roman louban - image processing of edge and surface defects

174 289 1
roman louban  -  image processing of edge and surface defects

Đang tải... (xem toàn văn)

Tài liệu hạn chế xem trước, để xem đầy đủ mời bạn chọn Tải xuống

Thông tin tài liệu

[...]... prove and ensure secure and precise edge detection 2.1.3 Multiple Edges A further complexity in edge detection arises with multiple edges, that is, composed of more than two single edges Analogously to the detection of a double edge, each one of them can be detected either beginning on the wood surface or on the background 2.1 Detection of an Edge 25 Fig 2.9 Detection of a double edge on non-square-edged... values at the edge, as with wood), sharpness of the edge representation (e.g., a cant), and the complexity of the edge (e.g., double edge as in a wood board with bark) 2.1 Detection of an Edge One of the most frequently used methods of direct edge detection from a grey scale image is based on a pre-determined edge model [3] and concerns the situation where the edge location must be known in advance Nevertheless,... Holmes Industrial image processing is gaining more and more importance as a testing methodology One of the most challenging and complex problems of industrial image processing is surface inspection, which is the process aimed at detecting a defect on a surface Often, the surface to be inspected is inhomogeneous and of high contrast Brightness fluctuations on the surface are common Still, all defects need... distance L0 and compared to the edgespecific minimum brightness I0 and to the edge- specific minimum brightness increase ΔI0 (minimum difference of the grey scale values) The length of the test distance L0 , the edge- specific minimum brightness value I0 , and the edgespecific brightness increase ΔI0 are calculated using the brightness values of the test area Fig 2.1 On -edge detection methods (scheme) 12 2 Edge. .. between a defective and a faultless surface? In case of a defect, there is always a boundary between the defect and the defect-free surface – a material edge For example, this edge can be identified on an angular grinding of a metallic surface that has a crack (Fig 1.3) [13] The roughness profile of the test surface shows the same result (Fig 1.4a) An intact surface cannot show such edges (Fig 1.4b) Fig... edge detection and evaluation and are used together as source data In this way, the physical background of edge formation is taken into consideration and with it the pre-conditions for dynamically determining edge- specific and imagecharacteristic parameters are created, which are adapted to global and local brightness conditions A technique for adaptive edge detection [20] is defined 2.1 Detection of. .. the edge between the background and the bark According to (2.4), it can be defined as follows: 22 2 Edge Detection Fig 2.6 Schematic presentation of the double edge detection technique: (a) plan view and (b) cross-sectional view 2.1 Detection of an Edge 23 Fig 2.7 Grey-scale profile across a double edge Fig 2.8 Histogram of the test area at the double edge Isurf virt = Ibgrd /ξ2 (2.31) The area-characteristic... calculated for the edge between the bark and the surface (Fig 2.7) In this case, the area-characteristic values for the edge- specific minimum brightness I0 surf and the edge- specific minimum brightness increase ΔI0 surf can be calculated on the basis of the surface brightness Isurf The calculation of these values can be carried out in the same way as was done with the edge between the background and the bark... recognition of defects This book will present an approach to this problem that allows the development of an algorithm suitable for the recognition of a surface defect This algorithm has been implemented as C-library functions for Seelector by hema electronic GmbH (a digital signal processing image processing system) [1] and as plug-ins for NeuroCheck (a PC image processing system) [2] and has been... irrespective of other problems and without identifying regular objects as defects There are a number of image processing systems that are able to carry out surface inspection more or less successfully However, the requirements of industry are growing so rapidly and on such a large scale that existing systems can no longer satisfy the demand The reason for this is not the computing capacity of an image processing . 093 3-0 33X ISBN 97 8-3 -6 4 2-0 068 2-1 e-ISBN 97 8-3 -6 4 2-0 068 3-8 DOI 10.1007/97 8-3 -6 4 2-0 068 3-8 Springer Dordrecht Heidelberg LondonNewYork Library of Congress Control Number: 2009929025 c  Springer-Verlag. http://www.springer.com/series/856 Roman Louban Image Processing of Edge and Surface Defects Theoretical Basis of Adaptive Algorithms with Numerous Practical Applications With118Figures 123 Dr. Roman Louban Thermosensorik. industrial image processing tried to imitate this ability by its own techniques. This book discusses the recognition of defects on free-form edges and in- homogeneous surfaces. My many years of experience

Ngày đăng: 05/06/2014, 12:05

Từ khóa liên quan

Mục lục

  • front-matter.pdf

    • Image Processing of Edge and Surface Defects

  • fulltext.pdf

    • 1 Introduction

      • 1.1 What Does an Image Processing Task Look Like?

