- •Getting Started
- •Product Description
- •Key Features
- •Configuration Notes
- •Related Products
- •Compilability
- •Image Import and Export
- •Introduction
- •Step 1: Read and Display an Image
- •Step 2: Check How the Image Appears in the Workspace
- •Step 3: Improve Image Contrast
- •Step 4: Write the Image to a Disk File
- •Step 5: Check the Contents of the Newly Written File
- •Image Enhancement and Analysis
- •Introduction
- •Step 1: Read Image
- •Step 2: Use Morphological Opening to Estimate the Background
- •Step 3: View the Background Approximation as a Surface
- •Step 4: Subtract the Background Image from the Original Image
- •Step 5: Increase the Image Contrast
- •Step 6: Threshold the Image
- •Step 7: Identify Objects in the Image
- •Step 8: Examine One Object
- •Step 9: View All Objects
- •Step 10: Compute Area of Each Object
- •Step 11: Compute Area-based Statistics
- •Step 12: Create Histogram of the Area
- •Getting Help
- •Product Documentation
- •Image Processing Examples
- •MATLAB Newsgroup
- •Acknowledgments
- •Introduction
- •Images in MATLAB
- •Expressing Image Locations
- •Pixel Indices
- •Spatial Coordinates
- •Intrinsic Coordinates
- •World Coordinates
- •Image Types in the Toolbox
- •Overview of Image Types
- •Binary Images
- •Indexed Images
- •Grayscale Images
- •Truecolor Images
- •Converting Between Image Types
- •Converting Between Image Classes
- •Overview of Image Class Conversions
- •Losing Information in Conversions
- •Converting Indexed Images
- •Working with Image Sequences
- •Overview of Toolbox Functions That Work with Image Sequences
- •Process Image Sequences
- •Process Multi-Frame Image Arrays
- •Image Arithmetic
- •Overview of Image Arithmetic Functions
- •Image Arithmetic Saturation Rules
- •Nesting Calls to Image Arithmetic Functions
- •Getting Information About a Graphics File
- •Reading Image Data
- •Writing Image Data to a File
- •Overview
- •Specifying Format-Specific Parameters
- •Reading and Writing Binary Images in 1-Bit Format
- •Determining the Storage Class of the Output File
- •Converting Between Graphics File Formats
- •Working with DICOM Files
- •Overview of DICOM Support
- •Reading Metadata from a DICOM File
- •Handling Private Metadata
- •Creating Your Own Copy of the DICOM Dictionary
- •Reading Image Data from a DICOM File
- •Viewing Images from DICOM Files
- •Writing Image Data or Metadata to a DICOM File
- •Writing Metadata with the Image Data
- •Understanding Explicit Versus Implicit VR Attributes
- •Removing Confidential Information from a DICOM File
- •Example: Creating a New DICOM Series
- •Working with Mayo Analyze 7.5 Files
- •Working with Interfile Files
- •Working with High Dynamic Range Images
- •Understanding Dynamic Range
- •Reading a High Dynamic Range Image
- •Creating a High Dynamic Range Image
- •Viewing a High Dynamic Range Image
- •Writing a High Dynamic Range Image to a File
- •Image Display and Exploration Overview
- •Displaying Images Using the imshow Function
- •Overview
- •Specifying the Initial Image Magnification
- •Controlling the Appearance of the Figure
- •Displaying Each Image in a Separate Figure
- •Displaying Multiple Images in the Same Figure
- •Dividing a Figure Window into Multiple Display Regions
- •Using the subimage Function to Display Multiple Images
- •Using the Image Tool to Explore Images
- •Image Tool Overview
- •Opening the Image Tool
- •Specifying the Initial Image Magnification
- •Specifying the Colormap
- •Importing Image Data from the Workspace
- •Exporting Image Data to the Workspace
- •Using the getimage Function to Export Image Data
- •Saving the Image Data Displayed in the Image Tool
- •Closing the Image Tool
- •Printing the Image in the Image Tool
- •Exploring Very Large Images
- •Overview
- •Creating an R-Set File
- •Opening an R-Set File
- •Using Image Tool Navigation Aids
- •Navigating an Image Using the Overview Tool
- •Starting the Overview Tool
- •Moving the Detail Rectangle to Change the Image View
- •Specifying the Color of the Detail Rectangle
- •Getting the Position and Size of the Detail Rectangle
- •Printing the View of the Image in the Overview Tool
- •Panning the Image Displayed in the Image Tool
