Pineapple quality grading using image processing and fuzzy logic based on Thai Agriculture Standards

Abstract

This research aimed to create a tool for pineapples quality grading according to the standard weight and size of Thai Agricultural Commodity Food Standard. The standard weights of pineapple are divided into 10 levels (A-J) and the standard sizes of pineapple are categorized into two classes (class I and class II). The developed tool consists of hardware components and a grading software program. The control light source box was constructed for camera and load cell installation to capture pineapple image and measure pineapple weight, respectively. The obtained image was sent to software program to change colors of the image into gray scale and to reduce noises in the image. The clearly edges of the image were employed to compute size of a pineapple and the data were transferred to fuzzy system. The inputs of fuzzy system determined the size and weight of pineapple which used to establish twenty fuzzy rules. The experiments performed by random selection size and weight of three pineapple kinds including Nanglae, Sriracha, Phuket. The experimental results reveal that classification of pineapple by the created tool exhibited high accuracy of size and weight detection equaled 87.5%. The average relative error performed 2.30% and 5.24% of size and weight, respectively.