imshow(I)

Grayscale Image pout.tif
Step 2: Check How the Image Appears in the Workspace
To see how the imread function stores the image data in the workspace, check the Workspace browser in the MATLAB desktop. The Workspace browser displays information about all the variables you create during a MATLAB session. The imread function returned the image data in the variable I, which is a 291-by-240 element array of uint8 data. MATLAB can store images as uint8, uint16, or double arrays. You can also get information about variables in the workspace by calling the

whos command. whos

MATLAB responds with

Name I

Size 291x240

Bytes 69840

Class uint8

Attributes

For more information about image storage classes, see Converting Between Image Classes on page 2-18.
Step 3: Improve Image Contrast
pout.tif is a somewhat low contrast image. To see the distribution of intensities in pout.tif, you can create a histogram by calling the imhist function. (Precede the call to imhist with the figure command so that the histogram does not overwrite the display of the image I in the current figure

window.)

figure, imhist(I)
Notice how the intensity range is rather narrow. It does not cover the potential range of [0, 255], and is missing the high and low values that would result in good contrast. The toolbox provides several ways to improve the contrast in an image. One way is to call the histeq function to spread the intensity values over the full range of the image, a process called histogram equalization.

I2 = histeq(I);

Display the new equalized image, I2, in a new figure window.

figure, imshow(I2)

Equalized Version of pout.tif
Call imhist again to create a histogram of the equalized image I2. If you compare the two histograms, the histogram of I2 is more spread out than the histogram of I1.

figure, imhist(I2)

The toolbox includes several other functions that perform contrast adjustment, including the imadjust and adapthisteq functions. See Adjusting Pixel Intensity Values on page 11-37 for more information. In addition, the toolbox includes an interactive tool, called the Adjust Contrast tool, that you can use to adjust the contrast and brightness of an image displayed in the Image Tool. To use this tool, call the imcontrast function or access the tool from the Image Tool. For more information, see Adjusting Image Contrast Using the Adjust Contrast Tool on page 4-42.

Specifying the Colormap

A colormap is a matrix that can have any number of rows, but must have three columns. Each row in the colormap is interpreted as a color, with the first element specifying the intensity of red, the second green, and the third blue. To specify the color map used to display an indexed image or a grayscale image in the Image Tool, select the Choose Colormap option on the Tools menu. This activates the Choose Colormap tool. Using this tool you can select one of the MATLAB colormaps or select a colormap variable from the MATLAB workspace. When you select a colormap, the Image Tool executes the colormap function you specify and updates the image displayed. You can edit the colormap command in the Evaluate Colormap text box; for example, you can change the number of entries in the colormap (default is 256). You can enter your own colormap function in this field. Press Enter to execute the command. When you choose a colormap, the image updates to use the new map. If you click OK, the Image Tool applies the colormap and closes the Choose Colormap tool. If you click Cancel, the image reverts to the previous colormap.
Choose from MATLAB colormaps. Specify colormap in the workspace.
Evaluate a colormap function. Click OK to select the colormap and close the dialog box.
Image Tool Choose Colormap Tool
Importing Image Data from the Workspace
To import image data from the MATLAB workspace into the Image Tool, use the Import from Workspace option on the Image Tool File menu. In the dialog box, shown below, you select the workspace variable that you want to import into the workspace. The following figure shows the Import from Workspace dialog box. You can use the Filter menu to limit the images included in the list to certain image types, i.e., binary, indexed, intensity (grayscale), or truecolor.
Image Tool Import from Workspace Dialog Box
Exporting Image Data to the Workspace
To export the image displayed in the Image Tool to the MATLAB workspace, you can use the Export to Workspace option on the Image Tool File menu. (Note that when exporting data, changes to the display range will not be preserved.) In the dialog box, shown below, you specify the name you want to assign to the variable in the workspace. By default, the Image Tool prefills the variable name field with BW, for binary images, RGB, for truecolor images, and I for grayscale or indexed images. If the Image Tool contains an indexed image, this dialog box also contains a field where you can specify the name of the associated colormap.

