--- title: "ImageAlign" language: "en" type: "Symbol" summary: "ImageAlign[ref, image] returns a version of image that is aligned with the reference image ref. ImageAlign[ref, {image1, ..., imagen}] gives the result of aligning each of the imagei with the reference image ref. ImageAlign[{image1, ..., imagen}] uses image1 as the reference image." keywords: - align - image align - transform - image registration - registration - correspondences - affine - rigid - perspective - similarity canonical_url: "https://reference.wolfram.com/language/ref/ImageAlign.html" source: "Wolfram Language Documentation" related_guides: - title: "Geometric Operations" link: "https://reference.wolfram.com/language/guide/ImageGeometry.en.md" - title: "Image Processing & Analysis" link: "https://reference.wolfram.com/language/guide/ImageProcessing.en.md" - title: "Computer Vision" link: "https://reference.wolfram.com/language/guide/ComputerVision.en.md" - title: "3D Images" link: "https://reference.wolfram.com/language/guide/3DImages.en.md" - title: "Image Composition" link: "https://reference.wolfram.com/language/guide/ImageComposition.en.md" - title: "Image Computation for Microscopy" link: "https://reference.wolfram.com/language/guide/ImageComputationForMicroscopy.en.md" - title: "Computational Photography" link: "https://reference.wolfram.com/language/guide/ComputationalPhotography.en.md" - title: "Optimization" link: "https://reference.wolfram.com/language/guide/Optimization.en.md" - title: "Image Computation: Update History" link: "https://reference.wolfram.com/language/guide/ImageComputation-UpdateHistory.en.md" related_functions: - title: "ImageStitch" link: "https://reference.wolfram.com/language/ref/ImageStitch.en.md" - title: "ImageCorrespondingPoints" link: "https://reference.wolfram.com/language/ref/ImageCorrespondingPoints.en.md" - title: "FindGeometricTransform" link: "https://reference.wolfram.com/language/ref/FindGeometricTransform.en.md" - title: "ImageKeypoints" link: "https://reference.wolfram.com/language/ref/ImageKeypoints.en.md" - title: "ImageTransformation" link: "https://reference.wolfram.com/language/ref/ImageTransformation.en.md" - title: "ImageCorrelate" link: "https://reference.wolfram.com/language/ref/ImageCorrelate.en.md" --- # ImageAlign ImageAlign[ref, image] returns a version of image that is aligned with the reference image ref. ImageAlign[ref, {image1, …, imagen}] gives the result of aligning each of the imagei with the reference image ref. ImageAlign[{image1, …, imagen}] uses image1 as the reference image. ## Details and Options * ``ImageAlign[image1, image2]`` finds a transformation of ``image2`` that registers it with ``image1`` and returns the result of applying this transformation to ``image2``. * ``ImageAlign`` works with arbitrary 2D and 3D images. * Resulting images have the same dimensions as the reference image. * Use ``FindGeometricTransform`` to compute the geometric transformation used for alignments. * The following options can be specified: | | | | | -------------------- | --------- | ----------------------------------- | | Background | Automatic | background value | | Method | Automatic | alignment method to use | | TransformationClass | Automatic | geometrical relation between images | * Possible settings for the ``Method`` option are: | | | | --------------------------- | ---------------------------------------------------------- | | Automatic | automatically choose a suitable method | | "Keypoints" | use corresponding keypoints | | {"Keypoints", method} | use an ImageCorrespondingPoints method | | "MeanSquareGradientDescent" | minimize the mean square distance of corresponding pixels | | "Fourier" | Fourier-based registration | | "FourierBlurInvariant" | blur invariant Fourier-based registration | * With the default setting ``TransformationClass -> Automatic``, ``ImageAlign`` attempts to find the simplest possible transformation. * Possible