Non-Destructive Prediction of Maturity from the Sound of Hand Hitting a Watermelon Using Machine Learning

Öz

Traditional methods for assessing the quality, taste, and ripeness of fruits and vegetables without cutting rely on attributes such as color, shape, surface patterns, and acoustic responses. The ripeness levels were verified by cutting the watermelons, and the corresponding sound data were examined using spectrogram analysis, extracting 120 features from each sample. Various machine learning algorithms, including Support Vector Classifier (SVC), Decision Trees (DTC), Random Forest Classifier (RFC), Multi-Layer Perceptron (MLP), and K-Nearest Neighbors Classifier (KNC), were applied to identify the most effective predictive model. The results indicate that the KNC model achieved the highest accuracy at 96.04%, followed by the RFC model with an accuracy of 95.47%. The RFC model classified ripe watermelons with 98.2% accuracy, while the KNC model distinguished overripe and underripe watermelons with accuracies of 96.3% and 96.2%, respectively. Despite the presence of background noise in the naturally recorded dataset, the system demonstrated high performance across all categories. The findings were compared with studies on acoustic pattern recognition in animals, environmental acoustic analysis, and healthcare applications. This study highlights that machine learning-based models provide a non-invasive approach to determining watermelon taste and ripeness, offering a practical solution for applications in the agriculture and food industries.

Anahtar Kelimeler

• Acoustic feature extraction
• Maturity prediction
• Mel Frequency Cepstral Coefficients
• Sound classification
• Supervised learning models

Makine Öğrenimi Kullanarak Elle Vurulan Karpuzun Sesinden Tahribatsız Olgunluk Tahmini

Öz

Sebze ve meyvelerin kesilmeden kalite, tat ve olgunluklarının değerlendirilmesine yönelik geleneksel yöntemler, renk, şekil, yüzey desenleri ve akustik tepkilere dayanmaktadır. Karpuzların olgunluk seviyeleri kesilerek doğrulanmış, ardından ses verileri spektrogram analizi ile incelenerek her örnekten 120 özellik çıkarılmıştır. Destek Vektör Sınıflandırıcısı (SVC), Karar Ağaçları (DTC), Rastgele Orman Sınıflandırıcısı (RFC), Çok Katmanlı Algılayıcı (MLP) ve K-En Yakın Komşu Sınıflandırıcısı (KNC) gibi çeşitli makine öğrenimi algoritmaları kullanılarak en etkili tahmin modeli belirlenmiştir. Sonuçlar, KNC modelinin %96,04 doğruluk oranı ile en iyi performansı sergilediğini, onu %95,47 doğruluk oranı ile RFC modelinin takip ettiğini göstermektedir. RFC modeli, olgun karpuzları %98,2 doğrulukla sınıflandırırken, KNC modeli aşırı olgun ve yeterince olgun olmayan karpuzları sırasıyla %96,3 ve %96,2 doğruluk oranlarıyla ayırt edebilmiştir. Doğal ortamlarda kaydedilen veri kümesinde arka plan gürültüsünün varlığına rağmen, sistem tüm kategorilerde yüksek bir performans sergilemiştir. Elde edilen bulgular, hayvanların ses desenleri, çevresel akustik analizler ve sağlık alanındaki uygulamalar üzerine yapılan çalışmalar ile karşılaştırılmıştır. Çalışma, makine öğrenimi tabanlı modellerin karpuzların tat ve olgunluk seviyelerini belirlemede invaziv olmayan bir yaklaşım sunduğunu ortaya koyarak, tarım ve gıda sektöründe pratik bir çözüm önerisi sunmaktadır.

Anahtar Kelimeler

• Akustik özellik çıkarımı
• Mel Frekans Kepstral Katsayıları
• Olgunluk tahmini
• Ses sınıflandırma
• Gözetimli öğrenme modelleri