Displaying the Target Image
As the foundation for any image processing GUI you create, use imshow to display the target image (or images) in a MATLAB figure window. (You can also use the MATLAB image or imagesc functions.) Once the image is displayed in the figure, you can associate any of the modular tools with the image displayed in the figure. This example uses imshow to display an image in a figure window.
himage = imshow('pout.tif');
Because some of the modular tools add themselves to the figure window containing the image, make sure that the Image Processing Toolbox ImshowBorder preference is set to 'loose', if you are using the imshow function. (This is the default setting.) By including a border, you ensure that the modular tools are not displayed over the image in the figure.
Creating the Modular Tools
In this section. Overview on page 5-11 Associating Modular Tools with a Particular Image on page 5-12 Getting the Handle of the Target Image on page 5-14 Specifying the Parent of a Modular Tool on page 5-15 Positioning the Modular Tools in a GUI on page 5-18 Example: Building a Pixel Information GUI on page 5-19 Adding Navigation Aids to a GUI on page 5-22
To associate a modular tool with a target image displayed in a MATLAB figure window, you must create the tool using the appropriate tool creation function. You specify a handle to the target image as an argument to the tool creation function. The function creates the tool and automatically sets up the interactivity connection between the tool and the target image. By default, most of the modular tool creation functions support a no-argument syntax that uses the image in the current figure as the target image. If the current figure contains multiple images, the tools associate themselves with the first image in the figure objects children (the last image created). impixelinfo, impixelinfoval and imdisplayrange can work with multiple images in a figure. For example, to use the Pixel Information tool with a target image, display the image in a figure window, using imshow, and then call the impixelinfo function to create the tool. In this example, the image in the current figure is the target image.

For example, imshow returns a handle to the image in this syntax.
hfig = figure; himage = imshow('moon.tif') himage = 152.0055
When you call the imhandles function, specifying a handle to the figure (or axes) containing the image, it returns a handle to the same image.
himage2 = imhandles(hfig) himage2 = 152.0055
Specifying the Parent of a Modular Tool
When you create a modular tool, in addition to specifying the target image, you can optionally specify the object that you want to be the parent of the tool. By specifying the parent, you determine where the tool appears on your screen. Using this syntax of the modular tool creation functions, you can add the tool to the figure window containing the target image, open the tool in a separate figure window, or create some other combination. Specifying the parent is optional; the modular tools all have a default behavior. Some of the smaller tools, such as the Pixel Information tool, use the parent of the target image as their parent, inserting themselves in the same figure window as the target image. Other modular tools, such as the Pixel Region tool or the Overview tool, open in separate figures of their own.
Tools With Separate Creation Functions
Two of the tools, the Pixel Region tool and the Overview tool, have a separate creation function to provide this capability. Their primary creation functions, imoverview and impixelregion, open the tools in a separate figure window. To specify a different parent, you must use the imoverviewpanel and impixelregionpanel functions.
Note The Overview tool and the Pixel Region tool provide additional capabilities when created in their own figure windows. For example, both tools include zoom buttons that are not part of their uipanel versions.
Example: Embedding the Pixel Region Tool in an Existing Figure
This example shows the default behavior when you create the Pixel Region tool using the impixelregion function. The tool opens in a separate figure window, as shown in the following figure.
himage = imshow('pout.tif') hpixelinfopanel = impixelinfo(himage); hdrangepanel = imdisplayrange(himage); hpixreg = impixelregion(himage);

Pixel Region tool

Pixel Region rectangle
Target Image with Pixel Region Tool in Separate Window (Default)
To embed the Pixel Region tool in the same window as the target image, you must specify the handle of the target images parent figure as the parent of the Pixel Region tool when you create it. The following example creates a figure and an axes object, getting handles to both objects. The example needs these handles to perform some repositioning of the objects in the figure to ensure their visibility. See Positioning the Modular Tools in a GUI on page 5-18 for more information. The example then creates the modular tools, specifying the figure containing the target image as the parent of the Pixel Region tool. Note that the example uses the impixelregionpanel function to create the tool.

mytform = cp2tform(input_points, base_points, 'projective');
Step 6: Transform the Unregistered Image
As the final step in image registration, transform the input image to bring it into alignment with the base image. You use imtransform to perform the transformation, passing it the input image and the TFORM structure, which defines the transformation. imtransform returns the transformed image. For more information about using imtransform, see Chapter 6, Spatial Transformations
registered = imtransform(unregistered, mytform);
The following figure shows the transformed image transparently overlaid on the base image to show the results of the registration. (To see how this is done, see Example: Performing Image Registration on page 6-22.

Transformation Types

The cp2tform function can infer the parameters for the following types of transformations, listed in order of complexity. 'nonreflective similarity' 'affine' 'projective' 'polynomial' (Order 2, 3, or 4) 'piecewise linear' 'lwm' The first four transformations, 'nonreflective similarity', 'affine', 'projective', and 'polynomial' are global transformations. In these transformations, a single mathematical expression applies to an entire image. The last two transformations, 'piecewise linear' and 'lwm' (local weighted mean), are local transformations. In these transformations, different mathematical expressions apply to different regions within an image. When exploring how different transformations affect the images you are working with, try the global transformations first. If these transformations are not satisfactory, try the local transformations: the piecewise linear transformation first, and then the local weighted mean transformation. Your choice of transformation type affects the number of control point pairs you need to select. For example, a nonreflective similarity transformation requires at least two control point pairs. A polynomial transformation of order 4 requires 15 control point pairs. For more information about these transformation types, and the special syntaxes they require, see cpselect.