- •Zooming In and Out on an Image in the Image Tool
- •Specifying the Magnification of the Image
- •Getting Information about the Pixels in an Image
- •Determining the Value of Individual Pixels
- •Saving the Pixel Value and Location Information
- •Determining the Values of a Group of Pixels
- •Selecting a Region
- •Customizing the View
- •Determining the Location of the Pixel Region Rectangle
- •Printing the View of the Image in the Pixel Region Tool
- •Determining the Display Range of an Image
- •Measuring the Distance Between Two Pixels
- •Using the Distance Tool
- •Exporting Endpoint and Distance Data
- •Customizing the Appearance of the Distance Tool
- •Adjusting Image Contrast Using the Adjust Contrast Tool
- •Understanding Contrast Adjustment
- •Starting the Adjust Contrast Tool
- •Using the Histogram Window to Adjust Image Contrast
- •Using the Window/Level Tool to Adjust Image Contrast
- •Example: Adjusting Contrast with the Window/Level Tool
- •Modifying Image Data
- •Saving the Modified Image Data
- •Cropping an Image Using the Crop Image Tool
- •Viewing Image Sequences
- •Overview
- •Viewing Image Sequences in the Movie Player
- •Example: Viewing a Sequence of MRI Images
- •Configuring the Movie Player
- •Specifying the Frame Rate
- •Specifying the Color Map
- •Getting Information about the Image Frame
- •Viewing Image Sequences as a Montage
- •Converting a Multiframe Image to a Movie
- •Displaying Different Image Types
- •Displaying Indexed Images
- •Displaying Grayscale Images
- •Displaying Grayscale Images That Have Unconventional Ranges
- •Displaying Binary Images
- •Changing the Display Colors of a Binary Image
- •Displaying Truecolor Images
- •Adding a Colorbar to a Displayed Image
- •Printing Images
- •Printing and Handle Graphics Object Properties
- •Setting Toolbox Preferences
- •Retrieving the Values of Toolbox Preferences Programmatically
- •Setting the Values of Toolbox Preferences Programmatically
- •Overview
- •Displaying the Target Image
- •Creating the Modular Tools
- •Overview
- •Associating Modular Tools with a Particular Image
- •Getting the Handle of the Target Image
- •Specifying the Parent of a Modular Tool
- •Tools With Separate Creation Functions
- •Example: Embedding the Pixel Region Tool in an Existing Figure
- •Positioning the Modular Tools in a GUI
- •Specifying the Position with a Position Vector
- •Build a Pixel Information GUI
- •Adding Navigation Aids to a GUI
- •Understanding Scroll Panels
- •Example: Building a Navigation GUI for Large Images
- •Customizing Modular Tool Interactivity
- •Overview
- •Build Image Comparison Tool
- •Creating Your Own Modular Tools
- •Overview
- •Create Angle Measurement Tool
- •Spatial Transformations
- •Resizing an Image
- •Overview
- •Specifying the Interpolation Method
- •Preventing Aliasing by Using Filters
- •Rotating an Image
- •Cropping an Image
- •Perform General 2-D Spatial Transformations
- •Spatial Transformation Procedure
- •Translate Image Using maketform and imtransform
- •Step 1: Import the Image to Be Transformed
- •Step 2: Define the Spatial Transformation
- •Step 3: Create the TFORM Structure
- •Step 4: Perform the Transformation
- •Step 5: View the Output Image
- •Defining the Transformation Data
- •Using a Transformation Matrix
- •Using Sets of Points
- •Creating TFORM Structures
- •Performing the Spatial Transformation
- •Specifying Fill Values
- •Performing N-Dimensional Spatial Transformations
- •Register Image Using XData and YData Parameters
- •Step 1: Read in Base and Unregistered Images
- •Step 2: Display the Unregistered Image
- •Step 3: Create a TFORM Structure
- •Step 4: Transform the Unregistered Image
- •Step 5: Overlay Base Image Over Registered Image
- •Step 6: Using XData and YData Input Parameters
- •Step 7: Using xdata and ydata Output Values
- •Image Registration
- •Image Registration Techniques
- •Control Point Registration
- •Using cpselect in a Script
- •Example: Registering to a Digital Orthophoto
- •Step 1: Read the Images
- •Step 2: Choose Control Points in the Images
- •Step 3: Save the Control Point Pairs to the MATLAB Workspace
- •Step 4: Fine-Tune the Control Point Pair Placement (Optional)
- •Step 6: Transform the Unregistered Image
