DSMAC

Digital Scene Matching Area Correlation

A small primer on a form of modern autonomous aerial guidance.

   

Conversion Factors

Example Reference Imagery

DSMAC stands for Digital Scene Matching Area Correlation (DSMAC). The premise of operation essentially boils down to running binary correlations on sequential binary converted greyscale images. Images are stored in binary array formats. This image represents a typical reference frame image before it is processed and converted to the correct correlation format.

Converted Imagery

 

The image on the left is the processed version of the image above, after all the steps have been performed to obtain a usable relatable image, with useful accuracy. The conversion involves spatial averaging, accounting for geometric distortion/errors and potential issues with brightness levels post-capture that can degrade accuracy of the correlation algorithm.

Reference imagery generation

This image corresponds to the process of generating reference imagery for use in DSMAC. High altitude cameras take a variety of images at different points which are synthesized into one reference image to account for distortion, weather, spectral response to the sun, noise in the sensor to generate a clean image that offers good feature geometry and contrast.

 

Photo Gallery

 

Contact

Binary Correlation
Ternary Correlation
Quarternion Correlation
Where am I?
Who am I?
What am I?
How do I get there?
How do I fly?
How do I work?
How do I make it happen?
How do I make what happen?
How do I make that happen?
One frame of data
Three frames of data?
Repackaged and compressed multi-frame timelapses



Check to sign up for the newsletter for more information