Computer Vision System Toolbox
Image registration and stereo vision are often used to combine information from multiple cameras and to derive 3D information from multiple 2D views.
Image registration is the transformation of images from different camera views to use a unified co-coordinate system. Computer Vision System Toolbox supports an automatic approach to image registration by using features. Typical uses include video mosaicking, video stabilization, and image fusion.
Feature detection, extraction, and matching are the first steps in the feature-based image registration workflow. With a pair of images, you can detect and extract features in each image, using one of several feature types available in the system toolbox. You can then determine putative matches between the two sets of features and visualize the matches. Typically, this workflow produces many interest points with matches that include outliers. You can remove the outliers with statistically robust methods such as RANSAC or least median of squares to compute a similarity, affine, or projective transformation. You can then apply the geometric transformation to align the two images.
Stereo vision is the process of reconstructing a 3D scene from two or more views of the scene. Using the system toolbox, you can perform uncalibrated stereo image rectification on a pair of stereo images and match individual pixels along epipolar lines to compute a disparity map. This builds on the feature-based registration workflow.
Stereo image rectification transforms a pair of stereo images so that a corresponding point in one image can be found in the corresponding row in the other image. You can rectify a pair of stereo images with the system toolbox by determining a set of matched interest points, estimating the fundamental matrix, and then deriving two projective transformations. This process reduces the 2D stereo correspondence problem to a 1D problem, which simplifies the process of determining the depth of each point in the scene from the camera.