Main Content

imregcorr

Estimate geometric transformation that aligns two 2-D images using phase correlation

所有的页面崩溃

Syntax

tform = imregcorr(moving,fixed)
tform = imregcorr(moving,Rmoving,fixed,Rfixed)
tform = imregcorr(___,transformType)
tform = imregcorr(___,'Window',window)
[tform,peakcorr] = imregcorr(___)

Description

example

tform= imregcorr(moving,fixed)estimates the geometric transformation that aligns an image,moving, with a reference image,fixed. The function returns a geometric transformation object,tform, that maps pixels inmovingto pixels infixed.

tform= imregcorr(moving,Rmoving,fixed,Rfixed)estimates the geometric transformation that aligns an image,moving, with a reference image,fixed.RmovingandRfixedare spatial referencing objects that contain spatial information about themovingandfixed图像,分别。的交易nsformation object returned,tform, defines the point mapping in the world coordinate system.

tform= imregcorr(___,transformType)also specifies the type of transformation,transformType.

tform= imregcorr(___,'Window',window)also specifies whether to perform windowing in the frequency domain. To increase the stability of registration results, specifywindowastrue. However, if the common features in your images are oriented along the edges, then settingwindowtofalsecan sometimes provide superior registration results.

[tform,peakcorr] = imregcorr(___)also returns the peak correlation,peakcorr, of the phase difference between the two images.

Examples

全部折叠

Open Live Script

Read a reference image into the workspace.

fixed = imread('cameraman.tif');

Create a synthetic moving image by scaling and rotating the fixed image.

theta = 20; S = 2.3; tform = affine2d([S.*cosd(theta) -S.*sind(theta) 0;...S.*sind(theta) S.*cosd(theta) 0;...0 0 1]); moving = imwarp(fixed,tform); moving = moving + uint8(10*rand(size(moving)));

Display the fixed and the moving image alongside each other.

imshowpair(fixed,moving,'montage')

Figure contains an axes object. The axes object contains an object of type image.

Estimate the transformation needed to align the images usingimregcorr.

tformEstimate = imregcorr(moving,fixed);

Apply estimated geometric transform to the moving image. This example uses the'OutputView'parameter to obtain a registered image the same size and with the same world limits as the reference image.

Rfixed = imref2d(size(fixed)); movingReg = imwarp(moving,tformEstimate,'OutputView',Rfixed);

View the original image and the registered image side-by-side to check the registration. Then view the registered image overlaid on the original using the'falsecolor'option to highlight any areas where the images differ.

imshowpair(fixed,movingReg,'montage')

Figure contains an axes object. The axes object contains an object of type image.

imshowpair(fixed,movingReg,'falsecolor');

Figure contains an axes object. The axes object contains an object of type image.

Input Arguments

全部折叠

Image to be registered, specified as a grayscale, binary, or RGB image. If you specify an RGB image,imregcorrconverts it to a grayscale image usingrgb2graybefore processing.

Note

The aspect ratio ofmovingaffects the output transformtform. For best results, use a square image.

Data Types:single|double|int8|int16|int32|uint8|uint16|uint32|logical

参考图像的target orientation, specified as a grayscale, binary, or RGB image. If you specify an RGB image,imregcorrconverts it to a grayscale image usingrgb2graybefore processing.

Note

The aspect ratio offixedaffects the output transformtform. For best results, use a square image.

Data Types:single|double|int8|int16|int32|uint8|uint16|uint32|logical

Type of transformation to estimate, specified as one of the following values.

Value Description
'translation' Translation
'rigid' Translation and rotation
'similarity'

Translation, rotation, and scaling

When using the'similarity'option, the phase correlation algorithm is only scale invariant within some range of scale difference between the fixed and moving images.imregcorrlimits the search space to scale differences within the range [1/4, 4].imregcorrdoes not detect scale differences less than 1/4 or greater than 4.

Data Types:char|string

Spatial referencing information associated with the image to be registered, specified as animref2dobject.

Spatial referencing information associated with the reference (fixed) image, specified as animref2dobject.

Use windowing to suppress spectral leakage effects in the frequency domain, specified asa numeric or logical1(true) or0(false). Whentrue,imregcorrfunction performs windowing using a Blackman filter.

Output Arguments

全部折叠

Geometric transformation, returned as anaffine2dobject.

Peak correlation value of the phase difference between the two images, returned as a numeric scalar.

Tips

  • If your image is of typedouble, you can achieve performance improvements by casting the image tosinglewithim2singlebefore registration. Input images of typedoublecause the algorithm to compute FFTs indouble.

References

[1] Reddy, B. S. and Chatterji, B. N.,An FFT-Based Technique for Translation, Rotation, and Scale-Invariant Image Registration, IEEE Transactions on Image Processing, Vol. 5, No. 8, August 1996

Extended Capabilities

Version History

Introduced in R2014a

See Also

Apps

Functions