. The interference of recognized objects repetitive detection, false detection.
12
Our research Analysis the vectorization model of existing methods. Propose an efficient vectorization model for engineering drawings. Propose a group of new graphical object recognition algorithms.
13
Common model of existing methods ( 2PV - 2 Phase Vectorization ) Raster image Low level vector form: Short line segments Graphical objects: Straight lines, circle/arc, curve Phase 1: Skeletonization or medial-axis approach Phase 2: Vector-based graphical object recognition
14
Motivation of 2PV Internal memory (RAM) –Used to be high price & limit capacity –High pixel access frequency cause swap Pixel tracking algorithm –No guide direction –Repetitive tracking
15
Object-Oriented Progressive-Simplification based Vectorization Model 1 phase model –Imitate the way that humans read drawings Recognize a graphical object in its entirety –Object-oriented feature Simplify the image data as the recognition goes on –Progressive-simplification feature
16
Workflow of OOPSV S-Line Recognition Arc Recognition Curve Recognition Symbol Recognition Text Recognition Line Symbol Text Arc Curve Symbol Text Curve Symbol Text Symbol Text S-LineArcCurveSymbolText
17
Graphical object recognition Get the intrinsic characteristic of individual type of graphical object. Use the characteristic as a guide to track the graphic object in complex environment.
18
Straight line recognition Seed segment detection A Seed Segment Irregular run Regular run
19
Direction guided tracking — based on the Bresenham algorithm Straight line recognition O Vo VpVp Seed segment Perpendicular runs Black segment White segment
20
Straight line recognition Dynamic adjustment to tracking direction P P’P’ VpVp O P P’P’ VpVp O PePe
21
Line net recognition A line net is a group of intersecting lines. Take advantage of the intersecting relationship to accelerate recognition. Example:
22
Circle/Arc recognition Arc segment detection –get initial arc center, radius, thickness Circular tracking –based on the Bresenham algorithm for circle
23
Circular tracking P2 P2 P1P1 O P P’ PEPE (a) Adjust succeeds(b) Adjust fails Legend: raster image tracking path medial axis testing path P2 P2 P1P1 O P P’ PEPE
24
Curve tracking Tracking result: a sequence of polyline. P5P5 P4P4 P1P1 P2P2 P3P3 P8P8 P9P9 P7P7 P6P6 PmPm P 10 P 11
25
Image simplification Intersection-preserving pixel deletion Based on the contour detection of the intersecting branches Branch at one side Middle Line Branches at both sides Recognized Line Contours of branches
26
Symbol recognition Common symbols –Cartography-based recognition Domain-specific symbols –Template-based recognition
27
Cartography
28
Symbol template
29
Text recognition Text segmentation –Difficulties: text touches line, similar size Character recognition –Stroke-based recognition algorithm
30
Image before the line recognition
31
Image after the line recognition
32
Suspension-Release mechanism Condition 1: Size( Box(li) ) < Tl Condition 2: p, p l C(p,L) >> C( FP(l,L), L) n i=1
33
Stroke-based character template Stroke definition Black position White position Aspect ratio scope Complexity level ① ② ④ ③ ⑤
34
Character recognition (UMNKLDREFBbhklI1)(BE) (E)Accept ‘E’, then cut image
35
Separate connected characters a. When the rightmost stroke is vertical b. When the rightmost stroke is not vertical f(x) x1 Base on the analysis of the rightmost stroke and the vertical projection.
36
Experimental result Implemented a complete vectorization system running on Windows platform using VC6.0. Automatic vectorization of an A0-size drawing (15M) takes about 5 minutes. ( PIII500/128M ) –Line vectorization takes less than 1 minute (1600 lines), faster than performing a thinning operation (3.5 mins).
37
CDI evaluation SizeDpDp FpFp PRIDvDv FvFv VRICDI A0 0.9660.0860.9400.9540.0820.9360.938 A1 0.9730.0770.9480.9650.0840.9410.945 A2 0.9750.0540.9610.9630.0680.9480.955 A3 0.9620.0490.9570.9590.0660.9470.952 This protocol was proposed in Machine Vision Application, (1997)
38
Manual editing cost evaluation Drawing Size False_alarmsMissesOne2manyMany2oneTotal Costs A0 Ⅰ 124802437354 Ⅱ 92621632263 A1 Ⅰ 63461118194 Ⅱ 4133916142 A2 Ⅰ 302668101 Ⅱ 22183667 A3 Ⅰ 16155769 Ⅱ 10133444 This protocol was proposed in LNCS V.1389, (1998)
39
Comparison of editing cost with VPStudio a. raster image b. our system c. VPStudio Conclusion: Object-oriented recognition algorithms produce less misrecognition, therefore the editing cost has decreased.
40
Conclusion Progressive simplification decreases both the complexity and workload of vectorization. The object-oriented recognition algorithms recognize graphical objects fast and entirely.
41
Related papers 7 journal papers & 4 conference papers IEEE Trans. PAMI reviewer’s comment: “ An efficient model is very important to recognize engineering drawings …. This paper suggested an object-oriented progressive-simplification based vectorization system for engineering drawings. It would bring an impact in this area.”
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.