Nondestructive discrimination of moldy pear core based on the recurrence plots of vibration acoustic signals and deep convolutional neural networks
DOI:
https://doi.org/10.25165/ijabe.v18i6.9916Keywords:
moldy pear core, deep convolutional neural network, vibration acoustic signals, recurrence plots, Markov transition field, nondestructive detectionAbstract
Moldy core is a serious internal defect in pears. Since there is no significant difference in appearance between the healthy pears and those with mild moldy core, it is still a great challenge for the early detection of moldy pear core. This study transformed the vibration acoustic signals (VA signal) of pears into recurrence plots and Markov transition field to enable image-based classification of moldy cores. In addition to traditional machine-learning baselines (Random Forest and k-Nearest Neighbors) trained on LBP-extracted texture features from RP/MTF, the deep models were constructed and compared, which include ResNet-101, DenseNet-121, SqueezeNet, Vision Transformer (ViT), and an improved SqueezeNet (ISqueezeNet). Hyperparameters were tuned via Bayesian optimization over optimizer type, learning rate, batch size, and L2 weight decay, yielding model-specific optimal settings. Under these configurations, the ISqueezeNet achieved the highest test accuracy of 93.05%, with class-wise accuracies of 89.28% (healthy), 96.15% (slight), and 94.44% (moderate and severe). Comparisons with lightweight networks (MobileNetV1 and ShuffleNetV2) further showed that ISqueezeNet attains superior accuracy with favorable parameter efficiency and inference speed. Grad-CAM visualizations confirmed that the model focuses on lesion-relevant regions, supporting interpretability and practical reliability. These results indicate that the proposed approach is promising for early, nondestructive detection of moldy pear cores. Key words: moldy pear core; deep convolutional neural network; vibration acoustic signals; recurrence plots; Markov transition field; nondestructive detection DOI: 10.25165/j.ijabe.20251806.9916 Citation: Yang Y, Zhao K, Zhao J, Song Y, Shen T. Nondestructive discrimination of moldy pear core based on the recurrence plots of vibration acoustic signals and deep convolutional neural networks. Int J Agric & Biol Eng, 2025; 18(6): 230–240.References
Pan T, Chyngyz E, Sun D W, Paliwal J, Pu H. Pathogenetic process monitoring and early detection of pear black spot disease caused by Alternaria alternata using hyperspectral imaging. Postharvest Biology and Technology, 2019; 154: 96–104.
Zhang Q, Huang W, Wang Q, Wu J, Li J. Detection of pears with moldy core using online full-transmittance spectroscopy combined with supervised classifier comparison and variable optimization. Computers and Electronics in Agriculture, 2022; 200: 107231.
Zhang Z, Liu H, Chen D, Zhang J, Li H, Shen M. SMOTE-based method for balanced spectral nondestructive detection of moldy apple core. Food Control, 2022; 141: 109100.
Liu H L, Wei Z Y, Lu M, Gao P, Li J K, Zhao J, Hu J. A Vis/NIR device for detecting moldy apple cores using spectral shape features. Computers and Electronics in Agriculture, 2024; 220: 108898.
Li Y P, Zhang X N, Nie J Y, Bacha S A S, Yan Z, Gao G W. Occurrence and co-occurrence of mycotoxins in apple and apple products from China. Food Control, 2020; 118: 107354.
Liu Z, Chen N, Le D X Lai Q R, Li B, Wu J, et al. Acoustic vibration multi-domain images vision transformer (AVMDI-ViT) to the detection of moldy apple core: Using a novel device based on micro-LDV and resonance speaker. Postharvest Biology and Technology, 2024; 211: 112838.
Tian S, Wang J, Xu H. Firmness measurement of kiwifruit using a self-designed device based on acoustic vibration technology. Postharvest Biology and Technology, 2022; 187: 111851.
Tempelaere A, Phan H M, Van De, Looverbosch T, Verboven P, Nicolai B. Non-destructive internal disorder segmentation in pear fruit by X-ray radiography and AI. Computers and Electronics in Agriculture, 2023; 212: 108142. doi: 10.1016/j.compag. 2023.108142
Van D, Looverbosch T, Raeymaekers E, Verboven P, Sijbers J, Nicolai B. Non-destructive internal disorder detection of Conference pears by semantic segmentation of X-ray CT scans using deep learning. Expert Systems with Applications, 2021; 176: 114925.
Velásquez C, Prieto F, Palou L, Cubero S, Blasco J, Aleixos N. New model for the automatic detection of anthracnose in mango fruits based on Vis/NIR hyperspectral imaging and discriminant analysis. Journal of Food Measurement and Characterization, 2024; 18(1): 560–570.
Mogollón M, R, Contreras C, De Freitas, S T, Zoffoli J P. NIR spectral models for early detection of bitter pit in asymptomatic ‘Fuji’ apples. Scientia Horticulturae, 2021; 280: 109945.
