Detection of Biochemical Variations in Natural and Artificially Ripened Fruits Using Spectral Analysis.

Abstract

As artificial fruit ripening can change a fruit\'s natural biochemical makeup, it is frequently challenging to tell the difference between naturally ripened and artificially ripened fruits by visual inspection. This study uses spectral reflectance analysis to examine biochemical differences between naturally ripened bananas, mangos, and apples and those that are artificially ripened. An ASD FieldSpec 4 spectroradiometer was used to collect spectral data between 350 and 2500 nm in wavelength. 180 banana, 330 mango, and 200 apple spectra made up the extensive library of 1,430 spectral characteristics. Preprocessing, such as noise reduction, second-derivative transformation, and normalisation, was applied to the gathered spectral data. For both the naturally ripened and artificially ripened groups, mean reflectance was computed separately at each wavelength to generate typical spectral patterns. The visible, near infrared, and short-wave infrared regions all showed distinct differences when the mean spectral fingerprints were compared. Significant variations were found in the following regions: 400–699 nm for pigments, 700–999 nm for internal tissue structure, roughly 970 nm for water, 1000–1300 nm for sugars and carbohydrates, roughly 1450 nm for moisture, roughly 1900 nm for the water combination band, and 2100–2300 nm for organic compounds. Around 970, 1450, 1900, and 2200 nm were the most noticeable and prevalent differences among the three fruits. Variations in water distribution, moisture, sugars, starch, carbohydrates, pigments, and other organic components related to the ripening process are indicated by these spectrum variances. The results show that broad-range spectrum analysis can offer a quick and non-destructive method for identifying biochemical differences and distinguishing between naturally ripened and artificially ripened fruits.

Introduction

Fruits are an essential part of the human diet, providing carbohydrates, vitamins, minerals, dietary fiber, antioxidants, and other bioactive compounds. During natural ripening, fruits undergo physiological and biochemical changes, including pigment transformation, starch-to-sugar conversion, changes in organic acids, tissue softening, and moisture variation. These changes alter the fruit's optical and spectral properties because light absorption and reflectance depend on its biochemical composition and physical structure.

Natural ripening is primarily regulated by ethylene, whereas artificial ripening often uses chemicals such as calcium carbide, which releases acetylene gas to accelerate ripening. Although artificially ripened fruits may appear similar externally, their internal biochemical development often differs, affecting moisture, pigments, sugars, starch, firmness, and other quality attributes. Conventional laboratory methods for detecting these differences are accurate but destructive, time-consuming, and require specialized equipment. Spectral reflectance analysis offers a rapid, non-destructive alternative by measuring how electromagnetic radiation interacts with fruit tissues across visible, near-infrared (NIR), and short-wave infrared (SWIR) wavelengths. Specific wavelength regions correspond to pigments, water, carbohydrates, sugars, proteins, and other organic compounds.

Previous studies have successfully used spectroscopy and hyperspectral imaging to evaluate fruit ripeness, firmness, soluble solids, acidity, and maturity in apples, mangoes, kiwifruit, strawberries, tomatoes, and other fruits. However, most research has focused on predicting quality parameters or classifying ripening stages rather than directly comparing the complete spectral fingerprints of naturally and artificially ripened fruits. Few studies have provided biochemical interpretations of wavelength-specific differences across multiple fruit species.

To address this gap, the present study compares naturally and artificially ripened apples, mangoes, and bananas using non-imaging spectral reflectance measurements collected with an ASD FieldSpec 4 spectroradiometer over the 350–2500 nm range. A total of 710 spectral signatures were analyzed (180 banana, 330 mango, and 200 apple samples). The spectral data were preprocessed using noise reduction, normalization, and second-derivative transformation, and representative mean spectral signatures were computed for each fruit and ripening condition. Comparative analysis focused on identifying wavelength regions associated with pigments (400–699 nm), internal tissue structure (700–999 nm), water absorption (970 nm), sugars and carbohydrates (1000–1300 nm), moisture (1450 nm), water combination absorption (1900 nm), and organic compounds such as sugars and starch (2100–2300 nm).

Results showed consistent spectral differences between naturally and artificially ripened fruits. In bananas, naturally ripened samples exhibited higher reflectance in the visible and near-infrared regions, indicating better pigment preservation and tissue integrity. Significant differences were also observed at wavelengths related to water, moisture, sugars, carbohydrates, starch, and other organic compounds, suggesting that artificial ripening alters multiple biochemical components rather than a single characteristic. Similar spectral trends were observed in apples, where differences across pigment, water, moisture, tissue structure, and carbohydrate-related wavelength regions indicated that artificial ripening affects both external appearance and internal biochemical composition.