Selecting Control Points

Although there is some loss of quality in the reconstructed image, it is clearly recognizable, even though almost 85% of the DCT coefficients were discarded.

Radon Transform

In this section. Radon Transformation Definition on page 9-20 Plotting the Radon Transform on page 9-23 Viewing the Radon Transform as an Image on page 9-25 Detecting Lines Using the Radon Transform on page 9-26
Note For information about creating projection data from line integrals along paths that radiate from a single source, called fan-beam projections, see Fan-Beam Projection Data on page 9-37. To convert parallel-beam projection data to fan-beam projection data, use the para2fan function.
Radon Transformation Definition
The radon function computes projections of an image matrix along specified directions. A projection of a two-dimensional function f(x,y) is a set of line integrals. The radon function computes the line integrals from multiple sources along parallel paths, or beams, in a certain direction. The beams are spaced 1 pixel unit apart. To represent an image, the radon function takes multiple, parallel-beam projections of the image from different angles by rotating the source around the center of the image. The following figure shows a single projection at a specified rotation angle.
Parallel-Beam Projection at Rotation Angle Theta
For example, the line integral of f(x,y) in the vertical direction is the projection of f(x,y) onto the x-axis; the line integral in the horizontal direction is the projection of f(x,y) onto the y-axis. The following figure shows horizontal and vertical projections for a simple two-dimensional function.
Horizontal and Vertical Projections of a Simple Function
Projections can be computed along any angle [[THETA]]. In general, the Radon transform of f(x,y) is the line integral of f parallel to the y-axis

R ( x) =

f ( x cos y sin , x sin + y cos ) dy

x cos y = sin

sin x cos y
The following figure illustrates the geometry of the Radon transform.
Geometry of the Radon Transform
Plotting the Radon Transform
You can compute the Radon transform of an image I for the angles specified in the vector theta using the radon function with this syntax.

[R,xp] = radon(I,theta);

The columns of R contain the Radon transform for each angle in theta. The vector xp contains the corresponding coordinates along the x-axis. The center pixel of I is defined to be floor((size(I)+1)/2); this is the pixel on the x-axis corresponding to x = 0. The commands below compute and plot the Radon transform at 0 and 45 of an image containing a single square object. xp is the same for all projection angles.

4 Reconstruct the head phantom image from the projection data created
in step 2 and display the results.
I1 = iradon(R1,10); I2 = iradon(R2,5); I3 = iradon(R3,2); imshow(I1) figure, imshow(I2) figure, imshow(I3)
The following figure shows the results of all three reconstructions. Notice how image I1, which was reconstructed from only 18 projections, is the least accurate reconstruction. Image I2, which was reconstructed from 36 projections, is better, but it is still not clear enough to discern clearly the small ellipses in the lower portion of the image. I3, reconstructed using 90 projections, most closely resembles the original image. Notice that when the number of projections is relatively small (as in I1 and I2), the reconstruction can include some artifacts from the back projection.
Inverse Radon Transforms of the Shepp-Logan Head Phantom

Fan-Beam Projection Data

In this section. Fan-Beam Projection Data Definition on page 9-37 Computing Fan-Beam Projection Data on page 9-38 Reconstructing an Image from Fan-Beam Projection Data on page 9-40 Example: Reconstructing a Head Phantom Image on page 9-41
Note For information about creating projection data from line integrals along parallel paths, see Radon Transform on page 9-20. To convert fan-beam projection data to parallel-beam projection data, use the fan2para function.
Fan-Beam Projection Data Definition
The fanbeam function computes projections of an image matrix along specified directions. A projection of a two-dimensional function f(x,y) is a set of line integrals. The fanbeam function computes the line integrals along paths that radiate from a single source, forming a fan shape. To represent an image, the fanbeam function takes multiple projections of the image from different angles by rotating the source around the center of the image. The following figure shows a single fan-beam projection at a specified rotation angle.
Fan-Beam Projection at Rotation Angle Theta
Computing Fan-Beam Projection Data
To compute fan-beam projection data, use the fanbeam function. You specify as arguments an image and the distance between the vertex of the fan-beam projections and the center of rotation (the center pixel in the image). The fanbeam function determines the number of beams, based on the size of the image and the settings of fanbeam parameters. The FanSensorGeometry parameter specifies how sensors are aligned. If you specify the value 'arc' for FanSensorGeometry (the default), fanbeam positions the sensors along an arc, spacing the sensors at 1 degree intervals. Using the FanSensorSpacing parameter, you can control the distance between sensors by specifying the angle between each beam. If you specify the value 'line' for FanSensorGeometry parameter, fanbeam position sensors along a straight line, rather than an arc. With 'line' geometry, the FanSensorSpacing parameter specifies the distance between the sensors, in pixels, along the x axis.

