Color Feature Extraction and Euclidean Distance for Classification of Oryza Sativa Nitrogen Adequacy Based on Leaf Color Chart (LCC)

Oryza sativa is a rice-producing plant which is one of the main commodities in various countries including Indonesia. In the process of maintaining the quality of rice plants in order to have good growth and high yields, an adequate supply of nitrogen (N) is needed. The most obvious and commonly seen symptom of N deficiency is a reduction in the green color of the leaves (chlorosis) [3]. Leaf color is an indocator that is useful for indicating the N fertilizer requirement of rice plants [3]. Currently a simple tool that can be used to measure the color of the leaves of rice plants as a determinant of the amount of N fertilizer is the Leaf Color Chart (LCC). However, the problem in this LCC is that the tool is still manual and the assessment / classification process is carried out using color estimates based on eye sight. This creates uncertainty because everyone has different estimates. Based on these problems, it is necessary to have an automatic classification system of rice leaf color that can help farmers in determining the category of rice plants based on LCC. In this study, the color classification system of rice leaf images was carried out by extracting RGB (Red, Green, Blue) image features of rice leaf images. While the classification process is done by finding the color similarity between the image of rice leaves with the LCC scale using the Euclidean Distance method. The results obtained from the color classification system of rice leaves in this study have an accuracy rate of 75%.