Abstract—Density is used to predict the internal quality of various fruits. For mangosteen, a floating technique based on density is used routinely as the sorting system for quality classification but it is inconvenient and time consuming. The applied image analysis technique was studied to evaluate the density of mangosteen. RGB digital 2-D pictures were converted to binary images. Pixels of mangosteen from eight pictures taken using different views were averaged and compared to the pixels of a reference circle plate. From the image analysis, the average volume of mangosteen was first calculated and then the average density of mangosteen was determined. The predicted densities of 80 fruits using the image analysis technique were compared with the measured densities and obtained a high correlation coefficient (R=0.91). The results showed the image analysis technique could be applied for nondestructive sorting to evaluate the internal quality of mangosteen based on the predicted density. 
 

Index Terms—nondestructive, image analysis, density, quality, prediction

Cite: Sontisuk Teerachaichayut, Wipawee Yokswad, Anupun Terdwongworakul, Piyamart Jannok, and Sofia Velez Fernandes, "Application of Image Analysis for Determination of Mangosteen Density," Journal of Advanced Agricultural Technologies, Vol. 2, No. 2, pp. 92-97, December 2015. Doi: 10.12720/joaat.2.2.92-97