1 Generate the test image and display it.
2 Compute fan-beam projection data of the test image, using the
FanSensorSpacing parameter to vary the sensor spacing. The example uses the fanbeam arc geometry, so you specify the spacing between sensors
by specifying the angular spacing of the beams. The first call spaces the beams at 2 degrees; the second at 1 degree; and the third at 0.25 degrees. In each call, the distance between the center of rotation and vertex of the projections is constant at 250 pixels. In addition, fanbeam rotates the projection around the center pixel at 1 degree increments.
D = 250; dsensor1 = 2; F1 = fanbeam(P,D,'FanSensorSpacing',dsensor1); dsensor2 = 1; F2 = fanbeam(P,D,'FanSensorSpacing',dsensor2); dsensor3 = 0.25 [F3, sensor_pos3, fan_rot_angles3] = fanbeam(P,D,. 'FanSensorSpacing',dsensor3);
3 Plot the projection data F3. Because fanbeam calculates projection data
at rotation angles from 0 to 360 degrees, the same patterns occur at an offset of 180 degrees. The same features are being sampled from both sides. Compare this plot to the plot of the parallel-beam projection data of the head phantom using 90 projections in Example: Reconstructing an Image from Parallel Projection Data on page 9-33.
figure, imagesc(fan_rot_angles3, sensor_pos3, F3)
colormap(hot); colorbar xlabel('Fan Rotation Angle (degrees)') ylabel('Fan Sensor Position (degrees)')
4 Reconstruct the image from the fan-beam projection data using ifanbeam.
In each reconstruction, match the fan sensor spacing with the spacing used when the projection data was created in step 2. The example uses the OutputSize parameter to constrain the output size of each reconstruction to be the same as the size of the original image |P|.
output_size = max(size(P)); Ifan1 = ifanbeam(F1,D,. 'FanSensorSpacing',dsensor1,'OutputSize',output_size); figure, imshow(Ifan1) Ifan2 = ifanbeam(F2,D,. 'FanSensorSpacing',dsensor2,'OutputSize',output_size); figure, imshow(Ifan2)
Ifan3 = ifanbeam(F3,D,. 'FanSensorSpacing',dsensor3,'OutputSize',output_size); figure, imshow(Ifan3)
The following figure shows the result of each transform. Note how the quality of the reconstruction gets better as the number of beams in the projection increases. The first image, Ifan1, was created using 2 degree spacing of the beams; the second image, ifan2, was created using 1 degree spacing of the beams; the third image, ifan3, was created using 0.25 spacing of the beams.
Reconstructions of the Head Phantom Image from Fan-Beam Projections
This chapter describes the Image Processing Toolbox morphological functions. You can use these functions to perform common image processing tasks, such as contrast enhancement, noise removal, thinning, skeletonization, filling, and segmentation. Morphology Fundamentals: Dilation and Erosion on page 10-2 Morphological Reconstruction on page 10-17 Distance Transform on page 10-36 Labeling and Measuring Objects in a Binary Image on page 10-39 Lookup Table Operations on page 10-44

Determining which pixels create a connected component depends on how pixel connectivity is defined. For example, this binary image contains one foreground object or two, depending on the connectivity. If the foreground is 4-connected, the image is all one object there is no distinction between a foreground object and the background. However, if the foreground is 8-connected, the pixels set to 1 connect to form a closed loop and the image has two separate objects: the pixels in the loop and the pixels outside the loop.
Defining Connectivity in an Image
The following table lists all the standard two- and three-dimensional connectivities supported by the toolbox. See these sections for more information: Choosing a Connectivity on page 10-22 Specifying Custom Connectivities on page 10-23 Supported Connectivities Two-Dimensional Connectivities 4-connected Pixels are connected if their edges touch. This means that a pair of adjoining pixels are part of the same object only if they are both on and are connected along the horizontal or vertical direction.
Supported Connectivities (Continued) 8-connected Pixels are connected if their edges or corners touch. This means that if two adjoining pixels are on, they are part of the same object, regardless of whether they are connected along the horizontal, vertical, or diagonal direction.
Three-Dimensional Connectivities 6-connected Pixels are connected if their faces touch.

18-connected

Pixels are connected if their faces or edges touch.

26-connected

Pixels are connected if their faces, edges, or corners touch.

Choosing a Connectivity

The type of neighborhood you choose affects the number of objects found in an image and the boundaries of those objects. For this reason, the results of many morphology operations often differ depending upon the type of connectivity you specify. For example, if you specify a 4-connected neighborhood, this binary image contains two objects; if you specify an 8-connected neighborhood, the image has one object.
Specifying Custom Connectivities
You can also define custom neighborhoods by specifying a 3-by-3-by-.-by-3 array of 0s and 1s. The 1-valued elements define the connectivity of the neighborhood relative to the center element. For example, this array defines a North/South connectivity which can be used to break up an image into independent columns.
CONN = [ 0; 0; 0 ] CONN = 0
Note Connectivity arrays must be symmetric about their center element. Also, you can use a 2-D connectivity array with a 3-D image; the connectivity affects each "page" in the 3-D image.

