Camouflaging my bison monitoring system
While waiting for the bison delivery, I've been busying myself working on the remote monitoring system. A major part of this project has been constructing the weather-proof, climate-controlled housing for the central system server. All of the aluminum and pvc exterior have been completed and much of the interior insulation and air handling systems (i.e. pvc tubing with a fan and heating element) are installed. One of the last touches I plan to add to this project is external camouflage. After all, it is a surveillance system.
To make the camouflage at least halfway decent, I did some research on the topic. Among the plethora of academic and military documents I reviewed, I found the concepts of background matching (specifically color matching) and disruptive camouflage to be most useful in the sense that they were relatively simple and could be implemented systematically. Given that I am completely inexpert in this very old and well developed field of study, I must apologize ahead of time for my simplification and probable complete misapplication of these principles. What follows is an account of how I applied a few hours of reading and a few more hours of playing around with GNU Image Manipulation Program (GIMP).
To start I took a relatively generic photograph (below) I had of the ranch during late summer. I chose late summer because it seems to incorporate colors from the greatest number of seasons. Some of the grasses have already matured and faded to a tan or brown color while others are still adding new growth. The photograph was taken using a white balance set to "auto" and was compressed to a lossy JPEG format. I also paid absolutely no attention to color spaces during this entire process.
Using the Color picker tool set to "sample average" with a wide radius I was able to derive an average color for specific areas in the photograph. The photo below show examples of the size of each area sampled and the resulting averaged color for each area. Averaged areas were selected using the non-random "this-looks-like-a-good-area" methodology.
For camouflage pattern colors I decided to use the darker green on the right side of the photo, the lighter green on the left side of the photo and the tannish color in the foreground. To quote a famous person (me), these colors were selected because they "seemed right."
Gimp has a camouflage generating tool under File > Create > Patterns > Camouflage... that allows you to generate a three color camo pattern (below). I have no idea what method was used to generate the pattern, but I would categorize it as possibly disruptive.
While a pattern like this would probably be okay at the ranch, it would be an incredible pain to paint. A much easier system would be a pattern of squares that could be easily transferred to whatever I wanted to paint.
To produce a gridded version of this pattern I adjusted the setting on the tool to have a granularity of around 7 and a size of 10. This produced a 10x10 pixel image that, when resized by 2300 percent with no interpolation, gave me the grid effect I wanted (below).
To get paints in these colors, I printed three small sample squares from this pattern on a color printer (again, using no screen or printer calibrations) and brought it into Home Depot where they read the colors and "matched them." Some time in the next few weeks Jess plans to paint the surveillance box using this pattern and we will give it a test on the ranch.
To make the camouflage at least halfway decent, I did some research on the topic. Among the plethora of academic and military documents I reviewed, I found the concepts of background matching (specifically color matching) and disruptive camouflage to be most useful in the sense that they were relatively simple and could be implemented systematically. Given that I am completely inexpert in this very old and well developed field of study, I must apologize ahead of time for my simplification and probable complete misapplication of these principles. What follows is an account of how I applied a few hours of reading and a few more hours of playing around with GNU Image Manipulation Program (GIMP).
To start I took a relatively generic photograph (below) I had of the ranch during late summer. I chose late summer because it seems to incorporate colors from the greatest number of seasons. Some of the grasses have already matured and faded to a tan or brown color while others are still adding new growth. The photograph was taken using a white balance set to "auto" and was compressed to a lossy JPEG format. I also paid absolutely no attention to color spaces during this entire process.
Using the Color picker tool set to "sample average" with a wide radius I was able to derive an average color for specific areas in the photograph. The photo below show examples of the size of each area sampled and the resulting averaged color for each area. Averaged areas were selected using the non-random "this-looks-like-a-good-area" methodology.
For camouflage pattern colors I decided to use the darker green on the right side of the photo, the lighter green on the left side of the photo and the tannish color in the foreground. To quote a famous person (me), these colors were selected because they "seemed right."
Gimp has a camouflage generating tool under File > Create > Patterns > Camouflage... that allows you to generate a three color camo pattern (below). I have no idea what method was used to generate the pattern, but I would categorize it as possibly disruptive.
While a pattern like this would probably be okay at the ranch, it would be an incredible pain to paint. A much easier system would be a pattern of squares that could be easily transferred to whatever I wanted to paint.
To produce a gridded version of this pattern I adjusted the setting on the tool to have a granularity of around 7 and a size of 10. This produced a 10x10 pixel image that, when resized by 2300 percent with no interpolation, gave me the grid effect I wanted (below).
To get paints in these colors, I printed three small sample squares from this pattern on a color printer (again, using no screen or printer calibrations) and brought it into Home Depot where they read the colors and "matched them." Some time in the next few weeks Jess plans to paint the surveillance box using this pattern and we will give it a test on the ranch.
Comments
Post a Comment