Online First

 Preview Page

Research Article

Relationships between External and Internal Quality Attributes in Mini Pumpkin for AI-Assisted Fruit Classification
Eun Ji Kim1†
,
Young Je Kim2†
,
Se Hun Ju1
,
Youngseok Kwon3
,
YoSup Park4
,
Inhoon Jang5
,
Jungin Kim6
,
Beom Seon Lee7
,
Haeyoung Na34*
1Interdisciplinary Program of Development and Utilization of Biological Resources, Graduate School of Mokpo National University, Muan 58554, Korea
2Department of Horticulture, Graduate School of Mokpo National University, Muan 58554, Korea
3Mokpo National University Nature Resource Institute, Muan 58554, Korea
4Department of Horticulture and Forestry, College of Life and Medical Sciences, Mokpo National University, Muan 58554, Korea
5School of AI & Semiconductor Convergence, Hankyong National University, Pyeongtaek 17738, Korea
6AI Research, Hankyong National University, Pyeongtaek 17738, Korea
7Department of Horticulture, School of Green Smart Farm, Sunchon National University, Sunchon 57922, Korea
*Corresponding Author
†These authors contributed equally to this work.
PDF XML Citation
Abstract

The mini pumpkin (Cucurbita maxima × Cucurbita moschata) has gained increasing amounts of attention in fresh produce markets owing to its high sugar content and suitability for single-serving consumption. Efficient and objective fruit grading is essential for maintaining consistent market quality in commercial production systems. This study evaluated the performance of an artificial-intelligence (AI)-based image classification model for the automated grading of mini pumpkins and investigated the relationships between external fruit traits and internal quality attributes. A total of 6,800 RGB images of mini pumpkins were collected and classified as marketable or immature fruits. A deep-learning model based on the ConvNeXt V2 architecture was trained and evaluated to classify mini pumpkin images. Additionally, morphological and physicochemical properties, in this case the fruit size, rind and flesh color, soluble solids content (SSC), and firmness, were measured to analyze the differences in fruit quality characteristics between the two fruit classes. Pearson’s correlation analysis was conducted to examine the relationships between external and internal fruit traits. The AI-based classification model achieved high overall accuracy of 99.0%, with F1-scores of 0.991 for both fruit classes. Marketable fruits showed significantly lower rind lightness (L*) and higher fruit weights, SSC values, and firmness levels relative to those of immature fruits. A correlation analysis revealed that darker rind color was associated with greater values for fruit weight, SSC, and firmness, indicating that external visual characteristics were associated with internal fruit quality attributes. Overall, the AI-based image analysis effectively classified mini pumpkin maturity and marketability based on external fruit characteristics. This approach provides a promising non-destructive method for automated fruit grading and quality evaluations in commercial pumpkin production systems.

Keywords
computer vision
deep learning
fruit maturity
image classification
non-destructive evaluation
postharvest quality
References
1

Aherwadi N, Mittal U, Singla J, Jhanjhi NZ, Yassine A, Hossain MS (2022) Prediction of fruit maturity, quality, and its life using deep learning algorithms. Electronics 11:4100. https://doi.org/10.3390/electronics11244100

10.3390/electronics11244100
2

Aline U, Bhattacharya T, Faqeerzada MA, Kim MS, Baek I, Cho BK (2023) Advancement of non-destructive spectral measurements for the quality of major tropical fruits and vegetables: a review. Front Plant Sci 14:1240361. https://doi.org/10.3389/fpls.2023.1240361

10.3389/fpls.2023.124036137662162PMC10471194
3

Apostolopoulos ID, Tzani M, Aznaouridis SI (2023) A general machine learning model for assessing fruit quality using deep image features. AI 4:812-830. https://doi.org/10.3390/ai4040041

10.3390/ai4040041
4

Ashraf M, Aslam MZ, Saeed N, Hussain SJ (2026) PalmNeXt: a ConvNeXt-based deep learning model for pest detection in date palm leaves. Front Plant Sci 16:1738129. https://doi.org/10.3389/fpls.2025.1738129

10.3389/fpls.2025.173812941658551PMC12872870
5

Bhargava A, Bansal A (2018) Fruits and vegetables quality evaluation using computer vision: A review. J King Saud Univ Comput Inf Sci 33:243-257. https://doi.org/10.1016/j.jksuci.2018.06.002

10.1016/j.jksuci.2018.06.002
6

Blasco J, Aleixos N, Moltó E (2003) Machine vision system for automatic quality grading of fruit. Biosyst Eng 85:415-423. https://doi.org/10.1016/S1537-5110(03)00088-6

10.1016/S1537-5110(03)00088-6
7

Chakraborty SK, Subeesh A, Potdar RR, Chandel NS, Jat D, Dubey K, Shelake P (2023) AI-enabled farm-friendly automatic machine for washing, image-based sorting, and weight grading of citrus fruits: Design optimization, performance evaluation, and ergonomic assessment. J Field Robot 40:1581-1602. https://doi.org/10.1002/rob.22193

