Automatic Color Mixer : A method for automated color recognition and replication

Detta är en Kandidat-uppsats från KTH/Maskinkonstruktion (Inst.)

Författare: Max Lindgren; Max Thiel; [2017]

Nyckelord: ;

Sammanfattning: Throughout history colors have been an important way forliving organisms to signal something to their surroundings.For instance, the color of a plant’s leaves or petals havebeen the way with which the plant communicates with itssurroundings for example to attract insects to spread itsDNA, or to signal that it is poisonous. Animals have alsobeen using colors to impress the other gender or to frightenother animals in order to stay alive and spread their DNA.One could argue that colors in some cases are vital andhave played a major role in evolution. This may not be thecase today for us humans, but the importance and use ofcolor is still very significant in todays society.The purpose of this project is to construct a demonstratorthat can mix a color that it gets as input from auser. Using automation control the machine can first mix acolor and then compensate for any errors just like a humanwould do to get the right result. To be able to do this acolor sensor that reads color in RGB is used. The differentlevels are then translated to a another color model calledHSV, which is then used for mixing the final color.The performance of the system was tested by lettinga sensor scan a sample color card, calculate how much ofeach primary color to pour into the mix. The mixed paintwas then painted onto a piece of paper that was left todry before being scanned. The difference in hue was notedafter the first mix and the system continued adjusting forany deviations until a good result was achieved. The resultof the tests was that it took a maximum amount of threemixes to get within 0,8% of the reference value. In half ofthe tests the demonstrator achieved a good result after thevery first mix.

  HÄR KAN DU HÄMTA UPPSATSEN I FULLTEXT. (följ länken till nästa sida)