Kaynakça

1. Gondchawar N, Kawitkar RS. IoT-based smart agriculture. Int J Adv Res Comput Commun Eng. 2016; 5(6): 838-42.
2. O'Grady MJ, Langton D, O'Hare GMP. Edge computing: A tractable model for smart agriculture? Artif Intell Agric. 2019; 3: 42-51.
3. Femling F, Olsson A, Alonso-Fernandez F. Fruit and vegetable identification using machine learning for retail applications. In: 2018 14th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS). IEEE; 2018 Nov. p. 9-15.
4. Barbole DK, Jadhav PM, Patil SB. A review on fruit detection and segmentation techniques in agricultural field. In: International Conference on Image Processing and Capsule Networks. Springer, Cham; 2021 Sep 10. p. 269-88.
5. Kyriacou MC, Leskovar DI, Colla G, Rouphael Y. Watermelon and melon fruit quality: The genotypic and agro-environmental factors implicated. Sci Hortic. 2018; 234: 393-408.
6. Jie D, Wei X. Review on the recent progress of non-destructive detection technology for internal quality of watermelon. Comput Electron Agric. 2018; 151: 156-64.
7. Kyriacou MC, Soteriou GA, Rouphael Y, Siomos AS, Gerasopoulos D. Configuration of watermelon fruit quality in response to rootstock‐mediated harvest maturity and postharvest storage. J Sci Food Agric. 2016; 96(7): 2400-9.
8. Soteriou GA, Kyriacou MC, Siomos AS, Gerasopoulos D. Evolution of watermelon fruit physicochemical and phytochemical composition during ripening as affected by grafting. Food Chem. 2014; 165: 282-9.

