Moving Forward: Color Recognition
After hours of rigorous internal discussion with the MARCO development team (myself), we have all agreed on an optimal path forward. To set the stage, this is a significant turning point in the project. Up until now I have just been developing the features that allow MARCO to work, but from here on out I will be developing what I consider MARCO itself.
This means a number of things. First and foremost, it means that the programming work is going to get far more complex, difficult, and mathy. Second, it means that the acquisition algorithms I wrote months ago will finally see the light of day for the first time.
Without giving away too much detail, these algorithms were developed to mathematically step through a target surface image and identify the targets with the lowest possible number of pixel value checks. This is a departure from traditional computer vision approaches that I believe may substantially increase the speed of target acquisition, but it comes with a few caveats, the most substantial of which is the early need for color recognition, which is a facet of the project I considered peripheral until I recently.
So what is the takeaway from this rambling, poorly-written thought dump?
My next task is to teach MARCO how to teach itself to recognize different colors in different lighting environments, this is a crucial part of the acquisition process, this will be substantially more difficult than anything I've done for this project thus far.
The next post will contain a proceedure.
This means a number of things. First and foremost, it means that the programming work is going to get far more complex, difficult, and mathy. Second, it means that the acquisition algorithms I wrote months ago will finally see the light of day for the first time.
Without giving away too much detail, these algorithms were developed to mathematically step through a target surface image and identify the targets with the lowest possible number of pixel value checks. This is a departure from traditional computer vision approaches that I believe may substantially increase the speed of target acquisition, but it comes with a few caveats, the most substantial of which is the early need for color recognition, which is a facet of the project I considered peripheral until I recently.
So what is the takeaway from this rambling, poorly-written thought dump?
My next task is to teach MARCO how to teach itself to recognize different colors in different lighting environments, this is a crucial part of the acquisition process, this will be substantially more difficult than anything I've done for this project thus far.
The next post will contain a proceedure.
Comments
Post a Comment