Topcon ImageMaster Photo

From IARC 207 Wiki
Jump to navigation Jump to search

Just some notes for myself.


General Tips

  1. Set DSLR to Aperture Priority when taking photos.
  2. Bright, flat, light is best for oblique photos
  3. Shoot a photo of the calibration print using the camera with aperture set to what is used in the field... you'll need one calibration print per lens & aperture combo.

Software Tips

  • I'll update this once I've done a total station survey but the software is tough to use, so some reminder type stuff for making arbitrary space surveys
  1. Add a camera to the project: Orientation Menu -> Camera -> Add Camera
  2. Browse to the calibration information for the camera & lens combo.
  3. Add images using the button Register Image
  4. After adding the images and specifying they use the camera in the previous step, pair them up by selecting two photos from the Image List and pressing the toolbar button Add Stereo Pair
  5. For me, with a simple case of four photos, 'pair' means 1 & 2 pair, 2&3 pair, 3&4 pair etc.
  6. start locating points on the first pair of images... to do this rightclick on an image and select Measure Tie Point...
  7. For nature scenes, I'm finding not enabling Center Detector or Corner Detector is fastest for the workflow.
  8. As you add points:
    1. Click a point in the left image
    2. click the same point in the right image
    3. name it
    4. click Measure in the measure tie point pop up window.
    5. repeat for the next points in the image.
  9. I found that it worked better to zoom in once more from the default. My precision went up quite a bit at the higher zoom setting. Do this for both images.
  10. when you're done with the first set of images, double-click on the second pair. You'll be prompted to save the point information and move to the second pair.
  11. On the second pair, typically the image on the left will show you points common between the first and second images. At the higher zoom level I found it easy enough to identify these points in the third image. First, identify the pre-existing point in both images. Then, click on the white label in the left image. Finally, click on the correct spot on the right image. Then, once again, click the measure button. Doing this across multiple images is the difference between a nice set of 2d merging of the image and a 3d computation. This is also why when you pair up, it is 1-2, 2-3, 3-4 etc. rather than 1-2, 2-3, 1-3 etc.