Feature Addition to the Color Harmonization Application in Two Dimensional Image

Abstract

– In the era of digital technology, the demand for good visual communication increases. It deals with how to harmonize the usage of colors on that visual communication. It is called as color harmony. Color harmony is a series of combinations of colors that are balanced and aesthetically for visual perception.  There is an application developed by Yang (2018) that can be used to design  a good visual communication based on color harmony. This paper discusses on how to add some additional features to that application so that it can be more easily used and understood for the general public. Additional features include manual harmonization operation, color palette recommendation using the K-means clustering algorithm, multithreading and GUI. This paper also measures the performance and expert judgment of the level of color harmony in the application output. The experiment data are 5 sample images with various resolution sizes of 640x480, 800x600, 960x720, 1024x768, 1280x960, 1440x1080, 1600x1200, and 1920x1440. The fastest prosses time for the best harmonic scheme is 201 seconds for 640x480 resolution and the longest time is 1350 seconds for 1600x1200 resolution. The rate of expert judgment in harmonious samples is 9.07 out of 12.00 and in non-harmonized samples is 6.02 out of 12.00.