Index > Interface > Menu Process | Previous | Next |
|
General | Matching | Calibration |
Allows the user to change the processing options related to the keypoints matching for step 1. Initial Processing. It contains 2 sections:
- Matching Image Pairs: Allows to select which pairs of images are matched.
- Matching Strategy: Allows to determine how the images are matched.
Allows the user to select which pairs of images are matched:
- Aerial Grid or Corridor: Optimizes the pair matching for Aerial Grid or Corridor flight paths.
- Free flight or Terrestrial: Optimizes the pair matching for Free-flight paths or Terrestrial images (for example, taking images around a house or a statue).
- Custom (for advanced users): Specific pair matching parameters useful in specific projects. Suggested if one of the above options does not provide the desired results. It contains the different pair matching parameters:
- Use Capture Time: Matches images considering the time on which they were taken.
- Number of Neighboring Images: It allows the user to set how many images (before and after in time) are used for the pair matching.
- Use Triangulation of Image Geolocation: Only available if the images have geolocation. It is only useful for aerial flights. The position of the images is triangulated. Each image is matched with images with which it is connected by a triangle.
- Use Distance: Only available if the images have geolocation. It is useful for oblique or terrestrial projects. Each image is matched with images within a relative distance.
- Relative Distance Between Consecutive Images: It allows the user to set the relative distance.
- Use Image Similarity: Uses the image content for pairs matching. Matches the n images with most similar content.
- Maximum Number of Pairs for Each Image Based on Similarity: Maximum number of image pairs with similar image content to be matched.
- Use MTPs: Images connected via a shared Manual Tie Point will be matched.
- Maximum Number of Image Pairs per MTP: Maximum number of image pairs connected by a given MTP.
- Use Time for Multiple Cameras: When having multiple flights without geolocation using the same flight plan over the same area, and having different camera models for each flight, it matches the images from one flight with the ones from the other flight using the time information.
Allows the user to determine how the images are matched:
- Use Geometrically Verified Matching: Slower but more robust. If selected, geometrically inconsistent matches are discarded. Useful when many similar features are present throughout the project: rows of plants in a farming field, window corners on a building's facade, etc.
Index > Interface > Menu Process | Previous | Next |
Hi,
when I read the log document with the default setting, I found the output of pairs like:
'Matched pair 10: <****_0001.JPG(0,35682),****_0386.JPG(385,14179)>: nrMatches 1763 | GPU 0 GeForce **** Design'
what does 'nrMatches' mean here? How do you determine the similarity between a pair of images? Maybe by SIFT features matching? How about the details such as 5 points method or RANSAC for outlier rejecting...
Sorry for so many questions, looking forward to your reply, thank you.
Best wishes,
Chenhao