FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF

DEVELOPMENT OF A PROTOTYPE VISION ASSISTED AUTOMATED CAD DRAWING

SYSTEM FOR 2D PLANE PROFILE BY

IV AN QUEK MING FOOK

A THESIS SUBMITTED IN PARTIAL

FULFILMENT OF THE REQUIREMENT FOR THE DEGREE OF

MASTER OF SCIENCE IN MECHATRONICS ENGINEERING

INTERNATIONAL ISLAMIC UNIVERSITY MALAYSIA

AUGUST 1999

ABSTRACT

Duplication is a process of making a copy of an existing object. To obtain a drawing of this object, the profile of the object has to be determined, followed by manual drawing to produce the drawing. Thus, this thesis presents a method that can help to automate the process of part drawing with the help of a vision system. The approach is useful and the time required to do shape detection of an arbitrary profile of an object can be reduced. By using the approach of a vision assisted automated CAD drawing, the tedious task of this process can be done easily.

In the developed prototype system, the object which its drawing is to be produced automatically, is placed under the camera frame of a vision acquisition system and the image of the object is acquired. Later, an image processing module performs the necessary image manipulation to obtain the information regarding the object in the acquired image. This information is used by the automatic drawing module to produce the drawing of the object automatically. Finally, the 2D drawing of the object is drawn automatically.

11

~ "~1 Ji ~11..Ll, i-') Js,

J ~

r-->-

r tL.;.;....~

.j ...1;;-L___;

.:.ii fa-~),

.:.iu

~

j..-.,J i j")\JI ..::.,.j')I

J_;:s;

9 ~

~~ ~J)) ..u-L...l\ _}lo

y

fa_

)LJ,)

~

~

~~

.:..LJ~ ~J ·"~1 } ~ I o ) ~ Js, J ~ 1 ~ ~ ~Jj)\ yL:5'~1 ~ ~

.r-.J4L'

I ' ~ \ J i ~ ~ ~\.;LI ..:-,Lo ..,.WI Js, Jr=J.} ~)J_raJI ~WI~ ^{J ~ I} ~~

o..l>-

r--"')

J.v

J

o-l>-

o

.~Lo_,;} " ~ 1 }

~

~ ~ lp>-iJ

iii

APPROVAL PAGE

I certify that I have supervised and read this study and in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science in Mechatronics Engineering.

Name:

~ JJr.

~/\/4s.s

Supervisor

I certify that I have read this study and in my opinion, it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a thesis for the degree of Master of Science in Mechatronics Engineering.

__, , r ~

Nf::11i ^{P,.. ff}

^{/J- /_.,,}

Examiner

This thesis was submitted to the Department of Mechatronics Engineering and is accepted as partial fulfilment of the requirements for the degree of Master of Science in Mechatronics Engineering.

This thesis was submitted to the Kulliyyah of Engineering and is accepted as partial fulfilment of the requirements for the degree of Master of Science in Mechatronics Engineering.

Name: !)IL AHMAD f'~tc !S1'V1AIL

Dean, Kulliyah of Engineering

IV

DECLARATION

I hereby declare that this thesis is the result of my own investigation, except where otherwise stated. Other sources are acknowledged by explicit references and a bibliography is appended.

Name IV AN QUEK MING FOOK

Signature

Date

V

© Copyright by Ivan Quek and International Islamic University Malaysia, 1999.

vi

ACKNOWLEDGMENT

This study would not have been possible without the help, advice and cooperation of many persons who were instrumental in the completion of this study.

First of all, I would like to record my sincere appreciation to my supervisor, Dr.

A.K.M. Mostafa Kamal, who had guided me through the completion of both, the proposal and thesis work. Despite his hectic schedule, he managed to scrutinize the drafts with patience and precision and imparted invaluable insights, advice and guidance to the shaping of final version of the thesis. Hence, a special sincere note of gratitude goes to him.

My deepest appreciation goes to all lecturers and staffs at the Kulliyyah of Engineering for their warmth, kindness, patience, sincerity and knowledge. I personally would like to thank Prof. El-Sayed El-Badawy (the postgraduate coordinator) for being an excellent lecturer and to whom I have gained a lot of knowledge, advice and who has equipped me with a good foundation of electronics, Dr. Ajmal I. Ansari and Dr. Nazim Neziri, my co-supervisors, for giving me good feedback on my research work.

