Pengolahan Citra untuk Pemeriksaan Mutu Buah Mangga

Abstract

Indonesia has many kinds of tropical fruits, including mango, that can be exported besides to supply the demand on the fruits in the country. To ensure the quality of mango, it is important to do sortation and grading on them, especially those for export, based on the quality requirement. In this case, the use of high technology such as image processing is necessary to put into practice with the aim to increase the consistence of sortation and grading processes as compare to the result of manual work. As a starting point in developing a sortation and grading method using image processing, physical (weight and firmness) and chemical (total soluble solid) properties of mango fruit were measured and visual appearance of the mango was analyzed, to find out whether they have relationship with one another. The results show that from the four criteria of visual appearance of Arumanis mango (area, shape factor roundness, color, and texture), only area that has relationship with the weight, while the rest of them do not show any relationship. But, the textural features extracted from the images of Arumanis mango show a significant - differences between defect mango on the surface and the good ones. From the results, sortation and grading of Arumanis mango using image processing based on the maturity index is not recommended, but sortation and grading can be performed based on the existence of defect on the surface in combination with the size. These criteria can be applied to do sortation and grading on Arumanis mango as they are two of many factors for quality measurement, especially for export.