The toolbox supports a set of functions that you can use for texture analysis. Texture analysis refers to the characterization of regions in an image by their texture content. Texture analysis attempts to quantify intuitive qualities described by terms such as rough, smooth, silky, or bumpy as a function of the spatial variation in pixel intensities. In this sense, the roughness or bumpiness refers to variations in the intensity values, or gray levels. Texture analysis is used in a variety of applications, including remote sensing, automated inspection, and medical image processing. Texture analysis can be used to find the texture boundaries, called texture segmentation. Texture analysis can be helpful when objects in an image are more characterized by their texture than by intensity, and traditional thresholding techniques cannot be used effectively.
Using Texture Filter Functions
The toolbox includes several texture analysis functions that filter an image using standard statistical measures, listed in the following table. Function
rangefilt stdfilt entropyfilt
Description Calculates the local range of an image. Calculates the local standard deviation of an image. Calculates the local entropy of a grayscale image. Entropy is a statistical measure of randomness.
These statistics can characterize the texture of an image because they provide information about the local variability of the intensity values of pixels in an
image. For example, in areas with smooth texture, the range of values in the neighborhood around a pixel will be a small value; in areas of rough texture, the range will be larger. Similarly, calculating the standard deviation of pixels in a neighborhood can indicate the degree of variability of pixel values in that region. The following sections provide additional information about the texture functions: Understanding the Texture Filter Functions on page 11-28 Example: Using the Texture Functions on page 11-29
Understanding the Texture Filter Functions
The functions all operate in a similar way: they define a neighborhood around the pixel of interest, calculate the statistic for that neighborhood, and use that value as the value of the pixel of interest in the output image. This example shows how the rangefilt function operates on a simple array.
A = [ 5; 10; 15; 20 ] A = 15 20
B = rangefilt(A) B = 11 6
The following figure shows how the value of element B(2,4) was calculated from A(2,4). By default, the rangefilt function uses a 3-by-3 neighborhood but you can specify neighborhoods or different shapes and sizes.

operation. This can reduce contrast-related ringing in the result. To exclude a pixel from processing, you create an array of the same size as the original image, and assign the value 0 to the pixels in the array that correspond to pixels in the original image that you want to exclude from processing. (See Accounting for Nonuniform Image Quality on page 13-14 for information about WEIGHT arrays.) To create a WEIGHT array, the example uses a combination of edge detection and morphological processing to detect high-contrast areas in the image. Because the blur in the image is linear, the example dilates the image twice. (For more information about using these functions, see Chapter 10, Morphological Operations.) To exclude the image boundary pixels (a
high-contrast area) from processing, the example uses padarray to assign the value 0 to all border pixels.
WEIGHT = edge(I,'sobel',.28); se1 = strel('disk',1); se2 = strel('line',13,45); WEIGHT = ~imdilate(WEIGHT,[se1 se2]); WEIGHT = padarray(WEIGHT(2:end-1,2:end-1),[2 2]); figure, imshow(WEIGHT) title('Weight Array')
6 Refine the guess at the PSF. The reconstructed PSF P returned by the first
pass at deconvolution shows a clear linearity, as shown in the figure in step 4. For the second pass, the example uses a new PSF, P1, which is the same as the restored PSF but with the small amplitude pixels set to 0.
P1 = P; P1(find(P1 < 0.01))=0;
7 Rerun the deconvolution, specifying the WEIGHT array and the modified
PSF. Note how the restored image has much less ringing around the sharp intensity contrast areas than the result of the first pass (step 4).
[J2 P2] = deconvblind(Blurred,P1,50,[],WEIGHT); figure, imshow(J2) title('Newly Deblurred Image'); figure, imshow(P2,[],'InitialMagnification','fit') title('Newly Reconstructed PSF')
The deconvblind function, by default, performs multiple iterations of the deblurring process. You can stop the processing after a certain number of iterations to check the result, and then restart the iterations from the point where processing stopped. To use this feature, you must pass in both the blurred image and the PSF as cell arrays, for example, {Blurred} and {INITPSF}. The deconvblind function returns the output image and the restored PSF as cell arrays. The output image cell array contains these four elements: Element