My sincere thank also goes to Dr. Ahmad Faris Ismail, the Dean of Kulliyyah of Engineering, Dr. Tasneem Pervez, the Head of Mechanical Engineering Department, Dr. Sufyan Al-Iryahim, the Head of Electrical and Computer Engineering Department, Dr. Momoh Jimoh Salami, Dr. Hossam Shallaby, Dr. Adznan Jantan, Dr. El-Hadi Laouar, Dr. Hamzah Mohd Salleh and other helpful lecturers, for making this university a perfect place to seek knowledge. Not to miss, the helpful staffs of Kulliyyah of Engineering; especially to Br. Ferhat, Br. Yusni, Br. Yahya, Sis.

Kamsiah, Sis. Azizah, Sis. Fatimah, and other helpful staffs for creating this kulliyyah, a warm place and nice environment to work with.

vii

A special note of appreciation also goes to my colleagues, friends and students, who have made my stay in IIUM enjoyable and memorable. Special mention should be made to the pioneer batch of postgraduate students, Br. Wajidi Al-Khateeb and Br.

Fong C.S. Also thanks to my graduate office mate, Br. Yousif, who has supported me in the struggle to obtain better facilities, such as personal computers and Internet facilities in our graduate office room as well as for the benefit of the future batch of postgraduate students.

Also not forgetting my family members, special thank to both my parents, my brothers and sisters for giving me all the support that I need to pursue my ambitious career in IIUM.

Finally, not to miss this grand opportunity, a special note of gratitude to my wonderful students of GEN 1230-Computer Aided Analysis II, C tutorial class, thanking all of you for giving me the opportunity to guide you. I hope that all of you have gained some knowledge in real-life C programming during this short period of time, and wishes all of you the best of luck in your future undertakings. Also, not forgetting the first batch of engineering undergraduates, I would like to congratulate all of you upon your completion of your engineering programme here. Thank you very much to all of you.

viii

TABLE OF CONTENTS

Abstract Approval Page Declaration Acknowledgment List of Tables List of Figures

List of Abbreviations

CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW

1.1 INTRODUCTION

1.2 LITERATURE REVIEW 1.3 STATEMENT OF PROBLEM 1.4 SOLUTION APPROACH

1.5 OBJECTIVES OF THE RESEARCH WORK 1.6 SCOPE OF THE RESEARCH WORK

1. 7 METHODOLOGY

CHAPTER 2: HARDWARE AND SOFTWARE FACILITIES

2.1 EQUIPMENT USED 2.1.1 Hardware Facilities 2.1.2 Software Facilities

2.2 GLOBAL LAB IMAGE ACQUISITION SYSTEM 2.3 GLOBAL LAB OT-IRIS IMAGE FILE FORMAT

CHAPTER3:THEPROTOTYPESYSTEM

3.1 GENERAL INFORMATION 3.2 IMAGE PROCESSING MODULE

3.2.1 Main Menu 3.2.2 File Utility Menu 3.2.3 Image Utility Menu 1 3.2.4 Image Utility Menu 2 3.2.5 Image Utility Menu 3 3.2.6 Shift Menu

3.3 AUTOMATIC DRAWING MODULE 3.3.1 About VISCAD Option

3.3.2 Sample AutoDraw Option 3.3.3 AutoDraw Option

3.4 THE DATA FILE FORMAT

CHAPTER 4: THE ALGORITHMS AND RESULT

4.1 THE OVERALL PROCEDURES INVOLVED 4.2 CALIBRATION PROCEDURE

4.3 THE ALGORITHMS 4.3.1 Read Image Algorithm 4.3.2 Write Image Algorithm

lX

11

IV V

vu

XI XU XVI

1 2 6 6 7 8 9 14 14 15 16 18 20 21 23 24 25 27 29 30 31 33 34 35 37

39 44 46 48 50

4.3.3 View Image Algorithm

4.3.4 Record Outline Data Algorithm

4.3.5 Record Calibration Parameter Algorithm 4.3.6 Negation Algorithm

4.3. 7 Thresholding Algorithm 4.3.8 Local Averaging Algorithm 4.3.9 Display Histogram Algorithm 4.3.10 Erosion Algorithm