transformations in order of increasing complexity are: | | | | ------------- | ------------------------------------- | | "Translation" | translation only | | "Rigid" | translation and rotation | | "Similarity" | translation, rotation, and scaling | | "Affine" | linear transformation and translation | | "Perspective" | linear fractional transformation | ## Examples (21) ### Basic Examples (2) Align a part of an image with the full image: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` --- Align two 3D volumes: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` ### Scope (2) Align a list of three images to a reference image: ```wl In[1]:= ImageAlign[[image], {[image], [image], [image]}, IconizedObject[«...»]] Out[1]= {[image], [image], [image]} ``` --- Align a list of three images, using the first image as the reference: ```wl In[1]:= ImageAlign[{[image], [image], [image]}, IconizedObject[«...»]] Out[1]= {[image], [image], [image]} ``` Three images are returned: ```wl In[2]:= Length[%] Out[2]= 3 ``` ### Options (11) #### Background (1) By default, a transparent background is used: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` Specify a different background color: ```wl In[2]:= ImageAlign[[image], [image], Background -> Gray] Out[2]= [image] ``` #### Method (5) By default, ``ImageAlign`` uses an automatic combination of methods and initializations: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` --- The ``"Keypoints"`` method works well with natural 2D photos: ```wl In[1]:= ref = [image];image = [image]; In[2]:= ImageAlign[ref, image, Method -> "Keypoints"] Out[2]= [image] ``` Use "KAZE" keypoints: ```wl In[3]:= ImageAlign[ref, image, Method -> {"Keypoints", "KAZE"}] Out[3]= [image] ``` Use all keypoints detected by multiple methods: ```wl In[4]:= ImageAlign[ref, image, Method -> {"Keypoints", {"ORB", "KAZE"}}] Out[4]= [image] ``` --- The ``"MeanSquareGradientDescent"`` method uses dense distances between images: ```wl In[1]:= ImageAlign[[image], [image], Method -> "MeanSquareGradientDescent", TransformationClass -> "Affine"] Out[1]= [image] ``` This method typically needs a large initial overlap between images: ```wl In[2]:= ImageAlign[[image], ImagePad[[image], 30], Method -> "MeanSquareGradientDescent", TransformationClass -> "Affine"] Out[2]= [image] ``` --- The ``"Fourier"`` method can align images even with a small overlap: ```wl In[1]:= {i1, i2} = {[image], [image]}; In[2]:= res = ImageAlign[i1, i2, Method -> "Fourier"] Out[2]= [image] In[3]:= HighlightImage[ImageCompose[i1, {res, .5}], {FaceForm[], EdgeForm[Thick], AlphaChannel[res]}] Out[3]= [image] ``` --- Use the ``"FourierBlurInvariant"`` method in the presence of noise and motion blur: ```wl In[1]:= {i1, i2} = {[image], [image]}; In[2]:= ImageAlign[i1, i2, Method -> "FourierBlurInvariant"] Out[2]= [image] In[3]:= ImageCompose[i1, {%, .5}] Out[3]= [image] ``` #### TransformationClass (5) Restrict the transformation to only translation: ```wl In[1]:= ImageAlign[[image], [image], TransformationClass -> "Translation"] Out[1]= [image] ``` --- ``"Rigid"`` transformation allows for rotations in addition to translations: ```wl In[1]:= ImageAlign[[image], [image], TransformationClass -> "Rigid"] Out[1]= [image] ``` --- ``"Similarity"`` transforms allow for translation, rotation, and scaling: ```wl In[1]:= ImageAlign[[image], [image], TransformationClass -> "Similarity"] Out[1]= [image] ``` --- Allow any affine transformation: ```wl In[1]:= ImageAlign[[image], [image], TransformationClass -> "Affine"] Out[1]= [image] ``` --- Perspective transformation: ```wl In[1]:= ImageAlign[[image], [image], TransformationClass -> "Perspective"] Out[1]= [image] ``` ### Applications (4) Align an image with its mirror image: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` --- Overlay the aligned images: ```wl In[1]:= i1 = [image];i2 = [image]; In[2]:= aligned = ImageAlign[i1, i2]; ImageCompose[aligned, {i1, .5}] Out[2]= [image] ``` --- Align two different images of the same object: ```wl In[1]:= ImageAlign[[image], [image]] Out[1]= [image] ``` --- Align images before performing focus stacking: ```wl In[1]:= images = {[image], [image], [image], [image]}; In[2]:= aligned = ImageAlign[images] Out[2]= [image] ``` Perform focus stacking: ```wl In[3]:= ImageFocusCombine[aligned] Out[3]= [image] ``` Compare with the stacking without alignment: ```wl In[4]:= ImageFocusCombine[images] Out[4]= [image] ``` ### Properties & Relations (2) For aligning images using a specific set of corresponding points, use ``FindGeometricTransform`` : ```wl In[1]:= {i1, i2} = {[image], [image]};pts1 = {{47.5, 151.5}, {149.5, 151.5}, {150.5, 50.5}, {47.5, 52.5}}; pts2 = {{64.5, 185.5}, {164.5, 173.5}, {184.5, 49.5}, {68.5, 54.5}}; In[2]:= {e, t} = FindGeometricTransform[pts1, pts2] Out[2]= {9.281079204581667`*^-13, TransformationFunction[(| | | | | ------------ | ------------ | -------- | | 0.694758 | -0.0245116 | -3.62636 | | -0.0264695 | 0.531574 | 19.9583 | | -0.000713898 | -0.000984443 | 1. |)]} ``` Transform the image with ``ImagePerspectiveTransformation`` : ```wl In[3]:= ImagePerspectiveTransformation[i2, t, DataRange -> Full, Padding -> 0] Out[3]= [image] ``` --- Use ``FindGeometricTransform`` to find the geometric transformation that aligns images: ```wl In[1]:= i1 = [image];i2 = [image]; In[2]:= {e, tr} = FindGeometricTransform[i1, i2] Out[2]= {0.341998, TransformationFunction[(| | | | | ---------- | ----------- | -------- | | 0.537851 | -0.0101117 | 153.442 | | -0.13068 | 0.916323 | 0.645807 | | -0.0016047 | 0.000214252 | 1. |)]} ``` Stitch images together by transforming one and composing on top of the other one: ```wl In[3]:= {w, h} = ImageDimensions[i2]; tmp = ImagePerspectiveTransformation[i2, tr, DataRange -> Full, PlotRange -> {{0, First@tr[{w, 0}]}, {0, h}}]; ImageCompose[tmp, {i1, .7}, Round@({w, h} / 2)] Out[3]= [image] ``` ## See Also * [`ImageStitch`](https://reference.wolfram.com/language/ref/ImageStitch.en.md) * [`ImageCorrespondingPoints`](https://reference.wolfram.com/language/ref/ImageCorrespondingPoints.en.md) * [`FindGeometricTransform`](https://reference.wolfram.com/language/ref/FindGeometricTransform.en.md) * [`ImageKeypoints`](https://reference.wolfram.com/language/ref/ImageKeypoints.en.md) * [`ImageTransformation`](https://reference.wolfram.com/language/ref/ImageTransformation.en.md) * [`ImageCorrelate`](https://reference.wolfram.com/language/ref/ImageCorrelate.en.md) ## Related Guides * [Geometric Operations](https://reference.wolfram.com/language/guide/ImageGeometry.en.md) * [Image Processing & Analysis](https://reference.wolfram.com/language/guide/ImageProcessing.en.md) * [Computer Vision](https://reference.wolfram.com/language/guide/ComputerVision.en.md) * [3D Images](https://reference.wolfram.com/language/guide/3DImages.en.md) * [Image Composition](https://reference.wolfram.com/language/guide/ImageComposition.en.md) * [Image Computation for Microscopy](https://reference.wolfram.com/language/guide/ImageComputationForMicroscopy.en.md) * [Computational Photography](https://reference.wolfram.com/language/guide/ComputationalPhotography.en.md) * [`Optimization`](https://reference.wolfram.com/language/guide/Optimization.en.md) * [Image Computation: Update History](https://reference.wolfram.com/language/guide/ImageComputation-UpdateHistory.en.md) ## History * [Introduced in 2010 (8.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn80.en.md) \| [Updated in 2014 (10.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn100.en.md) ▪ [2015 (10.2)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn102.en.md) ▪ [2016 (11.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn110.en.md) ▪ [2017 (11.1)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn111.en.md) ▪ [2021 (13.0)](https://reference.wolfram.com/language/guide/SummaryOfNewFeaturesIn130.en.md)