Abstract
Manual grading process for Fresh Fruit Bunches (FFB) at the oil palm mills lead to mistakenly grading the low quality fruits as the good ones. This will result on the low level of oil content during production work. The classification of the right category of fruits based on their color as the main indicator for fruit ripeness is extremely important. It is difficult to measure especially when one can interpret the colors differently by just looking into them visually. The most suitable color space represented should be able to determine the right color for the ripeness identification. HSV is proved to be a good choice because it has all the colors in the channel. Besides, it helps to choose colors which are familiar to the eyes especially when it comes to color intensity. This paper explores the use of Nearest Neighbor Distance for histogram-based fruit ripeness identification. Histogram features are extracted and tested using Nearest Neighbor Distance for the similarity matching. Results are very promising which Value provides the obvious distinctive score towards ripe or unripe category.