10.1002/rob.22193
8

Cubero S, Aleixos N, Moltó E, Gómez-Sanchis J, Blasco J (2011) Advances in machine vision applications for automatic inspection and quality evaluation of fruits and vegetables. Food Bioprocess Technol 4:487-504. https://doi.org/10.1007/s11947-010-0411-8

10.1007/s11947-010-0411-8
9

da Graça GA, Fontes PTN, de Jesus MS, Matos PN, Neta MTSL, Simões AN, de Carvalho LM, de Oliveira Júnior LFG, Carnelossi MAG, et al. (2024) Characterization of growth and development of pumpkin cv Mini Jack fruits. Food Chem Adv 5:100743. https://doi.org/10.1016/j.focha.2024.100743

10.1016/j.focha.2024.100743
10

de Almeida AB, de Lima TM, de Oliveira Filho JG, Santana RV, Lima DS, Moreira Alves E, Mariana E (2019) Relation between physicochemical characteristics and sensory profiles of cooked pumpkin varieties. Emir J Food Agric 31:697-707. https://doi.org/10.9755/ejfa.2019.v31.i9.2007

10.9755/ejfa.2019.v31.i9.2007
11

El-Mesery HS, Mao H, Abomohra AEF (2019) Applications of non-destructive technologies for agricultural and food products quality inspection. Sensors 19:846. https://doi.org/10.3390/s19040846

10.3390/s1904084630781709PMC6413199
12

Ermiş S, Ercan U, Kabaş A, Kabaş Ö, Moiceanu G (2025) Machine learning-based morphological classification and diversity analysis of ornamental pumpkin seeds. Foods 14:1498. https://doi.org/10.3390/foods14091498

10.3390/foods1409149840361582PMC12071245
13

Itle RA, Kabelka EA (2009) Correlation between L*a*b* color space values and carotenoid content in pumpkins and squash (Cucurbita spp.). HortScience 44:633-637. https://doi.org/10.21273/HORTSCI.44.3.633

10.21273/HORTSCI.44.3.633
14

Kader AA (2008) Flavor quality of fruits and vegetables. J Sci Food Agric 88:1863-1868. https://doi.org/10.1002/jsfa.3293

10.1002/jsfa.3293
15

Kamilaris A, Prenafeta-Boldú FX (2018) Deep learning in agriculture: A survey. Comput Electron Agric 147:70-90. https://doi.org/10.1016/j.compag.2018.02.016

10.1016/j.compag.2018.02.016
16

Karki S, Basak JK, Paudel B, Deb NC, Kim NE, Kook J, Kang MY, Kim HT (2024) Classification of strawberry ripeness stages using machine learning algorithms and colour spaces. Hortic Environ Biotechnol 65:337-354. https://doi.org/10.1007/s13580-023-00559-2

10.1007/s13580-023-00559-2
17

Lee P, Park SM (2025) Characteristics of hybrid individuals resulting from the hybridization of Cucurbita maxima and Cucurbita moschata. J Agric Life Environ Sci 37:373-385. https://doi.org/10.22698/jales.20250031

10.22698/jales.20250031
18

Lee SJ, Shin MH, Jayasooriya LSH, Wijethunga WMUD, Lee SK, Cho JG, Jang SH, Cho BK, Kim JG (2024) Prediction of fruit maturity of ‘Mihwang’ peach using deep learning. Hortic Sci Technol 42:80-93. https://doi.org/10.7235/HORT.20240007

10.7235/HORT.20240007
19

Li L, Hu DY, Tang TY, Tang YL (2022) Non-destructive detection of the quality attributes of fruits by visible-near infrared spectroscopy. J Food Meas Charact 17:1526-1534. https://doi.org/10.1007/s11694-022-01724-4

10.1007/s11694-022-01724-4
20

Liu Z, Wang P, Liu Z, Sun L, Li X, Liu J, Xing R, Zhang X, Deng H, Zhang Y (2024) Dynamic changes and correlation analysis of pumpkin quality and physiological indicators during postharvest storage. HortScience 59:1543-1549. https://doi.org/10.21273/HORTSCI18000-24

10.21273/HORTSCI18000-24
21

Ma J, Li M, Fan W, Liu J (2024) State-of-the-art techniques for fruit maturity detection. Agronomy 14:2783. https://doi.org/10.3390/agronomy14122783

10.3390/agronomy14122783
22

Mahendran R, Jayashree GC, Alagusundaram K (2011) Application of computer vision technique on sorting and grading of fruits and vegetables. J Food Process Technol. https://doi.org/10.4172/2157-7110.S1-001

