Design of Machine Learning for Limes Classification Based Upon Thai Agricultural Standard No. TAS 27-2017

Aug 4, 2026

Upload by Violet

Accurately classifying the limes quality under established Thai Agricultural Standard TAS 27-2017 is essential to build trust in farmer-to-market trading. Traditional machinery-based sorting aims to cut costs and reduce errors, yet it struggles to meet the strict TAS criteria for fresh limes sold to consumers. This research proposes a standard-aligned machine learning system that recognizes lime skin color and defects using convolutional neural networks with logistic regression. Results show lime image quality classification delivered through a computer GUI with accuracy above 90%.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages15

Reading Time38 min

AccessFree to download

Introduction

Global and Thai lime context

Nutritional and industrial uses of limes

Need for standardized quality regulations

Proposed System and Methodology

TAS 27-2017 driven classification targets

CNN-based image recognition workflow

Logistic regression integration

Results and Interface Implementation

Lime image quality presentation via GUI

Classification performance and accuracy

Discussion and Applications

Agricultural impact and standard compliance

Transferability to other agricultural products

Applied Science and Engineering Progress, Vol. 18, No. 1, 2025, 7322 1 Research Article Design of Machine Learning for Limes Classification Based Upon Thai Agricultural Standard No. TAS 27-2017 Athakorn Kengpol* and Alongkorn Klaiklueng Advanced Industrial Engineering Management Systems Research Center, Department of Industrial Engineering, Faculty of Engineering, King Mongkut’s University of Technology North Bangkok, Thailand * Corresponding author. E-mail: [athakorn@kmutnb.ac.th](athakorn@kmutnb.ac.th) DOI: 10 . 14416/j.asep.2024.01.005 Received: 29 September 2023; Revised: 20 November 2023;Accepted: 13 December 2023; Published online: 26 January 2024 © 2024 King Mongkut’s University of Technology North Bangkok. All Rights Reserved. Abstract Accurately classifying the limes quality of limes according to established standards is paramount for instilling trust in farmers' trading of agricultural produce. Historically, machinery has been employed to categorize the lime quality, with dual objectives of cost reduction and error mitigation, thereby facilitating the classifying process. Nevertheless, deploying such machinery to classify limes in their fresh produce form, intended for consumer sale, has encountered limitations imposed by the stringent criteria stipulated in Thai Agricultural Standards No. TAS 27-2017, a standard derived from the Codex Standard and widely adopted by numerous countries. Considering these constraints, the presented research aims to enhance the efficiency of limes classification, adhering to the standards. The Machine Learning System is designed to recognize and categorize limes based upon their skin color and defects to achieve this goal. This system employed convolutional neural network (CNN) models in conjunction with logistic regression equations, which are unavailable in the literature. The research findings indicate that this system is proficient in accurately presenting lime images and their corresponding quality classes via a Graphical User Interface on a computer screen, achieving an accuracy rate exceeding 90% . The implications of this research extend to the agricultural sector by augmenting the efficacy of Machine Learning for classifying limes in compliance with Thai Agricultural Standard No. TAS 27-2017. Furthermore, the methodology developed in this study can find applicability in classifying other agricultural products. Keywords: Agricultural standard, Convolutional neural network, Lime classification, Machine learning 1 Introduction The lime, citrus fruit of global cultivation, has exhibited a noteworthy trend of expanding harvested areas over the preceding decade, a phenomenon readily discernible in Figures 1 and 2. In 2021, the global lime harvested area amounted to 1.34 million hectares, yielding a lime production volume of 20.83 million tons. Approximately 43.71% of this prodigious production emanated from three lime-producing nations: India, Mexico,andChina.Thesenationscontributed3.55million tons, 2.98 million tons, and 2.57 million tons of lime, respectively, as delineated in Table 1 [1]. In Thailand, the lime harvested area in this nation encompassed a cumulative expanse of 0.02 million hectares, representing a modest 0.01% of the total global lime harvested area. Moreover, the lime production in Thailand amounted to a total of 0.16 million tons, signifying 0.77% of the world's lime production. Notably, a significant portion of the lime produced in Thailand is primarily earmarked for domestic consumption. Lime, a fruit of notable nutritional significance, is recognized for its ample supply of nutrients requisite for the sustenance of humans. Within the repertoire of lime cultivars, the cultivar Paen (Citrus aurantifolia (Christm) Swingle) holds a special place of favor among the Thai. Moreover, this versatile fruit finds application not only in the realm of dietary consumption but also extends its utility as a pivotal ingredient in pharmaceutical formula

FAQs

Why is lime quality classification under TAS 27-2017 important?

It ensures reliable trust between farmers and trading partners while meeting strict criteria for limes marketed to consumers under Thai Agricultural Standard TAS 27-2017.

What features does the proposed machine learning system use to classify limes?

The system classifies limes by recognizing skin color and defects, aligning the outputs with quality classes defined by TAS 27-2017.

How is the model implemented and evaluated in the study?

Convolutional neural network (CNN) models are combined with logistic regression, and results are presented through a graphical user interface on a computer screen, achieving accuracy above 90%.

Aug 4, 2026

Upload by Violet