FFTP or NCC method

For more consistently accurate results, FFTP is recommended. This method uses a larger template size than NCC and, because it works in the frequency domain, it looks at the patterns of details in the image rather than the gray values in a small neighborhood, which NCC uses. This makes FFTP more robust than NCC in cases where there is a large brightness difference between images or when a major land-use change has occurred between the images. FTTP can also better match images of the same area from different sensors or spectral bands.

Note: When working with radar data, and the input data contains image layers written as complex values, the total power in decibels is computed temporarily and used for matching. When possible, calibrate the data in sigma, beta, or gamma naught. When working with SAR data, Fast Fourier Transfer Phase (FFTP) matching is recommended.

To automatically collect TPs using FFTP or NCC

  1. In the Tie points per area box, type or select the number of TPs to use over the image or overlap area.

    The higher the number, the more TPs OrthoEngine will attempt to collect, and the more processing time will be required. The actual number of collected TPs may be higher or lower than the number you specify.

  2. In the Trials per point box, type or select a number representing how many attempts to make for each point before giving up on the match.

    The algorithm will try to match the primary sample point and, if that fails to match, a secondary point will be selected in the same grid cell as an additional sample.

    Note: The value you specify may slow the matching process; that is, a greater value slows the process more than a lesser value, because the higher number causes the system to try to match that many more times in each cell.
  3. In the Min. acceptance score box, type a value between zero and one.

    The value defines the minimum correlation score that will be considered a successful match. The best correlation score is one. Decreasing the score will generate more points.

  4. In the Search radius box, type the number representing the size of the area on the target image, and then from the list to the right, select the corresponding unit of measure.
  5. In the Input elevation list, click one of the following according to how you want height to factor in collecting the TPs:
    • To use a continuous elevation, click Constant value, and then in the Height box, below, type the approximate elevation of the terrain.
    • To use a digital elevation model (DEM), click DEM, and then in the File box, below, type the path and file name of the DEM file or, to select a file, click Browse.

      If necessary, you can make changes to attributes of the DEM you select. After you select the DEM file, click DEM Settings, and then proceed to Setting DEM file options.

    Note: The elevations specified by either the constant value or the DEM are used only to assist in finding matching TPs. Height information obtained from either source is not incorporated into the math model
  6. In the Distribution pattern list, do one of the following:
    • To evenly distribute the TPs over the entire image and match each TP in all of the overlapping images, click select Entire image.

      Typically, this is used to generate standard TP distributions for aerial photographs, such as the three-by-three pattern.

    • To evenly distribute the points only in the overlap area between any pair of overlapping images, click Overlap area.

      Typically, this is used with satellite images or aerial photographs with overlap of 60 percent or less.

  7. In the Search method list, do one of the following:
    • To find candidates by running a corner-detection algorithm on the image, which looks for corner-like features to use as candidates, click Susan.
    • To find candidates in a grid-like pattern, click Grid, and then in the Edge margin distance box, type or select the minimum distance you want, in pixels, between the edge of the image and the placement of the candidates.
      Note: If you do not specify a value, five percent of the overlap dimension or 256 pixels, whichever is less, will be applied.

    Grid does not preprocess the image; therefore, it tends to be faster than Susan, but it also finds fewer matches, because the grid point might fall on a featureless flat patch somewhere in the image that cannot be matched to anything in the overlapping images.

    Both Susan and Grid determine how to find the initial candidate positions in one image (the source image) for collecting sample points. During processing, a patch is built around each candidate position that it searches for in overlapping images.

    When collecting TPs, Susan is often preferred, because it facilitates performing quality assurance on the collected TPs: you can visualize a recognizable feature (a building corner or specific tree, for example) and look at the other image to see if it matches properly.

  8. In the Matching channels box, enter the channel or channels to use during collection.

    To use more than one channel, enter a comma-delimited list.

  9. Proceed to Selecting images, start time, and run.

Parent topic: Collecting tie points automatically

© PCI Geomatics Enterprises, Inc.®, 2026. All rights reserved.