4.3.11 Image Subtraction Algorithm 4.3.12 Shifting Operation Algorithm

4.3.13 Boundary Outline - N4 Mark And Kill Algorithm 4.3.14 Boundary Outline - HV Mark And Kill Algorithm 4.3.15 Boundary Outline - Morphology Technique Algorithm 4.3.16 Boundary Tracing Algorithm

4.3.17 AutoLISP Automatic Drawing Algorithm 4.4 SUMMARY OF RESULT

4.4.1 Comparison with Sobel Edge Detector 4.4.2 Comparison with Freeman Chain Coding 4.4.3 Drawing Comparison

4.4.4 Result on Additional Shapes

CHAPTER 5: CONCLUSION

51 54 56 58 60 63 66 69

72 75 79 84 90 95 103 106 106 109 110 112

5.1 CONCLUSION OF THE RESEARCH WORK 122

5.2 CAPABILITIES AND CONSTRAINTS OF THE PROTOTYPE 123

SYSTEM

5.3 CONTRIBUTIONS OF THE RESEARCH WORK 125

5.4 SUGGESTIONS FOR FUTURE WORK 125

APPENDIX 1: SOURCE CODE FOR THE IMAGE PROCESSING 128

MODULE

APPENDIX 2: SOURCE CODE FOR THE AUTOMATIC ORA WING 150

MODULE

BIBLIOGRAPHY 152

X

LIST OF TABLES

Table Description of Table Page

2.1 Header information of GLOBAL LAB DT-IRIS image file format. 19

3.1 Options available in Main Menu ofVISCAD. 23

3.2 Options available in File Utility Menu. 24

3 .3 Options available in Image Utility Menu 1. 26

3.4 Options available in Image Utility Menu 2. 28

3.5 Options available in Image Utility Menu 3. 29

3.6 Options available in Shift Menu. 30

3.7 Options available in VISCAD pull down menu m AutoCAD 33 envirorunent.

4.1 Grouping of grey level values according to color scheme 52 representation.

4.2 Operation time recorded for the three samples. 120

xi

LIST OF FIGURES

Figure Description of Figure Page

1.1 Methodology of vision assisted automated CAD drawing system. 11

2.1 GLOBAL LAB image acquisition system. 17

3.1 Menu hierarchy of VIS CAD image processing module. 22 3.2 Main screen of the developed VIS CAD application. 24

3.3 File Utility Menu screen. 25

3.4 Image Utility Menu I screen. 27

3.5 Image Utility Menu 2 screen. 28

3.6 Image Utility Menu 3 screen. 30

3.7 Shift Menu screen. 31

3.8 AutoCAD drawing environment with VISCAD implementation. 32 3.9 About VISCAD Automatic Drawing Module dialog screen. 34

3.10 Output screen of Sample AutoDraw option. 35

3.11 Prompt screen for data file selection in AutoDraw option. 36 3.12 A completed automatically drawn drawing of an object. 37

3.13 Data file format. 38

4.1 Setup of the vision system. 40

4.2 Flowchart of the overall steps involved. 43

4.3 Calibration procedure of the vision system. 44

4.4 Flowchart of Read Image algorithm. 49

4.5 Flowchart of Write Image algorithm. 51

4.6 Flowchart of View linage algorithm. 53

xii

4.7 Result of View Image in 16 Colors algorithm. 54 4.8 Flowchart of Record Outline Data algorithm. 55 4.9 Flowchart of Record Calibration Parameter algorithm. 57

4.10 Flowchart of Negation algorithm. 59

4.11 Result of Negation algorithm - image is partially negated. 60

4.12 Flowchart of Thresholding algorithm. 62

4.13 Result of Thresholding algorithm - image is partially thresholded. 63

4.14 Flowchart of Local Averaging algorithm. 64

4.15 Image before Local Averaging operation. 65

4.16 Result of Local Averaging algorithm. 66

4.17 Result of Display Histogram algorithm. 67

4.18 Flowchart of Display Histogram algorithm. 68

4.19 Erosion procedure. 70

4.20 Flowchart of Erosion algorithm. 71

4.21 Result of Erosion algorithm. 72

4.22 Flowchart oflmage Subtraction algorithm. 74

4.23 Right shifting of an image. 76

4.24 Flowchart of Shifting Operation algorithm. 77

4.25 Result of Shifting algorithm - before RIGHT shifting. 78 4.26 Result of Shifting algorithm - after RIGHT shifting 80 pixels. 79 4.27 N4 Mark and Kill boundary detection method. 80 4.28 Flowchart of Boundary Outline using N4 Mark and Kill algorithm. 81 4.29 Result of N4 Mark and Kill Boundary Detection algorithm - 82

marking boundary pixels.

