An Automated Tomato Quality Grading using Clustering based Support Vector Machine

This paper focus on representing the technique of fruit grade classification, automated machine vision based technology has become more potential and important to many areas like agricultural sector and food processing industry. The proposed system which calculates the grade of fruit based on its external features. Grading of fruits is one of the most important processes in post harvesting, but this procedure is mostly carried out manually which is not efficient as it requires enormous number of employment, and tends to human error. The grading process is carried out by capturing the fruit image using digital camera and this image is interpreted using image various processing techniques. Color is very prominent feature for recognizing defect and ripeness of the fruit. The major objective is to check the fruit quality with high speed for analyzing maximum number of fruits in least amount of time. The conventional process of fruit quality assessment needs new tools deciding the quality of fruit. This system performs color features and size of fruit and captures the fruit side view image. The Otsu thresholding and K-Means clustering algorithms are used to extract the features of fruit. This system achieve the fruit quality sorting using Support Vector Machine and gives a few advantages over conventional techniques. This system will help in the advancement of an automated non destructive grading system with high speed, high accuracy and low cost. Implementation of this system will have applications in fruit quality grading in field like food science and trades where standardization is essential.