In-Line Sorting of Harumanis Mango Based on External Quality Using Visible Imaging

Abstract

The conventional method of grading Harumanis mango is time-consuming, costly and affected by human bias. In this research, an in-line system was developed to classify Harumanis mango using computer vision. The system was able to identify the irregularity of mango shape and its estimated mass. A group of images of mangoes of different size and shape was used as database set. Some important features such as length, height, centroid and parameter were extracted from each image. Fourier descriptor and size-shape parameters were used to describe the mango shape while the disk method was used to estimate the mass of the mango. Four features have been selected by stepwise discriminant analysis which was effective in sorting regular and misshapen mango. The volume from water displacement method was compared with the volume estimated by image processing using paired t-test and Bland-Altman method. The result between both measurements was not significantly different (P > 0.05). The average correct classification for shape classification was 98% for a training set composed of 180 mangoes. The data was validated with another testing set consist of 140 mangoes which have the success rate of 92%. The same set was used for evaluating the performance of mass estimation. The average success rate of the classification for grading based on its mass was 94%. The results indicate that the in-line sorting system using machine vision has a great potential in automatic fruit sorting according to its shape and mass.

Real-Time Size and Mass Estimation of Slender Axi-Symmetric Fruit/Vegetable Using a Single Top View Image

“…Due to the limitation of the load cell, many researchers try to find other methods to estimate weight/volume without using a load cell [ 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 ]. One of the most accurate ways is to estimate the volume of the product and using a pre-determined “density” to calculate the actual weight of the object because mass and volume are correlated to a certain degree.…”

“…One of the most accurate ways is to estimate the volume of the product and using a pre-determined “density” to calculate the actual weight of the object because mass and volume are correlated to a certain degree. For instance, the correlation coefficient (R 2 ) between mass and volume of mango is 0.9978 [ 15 ], tangerine, lime, lemon, and orange are 0.9045, 0.9592, 0.9641, and 0.9232, respectively [ 28 ].…”

“…On the other hand, there are some proposed that used only one two-dimensional image of an object to estimate its volume by using the disk method [ 29 , 30 , 31 , 32 , 33 ]. They slice the fruit into many disks with the properties shown in Figure 1 [ 15 , 28 , 29 , 30 , 31 , 32 , 33 ].…”

“…Ibrahim et al [ 15 ] also modified the disk method to fit with unpredictably shaped objects. Once they have two view images (top and side) of a mango, the area of each disk is defined in Equation (8).…”

“…Based on the results of those previously mentioned steps, we can estimate the fruit/vegetable volume. In addition, since we know there is a high correlation between fruit/vegetable weight and its volume [ 15 , 19 , 28 ], we can infer the mass based on volume information. Once we get the volume of the top, the middle, and the bottom parts, we estimate the volume and the density, from which we can calculate the object’s weight.…”