- •Geometric Transformation Types
- •Selecting Control Points
- •Specifying Control Points Using the Control Point Selection Tool
- •Starting the Control Point Selection Tool
- •Using Navigation Tools to Explore the Images
- •Using Scroll Bars to View Other Parts of an Image
- •Using the Detail Rectangle to Change the View
- •Panning the Image Displayed in the Detail Window
- •Zooming In and Out on an Image
- •Specifying the Magnification of the Images
- •Locking the Relative Magnification of the Input and Base Images
- •Specifying Matching Control Point Pairs
- •Picking Control Point Pairs Manually
- •Using Control Point Prediction
- •Moving Control Points
- •Deleting Control Points
- •Exporting Control Points to the Workspace
- •Saving Your Control Point Selection Session
- •Using Correlation to Improve Control Points
- •Intensity-Based Automatic Image Registration
- •Registering Multimodal MRI Images
- •Step 1: Load Images
- •Step 2: Set up the Initial Registration
- •Step 3: Improve the Registration
- •Step 4: Improve the Speed of Registration
- •Step 5: Further Refinement
- •Step 6: Deciding When Enough is Enough
- •Step 7: Alternate Visualizations
- •Designing and Implementing 2-D Linear Filters for Image Data
- •Overview
- •Convolution
- •Correlation
- •Performing Linear Filtering of Images Using imfilter
- •Data Types
- •Correlation and Convolution Options
- •Boundary Padding Options
- •Multidimensional Filtering
- •Relationship to Other Filtering Functions
- •Filtering an Image with Predefined Filter Types
- •Designing Linear Filters in the Frequency Domain
- •FIR Filters
- •Frequency Transformation Method
- •Frequency Sampling Method
- •Windowing Method
- •Creating the Desired Frequency Response Matrix
- •Computing the Frequency Response of a Filter
- •Transforms
- •Fourier Transform
- •Definition of Fourier Transform
- •Visualizing the Fourier Transform
- •Discrete Fourier Transform
- •Relationship to the Fourier Transform
- •Visualizing the Discrete Fourier Transform
- •Applications of the Fourier Transform
- •Frequency Response of Linear Filters
- •Fast Convolution
- •Locating Image Features
- •Discrete Cosine Transform
- •DCT Definition
- •The DCT Transform Matrix
- •DCT and Image Compression
- •Radon Transform
- •Radon Transformation Definition
- •Plotting the Radon Transform
- •Viewing the Radon Transform as an Image
- •Detecting Lines Using the Radon Transform
- •The Inverse Radon Transformation
- •Inverse Radon Transform Definition
- •Improving the Results
- •Reconstruct Image from Parallel Projection Data
- •Fan-Beam Projection Data
- •Fan-Beam Projection Data Definition
- •Computing Fan-Beam Projection Data
- •Image Reconstruction Using Fan-Beam Projection Data
- •Reconstruct Image From Fanbeam Projections
- •Morphological Operations
- •Morphology Fundamentals: Dilation and Erosion
- •Understanding Dilation and Erosion
- •Processing Pixels at Image Borders (Padding Behavior)
- •Understanding Structuring Elements
- •The Origin of a Structuring Element
- •Creating a Structuring Element
- •Structuring Element Decomposition
- •Dilating an Image
- •Eroding an Image
- •Combining Dilation and Erosion
- •Morphological Opening
- •Skeletonization
- •Perimeter Determination
- •Morphological Reconstruction
- •Understanding Morphological Reconstruction
- •Understanding the Marker and Mask
- •Pixel Connectivity
- •Defining Connectivity in an Image
- •Choosing a Connectivity
- •Specifying Custom Connectivities
- •Flood-Fill Operations
- •Specifying Connectivity
- •Specifying the Starting Point
- •Filling Holes
- •Finding Peaks and Valleys
- •Terminology
- •Understanding the Maxima and Minima Functions
- •Finding Areas of High or Low Intensity
- •Suppressing Minima and Maxima
- •Imposing a Minimum
- •Creating a Marker Image
- •Applying the Marker Image to the Mask
- •Distance Transform
- •Labeling and Measuring Objects in a Binary Image
- •Understanding Connected-Component Labeling
- •Remarks
- •Selecting Objects in a Binary Image
- •Finding the Area of the Foreground of a Binary Image
- •Finding the Euler Number of a Binary Image
- •Lookup Table Operations
- •Creating a Lookup Table
- •Using a Lookup Table