function [] = writeRegion(obj, region_start, region_data)
The first argument, region_start, indicates the first pixel of the block that the writeRegion method writes. The second argument, region_data, contains the new data that the method writes to the file. Classes that implement the writeRegion method can be more complex than LanAdapter. When creating a writable Image Adapter object, classes often have the additional responsibility of creating new files in the class constructor. This file creation requires a more complex syntax in the constructor, where you potentially need to specify the size and data type of a new file you want to create. Constructors that create new files can also encounter other issues, such as operating system file permissions or potentially difficult file-creation code.
Using the LanAdapter Class with blockproc
Now that you understand how the LanAdapter class works, you can use it to enhance the visible bands of a LAN file. Run the Computing Statistics for Large Images (ipexblockprocstats) demo to see how the blockproc function works with the LanAdapter class.
Using Columnwise Processing to Speed Up Sliding Neighborhood or Distinct Block Operations
In this section. Understanding Columnwise Processing on page 15-24 Using Column Processing with Sliding Neighborhood Operations on page 15-24 Using Column Processing with Distinct Block Operations on page 15-26
Understanding Columnwise Processing
Performing sliding neighborhood and distinct block operations columnwise, when possible, can reduce the execution time required to process an image. For example, suppose the operation you are performing involves computing the mean of each block. This computation is much faster if you first rearrange the blocks into columns, because you can compute the mean of every column with a single call to the mean function, rather than calling mean for each block individually. To use column processing, use the colfilt function. This function
1 Reshapes each sliding or distinct block of an image matrix into a column in

a temporary matrix

2 Passes the temporary matrix to a function you specify 3 Rearranges the resulting matrix back into the original shape

example 13-6 decorrelation stretching 11-45 See also contrast adjustment detail rectangle getting position of 4-25 in Control Point Selection Tool 7-16 specifying color in Image Tool 4-25 DICOM files reading and writing 3-9 DICOM unique identifier generating 3-17 Digital Imaging and Communications in Medicine. See DICOM files dilation 10-2 discrete cosine transform 9-16 image compression 9-18 transform matrix 9-18 discrete Fourier transform 9-7 display depth 14-2 See also screen color resolution display range getting information about 4-35 Display Range tool creating 5-5 in Image Tool 4-35 displaying images
adding a colorbar 4-74 binary 4-70 binary images with different colors 4-70 comparison of functions 4-2 controlling figure window 4-6 grayscale images 4-68 indexed images 4-67 initial size 4-6 multiple images 4-55 multiple images in figure 4-8 toolbox preferences for 4-78 truecolor 4-72 unconventional ranges of data 4-68 using imshow 4-4 using the Image Tool 4-11 distance between pixels 4-37 Euclidean 4-37 Distance tool creating 5-5 distance transform 10-36 distinct block operations 15-8 overlap 15-11 zero padding 15-8 distortion operator 13-2 dithering 14-11 example 14-11 dropper button in Adjust Contrast tool 4-46
edge example 11-12 edge detection 11-11 Canny method 11-11 example 11-12 Sobel method 11-12

edgetaper

avoiding boundary ringing 13-27

Index-4

enhancing images decorrelation stretching 11-45 intensity adjustment 11-38 noise removal 11-50 erosion 10-2 Euclidean distance 4-37 Euler number calculating 10-43 exploring images using Image Tool 4-23 exporting data in Image Tool 4-18 eye dropper button in Adjust Contrast tool 4-46
figure controlling borders 4-6 files displaying images from disk 4-4 filling a region 12-8 filling holes in images 10-25 filter design 8-15 frequency sampling method 8-18 frequency transformation method 8-16 windowing method 8-19

filter2

fan-beam projection data arc geometry 9-38 computing 9-37 line geometry 9-39 reconstructing image from 9-40

fanbeam

using 9-37 fast Fourier transform 9-7 zero padding 9-9 See also Fourier transform feature measurement 1-20 area 11-10 binary images 10-39 bounding box 11-10 center of mass 11-10 fft 9-7 fft2 9-7 example 9-8 using 9-10 fftn 9-7 fftshift example 8-17 using 9-10
compared to imfilter 8-13 example 11-52 filtering a region 12-5 masked filtering 12-5 filters adaptive 11-54 binary masks 12-5 computing frequency response 9-10 designing 8-15 finite impulse response (FIR) 8-16 frequency response 8-22 imfilter 8-5 Infinite Impulse Response (IIR) 8-16 linear 8-2 median 11-51 multidimensional 8-12 unsharp masking 8-13 FIR filters 8-16 transforming from one-dimensional to two-dimensional 8-16 flat-field correction 13-14 flood-fill operation 10-23 Fourier transform 9-2 applications of the Fourier transform 9-10 centering the zero-frequency coefficient 9-10 computing frequency response 9-10 convolution and 9-11 correlation 9-12

