SEARCHING SIMILAR OF PAINT COLOR BASED ON RGB AND HSV SPECTRUM USING K-MEANS METHOD
Rian Tri Cahyono, Eko Agus Prasetyo, Rifki Fahrial · 2019
Currently the community in determining the desired paint color choice still manually searches by selecting the drawings one by one in the manual paint catalog, the more color choices available in the catalog take a long time to determine the desired color choice. This research replaces the search manually by looking at the catalog with a computer based system. Similar paint color search uses image processing, ie reading the intensity of the Red, Green, Blue (RGB) value of the image, the value of the RGB is then calculated to get the HSV value, so as to get the RGB and HSV features as the input calculation method kmeans clustering, the data used as training data obtained from dulux wall paint catalog of 317 data. The results in this study average data can be grouped into 11 groups and 15 iterations, also calculated by a method to check the cluster validity of Davies Bouldin Index (DBI) sebanya 3 times the experiment got an average value of 0879. Keywords: Image, RGB, HSV, K-means, color, paint