Calculating Distance Using Stereographic Principles: A PCOT Implementation

As part of my final year at university, I developed a plugin for the PanCam Operational Toolkit (PCOT).

PCOT is part of Aberystwyth University's work on the ESA Rosalind Franklin Exomars Rover Mission. You can read more about PCOT itself here.

The project involved creating two plugins. One to rectify the pair of images, the other to calculate the distance from the x-disparity and known parameters about the camera system (distance between the two cameras)

Rectification Plugin

Rectification is the process of adjusting a pair of images so that they share a common image plane. This means warping the image so that both image appear to have been taken from cameras which are perfectly parallel to one another. 

This process causes points within the image to become horizonal. A point in the left image can be found on the exact same Y value in the right image.

This is paramount for all stereographic image processing.

Distance Estimation

This plugin takes in a set of stereographic images (1 from the left, 1 from the right) which have been marked by the user with matched points of interest. (Point in the left image is named the same in the right image)

The disparity (difference in x value) is then compared to known information about the camera system. The distance to the point can then be estimated.