Coarse Acquisition

As you may have (but probably haven't) read in the pages I wrote a few months ago, coarse acquisition is MARCO's first stab at identifying the target surface in its field of view.  The single purpose of coarse acq. is to find the colored targets on the target surface using input from the machine-learned color database.  With this information, MARCO can verify its target's identity based on the position of each target.  Furthermore, coarse acq. passes the rough location of each target on the image to the fine acquisition process, which is currently under development.

Coarse Acquisition visualization
Without divesting too much detail, coarse acquisition uses a unique algorithm to comb the target surface for pixels meeting the color bounds specified in the polo file and retrieved from the color database.  If coarse acquisition fails, it's usually because the color bounds are too strict and MARCO fails to recognize the targets, but if this happens MARCO recursively widens its color bounds until coarse acq. succeeds, as it did with the image above.  It usually takes 2-4 iterative re-acquisitions to pass coarse acquisition.

Coarse acq. is really a special part of MARCO, because certain nuances of the system architecture allow me to cheat my way out of traditional, expensive computer vision/image processing techniques and very quickly extract the information I need from the image.  Coarse acquisition happens fast, a lot faster than even I expected, and in its current state it's a little broken so it could actually happen a little faster.  If anyone is familiar with image processing, these images are taken and ingested at 7.68 Megapixels (3200 x 2400), and if coarse acquisition succeeds on the first loop I can have the pixel locations of those targets in less than a half-second, excluding time spent imaging.

A time-based analysis of MARCO's ability to chug through images within a reasonable period of data-relevancy may be forthcoming, but for now I am moving on to fine acq., which in theory will be quicker and easier to build.

Comments

Post a Comment

Popular posts from this blog

The Future of MARCO

So I took some time, got a real job at NASA and I've been out traveling a bit, and I only just recently had some time to settle down and take a good look at this project from a more seasoned perspective.  For reference, when I set out to make this happen, I didn't know python, I had little experience with linux, and "computer vision" was little more than a buzzword to me.  I knew the math worked out, however I had a hard time translating the principles of MARCO from the realm of math where everything is clean and orderly to the real world where colors sometimes don't look like they should and the sun shines more brightly on some days than on others. I was not well-prepared to take on this project, and it's a good thing I started as early as I did because there's no way in hell I would have been able to achieve the fairly-positive results I got with less time. Be that as it may, I'm now in an incrementally better position to tackle this task, and I ...
Keep reading

Preliminary Results and higher-order error sources

Image
I have some test results today, and I'm quite happy with how they turned out.  Quick refresher of the parameters: range refers to the target distance in centimeters, a_ang is the rotation of the major axis of the projected ellipse, and b_ang is a measurement of the target's rotation about the major axis, both in radians. The convention when dealing with rotated ellipses is to measure the angle of rotation of the major axis from +j (beh) clockwise (BEH!) but also, since I went through the trouble of rotating these images so that they're not sideways, a_ang is measured clockwise from -i.  The elliptical fit is programmed to run along the inside edge of the larger target circle. {'range': 42.27010789888579, 'a_ang': -0.04591864992309163, 'b_ang': 0.3644755561635681} Standard up front shot, I'm impressed that the software picked up that the target is actually tilted upwards relative to the camera though I think .36 radians (20 ish degree...
Keep reading

A Working Filter Actuator

It may have taken a week and an unwieldy hardware solution, but I now have an operational filter actuator.  hooray The transistors I tried just simply could not supply enough current in series to the solenoid when biased by the 3.3V digital output of a GPIO pin, so I threw together a servo-operated switch that flicks the filters back and forth pretty reliably.  I've been testing the switch and actuator by setting the scripts that operate them on an infinite loop (see video) and after a couple of solder joint breaks and design revisions the assembly finally seems to be straightened out. In the meantime, I have continued working on the startup scripts. Of course I have scripts operating both lights and the actuator, and I am integrating them all into a sort of pre-mission check to verify functionality.  I have managed to take images using both the IR and visible light filters, and the pictures look good.  All that remains before the real hard part of this project i...
Keep reading