xiii

4.30 Result of N4 Mark and Kill Boundary Detection algorithm - 83 deletion of non-boundary pixels.

4.31 Result of N4 Mark and Kill Boundary Detection algorithm - image 84 outline.

4.32 HV Mark and Kill boundary detection method. 85 4.33 Flowchart of Boundary Outline using HV Mark and Kill 86

algorithm.

4.34 Result of HV Mark and Kill Boundary Detection algorithm - 87 marking boundary pixels in horizontal direction.

4.35 Result of HV Mark and Kill Boundary Detection algorithm - 88 marking boundary pixels in vertical direction after horizontal direction.

4.36 Result of HV Mark and Kill Boundary Detection algorithm - 89 deletion of non-boundary pixels.

4.37 Result of HV Mark and Kill Boundary Detection algorithm - 90 image outline.

4.38 Image outline obtained using Morphology Technique of Erosion 91 and Subtraction operation.

4.39 Flowchart of Boundary Outline using Morphology Technique 92 algorithm.

4.40 Result of Boundary Outlining algorithm using Morphology 93 Technique of Erosion and Subtraction - erosion on image.

4.41 Result of Boundary Outlining algorithm using Morphology 94 Technique of Erosion and Subtraction - subtracting original image with eroded image.

4.42 Result of Boundary Outlining algorithm using Morphology 95 Technique of Erosion and Subtraction - image outline.

4.43 Direction of boundary tracing. 96

4.44 Special cases with 2 neighbouring pixels. 97

4.45(a) Flowchart of Boundary Tracing algorithm. 99

4.45(b) Flowchart of Boundary Tracing algorithm. 100

xiv

4.45( c) Flowchart of Boundary Tracing algorithm. 1 O 1

4.46 Result of Boundary Tracing algorithm. 102

4.4 7 Flowchart of AutoLISP Automatic Drawing algorithm. 104 4.48 Result of AutoLISP Automatic Drawing algorithm. 105 4.49 Outline of an image produced using Sobel operator. 107 4.50 Outline of an image produced using the developed boundary 108

outlining algorithms.

4.51 Simulated object with its dimensions shown. 110 4.52 Automatically drawn drawing of an object with its dimensions 111

shown.

4.53 Binarized image of a paper shape. 113

4.54 Binarized image of a leadframe. 113

4.55 Outline of the paper shape produced using the boundary outlining 114 method.

4.56 Outline of the leadframe produced using the boundary outlining 115 method.

4.57 Result of the paper shape drawn using the Automatic Drawing 116 module.

4.58 Result of the leadframe drawn using the Automatic Drawing 117 module.

4.59 4.60

Outline of the paper shape produced using Sobel Edge Detector.

Outline of the leadframe produced using Sobel Edge Detector.

xv

118 119

LIST OF ABBREVIATIONS

Abbreviation Description of Abbreviation

2D Two Dimensional

3D Three Dimensional

e.g. (exempligratia); for example etc. (et cetera); and so forth AVI Automated Visual Inspection CAD Computer Aided Drafting

CAD/CAM Computer Aided Drafting / Computer Aided Manufacturing CCD Charge Coupled Device

CMM Coordinates Measuring Machine DXF Drawing Interchange File

IC Integrated Circuit

IGES Initial Graphics Exchange Specification STL Stereolithography

xvi

CHAPTER I

INTRODUCTION AND LITERATURE REVIEW

1.1 INTRODUCTION

Machine vision system has been used in a wide variety of fields, ranging from pattern recognition, image processing for medical purposes, industrial based application such as assembly process, components, robotics vision and others. But, machine vision system is widely employed in the field of inspection and measurement as well as in object or pattern recognition.

In sho1i, machine vision involves automatic image interpretation for the purpose of control; process control, quality control, machinery control, robotics control and etc.

