Manuals / Brands / Computer Equipment / Network Card / National Instruments / Computer Equipment / Network Card

National Instruments IMAQ Vision for Measurement Studio Trainingthe Color Pattern Matching Algorithm

1 114

Download 114 pages, 1.93 Mb

Chapter5 MachineVision

IMAQVision for LabWindows/CVI User Manual 5-20 ni.com

Feature Detail

A templatewith relatively coarse features is less sensitive to variations in

sizeand rotation than a model with fine features. However, the model must

contain enough detail to identify it.

Positional Information

A templatewith strong edges in both the x and y directions is easier to

locate.

Background Information

Unique background information ina templateimproves search

performanceand accuracy during the grayscale pattern matching phase.

This requirement could conflict with the “color information”requirement

becauseb ackground colors may not be desirable during the color location

phase.A void this problem by choosing a template with sufficient

background information for grayscale pattern matching whilespecifying

the exclusion of the background color during the color location phase.

Refer to the Training the Color Pattern Matching Algorithm section for

more information about how to ignore colors.

Trainingthe Color Pattern Matching Algorithm

Afteryou have created a good template image, the color pattern

matching algorithm learns the important features of the template. Use

imaqLearnColorPattern()to learn the template.The learning process

depends on the type of matching that you expect to perform. By default,

the color pattern matching algorithm learns only those features from the

templatethat are necessaryfor shift-invariant m atching. However, if you

want to match the template at any orientation, the learning process must

consider the possibility of arbitrary orientations. Use the learnMode

elementof t he imaqLearnColorPattern()options parameter to

specify which type of learning mode to use.

Exclude colors in the template that you are not interested in using during

the search phase. Typicallyyou should ignore colors that either belong to

thebackground of the object or are not unique to th e template, reducing the

potential for incorrect matches during the color location phase. Youcan

ignore certain predefinedcolor susin g the ignoreModeelement of the

options parameter.To ignore other colors, first learn the colors to ignore

usingimaqLearnColor(). Then set the colorsToIgnore element of the

options parameter to the resulting ColorInformation structure from

imaqLearnColor().

Contents

Main IMAQ Page Important Information Warranty Copyright Trademarks WARNINGREGARDING USE OF NATIONAL INSTRUMENTS PRODUCTS Conventions Contents Chapter 1 Introduction to IMAQ Vision Chapter 2 Getting Measurement-Ready Images Chapter 3 Grayscale and Color Measurements Chapter 4 Blob Analysis Chapter 5 Machine Vision Page Introduction to IMAQ Vision About IMAQ Vision Documentation and Examples Application Development Environments IMAQ Vision Function Tree Page IMAQ Machine Vision Function Tree Creating IMAQ Vision Applications Chapter1 Introductionto IMAQ Vision Figure1-2. InspectionSteps for Building a Vision Application IMAQVision for LabWindows/CVI User Manual 1-6 ni.com Getting Measurement-Ready Images Set Up Your Imaging System Calibrate Your Imaging System Create an Image Page Source and Destination Images AcquireorReadanImage Acquiring an Image Reading a File Converting an Array to an Image Display an Image Attach Calibration Information Analyze an Image Improve an Image Lookup Tables Filters Convolution Filter Nth Order Filter Grayscale Morphology FFT Page Complex Image Operations Grayscale and Color Measurements Define Regions of Interest Interactively Defining Regions Page Page Page Page Programmatically Defining Regions Defining Regions with Masks Measure Grayscale Statistics Measure Color Statistics Figure3-4. Primary Components of a Color Image Comparing Colors Learning Color Information Choosing the Right Color Information Specifying the Color Information to Learn Using the Entire Image Using a Region in the Image Using Multiple Regions in the Image Choosing a Color Representation Sensitivity Ignoring Learned Colors Blob Analysis Correct Image Distortion Create a Binary Image Improve the Binary Image Removing Unwanted Blobs Separating TouchingBlobs Improving Blob Shapes Make Particle Measurements Page Page Convert Pixel Coordinates to Real-World Coordinates Machine Vision Locate Objects to Inspect UsingEdgeDetectiontoBuildaCoordinateTransform Page Using Pattern Matching to Build a Coordinate Transform Page Chapter5 MachineVision NationalInstruments Corporation 5-7 IMAQVision for LabWindows/CVI User Manual ChoosingaMethodtoBuildtheCoordinateTransform Figure5-4. Buildinga Coordinate Transform Set Search Areas Interactively Defining Regions Programmatically Defining Regions Find Measurement Points Finding Features Using Edge Detection Finding Lines or Circles Figure5-5. Findinga Straight Feature Finding Edge Points Along One Search Contour Finding Edge Points Along Multiple Search Contours Finding Points Using Pattern Matching Defining and Create Good TemplateImages Symmetry Feature detail Positional information Background information Trainingthe Pattern Matching Algorithm Defining a Search Area Setting Matching Parameters and Tolerances Match Mode Minimum Contrast Rotation Angle Ranges Testingthe Search Algorithm on Test Images Using a Ranking Method to Verify Results Finding Points Using Color Pattern Matching Defining and Creating Good Color TemplateImages Color Information Symmetry Feature Detail Positional Information Trainingthe Color Pattern Matching Algorithm Defining a Search Area Setting Matching Parameters and Tolerances Color Sensitivity Search Strategy Color Score Weight Minimum Contrast Rotation Angle Ranges Testingthe Search Algorithm on Test Images Finding Points Using Color Location Convert Pixel Coordinates to Real-World Coordinates Make Measurements Distance Measurements Analytic Geometry Measurements Instrument Reader Measurements Display Results Page Calibration Perspective and Nonlinear Distortion Calibration Defining a Calibration Template Defining a Reference Coordinate System Page Page Learning Calibration Information Specifying Scaling Factors Choosing a Region of Interest Choosing a Learning Algorithm Using the Learning Score Learning the Error Map Learning the Correction Table Setting the Scaling Method Calibration Invalidation Simple Calibration Save Calibration Information Attach Calibration Information A Technical Support Resources Web Support NI Developer Zone Customer Education System Integration Worldwide Support Glossary Numbers/Symbols A B C Page D E F G H I Page J K L M N O P Page Q R S T Page Index A B C D E F G H I Page Page L M N O P R S T V W Z