Oranges are one of the many fruits that produce vitamin C. The size of oranges will affect the selling price in the market. Large oranges will be sold at a higher price and even become an export commodity. Oranges are valued by two factors; size and quality. This research aims to develop an automated system to determine the size of oranges using the requirements of the Indonesian National Standard (SNI 3932:2008) on the quality of Kepro oranges. This process uses image processing techniques, specifically segmentation by finding the area of the orange diameter. Orange size is measured by its diameter, and there are four levels of size based on SNI, namely first (70 mm), second (61-70 mm), third (51-60 mm), and fourth (40-50 mm). This size determination is usually done visually, but due to its subjectivity, this research aims to create a more objective automated system. The image processing includes testing several edge detection methods such as Prewitt, Canny, Roberts, and Sobel. In addition, the use of RGB coloring was also explored to improve the clarity of orange edges. The results show that the developed system is successful in acquiring images of oranges and identifying their size according to the requirements of the Indonesian National Standard.
Copyrights © 2023