Non-Destructive Assessment of Fruit Quality Using Image Texture Features and Minimum Distance Classification

Authors

  • Sukanya Phongsuphap Faculty of Information and Communication Technology, Mahidol University, Bangkok
  • Wimon Utanon Information Communication and Data Science Program, Faculty of Science and Technology, Bansomdejchaopraya Rajabhat University, Bangkok
  • Thannicha Thongyoo Faculty of Information Technology, Thepsatri Rajabhat University Rajabhat University, Bangkok
  • Eka Utanon Information Communication and Data Science Program, Faculty of Science and Technology, Bansomdejchaopraya Rajabhat University, Bangkok

Keywords:

Fruit Quality Assessment, Image Processing Technique, Texture Features

Abstract

This research aims to study non-destructive fruit quality assessment using guava-kimju as a case study. It presents an image processing technique by analyzing the guava's texture features obtained from images. The study explores the attributes related to taste and flesh quality of guava. The experiments are conducted using texture features (1) Single feature, (2) Multiple features, and classification is performed using the minimum distance classification method.

The results have shown that, (1) classifying the correlation between guava taste (sweet/not sweet) using the minimum distance method and the "smoothness" texture feature is more accurate than other texture features, with an accuracy rate of 66.67%. (2) Classifying the correlation between guava flesh texture (soft/not soft) using the minimum distance method and the "clumpiness" and "missibility" texture features is more accurate than other texture features, with an accuracy rate of 80.00%.

References

ณัฐพล ชัยทวิชธานันท์. (2555). การตรวจสอบฝักมะขามหวานเสียโดยวิธีการประมวลผลภาพ. มหาวิทยาลัยราชภัฏเพชรบูรณ์ สถาบันวิจัยและพัฒนา.

วิมล อุทานนท์, สุกัญญา พงษ์สุภาพ, วันทนี สว่างอารมณ์ และณัฐดนัย สิงห์คลีวรรณ. (2563). การประเมินคุณภาพผลไม้แบบไม่ทำลายโดยใช้เทคนิคการประมวลผลภาพและการจำแนกด้วยฐานกฎ. ว.สารสนเทศ. 19(1), 129-138.

ศิริลักษณ์ วงศ์เกษม. (2555). การใช้คอมพิวเตอร์วิทัศน์ตรวจสอบคุณภาพและคัดแยกผลผลิตทางการเกษตร. ว.มหาวิทยาลัยราชภัฏพิบูลสงคราม. 13(2), 10-21.

Dae Gwan Kim, Burks, Thomas F., Jianwei Qin & Bulanon, D. M. (2009). Classification of grapefruit peel diseases using color texture feature analysis. International Journal of Agric & Biol Eng. 2(3), 41-50.

Gonzalez, R. C. & Woods, R. E. (2009). Digital image processing (3rd ed.). Prentice Hall.

Haralick, R. M., Shanmugam, K. & Dinstein, I. (1973). Textural features for image classification. IEEE Transactions on Systems, Man, and Cybernetics. 3(6), 610-621, USA.

Leemans, V. & Destain, M.-F. (2004). A real-time grading method of apples based on features extracted from defects. Journal of Food Engineering. 61(1), 83-89.

Mirzaei, H., & Saraee, M. (2007). A new and robust apple evaluation method using image processing. Presented at First Joint Congress on Fuzzy and Intelligent Systems. Ferdowsi University of Mashhad, Iran.

Sukanya, P., Takamatsu R. & Sato M. (1998). Image classification using the surface-shape operator and multiscale features. Proc. of fourteen international conference on pattern recognition. 334-337.

Tamura, H., Mori, S., & Yamawaki, T. (1978). Textural features corresponding to visual perception. IEEE Transactions on Systems, Man, & Cybernetics. 8(6), 460–473.

Unay D. & Gosselin B. (2004). Apple Defect Detection and Quality Classification with MLP- Neural Networks. Retrieved March, 12 2018. https://www.researchgate.net/publication/228952374.

Downloads

Published

2023-12-22

Issue

Section

บทความวิจัย