Two distinct areas of interest in machine vision are robot vision and automated visual inspection.

With advancement in science and teclmology, especially in the field of electronics and computers graphics technology, machine vision has gone beyond the field of visual inspection, and has expanded the field of image processing into other new dimension.

New field of research has emerged such as data visualization, object reconstruction, surface modeling, computer tomography and etc.

1.2 LITERATURE REVIEW

Many research areas involving machine vision and image processing has grown and expanded in the last couple of years. With the help of today's technology, machine vision and image processing has grown at a rapid speed. Most of the application of machine vision is focused on automated visual inspection (A VI) or pattern or object recognition. For the case of visual inspection, there are numerous literatures dealing with this field of research, especially in the field of quality control, inspection and measurement. There are also huge volumes of literature on pattern or object recognition as well.

For example, in the last five years, intensive research on quality inspection has been actively involved using machine vision system. Ang, Oh and Teo ^[IJ has employed vision system for the inspection of keypad ink printed graphics, while Ching and Ho

[2

l uses vision system to inspect printed characters on specularly reflective light bulb.

Ye and Ong [³l uses vision system for the inspection of IC dies. These methods of quality inspection have greatly increased the productivity rate in the manufacturing field.

Extensive research area also goes on image conversion from raster images to vector images and vice versa which is more related to the field of computer graphics, but the proposed research does not attempt to this extent. For clarification purposes, the proposed research does not concentrate on this area nor does it uses any file

2

translation format; for example, DXF file or IGES file format, for importing or exporting images into or from CAD drawing module.

Most of the research on 3D objects or models construction in machine vision is used for recognition purposes, mainly in the field of robot vision. This technique involved training or learning phase for the vision system to recognize the object. The primitive models of the object are stored in a database system. By performing the degree of matching between the image scene with the database model, the automatic construction of the object can be done. This is the method employed by Gros [⁴l.

Similar technique is also employed by Eklundh, Olofsson and Li 1⁵l _ Shapiro and Costa [⁶l used a database of appearance-based object models where each object is represented by a small set of view classes which consists of a set of features that have been detected in one or more training images of the view classes. By performing the degree of matching, it can generate the appearance of the 3D object. Another area of research concentrates on surface geometric generation of a 3D model or the object surface contour. Koivunen and Bajcsy ^[?J uses spline representation, while Besl [SJ

employed triangularization to generate the surface of the model. Startchik, Bost, Rauber, Milanese and Pun [⁹l focus on the extraction and grouping of image primitives into geometric models for object recognition. Thomas and Lim [!OJ attempted using constant contours of constant intensity as matching primitives in a stereo reconstruction system to obtain an initial 3D model of an object.

3

Another technique called stereo imaging maps every world (scene) coordinates system to image coordinates system to generate the model of the object. This method requires different angle of viewing to obtain the images. Lu and Hernandez (l IJ uses stereo pair of CCD cameras to obtain surface data to construct the 3D human face. Koivunen ^[I2J uses data provided by 3D imaging sensors to construct the geometric models of an object. Other method employed structured lighting; projection of a known light mask on the object to obtain the depth information before proceeding to construct the model of the object.

Other technique such as STL (stereo lithography) method is used to construct 3D model. This method is used widely in CAD/CAM rapid prototyping field. Others employed scanning the depth information through ascoutic wave reflection and measuring the reflected return wave to determine the depth information, while some researchers use CMM ( coordinates measuring machine) to obtain the surface information. The approach of using a CMM is used by Lin and Chen [I3

l to gather surface information before reconstructing the model of the object.

Another area of research called image visualization exists. This deals with object or model construction for visualization purposes. This method involves the construction of 3D model based solely on primitive models. Such visualization are used in computer tomography purposes, terrain surface mapping or remote sensing based on range images and surface triangularization for surface modeling. All of these methods go beyond the scope of image processing and machine vision into another area called

4

computer graphics.

In image processing alone, there are huge volumes of literature on feature detection, part recognition, object matching and etc. Various types of detection methods are developed and implemented, such as boundary detection, edge detection or comer detection.