- •Getting Image Pixel Values Using impixel
- •Creating an Intensity Profile of an Image Using improfile
- •Displaying a Contour Plot of Image Data
- •Creating an Image Histogram Using imhist
- •Getting Summary Statistics About an Image
- •Computing Properties for Image Regions
- •Analyzing Images
- •Detecting Edges Using the edge Function
- •Detecting Corners Using the corner Function
- •Tracing Object Boundaries in an Image
- •Choosing the First Step and Direction for Boundary Tracing
- •Detecting Lines Using the Hough Transform
- •Analyzing Image Homogeneity Using Quadtree Decomposition
- •Example: Performing Quadtree Decomposition
- •Analyzing the Texture of an Image
- •Understanding Texture Analysis
- •Using Texture Filter Functions
- •Understanding the Texture Filter Functions
- •Example: Using the Texture Functions
- •Gray-Level Co-Occurrence Matrix (GLCM)
- •Create a Gray-Level Co-Occurrence Matrix
- •Specifying the Offsets
- •Derive Statistics from a GLCM and Plot Correlation
- •Adjusting Pixel Intensity Values
- •Understanding Intensity Adjustment
- •Adjusting Intensity Values to a Specified Range
- •Specifying the Adjustment Limits
- •Setting the Adjustment Limits Automatically
- •Gamma Correction
- •Adjusting Intensity Values Using Histogram Equalization
- •Enhancing Color Separation Using Decorrelation Stretching
- •Simple Decorrelation Stretching
- •Adding a Linear Contrast Stretch
- •Removing Noise from Images
- •Understanding Sources of Noise in Digital Images
- •Removing Noise By Linear Filtering
- •Removing Noise By Median Filtering
- •Removing Noise By Adaptive Filtering
- •ROI-Based Processing
- •Specifying a Region of Interest (ROI)
- •Overview of ROI Processing
- •Using Binary Images as a Mask
- •Creating a Binary Mask
- •Creating an ROI Without an Associated Image
- •Creating an ROI Based on Color Values
- •Filtering an ROI
- •Overview of ROI Filtering
- •Filtering a Region in an Image
- •Specifying the Filtering Operation
- •Filling an ROI
- •Image Deblurring
- •Understanding Deblurring
- •Causes of Blurring
- •Deblurring Model
- •Importance of the PSF
- •Deblurring Functions
- •Deblurring with the Wiener Filter
- •Refining the Result
- •Deblurring with a Regularized Filter
- •Refining the Result
- •Deblurring with the Lucy-Richardson Algorithm
- •Overview
- •Reducing the Effect of Noise Amplification
- •Accounting for Nonuniform Image Quality
- •Handling Camera Read-Out Noise
- •Handling Undersampled Images
- •Example: Using the deconvlucy Function to Deblur an Image
- •Refining the Result
- •Deblurring with the Blind Deconvolution Algorithm
- •Example: Using the deconvblind Function to Deblur an Image
- •Refining the Result
- •Creating Your Own Deblurring Functions
- •Avoiding Ringing in Deblurred Images
- •Color
- •Displaying Colors
- •Reducing the Number of Colors in an Image
- •Reducing Colors Using Color Approximation
- •Quantization
- •Colormap Mapping
- •Reducing Colors Using imapprox
- •Dithering
- •Converting Color Data Between Color Spaces
- •Understanding Color Spaces and Color Space Conversion
- •Converting Between Device-Independent Color Spaces
- •Supported Conversions
- •Example: Performing a Color Space Conversion
- •Color Space Data Encodings
- •Performing Profile-Based Color Space Conversions
- •Understanding Device Profiles
- •Reading ICC Profiles
- •Writing Profile Information to a File
- •Example: Performing a Profile-Based Conversion
- •Specifying the Rendering Intent
- •Converting Between Device-Dependent Color Spaces
- •YIQ Color Space
- •YCbCr Color Space
- •HSV Color Space
- •Neighborhood or Block Processing: An Overview
- •Performing Sliding Neighborhood Operations
- •Understanding Sliding Neighborhood Processing
- •Determining the Center Pixel
- •General Algorithm of Sliding Neighborhood Operations
- •Padding Borders in Sliding Neighborhood Operations
- •Performing Distinct Block Operations
- •Understanding Distinct Block Processing
- •Implementing Block Processing Using the blockproc Function
- •Applying Padding
- •Block Size and Performance
- •TIFF Image Characteristics
- •Choosing Block Size
- •Using Parallel Block Processing on large Image Files
- •What is Parallel Block Processing?
