Neural network training in SCILAB for classifying mango (Mangifera indica) according to maturity level using the RGB color model

Eduardo Castillo-Castaneda

doi:10.3934/steme.2021014

Article Contents

2021, Volume 1, Issue 3: 186-198. Doi: 10.3934/steme.2021014 open access

This issue Previous Article Reimagining multiplication as diagrammatic and dynamic concepts via cutting, pasting and rescaling actions Next Article A laser-cutting-centered STEM course for improving engineering problem-solving skills of high school students in China

Neural network training in SCILAB for classifying mango (Mangifera indica) according to maturity level using the RGB color model

Eduardo Castillo-Castaneda^1, ,

1.
Department of Mechatronics, Instituto Politecnico Nacional, CICATA Unidad Queréro, CP 76090, México

* Correspondence: ecastilloca@ipn.mx; Tel: +52-442-229-0804
* Correspondence: ecastilloca@ipn.mx; Tel: +52-442-229-0804

Academic Editor: Med Amine Laribi

Received: May 31, 2021

Revised: June 30, 2021

Published: August 2021

Abstract Full Text(HTML) Figure(16) Related Papers Cited by

Abstract

Industries that use fruits as raw materials must, at some point in the process, classify them to discard the unsuitable ones and thus ensure the quality of the final product. To produce mango nectar, it is necessary to ensure that the mango is mature enough to start the extraction of the nectar; however, sorting thousands of mangoes may require many people, who can easily lose attention and reduce the accuracy of the result. Such kind of decision can be supported by current Artificial Intelligence techniques. The theoretical details of the processing are presented, as well as the programming code of the neural network using SCILAB as a computer language; the code includes the color extraction from mango images. SCILAB programming is simple, efficient and does not require computers with large processing capacity. The classification was validated with 30 images (TIF format) of Manila variety mango; the mangoes were placed on a blue background to easily separate the background from the object of interest. Four and six mangoes were used to train the neural network. This application of neural networks is part of an undergraduate course on artificial intelligence, which shows the potential of these techniques for solving real and concrete problems.

Keywords:

Mathematics Subject Classification: Article.

Citation:

FullText(HTML)

Figure 1. Manual mango classification [2].

Download: Full-size image PowerPoint slide

Figure 2. Stages for the implementation of an AI computer program.

Download: Full-size image PowerPoint slide

Figure 3. Levels of mango maturity.

Download: Full-size image PowerPoint slide

Figure 4. Segmentation applied to separate the airplane from the sky (including clouds).

Download: Full-size image PowerPoint slide

Figure 5. Some colors and the corresponding RGB values.

Download: Full-size image PowerPoint slide

Figure 6. a), b), c), d) Color extraction images, e) SCILAB code.

Download: Full-size image PowerPoint slide

Figure 7. Model of an artificial neuron, named perceptron, with 3 input variables.

Download: Full-size image PowerPoint slide

Figure 8. An ANN with 3 layers.

Download: Full-size image PowerPoint slide

Figure 9. SCILAB code to calculate de ANN output.

Download: Full-size image PowerPoint slide

Figure 10. Training of an ANN from known values of $ X $ and $ Y $ vectors.

Download: Full-size image PowerPoint slide

Figure 11. SCILAB program for ANN training.

Download: Full-size image PowerPoint slide

Figure 12. Mangoes selected for ANN training.

Download: Full-size image PowerPoint slide

Figure 13. Input and output variables for ANN training.

Download: Full-size image PowerPoint slide

Figure 14. Preliminary result of the classification of 30 mangoes.

Download: Full-size image PowerPoint slide

Figure 15. Green, medium, and mature mangoes for ANN training.

Download: Full-size image PowerPoint slide

Figure 16. Result with improved training using 6 mangoes.

Download: Full-size image PowerPoint slide

Related Papers

Cited by

References

[1]	S.N. Jha, S. Chopra and A.R.P. Kingsly, Modeling of color values for nondestructive evaluation of maturity of mango, Journal of Food Engineering, 78 (2007), 22-26.
[2]	In: https://www.magmar.com.mx/, available 4th July 2021.
[3]	S.J. Russell and P. Norvig, Artificial Intelligence: A modern Approach, Prentice Hall, New Jersey, USA, 1995.
[4]	Bughin, J., Seong, J., Manyika, J., Chui, M. and Joshi, R., Notes from AI frontier: Modeling the impact of AI on the world economy. Discussion paper, 2018, McKinsey Global Institute.
[5]	S. Lucci and D. Kopec, Artificial Intelligence in the 21st Century, Mercury Learning and Information, VA, USA, 2015.
[6]	D.S. Prabha and J.S. Kumar, Assessment of banana fruit maturity by image processing technique, J Food Sci Technol, 52 (2015), 1216-1327.
[7]	R. Torres, E.J. Montes, O.A. Pérez and R.D. Andrade, Relación del Color y del Estado de Madurez, con las Propiedades FisicoquȪmica de Frutas Tropicales, Información Tecnológica, 24 (2013), 51-56.
[8]	J.C. Russ, The Image Processing Handbook, Prentice Hall, 1999.
[9]	R.C. González and R.E. Woods, Digital Image Processing, Prentice Hall, 2002.
[10]	F. Rosenblatt, The perceptron: A probabilistic model for information storage and organization in the brain, Psychological Review, 65 (1958).
[11]	Rojas, R., The Backpropagation Algorithm. Neural Networks, 1996, pp. 149-182. Springer-Verlag.
[12]	D.P. Mandic, A generalized normalized gradient descent algorithm, IEEE Signal Processing Letters, 11 (2004), 115-118.