R2023b

New Features, Bug Fixes, Compatibility Considerations

Image Display and Exploration: Enhancements to Image Viewer app

The enhanced Image Viewer app provides more functionality and better ease of use. For example, you can now use the app to interactively draw polygon ROIs and measure their areas. Open the app using the new imageViewer function.

The enhanced Image Viewer app and the imageViewer function are recommended over the imtool function, which will be removed in a future release. For more information, see Functionality being removed or changed.

Volume Visualization: Cinematic rendering, volumetric light scattering, advanced lighting controls, and interactive cropping

The Volume object has new properties and rendering styles for displaying 3-D image volumes. To set property values at object creation, specify them to the volshow function as name-value arguments.

• Control the amount of light reflected by a volume using the new SpecularReflectance property. Increase the specular reflectance to make a volume appear shinier.

• Specify the RenderingStyle property as "CinematicRendering" to display a photorealistic volume using iterative postprocessing. Specify the number of postprocessing iterations using the new CinematicNumIterations property. For an example, see Display Volume Using Cinematic Rendering.

• Specify the RenderingStyle property as "LightScattering" to display the object with volumetric light scattering, including light absorption, inscattering, and outscattering. Specify the balance between rendering quality and speed using the new LightScatteringQuality property. For an example, see Display Translucent Volume with Advanced Light Scattering.

The Viewer3D object has new properties and tools to control the lighting and remove objects in a scene. To set properties at object creation, specify them to the viewer3d function as name-value arguments.

• Control the strength of ambient and diffuse light within a viewer using the new AmbientLight and DiffuseLight properties, respectively.

• Apply denoising to the objects in a viewer using the new Denoising property. Specify the amount of smoothing and standard deviation of the Gaussian smoothing kernel using the new DenoisingDegreeOfSmoothing and DenoisingSigma properties, respectively.

• Use a crop box to crop the objects in a viewer to a rectangular subregion. For an example, see the Crop Blocked Volume section of Display Large 3-D Images Using Blocked Volume Visualization.

• Use the 3-D scissors tool to remove regions from of a viewer. For an example, see Remove Objects from Volume Display Using 3-D Scissors.

Border selection and deletion: Remove or retain specified borders and border structures

The imclearborder function has a new Borders name-value argument that enables you to specify which image borders to remove or from which to remove connected structures.

The imkeepborder function retains only the light structures in an image that are connected to the image borders. You can specify a subset of image borders which to retain or for which to retain structures.

imclearborder: Improved performance

The imclearborder function shows improved performance. For example, in this code, the call to imclearborder is about 1.7x faster than in the previous release.

function t = imclearborderTimingTest 
    A = imbinarize(imread("rice.png"));
    f = @() imclearborder(A);
    t = timeit(f);
end

The approximate execution times are:

R2023a: 0.85 ms

R2023b: 0.51 ms

The code was timed on a macOS 12.5.1, Intel^® Core i9 CPU @ 3.6 GHz test system.

C Code Generation: Generate code from additional functions using MATLAB Coder

This list indicates the Image Processing Toolbox™ functions that have been enabled for code generation in this release. For all target platforms, these functions generate C code. For a complete list of Image Processing Toolbox functions that support code generation, see Functions Supporting Code Generation.

• axes2pix

• dct2

• decorrstretch

• idct2

• imimposemin

• roifilt2

• roipoly

The edge function now supports code generation for the "approxcanny" edge detection method.

Hyperspectral Image Processing: Image Processing Toolbox Hyperspectral Imaging Library (Version 23.2)

• Read hyperspectral data from .bsq, .bil, .bip, and multipage TIFF files using the new capabilities of the hypercube object.

• View statistical data of spectral signatures, and export the spectral signatures using the new capabilities in the Hyperspectral Viewer app.

• Use the hyperSlic function to perform 2-D superpixel oversegmentation of hyperspectral cubes using the simple linear iterative clustering (SLIC) algorithm.

• The Change Detection in Hyperspectral Images example shows how to detect changes in land cover from hyperspectral images taken at different times.

• The Ship Detection from Sentinel-1 C Band SAR Data Using YOLO v2 Object Detection example shows how to detect ships from Sentinel-1 C Band SAR Data using YOLO v2 object detection.

• The Automate Pixel Labeling of Hyperspectral Images Using ECOSTRESS Spectral Signatures in Image Labeler example shows how to load hyperspectral images into the Image Labeler (Computer Vision Toolbox) and automatically label pixels.

To use these functions and tools, you must download the Image Processing Toolbox Hyperspectral Imaging Library from the Add-On Explorer. See Get and Manage Add-Ons.

 Functionality being removed or changed

imtool will be removed

Still runs

The imtool function will be removed in a future release. In most situations, use the Image Viewer app instead. The app has more functionality and is easier to use than the Image Tool. To update your code, replace instances of imtool with imageViewer.

Unlike Image Tool, the Image Viewer app does not return a figure. If you want to display an image in Image Tool and return the figure that contains the tool, use the images.compatibility.imtool.r2023b.imtool function instead. To update your code that displays an image and returns the Image Tool figure, replace instances of imtool with images.compatibility.imtool.r2023b.imtool.

Discouraged Usage	Recommended Replacement
This example uses the imtool function to display a grayscale image in the Image Tool. I = imread("cameraman.tif"); imtool(I)	Here is equivalent code that uses the imageViewer function to display the grayscale image in the Image Viewer app. I = imread("cameraman.tif"); imageViewer(I)
This example uses the imtool function to display a grayscale image in the Image Tool, setting the initial display range and interpolation method. I = imread("cameraman.tif"); imtool(I,[5 250],Interpolation="bilinear")	Here is equivalent code that uses the imageViewer function to display the grayscale image and set the initial display range and interpolation method. To set the display range, you must use the DisplayRange name-value argument. I = imread("cameraman.tif"); imageViewer(I,DisplayRange=[5 250],Interpolation="bilinear")
This example uses the imtool function to display an indexed image in the Image Tool, specifying the colormap. [I,map] = imread("trees.tif"); imtool(I,map)	Here is equivalent code that uses the imageViewer function to display the indexed image. To specify the colormap, you must use the Colormap name-value argument. [I,map] = imread("trees.tif"); imageViewer(I,Colormap=map)
This example uses the imtool function to display a grayscale image in the Image Tool and returns the figure containing the Image Tool. I = imread("cameraman.tif"); hTool = imtool(I);	Here is equivalent code that uses the images.compatibility.imtool.r2023b.imtool function to display a grayscale image in the Image Tool and return the figure containing the Image Tool. I = imread("cameraman.tif"); hTool = images.compatibility.imtool.r2023b.imtool(I);

bigimage and bigimageDatastore will be removed

Still runs

The bigimage object and the bigimageDatastore object will be removed in a future release. To read and process large or multiresolution images, use the blockedImage object and the blockedImageDatastore object instead.

Discouraged Usage	Recommended Replacement
This example creates a bigimage object. bigIm = bigimage("tumor_091R.tif");	Here is equivalent code, replacing the bigimage object with the blockedImage object. blockedIm = blockedImage("tumor_091R.tif");
This example creates a bigimageDatastore object from the bigimage object bigIm. bigImds = bigimageDatastore(bigIm);	Here is equivalent code that creates a blockedImageDatastore object from the blockedImage object blockedIm. blockImds = blockedImageDatastore(blockedIm);

Continue to display large or multiresolution image data using the bigimageshow function.