Identify Toxin Contamination in Peanuts Using the Development of Machine Vision Based on Image Processing Technique

Abstract

This research aimed at identifying the use of image processing technique and classifying the use of K-mean clustering of contaminated and uncontaminated peanuts. A machine vision system was made of a small aluminum box, equipped with a camera, a petri for placing the sample, USB connector, UV lamp, and a computer. Image processing methods consisted of analysis of the average color of RGB in the region of interest, convertion of RGB into HSV, segmentation process, as well as convertion image into grayscale and binary objects in order to obtain the total number of pixels value in the object area so that the mean value of the pixels of the area can be calculated. K-means algorithm was used to classify the contaminated and uncontaminated peanuts based on the average pixel value of R,G, B color parameters. The accuracy of a system was 100% meaning that the performance of machine vision can be used to identify the contaminated and uncontaminated peanuts. Keywords : aflatoxin, K-mean clustering, machine vision, image processing, peanuts