Image Registration

Align two images using intensity correlation, feature matching, or control point mapping

Together, Image Processing Toolbox™ and Computer Vision Toolbox™ offer four image registration solutions: interactive registration with a Registration Estimator app, intensity-based automatic image registration, control point registration, and automated feature matching. For help in choosing among the approaches, see Approaches to Registering Images.

Apps

Registration EstimatorRegister 2-D grayscale images

Functions

expand all

imregisterIntensity-based image registration
imregconfigConfigurations for intensity-based registration
imregtform Estimate geometric transformation that aligns two 2-D or 3-D images
imregcorrEstimate geometric transformation that aligns two 2-D images using phase correlation
imregdemonsEstimate displacement field that aligns two 2-D or 3-D images
imregmtbRegister 2-D images using median threshold bitmaps
normxcorr2Normalized 2-D cross-correlation
MattesMutualInformationMattes mutual information metric configuration
MeanSquaresMean square error metric configuration
RegularStepGradientDescentRegular step gradient descent optimizer configuration
OnePlusOneEvolutionaryOne-plus-one evolutionary optimizer configuration
cpselect Control Point Selection tool
fitgeotransFit geometric transformation to control point pairs
cpcorrTune control point locations using cross-correlation
cpstruct2pairsExtract valid control point pairs from cpstruct structure
imwarpApply geometric transformation to image

Objects

expand all

imref2dReference 2-D image to world coordinates
imref3dReference 3-D image to world coordinates
affine2d2-D affine geometric transformation
affine3d 3-D affine geometric transformation
projective2d 2-D projective geometric transformation

Topics

Register Images Interactively

Register Images Using the Registration Estimator App

This example shows how to use the Registration Estimator app to align a pair of images.

Techniques Supported by Registration Estimator App

Registration Estimator app provides ten algorithms for feature-based, intensity-based, and nonrigid registration.

Register Images Using Intensity-Based Optimization

Intensity-Based Automatic Image Registration

Intensity-based automatic image registration uses a similarity metric, an optimizer, and a transformation type to register two images iteratively.

Create an Optimizer and Metric for Intensity-Based Image Registration

Select an image metric and an optimizer suitable for either monomodal or multimodal images.

Use Phase Correlation as Preprocessing Step in Registration

Phase correlation is useful to estimate an initial transformation when images are severely misaligned.

Register Multimodal MRI Images

This example shows how to align two multimodal MRI images to a common coordinate system using automatic intensity-based image registration.

Register Multimodal 3-D Medical Images

This example shows how to align two volumetric images using automatic intensity-based image registration.

Registering an Image Using Normalized Cross-Correlation

This example shows how to determine the translation needed to align a cropped subset of an image with the larger image.

Register Images Using Control Point Mapping

Control Point Registration

To determine the parameters of a transformation, you can pick corresponding points in a pair of images.

Geometric Transformation Types for Control Point Registration

Control point registration can infer the parameters for nonreflective similarity, affine, projective, polynomial, piecewise linear, and local weighted mean transformations.

Control Point Selection Procedure

To specify control points in a pair of images interactively, use the Control Point Selection Tool.

Find Image Rotation and Scale

This example shows how to use control points to align two images that differ by a rotation and scale change.

Use Cross-Correlation to Improve Control Point Placement

Fine-tune your control point selections using cross-correlation.

Registering an Aerial Photo to an Orthophoto

This example shows how to use control point mapping to register two images with a projective transformation.