Ghooshkhaneh N G, Golzarian M R, Mollazade K. VIS-NIR spectroscopy for detection of citrus core rot caused by Alternaria alternata. Food Control, 2023; 144: 109320.
Tang Y, Yang J P, Zhuang J J, Hou C J, Miao A M, Ren J C. Early detection of citrus anthracnose caused by Colletotrichum gloeosporioides using hyperspectral imaging. Computers and Electronics in Agriculture, 2023; 214: 108348.
Chun S W, Song D J, Lee K H, Kim M J, Kim M S, Kim K S. Deep learning algorithm development for early detection of Botrytis cinerea infected strawberry fruit using hyperspectral fluorescence imaging. Postharvest Biology and Technology, 2024; 214: 112918.
Qin Y, Jia W, Sun X, Lv H. Development of electronic nose for detection of micro-mechanical damages in strawberries. Frontiers in Nutrition, 2023; 10: 1222988.
Zhang J C, Zhang P, Xue Y L, Jia X Y, Li J K. Characterization of characteristic odor of rotten core apples based on electronic nose and establishment of non-destructive detection model. Food and Fermentation Industries, 2022; 48(2): 267–273.
Zhang H, Zha Z H, Kulasiri D, Wu J. Detection of early core browning in pears based on statistical features in vibro-acoustic signals. Food and Bioprocess Technology, 2021; 14: 887–897.
Han Q L, Long B X, Yan X J, Wang W, Liu F, Chen X L. Exploration of using acoustic vibration technology to non-destructively detect moldy kernels of in-shell hickory nuts (Carya cathayensis Sarg. ). Computers and Electronics in Agriculture, 2023; 212: 108137.
Mohammed A, Kora R. A comprehensive review on ensemble deep learning: Opportunities and challenges. Journal of King Saud University-Computer and Information Sciences, 2023; 35(2): 757–774.
Al-Fraihat D, Sharrab Y, Alzyoud F, Qahmash A, Tarawneh M, Maaita A. Speech recognition utilizing deep learning: A systematic review of the latest developments. Human-centric Computing and Information Sciences, 2024; 14. doi: 10.22967/HCIS.2024.14.015
Jiang H, Diao Z, Shi T, Zhou Y, Wang F, Hu W. A review of deep learning-based multiple-lesion recognition from medical images: classification, detection and segmentation. Computers in Biology and Medicine, 2023; 157: 106726.
Zhu Z, Lei Y, Qi G, Chai Y, Mazur N, An Y, Huang X. A review of the application of deep learning in intelligent fault diagnosis of rotating machinery. Measurement, 2023; 206: 112346.
Cai Z, Sun C, Zhang H, Zhang Y, Li J. Developing universal classification models for the detection of early decayed citrus by structured-illumination reflectance imaging coupling with deep learning methods. Postharvest Biology and Technology. 2024; 210: 112788. doi: 10.1016/j.postharvbio.2024.112788
Ünal Z, Kızıldeniz T, Özden M, Aktaş H, Karagöz Ö. Detection of bruises on red apples using deep learning models. Scientia Horticulturae, 2024; 329: 113021.
Shin J, Chang Y K, Heung B, Nguyen-Quang T, Price G W, Al-Mallahi A. A deep learning approach for RGB image-based powdery mildew disease detection on strawberry leaves. Computers and electronics in agriculture, 2021; 183: 106042.
Seo D, Nam H. Deep rp-cnn for burst signal detection in cognitive radios. IEEE Access, 2020; 8: 167164–167171.
Zhou Y, Wang Z, Zuo X, Zhao H. Identification of wear mechanisms of main bearings of marine diesel engine using recurrence plot based on CNN model. Wear, 2023; 520: 204656.
Petrauskiene V, Pal M, Cao M, Wang J, Ragulskis M. Color recurrence plots for bearing fault diagnosis. Sensors, 2022; 22(22): 8870.
Li H, Zhao K, Zha Z H, Zhang H, Wu J. Early nondestructive detection of blackheart disease in pear based on empirical mode decomposition of acoustic vibration signals. Food Science, 2023; 44(20): 357–371.
He K M, Zhang X Y, Ren S Q, Sun J. Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016; pp.770–778. doi: 10.48550/arXiv.1512.03385
Huang G, Liu Z, Van Der, Maaten L, Weinberger K Q. Densely connected convolutional networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2017; pp.4700–4708. doi: 10.1109/CVPR.2017.243
Wang N, Yu S K, Qi Z P, Ding X Y, Wu X, Hu N. Pears classification by identifying internal defects based on X-ray images and neural networks. Advances in Manufacturing, 2025; 13: 552–561.
Seo H-J, Song J-H. Detection of internal browning disorder in ‘Greensis’ pears using a portable non-destructive instrument. Horticulturae, 2023; 9(8): 944.
Downloads
Published
How to Cite
Issue
Section
License
IJABE is an international peer reviewed open access journal, adopting Creative Commons Copyright Notices as follows.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
