Orthorectified images have removed relief displacements and tilt displacements. Since the assumption of orthorectification is to keep the light source at infinite distance so that the light rays are parallel to each other and perpendicular to ground. They should not have shadows. Is it true?


If there was a shadow in the raw image, say from a tree or tall building, it will still be in the image after the rectification process. Essentially, parts the photo are being stretched or warped to match the position of visible objects in the photo to known places on a map; this will correct for the pitch, yaw, and roll of the plane during the flight, and also to eliminate the effects of ground relief, leaving an image which is "top down" and uniform scale. It does not change the content of the image.

Here's an example of a raw image of Toronto taken in 2007 by First Base Solutions (http://maps.firstbasesolutions.com)Before Orthorectification

Compare to the same section of photo after orthorectification. Note the shadow and image content are basically the same.After orthorectification

  • I agree with the answer, but not with the illustration. If the second image had been correctly orthorectified, you should not see the sides of the buildings. The light rays that are parallel to each others are those coming from the observed objects. The shadows are determined by the position of the sun. – radouxju Dec 26 '13 at 20:52
  • 2
    The image above was captured by aircraft at low altitude (short focal length), which causes far more radial perspective than say, satellite capture. Sides of very tall buildings or relief from mountains are generally visible in this type of image acquisition unless the camera is directly overhead of the particular building. Flying a tighter grid and then mosaicking the image can help eliminate radial perspective in urban areas, but again, rectification itself does not affect the content of the image. – Salina Mar 26 '14 at 16:34
  • 1
    the top of the building should have been moved to hide the sides, leaving a no data area on the other side of the building, which can filled by another image. Only then can you talk of a true orthorectification. (otherwise it is a good geometric correction, but if you don't take the height into account it is less rigourous) – radouxju Mar 26 '14 at 19:06
  • Correct, you're referring to mosaicking, and applying a 3D model to correct for terrain variations, two processes often performed along with rectification. In this case, a 40m spot height grid (+/- 40cm) of bare earth features was used to represent terrain. This is a tiny clip out of a much larger mosaicked photo chosen deliberately because it does not show effects of the mosaic. The question here is, after all, about how shadows are affected by rectification specifically, not about other processes. You can contact me directly if you have further questions about this image. – Salina Mar 27 '14 at 20:27

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.