Copyright © 2016 IJECCE, All right reserved 126 International Journal of Electronics Communication and Computer Engineering Volume 7, Issue 2, ISSN (Online): 2249–071X Grading of Apples and Oranges by Image Processing Osama A. Alhashi Lecturer Assistant, College of Electronic Technology, Libya osamahashi@yahoo.com Fathey S. Almahjob Lecturer Assistant, College of Electronic Technology, Libya almahjob@ymail.com Abdelsalam A. Almarimi Associate Professor, College of Electronic Technology Baniwalid, Libya belgasem2000@yahoo.com Abdosllam M. Abobaker Associate Professor, College of Electronic Technology Baniwalid, Libya almahjub11@yahoo.com Abstract – The aim of this paper is to assess and determine the ripeness and quality of apples. To meet the goals, we propose and implement certain methodologies and algorithms that are based on digital fuzzy image processing, content predicated analysis, and statistical analysis. We found that the proposed algorithm is an efficient one as it is able to detect and sort the apples with more accuracy in grading compared to human expert sorting. The textures on apple skin are captured using digital camera. These images are filtered using image processing technique. All the information gathered is processed using MATLAB to determine the apple ripeness accuracy. In MATLAB, first we find the RGB component of a good apple and a ripen apple. Then, the image is converted to a grayscale image in order to obtain the histogram graph for analyzing the results. Besides, we also apply the same algorithm to orange fruit to verify the validity. The results presented in this paper corroborate that the automated grading system helps to minimize the processing time as well as the assessment error. Keywords – Digital Image Processing, RGB Components, Histogram, Apple Ripeness. I. INTRODUCTION It is impossible to overestimate the importance of the agricultural products as they form the basic requirements for living beings, more specifically, human beings. In the previous years, the grading of the agricultural products was carried out by humans. Therefore, it has become the need of the hour to propose an automated grading to speed up the grading system of the agricultural products to fulfil the demands posed by the society in the recent years [1], [2]. In this line, in the recent past, researchers have proposed several models using image processing techniques [3],[4], [5], [6]. Over the last two centuries, fruit categorizations in agriculture have changed from traditional grading by humans to automatic grading. Many companies have started moving to automated grading in many crops such as grading on peaches, oranges, etc. In order to classify apples, we need to be aware of the apple grading standard. Color and size are the most significant criteria that are used to sort out the fruits. However, for sorting of apples, the skin texture of apples is another major factor that could improve the accuracy of the classification system. Traditional methods for assessing fruit ripeness are unfortunately destructive. Hence, they cannot be so readily applied, particularly in mass production. This paper presents a non-destructive method for determining the ripeness of apple based on its color. We take into account the RGB components of good and ripened apples. Then, the image is converted to grayscale image to obtain the histogram graph with which the results are analyzed [7], [8]. The objective of this work is to classify between the ripeness indexes of the apples based on RGB and histogram process. The significance of this study is to determine the ripeness stages by using color space of the apples. II. PROPOSED METHOD: APPLE GRADING This study proposes an apple grading method for classifying the quality of apples by using image analysis. The algorithm of the proposed method consists of the following five steps. Step 1: Detect the color of an apple by determining the mean of three color arrays for red, green and blue. Mean image = [Red value (Find size image) + Green value (Find size image) + Blue value (Find size image)]/3 Step 2: Determine the size of the apple by calculating the area of image object. Step 3: Apply edge detection algorithm to determine the skin of image apple. Mean skin = [edge (Red value) + edge (Green value) + edge (Blue value)]. Step 4: Histogram results are compared using size, color and skin for grading the apples. Step 5: Rank the quality of apples based on apple grade. The flow and framework of this study are shown in Figs. 1 and 2. Fig. 1. Flow chart of the proposed work