For line detection or other features from imperfect edges, the Hough [l4

J transform method is an efficient method. Ballard ^{[l 5J} discusses the generalized Hough transform to detect arbitrary shapes. Stockman £¹⁶¹ used Hough transform techniques for pose clustering in 2D and 3D problems in object recognition and localization. Illingworth and Kittler [l7

J presented the adaptive Hough transform method, while Sklansky [l&J

used Hough transform technique for curve detection. Brady and Asada [l9

l present a rigorous approach to contour and region descriptors. Curve fitting using line segments (poly lines) and circular arcs is adapted by Pavlidis [^201. The method for fitting cubic splines to edge points using orientation was published by Tehrani, Weymouth and Schunck l²¹¹•

On comer detection techniques, for example, there are many detection methods presented by past researchers. These comer detection methods include Medioni- y asumoto rnJ comer detection, Beus-Tiu l²³l comer detection, Rosenfeld-Johnston [²⁴l

comer detector, Rosenfeld-Weska l²⁵J comer detector and Cheng-Hsu [²⁶¹ comer detector . Liu and Srinath [²⁷¹, made the comparison between various comer detection schemes available on 2D images.

5

Most of the edge/boundary detection methods employed in image processing application are based on mask operator such as Canny (²⁸¹, Marr-Hildreth r²⁹l, Frei- Chen (³⁰¹, Roberts and Sobel (³¹¹. Haralick (³²¹, Davies l³³l and Hueckel [³⁴l present various methods using mask operator on image processing application.

Some edge detection methods employ statistical technique. For example, Y akimovsky

l³⁵l, who treated edge detection as an exercise in hypothesis testing. While, Haralick

[32

l presented an edge detection scheme based on the second directional derivative. His scheme incorporated a form of image smoothing based on approximating the image with local surface patches.

Though many edge detectors have been developed, there is still no well-defined metric to help in selecting the appropriate edge detector for certain application.

Certain edge detectors may be more appropriate or more successful in ce1iain circumstances.

For example, Kamal (³⁶• 371

, has developed boundary detection method based on Freeman Chain Coding principle for the use in robotics application. His boundary detection method was implemented on 2D images for detection of automobile light cluster base. Koller, Gerig, Szekely and Dettwiler [³⁸¹, presented a parameter-free teclmique for the segmentation and local description of line structures on multiple scales, both in 2D and 3D on image information.

6

Henricsson, Streilein and Gruen l³⁹l, presented a method which combines image processing application and CAD application system, in their work of automated 3D reconstruction of buildings and photo realistic visualization within the context of 3D city model. While, Lin and Lu [⁴0J had integrated a CAD package and a vision system to attain a higher level of intelligence and automation in the context of manufacturing environments.

From the survey in the literature review, it can be concluded that the field of image processing and machine vision system plays a very important role today. Vision system not only is used in machine inspection but it has expands into wider fields.

More and more applications can be generated with the integration of vision system, such automated object recognition, object reconstruction, robotic vision application and etc.

1.3 STATEMENT OF PROBLEM

Duplication is a process of making a copy of an existing object. To obtain a drawing out of an object, first it requires that the profile of the object to be determined, then followed by measurement procedure to obtain appropriate dimension of the object and later draw the object profile according to the dimension on a CAD system. This way of obtaining the 2D drawing is tedious, time consuming and laborious.

7

1.4 SOLUTION APPROACH

From the statement of problem, there is a need to solve this non-productive activity.

Thus, the aim of this research work is to come up with a solution or a method of producing 2D drawing automatically through the help of a vision system. Since the area of image processing is so wide, the research work concentrates on 2D image processing and attempts to come up with a prototype system that is capable of detecting the object's 2D profile and performing automatic drawing so as to produce the 2D drawing of the object automatically. Without using much human effort, the system will be able to detect and trace the profile of the object to be duplicated and transfer the obtained profile information into a data file, which will be used later by a CAD automatic drawing module. The automatic drawing module interprets the data stored in the data file and performs automatic drawing so that a 2D drawing of the object can be obtained. By doing this, the system will provide a solution or a way to simplify the task of producing 2D drawing of an object.

1.5 OBJECTIVES OF THE RESEARCH WORK

The sole purpose of this research work is to come up with a system that can provide a solution to simplify the task of producing 2D drawing of an object. This method can greatly reduce the time required and eliminating the task of object measurement and manual drawing by means of automating these tasks with the help of a vision system and CAD system. The following objectives are the prime aims of this research work.

8