- •When to Use Parallel Block Processing
- •How to Use Parallel Block Processing
- •Working with Data in Unsupported Formats
- •Learning More About the LAN File Format
- •Parsing the Header
- •Reading the File
- •Examining the LanAdapter Class
- •Using the LanAdapter Class with blockproc
- •Understanding Columnwise Processing
- •Using Column Processing with Sliding Neighborhood Operations
- •Using Column Processing with Distinct Block Operations
- •Restrictions
- •Code Generation for Image Processing Toolbox Functions
- •Supported Functions
- •Examples
- •Introductory Examples
- •Image Sequences
- •Image Representation and Storage
- •Image Display and Visualization
- •Zooming and Panning Images
- •Pixel Values
- •Image Measurement
- •Image Enhancement
- •Brightness and Contrast Adjustment
- •Cropping Images
- •GUI Application Development
- •Edge Detection
- •Regions of Interest (ROI)
- •Resizing Images
- •Image Registration and Alignment
- •Image Filtering
- •Fourier Transform
- •Image Transforms
- •Feature Detection
- •Discrete Cosine Transform
- •Image Compression
- •Radon Transform
- •Image Reconstruction
- •Fan-beam Transform
- •Morphological Operations
- •Binary Images
- •Image Histogram
- •Image Analysis
- •Corner Detection
- •Hough Transform
- •Image Texture
- •Image Statistics
- •Color Adjustment
- •Noise Reduction
- •Filling Images
- •Deblurring Images
- •Image Color
- •Color Space Conversion
- •Block Processing
- •Index
- •Summary of Modular Tools
- •Rules for Dilation and Erosion
- •Rules for Padding Images
- •Supported Connectivities
- •Distance Metrics
- •File Header Content
Performing Sliding Neighborhood Operations
Performing Sliding Neighborhood Operations
In this section...
“Understanding Sliding Neighborhood Processing” on page 15-3
“Implementing Linear and Nonlinear Filtering as Sliding Neighborhood Operations” on page 15-5
Understanding Sliding Neighborhood Processing
A sliding neighborhood operation is an operation that is performed a pixel at a time, with the value of any given pixel in the output image being determined by the application of an algorithm to the values of the corresponding input pixel’s neighborhood. A pixel’s neighborhood is some set of pixels, defined by their locations relative to that pixel, which is called the center pixel. The neighborhood is a rectangular block, and as you move from one element to the next in an image matrix, the neighborhood block slides in the same direction. (To operate on an image a block at a time, rather than a pixel at a time,
use the distinct block processing function. See “Performing Distinct Block Operations” on page 15-8 for more information.)
The following figure shows the neighborhood blocks for some of the elements in a 6-by-5 matrix with 2-by-3 sliding blocks. The center pixel for each neighborhood is marked with a dot. For information about how the center pixel is determined, see “Determining the Center Pixel” on page 15-4.
Neighborhood Blocks in a 6-by-5 Matrix
15-3
15 Neighborhood and Block Operations
Determining the Center Pixel
The center pixel is the actual pixel in the input image being processed by the operation. If the neighborhood has an odd number of rows and columns, the center pixel is actually in the center of the neighborhood. If one of the dimensions has even length, the center pixel is just to the left of center or just above center. For example, in a 2-by-2 neighborhood, the center pixel is the upper left one.
