i2k PinPoint Technology
i2k PinPoint provides the functionality users need to automatically establish correspondence between images as the basis for georeferencing, orthorectification, location recognition, and deformable image registration. Unlike the i2k Align product and its core i2Align algorithm, i2k PinPoint matches images in small regions, providing reliable inter-image correspondences despite parallax, differences in viewpoint, physical changes in the imaged areas, and differences in image modalities.
i2k PinPoint works with one or more base images, matching these against one or more target images.
- Matches Control Points
- Accurately finds the matches for pre-specified control point locations from the base image in a target image.
- Generates Correspondences
- Determines the most accurate matches between the base image and the target image, spreading the points across the base image as much as possible.
- Evaluates the Matches
- Provides a measure of the reliability of each match. As one instance of the utility of this, control point locations that do not appear in the target image may be easily and automatically discounted.
- Displays Matches
- Shows users the matching point locations between the base and target images to allow rapid verification that the points are indeed well-matched.
- Aggressive Mode
- Allows more aggressive settings of the software to handle many of the most extremely difficult cases.
- Works in Seconds
- Correspondence generation for two images is completed in a matter of seconds.
- Rejects Incorrect Images
- Automatically determines when a base and target image can not be matched. Most often this occurs when the two images do not show overlapping areas.
- Multiple Base Images
- Allows users to specify multiple base images. The software automatically determines which of these to match against each target image.
- Multiple Target Images
- Allows users to specify multiple target images and use the results of matching one target image to guide the matching process for subsequent target images.
- Scale Differences
- The software automatically handles scale differences of up to 5x between images. When the (approximate) ground resolutions of the base and target images are provided, the software accounts for these differences, allowing matching across widely differing scales.
- Location Recognition
- By providing a small target image or image region, this region is automatically located in the base image or images.
- Push Broom and Whisk Broom
- Even generates correspondences between push broom and whisk broom images.