Conclusion

This study used spectral reflectance analysis to examine biochemical differences between naturally occurring and artificially ripened bananas, mangos, and apples. 180 banana, 330 mango, and 200 apple spectra were among the 1,430 spectral signatures that were examined. An ASD FieldSpec 4 spectroradiometer was used to provide spectral observations over the wide wavelength range of 350–2500 nm. To increase the visibility and comparability of spectrum fluctuations, the obtained data were pre-processed using noise reduction, second-derivative transformation, and normalisation. For both the naturally ripened and artificially ripened groups, mean spectral signatures were computed wavelength by wavelength. This method preserved the entire spectral pattern while reducing random sample-to-sample variance. For all three fruits, a comparison of the mean spectra revealed distinct changes between the two ripening regimes. These variations were focused in particular wavelength ranges linked to significant biological components rather than being evenly distributed over the entire spectral spectrum. Variations linked to chlorophyll and carotenoid pigments were obvious in the 400–699 nm range, whereas variations in internal cell structure and tissue integrity were visible in the 700–999 nm range. Water absorption properties were linked to variations at 970 nm. Variations related to sugars, starch, and carbs were observed in the 1000–1300 nm range. Changes in moisture distribution and water-related absorption behaviour were suggested by notable fluctuations between 1450 and 1900 nm. Variations related to sugars, starch, phenolic compounds, and other organic components were observed in the 2100–2300 nm range. The most noticeable and prevalent differences between bananas, mangoes, and apples were found in the spectral areas of 970, 1450, 1900, and 2200 nm. These results suggest that the ripening process is linked to quantifiable alterations in fruit\'s water, moisture, carbohydrates, sugars, pigments, and other organic components. Common constituent-related wavelength areas were found, despite the fact that the three fruit kinds\' spectral differences differed in magnitude. The findings show that spectral reflectance analysis offers valuable insights beyond fruit\'s outward look. Without harming the material, the suggested method makes it possible to identify and analyse internal biochemical changes. As a result, broad-range spectral analysis has great promise as a quick, non-destructive, and trustworthy method for evaluating fruit quality and distinguishing between naturally ripened and artificially ripened fruits. Future research can concentrate on creating reduced-wavelength sensing systems for useful and real-time fruit quality assessment as well as confirming the identified spectral regions using laboratory-based biochemical tests.

References

[1] K. O. Odongo et al., “Direct and rapid screening of calcium carbide in ripened bananas using chemometric-assisted laser Raman spectroscopy,” Applied Physics B, 2023. [2] “Calcium carbide (CaC?) ripening in fruits: Health risks, non-destructive detection, and future perspectives,” Comprehensive Reviews in Food Science and Food Safety, 2025. [3] R. K. Chaudhary et al., “Mango quality assessment using near-infrared spectroscopy and hyperspectral imaging,” Agronomy, vol. 15, no. 10, 2025. The study explains the relationship between NIR spectral responses and molecular bonds associated with water, sugars, and proteins. [4] R. Lu, “Multispectral imaging for predicting firmness and soluble solids content of apple fruit,” Postharvest Biology and Technology, 2004. [5] P. Rungpichayapichet et al., “Prediction mapping of physicochemical properties in mango by hyperspectral imaging,” Biosystems Engineering, 2017. [6] A. Benelli et al., “Ripeness evaluation of kiwifruit by hyperspectral imaging,” Biosystems Engineering, 2022. [7] M. Zhao et al., “Determination of quality and maturity of processing tomato using near-infrared hyperspectral imaging,” LWT, 2023. [8] R. Lu, “Multispectral imaging for predicting firmness and soluble solids content of apple fruit,” Postharvest Biology and Technology, vol. 31, no. 2, pp. 147–157, 2004. [9] R. Ullah, S. Khan, S. Bilal, and M. Nurjis, “Non-invasive assessment of mango ripening using fluorescence spectroscopy,” Optik, vol. 127, no. 13, pp. 5186–5189, 2016. [10] C. Guo, F. Liu, W. Kong, Y. He, and B. Lou, “Hyperspectral imaging analysis for ripeness evaluation of strawberry with support vector machine,” Journal of Food Engineering, vol. 179, pp. 11–18, 2016. [11] P. Rungpichayapichet, B. Mahayothee, M. Nagle, P. Khuwijitjaru, and J. Müller, “Prediction mapping of physicochemical properties in mango by hyperspectral imaging,” Biosystems Engineering, vol. 159, pp. 109–120, 2017. [12] Z. Gao, Y. Shao, G. Xuan, Y. Wang, Y. Liu, and X. Han, “Real-time hyperspectral imaging for the in-field estimation of strawberry ripeness with deep learning,” Artificial Intelligence in Agriculture, vol. 4, pp. 31–38, 2020. [13] A. Benelli, C. Cevoli, L. Ragni, and A. Fabbri, “Ripeness evaluation of kiwifruit by hyperspectral imaging,” Biosystems Engineering, vol. 223, pp. 42–52, 2022. [14] N. M. T. Nguyen et al., “Ripeness evaluation of Achacha fruit using hyperspectral imaging,” Agriculture, vol. 12, no. 12, Art. no. 2145, 2022. [15] P. Tian et al., “Detection of mango soluble solid content using hyperspectral imaging technology,” Infrared Physics & Technology, 2023. [16] N. Renugadevi et al., “An analysis on detection of artificially ripened fruits,” 2025. [17] R. K. Chaudhary et al., “Mango quality assessment using near-infrared spectroscopy and hyperspectral imaging,” Agronomy, vol. 15, no. 10, Art. no. 2271, 2025.

Copyright

Copyright © 2026 Snehal S. Datwase, Dr. Charansing N. Kayte, Dr. R. R. Deshmukh. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.