Image Processing Toolbox 7.1
Perform image processing, analysis, and algorithm development
Image Processing Toolbox provides a comprehensive set of reference-standard algorithms and graphical tools for image processing, analysis, visualization, and algorithm development. You can perform image enhancement, image deblurring, feature detection, noise reduction, image segmentation, spatial transformations, and image registration. Many functions in the toolbox are multithreaded to take advantage of multicore and multiprocessor computers. Image Processing Toolbox supports a diverse set of image types, including high dynamic range, gigapixel resolution, ICC-compliant color, and tomographic images. Graphical tools let you explore an image, examine a region of pixels, adjust the contrast, create contours or histograms, and manipulate regions of interest (ROIs). With the toolbox algorithms you can restore degraded images, detect and measure features, analyze shapes and textures, and adjust the color balance of images. Key Features Image enhancement, filtering, and deblurring Image analysis, including segmentation, morphology, feature extraction, and measurement Spatial transformations and image registration Image transforms, including FFT, DCT, Radon, and fan-beam projection Workflows for processing, displaying, and navigating arbitrarily large images Modular interactive tools, including ROI selections, histograms, and distance measurements ICC color management Multidimensional image processing Image-sequence and video display DICOM import and export Importing and Exporting Images Image Processing Toolbox supports images generated by a wide range of devices, including digital cameras, satellite and airborne sensors, medical imaging devices, microscopes, telescopes, and other scientific instruments. You can visualize, analyze, and process these images in many data types, including single- and double-precision floating-point and signed and unsigned 8-, 16-, and 32-bit integers. There are several ways to import and export images into and out of the MATLAB environment for processing. You can use Image Acquisition Toolbox to acquire live images from Web cameras, frame grabbers, DCAM-compatible cameras, and other devices. Using Database Toolbox, you can access images stored in ODBC/JDBC-compliant databases. MATLAB supports standard data and image formats, including JPEG, JPEG-2000, TIFF, PNG, HDF, HDF-EOS, FITS, Microsoft Excel, ASCII, and binary files. It also supports the multiband image formats BIP and BIL, as used by LANDSAT for example. Low-level I/O and memory mapping functions enable you to develop custom routines for working with any data format.
Image Processing Toolbox supports a number of specialized image file formats. For medical images, it supports the DICOM file format, including associated metadata, as well as the Analyze 7.5 and Interfile formats. The toolbox can also read geospatial images in the NITF format and high dynamic range images in the HDR format.
A display of MRI slices from a series of DICOM files, enabling visualization and exploration of medical images in MATLAB.
Displaying and Exploring Images Image Processing Toolbox extends MATLAB graphics to provide image display capabilities that are highly customizable. You can create displays with multiple images in a single window, annotate displays with text and graphics, and create specialized displays such as histograms, profiles, and contour plots. In addition to display functions, the toolbox provides a suite of interactive tools for exploring images and building GUIs. You can view image information, zoom and pan around the image, and closely examine a region of pixels. You can interactively place and manipulate ROIs, including points, lines, rectangles, polygons, ellipses, and freehand shapes. You can also interactively crop, adjust the contrast, and measure distances. The suite of tools is available within Image Tool or from individual functions that can be used to create customized GUIs.

A typical interactive session using Image Tool. The Overview window (left) is used to navigate when looking at magnified views in the Image Tool. The Pixel Region window (right) superimposes pixel values on a highly magnified view. LANDSAT image of Paris courtesy of Space Imaging, LLC.
Using region-of-interest tools to create a mask. Items in the original image (top) are selected to create the mask (bottom).
The toolbox includes tools for displaying video and sequences in either a time-lapsed video viewer or an image montage. Volume visualization tools in MATLAB let you create isosurface displays of multidimensional image data sets.
Video viewer paused on an individual frame of a video sequence.
Preprocessing and Postprocessing Images Image Processing Toolbox provides reference-standard algorithms for preprocessing and postprocessing tasks that solve frequent system problems, such as interfering noise, low dynamic range, out-of-focus optics, and the difference in color representation between input and output devices. Image enhancement techniques in Image Processing Toolbox enable you to increase the signal-to-noise ratio and accentuate image features by modifying the colors or intensities of an image. You can: Perform histogram equalization Perform decorrelation stretching Remap the dynamic range Adjust the gamma value Perform linear, median, or adaptive filtering The toolbox includes specialized filtering routines and a generalized multidimensional filtering function that handles integer image types, offers multiple boundary-padding options, and performs convolution and correlation. Predefined filters and functions for designing and implementing your own linear filters are also provided.
Performing connected components analysis on an image with nonuniform background intensity using MATLAB and Image Processing Toolbox.
Image deblurring algorithms in Image Processing Toolbox include blind, Lucy-Richardson, Wiener, and regularized filter deconvolution, as well as conversions between point spread and optical transfer functions. These functions help correct blurring caused by out-of-focus optics, movement by the camera or the subject during image capture, atmospheric conditions, short exposure time, and other factors. All deblurring functions work with multidimensional images.
Image of the sun using deblurring algorithms. Image courtesy of the SOHO EIT Consortium.
Device-independent color management in Image Processing Toolbox enables you to accurately represent color independently from input and output devices. This is useful when analyzing the characteristics of a device, quantitatively measuring color accuracy, or developing algorithms for several different devices. With specialized functions in the toolbox, you can convert images between device-independent color spaces, such as sRGB, XYZ, xyY, L*a*b*, uvL, and L*ch. For more flexibility and control, the toolbox supports profile-based color space conversions using a color management system based on ICC version 4. For example, you can import n-dimensional ICC color profiles, create new or modify existing ICC color profiles for specific input and output devices, specify the rendering intent, and find all compliant profiles on your machine. Image transforms such as FFT and DCT play a critical role in many image processing tasks, including image enhancement, analysis, restoration, and compression. Image Processing Toolbox provides several image transforms, including Radon and fan-beam projections. You can reconstruct images from parallel-beam and fan-beam projection data (common in tomography applications). Image transforms are also available in MATLAB and Wavelet Toolbox. Image conversions between data classes and image types are a common requirement for imaging applications. Image Processing Toolbox provides a variety of utilities for conversion between data classes, including single- and double-precision floating-point and signed or unsigned 8-, 16-, and 32-bit integers. The toolbox includes algorithms for conversion between image types, including binary, grayscale, indexed color, and truecolor. Specifically for color images, the toolbox supports a variety of color spaces (such as YIQ, HSV, and YCrCb) as well as Bayer pattern encoded and high dynamic range images.