      • 1.2 Conventional Methods of Defect Recognition

        • 1.2.1 Structural Analysis

        • 1.2.2 Edge-Based Segmentation with Pre-definedThresholds

      • 1.3 Adaptive Edge-Based Object Detection

  • fulltext_001.pdf

    • 2 Edge Detection

      • 2.1 Detection of an Edge

        • 2.1.1 Single Edge

        • 2.1.2 Double Edge

        • 2.1.3 Multiple Edges

      • 2.2 Non-Linear Approximation as Edge Compensation

  • fulltext_002.pdf

    • 3 Defect Detection on an Edge

      • 3.1 Defect Recognition on a Regular Contour

      • 3.2 Defect Detection on a Dented Wheel Contour

      • 3.3 Recognition of a Defect on a Free-Form Contour

        • 3.3.1 Fundamentals on Morphological Enveloping Filtering

        • 3.3.2 Defect Recognition on a Linear Edge Using an Envelope Filter

        • 3.3.3 Defect Recognition on a Free-Form Edge Using an Envelope Filter

  • fulltext_003.pdf

    • 4 Defect Detection on an Inhomogeneous High-Contrast Surface

      • 4.1 Defect Edge

      • 4.2 Defect Recognition

        • 4.2.1 Detection of Potential Defect Positions

        • 4.2.2 100% Defect Positions

        • 4.2.3 How Many 100% Defect Positions Must a Real Defect Have?

        • 4.2.4 Evaluation of Detected Defects

      • 4.3 Setup of Adaptivity Parameters of the SDD Algorithm

      • 4.4 Industrial Applications

        • 4.4.1 Surface Inspection of a Massive Metallic Part

        • 4.4.2 Surface Inspection of a Deep-Drawn Metallic Part

        • 4.4.3 Inspection of Non-Metallic Surfaces

        • 4.4.4 Position Determination of a Welded Joint

        • 4.4.5 Robot-Assisted Surface Inspection

  • fulltext_004.pdf

    • 5 Defect Detection on an Inhomogeneous Structured Surface

      • 5.1 How to Search for a Blob?

      • 5.2 Adaptive Blob Detection

        • 5.2.1 Adaptivity Level 1

        • 5.2.2 Further Adaptivity Levels

      • 5.3 Setup of Adaptivity Parameters of the ABD Algorithm

      • 5.4 Industrial Applications

        • 5.4.1 Cell Inspection using Microscopy

        • 5.4.2 Inspection of a Cold-Rolled Strip Surface

        • 5.4.3 Inspection of a Wooden Surface

  • fulltext_005.pdf

    • 6 Defect Detection in Turbo Mode

      • 6.1 What is the Quickest Way to Inspect a Surface?

      • 6.2 How to Optimize the Turbo Technique?

  • fulltext_006.pdf

    • 7 Adaptive Edge and Defect Detection as a basis for Automated Lumber Classification and Optimisation

      • 7.1 How to Grade a Wood Cutting?

        • 7.1.1 Boundary Conditions

        • 7.1.2 Most Important Lumber Terms

      • 7.2 Traditional Grading Methods

        • 7.2.1 Defect-Related Grading

        • 7.2.2 Grading by Sound Wood Cuttings

      • 7.3 Flexible Lumber Grading

        • 7.3.1 Adaptive Edge and Defect Detection

        • 7.3.2 Defect-Free Areas: From ``Spaghetti'' to ``Cutting''

        • 7.3.3 Simple Lumber Classification Using only Four Parameters

        • 7.3.4 The 3-Metres Principle

        • 7.3.5 Grading of Lumber with Red Heart

      • 7.4 The System for Automatic Classification and Sorting of Hardwood Lumber

        • 7.4.1 Structure of the Vision system

        • 7.4.2 User Interface

  • fulltext_007.pdf

    • 8 Object Detection on Images Captured Using a Special Equipment

      • 8.1 Evaluation of HDR Images

      • 8.2 Evaluation of X-ray Images

  • fulltext_008.pdf

    • 9 Before an Image Processing System is Used

      • 9.1 Calibration

        • 9.1.1 Evaluation Parameters

        • 9.1.2 Industrial Applications

      • 9.2 Geometrical Calibration

        • 9.2.1 h-Calibration

        • 9.2.2 l-Calibration

      • 9.3 Smallest Detectable Objects

        • 9.3.1 Technical Pre-Condition for Minimal Object Size

        • 9.3.2 Minimum Detectable Objects in Human Perception

  • back-matter.pdf

    • References

    • Index

Tài liệu cùng người dùng

  • Đang cập nhật ...

Tài liệu liên quan