9. Arendse E, Fawole OA, Magwaza LS, Opara UL. Non-destructive characterization and volume estimation of pomegranate fruit external and internal morphological fractions using X-ray computed tomography. J Food Eng. 2016; 186: 42-9.
10. Wiktor A, Gondek E, Jakubczyk E, Nowacka M, Dadan M, Fijalkowska A, Witrowa-Rajchert D. Acoustic emission as a tool to assess the changes induced by pulsed electric field in apple tissue. Innov Food Sci Emerg Technol. 2016; 37: 375-83.
11. Zhang B, Huang W, Li J, Zhao C, Fan S, Wu J, Liu C. Principles, developments and applications of computer vision for external quality inspection of fruits and vegetables: A review. Food Res Int. 2014; 62: 326-43.
12. Nelson SO, Guo WC, Trabelsi S, Kays SJ. Dielectric spectroscopy of watermelons for quality sensing. Meas Sci Technol. 2007; 18(7): 1887.
13. Wang L, Sun DW, Pu H, Zhu Z. Application of hyperspectral imaging to discriminate the variety of maize seeds. Food Anal Methods. 2016; 9: 225-34.
14. Fan Y, Lai K, Rasco BA, Huang Y. Determination of carbaryl pesticide in Fuji apples using surface-enhanced Raman spectroscopy coupled with multivariate analysis. LWT Food Sci Technol. 2015; 60(1): 352-7.
15. Mohd Ali M, Hashim N, Bejo SK, Shamsudin R. Quality evaluation of watermelon using laser-induced backscattering imaging during storage. Postharvest Biol Technol. 2017; 123: 51-9.
16. Mao J, Yu Y, Rao X, Wang J. Firmness prediction and modeling by optimizing acoustic device for watermelons. J Food Eng. 2016; 168: 1-6.
17. Chawgien K, Kiattisin S. Machine learning techniques for classifying the sweetness of watermelon using acoustic signal and image processing. Comput Electron Agric. 2021; 181: 105938.
18. Chen M, Challita U, Saad W, Yin C, Debbah M. Artificial neural networks-based machine learning for wireless networks: A tutorial. IEEE Commun Surv Tutor. 2019; 21(4): 3039-71.
19. Probst P, Wright MN, Boulesteix AL. Hyperparameters and tuning strategies for random forest. WIREs Data Min Knowl Discov. 2019; 9(3): e1301.
20. Rani KV, Jawhar SJ. Lung lesion classification scheme using optimization techniques and hybrid (KNN-SVM) classifier. IETE J Res. 2022; 68(2): 1485-99.
21. Keikhosrokiani P, Naidu AB, Iryanti Fadilah S, Manickam S, Li Z. Heartbeat sound classification using a hybrid adaptive neuro-fuzzy inference system (ANFIS) and artificial bee colony. Digit Health. 2023; 9: 20552076221150741.
22. Xie J, Hu K, Zhu M, Yu J, Zhu Q. Investigation of different CNN-based models for improved bird sound classification. IEEE Access. 2019; 7: 175353-61.
23. Malfante M, Mars JI, Dalla Mura M, Gervaise C. Automatic fish sounds classification. J Acoust Soc Am. 2018;143(5): 2834-46.
24. Salau AO, Jain S. Feature extraction: A survey of the types, techniques, applications. In: 2019 International Conference on Signal Processing and Communication (ICSC). IEEE; 2019 Dec 23. p. 158-64.
25. Di N, Sharif MZ, Hu Z, Xue R, Yu B. Applicability of VGGish embedding in bee colony monitoring: Comparison with MFCC in colony sound classification. PeerJ. 2023; 11: e14696.
26. Sharma G, Umapathy K, Krishnan S. Trends in audio signal feature extraction methods. Appl Acoust. 2020; 158: 107020.
27. Jie D, Zhou W, Wei X. Nondestructive detection of maturity of watermelon by spectral characteristic using NIR diffuse transmittance technique. Scientia Horticulturae, 2019; 257: 108718.
28. Gu Q, Li T, Hu Z, Zhu Y, Shi J, Zhang L, Zhang X. Quantitative analysis of watermelon fruit skin phenotypic traits via image processing and their potential in maturity and quality detection. Computers and Electronics in Agriculture, 2025; 230:109960.
29. Zhang YX, Han JL, Yao W. Non-destructive watermelon maturity detection by acoustic response. In 2010 2nd International Conference on Information Engineering and Computer Science, 2010; 1-4. IEEE.
30. Garvin J, Abushakra F, Choffin Z, Shiver B, Gan Y, Kong L, Jeong N. Microwave imaging for watermelon maturity determination. Current Research in Food Science, 2023; 6: 100412.
31. Vega-Castellote M, Sánchez MT, Torres I, De la Haba MJ, Pérez-Marín D. Assessment of watermelon maturity using portable new generation NIR spectrophotometers. Scientia Horticulturae, 2022; 304: 111328.
32. Pamungkas WA, Bintoro N. Evaluation of watermelon ripeness using self-developed ripening detector. In IOP Conference Series: Earth and Environmental Science, 2021; 653:1, 012020. IOP Publishing.
33. Babu PA, Nagaraju VS, Vallabhuni RR. Speech emotion recognition system with Librosa. In: 2021 10th IEEE International Conference on Communication Systems and Network Technologies (CSNT). IEEE; 2021 Aug 12. p. 421-4.
34. Wu CH, Ho JM, Lee DT. Travel-time prediction with support vector regression. IEEE Trans Intell Transp Syst. 2004; 5(4): 276-81.
35. Pekel E. Estimation of soil moisture using decision tree regression. Theor Appl Climatol. 2020; 139(3-4): 1111-9.
36. Breiman L. Random forests. Mach Learn. 2001; 45: 5-32.
37. Soucy P, Mineau GW. A simple KNN algorithm for text categorization. In: Proceedings of the 2001 IEEE International Conference on Data Mining. IEEE; 2001; p. 647-8.
38. Han HG, Qiao JF. A structure optimisation algorithm for feedforward neural network construction. Neurocomputing. 2013; 99: 347-57.
39. Liao J, Li H, Feng A, Wu X, Luo Y, Duan X, Li J. Domestic pig sound classification based on TransformerCNN. Appl Intell. 2023; 53(5): 4907-23.
40. Kim CI, Cho Y, Jung S, Rew J, Hwang E. Animal sounds classification scheme based on multi-feature network with mixed datasets. KSII Trans Internet Inf Syst. 2020; 14(8): 3384-98.
41. Demir F, Abdullah DA, Sengur A. A new deep CNN model for environmental sound classification. IEEE Access. 2020; 8: 66529-37.
42. Monaco A, Amoroso N, Bellantuono L, Pantaleo E, Tangaro S, Bellotti R. Multi-time-scale features for accurate respiratory sound classification. Appl Sci. 2020; 10(23): 8606.
43. Khan MI, Acharya B, Chaurasiya RK. iHearken: Chewing sound signal analysis-based food intake recognition system using Bi-LSTM Softmax network. Comput Methods Programs Biomed. 2022; 221: 106843.
44. Akbal E, Tuncer T, Dogan S. Vehicle interior sound classification based on local quintet magnitude pattern and iterative neighborhood component analysis. Appl Artif Intell. 2022; 36(1): 2137653.