Analyzing Images Image Processing Toolbox provides a comprehensive suite of reference-standard algorithms and graphical tools for image analysis tasks such as statistical analysis, feature extraction, and property measurement. Statistical functions let you analyze the general characteristics of an image by: Computing the mean or standard deviation Determining the intensity values along a line segment Displaying an image histogram Plotting a profile of intensity values
Image with histogram for the red channel.
Edge-detection algorithms let you identify object boundaries in an image. These algorithms include the Sobel, Prewitt, Roberts, Canny, and Laplacian of Gaussian methods. The powerful Canny method can detect true weak edges without being "fooled" by noise. Image segmentation algorithms determine region boundaries in an image. You can explore many different approaches to image segmentation, including automatic thresholding, edge-based methods, and morphology-based methods such as the watershed transform, often used to segment touching objects.
Detection and outlining of an aircraft using segmentation and morphology.
Morphological operators enable you to detect edges, enhance contrast, remove noise, segment an image into regions, thin regions, or perform skeletonization on regions. Morphological functions in Image Processing Toolbox include: Erosion and dilation Opening and closing Labeling of connected components Watershed segmentation Reconstruction Distance transform Image Processing Toolbox also contains advanced image analysis functions that let you: Measure the properties of a specified image region, such as the area, center of mass, and bounding box Detect lines and extract line segments from an image using the Hough transform Measure properties, such as surface roughness or color variation, using texture analysis functions Spatial Transformations and Image Registration Spatial transformations modify the spatial relationships between pixels in an image and are useful for tasks such as rotating an image, creating thumbnails, correcting geometric distortions, and performing image registration. Image Processing Toolbox supports common transformational operations, such as resizing, rotating, and interactive cropping of images, as well as generalized transformations for arbitrary-dimensional arrays. Image registration is important in remote sensing, medical imaging, and other applications where images must be aligned to enable quantitative analysis or qualitative comparison. Using Image Processing Toolbox, you can interactively select points in a pair of images and align the two images by performing a spatial transformation, such as linear conformal, affine, projective, polynomial, piecewise linear, or local weighted mean. You can also perform image registration using normalized 2D cross-correlation.

Choosing control points to register an aerial photo to an orthophoto. The Control Point Selection Tool helps you select landmark points and align images (result, bottom). Color aerial photo (top left) courtesy of mPower3/Emerge. Grayscale orthophoto (top right) courtesy of MassGIS.
Working with Large Images Some images are so large that they are difficult to process and display with standard methods. Image Processing Toolbox provides specific workflows for working with larger images than otherwise possible. Without loading a large image entirely into memory, you can create a reduced-resolution data set (R-Set) that divides an image into spatial tiles and resamples the image at different resolution levels. This workflow improves performance in image display and navigation. You can use a block processing workflow to apply a function to each distinct block of a large image, which significantly reduces memory use. An additional option for working with large images is to use the Parallel Computing Toolbox.
A large multispectral image of Paris. Dividing the image into blocks for processing reduces memory usage. Image courtesy Space Imaging, LLC.

Resources

Product Details, Demos, and System Requirements www.mathworks.com/products/image Trial Software www.mathworks.com/trialrequest Sales www.mathworks.com/contactsales Technical Support www.mathworks.com/support Online User Community www.mathworks.com/matlabcentral Training Services www.mathworks.com/training Third-Party Products and Services www.mathworks.com/connections Worldwide Contacts www.mathworks.com/contact
2010 The MathWorks, Inc. MATLAB and Simulink are registered trademarks of The MathWorks, Inc. See www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand names may be trademarks or registered trademarks of their respective holders.