Color Features Extraction Based on Min-Max Value from RGB, HSV, and HCL on Medan Oranges Image

Abstract

Research on plantation products has now turned to non-destructive research, this is because the quality of plantation products still uses the manual method of relying on sight or hand size to distinguish which is good, damaged, ripe, raw, large or small. Of course, the results are inconsistent, due to differences in perceptions of sight and the size of the hand between farmers with each other. Now the researcher conduct research based on the analysis of image processing. Where color extraction features (other than shape and texture) which is the stage of extracting the information contained in an object in a digital image. This information is used to distinguish between one object and another object at the stage of grouping/identification analysis based on color. In this case, the author extracts color features based on the minimum and maximum values for each component of the values R, G, B, H, S, V, H, C and L using the RGB, HSV and HCL methods. Thus, it can be seen the differences in the results of color extraction that characterize the object: Medan oranges of the three methods. The conclusion resulted from this research can't be used as a basis for determining specific the characteristics of each oranges class, because there is any overlap minimum-maximum value