10.4172/2157-7110.S1-001
23

Moreno SR, Yazdanpanah M, Huang T, Sims CA, Chase CA, Meru G, Simonne A, MacIntosh AJ (2023) Comparative analysis of qualitative attributes for selection of calabaza genotypes in the southeast united states. Horticulturae 9:409. https://doi.org/10.3390/horticulturae9030409

10.3390/horticulturae9030409
24

Naranjo-Torres J, Mora M, Hernández-García R, Barrientos RJ, Fredes C, Valenzuela A (2020) A review of convolutional neural network applied to fruit image processing. Appl Sci 10:3443. https://doi.org/10.3390/app10103443

10.3390/app10103443
25

Pathare PB, Opara UL, Al-Said FAJ (2013) Colour measurement and analysis in fresh and processed foods: A review. Food Bioprocess Technol 6:36-60. https://doi.org/10.1007/s11947-012-0867-9

10.1007/s11947-012-0867-9
26

Ren Y, Li B, Jia H, Yang X, Sun Y, Shou J, Jiang G, Shi Y, Chen K (2023) Comparative analysis of fruit firmness and genes associated with cell wall metabolisms in three cultivated strawberries during ripening and postharvest. Food Qual Saf 7:fyad020. https://doi.org/10.1093/fqsafe/fyad020

10.1093/fqsafe/fyad020
27

Rojas Santelices I, Cano S, Moreira F, Peña Fritz Á (2025) Artificial vision systems for fruit inspection and classification: systematic literature review. Sensors 25:1524. https://doi.org/10.3390/s25051524

10.3390/s2505152440096367PMC11902675
28

Ropelewska E, Cai X, Zhang Z, Sabanci K, Aslan MF (2022) Benchmarking machine learning approaches to evaluate the cultivar differentiation of plum (Prunus domestica L.) Kernels. Agriculture 12:285. https://doi.org/10.3390/agriculture12020285

10.3390/agriculture12020285
29

Rural Development Administration (RDA) (2023) Organic cultivation manual for sweet pumpkin. Naju, Korea

30

Ruwanthika KOGH, Munasinghe MLAMS, Marapana RAUJ (2023) Overview of Cucurbita spp. (pumpkin) and development of value-added products emphasizing its nutritional and chemical composition. World J Adv Res Rev 18:1215-1226. https://doi.org/10.30574/wjarr.2023.18.2.0938

10.30574/wjarr.2023.18.2.0938
31

Schober P, Boer C, Schwarte LA (2018) Correlation coefficients: Appropriate use and interpretation. Anesth Analg 126:1763-1768. http://doi.org/10.1213/ANE.0000000000002864

10.1213/ANE.0000000000002864
32

Tapia-Mendez E, Cruz-Albarran IA, Tovar-Arriaga S, Morales-Hernandez LA (2023) Deep learning-based method for classification and ripeness assessment of fruits and vegetables. Appl Sci 13:12504. https://doi.org/10.3390/app132212504

10.3390/app132212504
33

Vilhena NQ, Quiles A, Gil R, Llorca E, Fernández-Serrano P, Vendrell M, Salvador A (2022) Slight changes in fruit firmness at harvest determine the storage potential of the ‘Rojo Brillante’ persimmon treated with gibberellic acid. Horticulturae 8:140. https://doi.org/10.3390/horticulturae8020140

10.3390/horticulturae8020140
34

Walsh KB, Blasco J, Zude-Sasse M, Sun X (2020) Visible-NIR ‘point’ spectroscopy in postharvest fruit and vegetable assessment: The science behind three decades of commercial use. Postharvest Biol Technol 168:111246. https://doi.org/10.1016/j.postharvbio.2020.111246

10.1016/j.postharvbio.2020.111246
35

Woo S, Debnath S, Hu R, Chen X, Liu Z, Kweon IS, Xie S (2023) ConvNeXt V2: Co-designing and scaling convnets with masked autoencoders. arXiv:2301.00808. https://doi.org/10.1109/CVPR52729.2023.01548

10.1109/CVPR52729.2023.01548
36

Zhao C, Yin Z, Tan Y, Zhang W, Guo P, Ma Y, Wu H, Hu D, Lu Q (2025) Apple watercore grade classification method based on convNeXt and visible/near-infrared spectroscopy. Agriculture 15:756. https://doi.org/10.3390/agriculture15070756

10.3390/agriculture15070756
Information
  • Publisher :KOREAN SOCIETY FOR HORTICULTURAL SCIENCE
  • Publisher(Ko) :한국원예학회
  • Journal Title :Horticultural Science and Technology
  • Journal Title(Ko) :원예과학기술지
  • Received Date : 2026-04-07
  • Revised Date : 2026-05-29
  • Accepted Date : 2026-06-17
  • DOI :https://doi.org/10.7235/HORT.20260021
Journal Informaiton Horticultural Science and Technology Horticultural Science and Technology
Online Submission
  • scopus
  • NRF
  • KOFST
  • crossref crossmark
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close