A Machine Learning–Based Quality Control Algorithm for Tangerines Using Descriptors Derived from Peel Sectorization

Abstract

Tangerines must fulfill specific quality control requirements established by major importing countries such as peel tone and ripeness, which are analyzed through visual inspection. This conventional evaluation is subjective since it depends on human perception; hence, automated solutions based on computer vision are considered, using different types of cameras, color models, adaptive thresholds, transforms, and convolutional neural networks (CNNs). In the state of the art, the proposed solutions are focused on determining peel defects by applying image processing and machine learning techniques with performance levels above 85%. In this work, an alternative solution is presented using RGB cameras with a resolution of 640 × 480 pixels and descriptors designed to characterize peel-tone variations by sectors, simulating human visual assessment for quality determination. These descriptors are used as inputs to classifiers based on support vector machines and multilayer perceptron neural networks, achieving accuracy levels above 92%. Finally, a comparative analysis between the proposed algorithm and CNN-based quality classifiers such as YOLOv11 and ResNet50 is done, proving that the proposed method outperforms CNN-based classifiers.



Received: 30 December 2025 | Revised: 23 April 2026 | Accepted: 23 July 2026



Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.



Data Availability Statement

The data that support the findings of this study are openly available in Mendeley Data at https://doi.org/10.17632/9tfbpxpyjz.1.



Author Contribution Statement

Guillermo Kemper: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration, Funding acquisition. Santiago Huamani: Conceptualization, Resources, Writing – original draft. Sebastián Rojas: Conceptualization, Resources, Writing – original draft. Enrique Tabory: Methodology, Software, Validation, Formal analysis, Investigation, Writing – original draft, Writing – review & editing. Enori Cardenas: Formal analysis, Investigation, Writing – original draft. Kevin Guerra: Methodology, Software, Validation, Formal analysis, Investigation, Writing – original draft.