|Meeting the needs of imaging research community
|Image Registration and Fusion Systems, Beavercreek, Ohio, U.S.A., Copyright 2015
Most Popular Software
Use this software to remove haze from aerial images. This software can also be used to enhance very large
locally over-exposed or under-exposed images to enhance local image details. For more information and
examples, visit the Haze remover page.
Use this software to register images that show different views of a terrain scene. If the images to be
registered have large rotation and scaling differences, this software will bring the images into approximate
alignment by allowing the user to interactively select two corresponding points in the images. Once the
images are in approximate alignment, the software will then register the images automatically. An example
of nonrigid image registration by this software is shown above. For more information about this software,
visit the Semi-automatic same modality and multi-modality image registration pages.
Use this software to fuse multi-exposure images into a single well-exposed image. The software replaces
under-exposed and over-exposed areas in some images with well-exposed same areas in other images to
create an image that is well-exposed everywhere. The fusion process maximizes information content in the
created image. For more information about this software, visit the multi-exposure fusion page.
Two sets of images are shown. Each set contains two images of a scene taken from different view angles. The dominant
orientation detected for each image is shown with a yellow line. If there is sufficient overlap between the contents of two
images, the images will have the same dominant orientation. If different orientations are detected for the two images, one
image can be rotated with respect to the other to align their orientations. Images that are in the same orientation are easier
to register and analyze. Rotating images in the right to the orientation of images in the left, the images shown below are
obtained. An image in the right after rotation maintains its scale but takes the dimensions of the image in the left.