Skip to content

Latest commit

 

History

History
52 lines (31 loc) · 2.2 KB

README.md

File metadata and controls

52 lines (31 loc) · 2.2 KB

##########################################################################################

                             PROJECT ADAM SCANNER CODE

##########################################################################################

ABOUT: Project ADAM is the name I give to my Honours Project as it is more catchy than "3D Scanner-Printer Hybrid for the creation of prosthetic limbs", but that's what this is. The physical model is under construction but the scanning code is still needing a lot of work done. Imagine, if you will, a box with sides FRONT, BACK, LEFT and RIGHT. If lasers on sides FRONT and BACK project a laser line through the centre of the box a stump can be stuck down into the box, interrupting the laser line and creating a silhouette of laser light around the stump. If then, the lasers move along the FRONT and BACK sides of the box and there is a camera in the LEFT and RIGHT these silhouettes can be captured for different distances along the LEFT-RIGHT axis. By then using OpenCV to extract the laser line from each image we can turn it into an array of X, Y points and append a Z value that's derived from the distance along the axis (which can be read from the image name where "1img34" refers to "camera 1 image 34"). These different layers are then appended to create an array of points (a Point Cloud) and the points can then be put through a simple tessellation to create a 3D STL.

This STL is then used by a prosthetics technician to develop a well-fitting prosthesis

and the design is printed on the same machine.

##########################################################################################

CURRENT STATE AT UPLOAD TO GITHUB: As it stands the code is currently in several smaller test files.

  • PhotoTest holds the main code that carries out the scanning itself
  • I2LTEST1 works up to the Threshold operation
  • I2LTEST4 is currently under development and works up to the extraction of the points and subsequent writing to new array. This is what's currently under development (16.02.18)
  • imgToContourDEMO just proves that the program works up to the point of readContours

I PLAN TO IMPROVE THE IN-CODE DOCUMENTATION SOON