For any m-by-n neighborhood, the center pixel is
floor(([m n]+1)/2)
In the 2-by-3 block shown in the preceding figure, the center pixel is (1,2), or the pixel in the second column of the top row of the neighborhood.
General Algorithm of Sliding Neighborhood Operations
To perform a sliding neighborhood operation,
1Select a single pixel.
2Determine the pixel’s neighborhood.
3Apply a function to the values of the pixels in the neighborhood. This function must return a scalar.
4Find the pixel in the output image whose position corresponds to that of the center pixel in the input image. Set this output pixel to the value returned by the function.
5Repeat steps 1 through 4 for each pixel in the input image.
For example, the function might be an averaging operation that sums the values of the neighborhood pixels and then divides the result by the number of pixels in the neighborhood. The result of this calculation is the value of the output pixel.
Padding Borders in Sliding Neighborhood Operations
As the neighborhood block slides over the image, some of the pixels in a neighborhood might be missing, especially if the center pixel is on the border of
15-4
Performing Sliding Neighborhood Operations
the image. For example, if the center pixel is the pixel in the upper left corner of the image, the neighborhoods include pixels that are not part of the image.
To process these neighborhoods, sliding neighborhood operations pad the borders of the image, usually with 0’s. In other words, these functions process the border pixels by assuming that the image is surrounded by additional rows and columns of 0’s. These rows and columns do not become part of the output image and are used only as parts of the neighborhoods of the actual pixels in the image.
Implementing Linear and Nonlinear Filtering as Sliding Neighborhood Operations
You can use sliding neighborhood operations to implement many kinds of filtering operations. One example of a sliding neighbor operation is
convolution, which is used to implement linear filtering. MATLAB provides the conv and filter2 functions for performing convolution, and the toolbox provides the imfilter function. See “2-D Filtering” for more information about these functions.
In addition to convolution, there are many other filtering operations you can implement through sliding neighborhoods. Many of these operations are nonlinear in nature. For example, you can implement a sliding neighborhood operation where the value of an output pixel is equal to the standard deviation of the values of the pixels in the input pixel’s neighborhood.
To implement a variety of sliding neighborhood operations, use the nlfilter function. nlfilter takes as input arguments an image, a neighborhood size, and a function that returns a scalar, and returns an image of the same size as the input image. nlfilter calculates the value of each pixel in the output image by passing the corresponding input pixel’s neighborhood to the function.
Note Many operations that nlfilter can implement run much faster if the computations are performed on matrix columns rather than rectangular neighborhoods. For information about this approach, see “Using Columnwise Processing to Speed Up Sliding Neighborhood or Distinct Block Operations” on page 15-26.
15-5
15 Neighborhood and Block Operations
For example, this code computes each output pixel by taking the standard deviation of the values of the input pixel’s 3-by-3 neighborhood (that is, the pixel itself and its eight contiguous neighbors).
I = imread('tire.tif');
I2 = nlfilter(I,[3 3],'std2');
You can also write code to implement a specific function, and then use this function with nlfilter. For example, this command processes the matrix I in 2-by-3 neighborhoods with a function called myfun.m. The syntax @myfun is an example of a function handle.
I2 = nlfilter(I,[2 3],@myfun);
If you prefer not to write code to implement a specific function, you can use an anonymous function instead. This example converts the image to class double because the square root function is not defined for the uint8 datatype.
I = im2double(imread('tire.tif')); f = @(x) sqrt(min(x(:)));
I2 = nlfilter(I,[2 2],f);
(For more information on function handles and anonymous functions, see function_handle in the MATLAB Function Reference documentation.)
The following example uses nlfilter to set each pixel to the maximum value in its 3-by-3 neighborhood.
Note This example is only intended to illustrate the use of nlfilter. For a faster way to perform this local maximum operation, use imdilate.
I = imread('tire.tif'); f = @(x) max(x(:));
I2 = nlfilter(I,[3 3],f); imshow(I);
figure, imshow(I2);
15-6
Performing Sliding Neighborhood Operations
Each Output Pixel Set to Maximum Input Neighborhood Value
15-7
