A novel approach of quality assessment of potatoes using artificial neural network

Abstract

A novel approach for quality assessment of potatoes using an artificial neural network (ANN) technique is developed and integrated with thermal imaging system to analyze the decay of tissues in potatoes. The merger of ANN with acquired thermal images leads to increased system efficiency, accuracy, reliability, and speed. A sample of 480 thermal images of potatoes was generated for training and testing the dataset. Categorization of potatoes as healthy or defective is performed based on seven critical parameters, such as standard deviation, energy, entropy, skewness, kurtosis, means, and variance. A classifier based on ANN is employed, incorporating two hidden layers with 200 neurons in the first layer and 50 neurons in the second layer. The input layer size is feature-dependent, while the output layer employs two neurons for binary classification. Among the seven features evaluated individually, entropy and energy achieved the best results, with classification accuracies of 79.81% and 81.27%, respectively. The highest overall performance, 94.89% accuracy with 455 correctly classified samples, was obtained by combining all features except variance. It is observed that integrating thermal imaging with ANNs is a promising approach for quality assessment. Selecting appropriate features is crucial for optimizing the classification model.

Keywords

Quality assessment of agricultural products / Artificial neural network / Smart assessment technique / Thermal imaging / Feature extraction / Image processing

Acknowledgments

None.

Funding

None.

Conflict of interest

The authors declare that they have no competing interests.

Author contributions

Conceptualization: Zeeshan Ali Haq, Zainul Abdin Jaffery

Formal analysis: Zainul Abdin Jaffery, Shabana Mehfuz

Investigation: All authors

Methodology: Zeeshan Ali Haq, Zainul Abdin Jaffery

Writing-original draft: Zeeshan Ali Haq

Writing-review & editing: Zainul Abdin Jaffery, Shabana Mehfuz

References

[1] Kumar RA, Rajpurohit VS, Bidari KY. Multi class grading and quality assessment of pomegranate fruits based on physical and visual parameters. Int J Fruit Sci. 2019; 19(4):372-396. doi: 10.1080/15538362.2018.1552230

[2] Wani JA, Sharma S, Muzamil M, Ahmed S, Sharma S, Singh S. Machine learning and deep learning based computational techniques in automatic agricultural diseases detection: Methodologies, applications, and challenges. Arch Comput Methods Eng. 2022; 29(1):641-677. doi: 10.1007/s11831-021-09588-5

[3] Steinbrener J, Posch K, Leitner R. Hyperspectral fruit and vegetable classification using convolutional neural networks. Comput Electron Agric. 2019; 162:364-372. doi: 10.1016/j.compag.2019.04.019

[4] Bird JJ, Barnes CM, Manso LJ, Ekárt A, Faria DR. Fruit quality and defect image classification with conditional GAN data augmentation. Sci Hortic. 2022;293:110684. doi: 10.1016/j.scienta.2021.110684

[5] Pathmanaban P, Gnanavel BK, Anandan SS, SathiyamurthyS. Advancing post-harvest fruit handling through AI-based thermal imaging: Applications, challenges, and future trends. Discov Food. 2023; 3(1):27. doi: 10.1007/s44187-023-00068-2

[6] Qingwang W, Jin P, Wu Y, Zhou L, Shen T. Infrared image enhancement: A review. IEEE J Sel Top Appl Earth Obs Remote Sens. 2024; 18:1-22. doi: 10.1109/JSTARS.2024.3523418

[7] Kingston RH, editor. Blackbody radiation, image plane intensity, and units. In: Optics and Photonics, Optical Sources, Detectors, and Systems. Ch. 1. Amsterdam:Elsevier; 1995. p. 1-32.

[8] Jaffery ZA. Thermal image based monitoring of PV modules and solar inverters. In: Haque A, Blaabjerg F, Wang H, Yang Y, Jaffery ZA. Reliability of Power Electronics Converters for Solar Photovoltaic Applications. England: The Institution if Engineers and Technology; 2021. doi: 10.1049/pbpo170e_ch7

[9] Singh L, Alam A, Kumar KV, Kumar D, Kumar P, Jaffery ZA. Design of thermal imaging-based health condition monitoring and early fault detection technique for porcelain insulators using Machine learning. Environ Technol Innov. 2021;24:102000. doi: 10.1016/j.eti.2021.102000

[10] Sabah AN, Jaffery ZA. Fault Detection of Induction Motor using Thermal Imaging. In: 2022 IEEE IAS Global Conference on Emerging Technologies (GlobConET). IEEE; 2022. p. 84-90. doi: 10.1109/globconet53749.2022.9872516

[11] Oppenheim D, Shani G, Erlich O, Tsror L. Using deep learning for image-based potato tuber disease detection. Phytopathology. 2019; 109(6):1083-1087. doi: 10.1094/phyto-08-18-0288-r

[12] Afzaal H, Farooque AA, Schumann AW, et al. Detection of a potato disease (early blight) using artificial intelligence. Remote Sens. 2021; 13(3):411. doi: 10.3390/rs13030411

[13] Afonso MV, Blok PM, Polder G, Van der Wolf JM, Kamp J. Blackleg detection in potato plants using convolutional neural networks. IFAC PapersOnline. 2019; 52(30):6-11. doi: 10.1016/j.ifacol.2019.12.48

[14] Osdaghi E, Van der Wolf JM, Abachi H, Li X, De Boer SH, Ishimaru CA. Bacterial ring rot of potato caused by Clavibacter sepedonicus: A successful example of defeating the enemy under international regulations. Mol Plant Pathol. 2022; 23(7):911-932. doi: 10.1111/mpp.13191

[15] Madhiarasan M, Louzazni M. Analysis of artificial neural network: Architecture, types, and forecasting applications. J Electr Comput Eng. 2022; 2022(1): 5416722. doi: 10.1155/2022/5416722

[16] Pholpho T, Pathaveerat S, Sirisomboon P. Classification of longan fruit bruising using visible spectroscopy. J Food Eng. 2011; 104(1):169-172. doi: 10.1016/j.jfoodeng.2010.12.011

[17] Wu G, Wang C. Investigating the effects of simulated transport vibration on tomato tissue damage based on vis/ NIR spectroscopy. Postharvest Biol Technol. 2014; 98:41-47. doi: 10.1016/j.postharvbio.2014.06.016

[18] Jiménez-Jiménez F, Castro-García S, Blanco-Roldán GL, Agüera-Vega J, Gil-Ribes JS. Non-destructive determination of impact bruising on table olives using Vis-NIR spectroscopy. Biosyst Eng. 2012; 113(4):371-378. doi: 10.1016/j.biosystemseng.2012.09.007

[19] Luo X, Takahashi T, Kyo K, Zhang S. Wavelength selection in vis/NIR spectra for detection of bruises on apples by ROC analysis. J Food Eng. 2012; 109(3):457-466. doi: 10.1016/j.jfoodeng.2011.10.035

[20] Zhang W, Lv Z, Xiong S. Nondestructive quality evaluation of agro-products using acoustic vibration methods-a review. Crit Rev Food Sci Nutr. 2018; 58(14):2386-2397. doi: 10.1080/10408398.2017.1324830

[21] Baranowski P, Mazurek W, Pastuszka-Woźniak J. Supervised classification of bruised apples with respect to the time after bruising on the basis of hyperspectral imaging data. Postharvest Biol Technol. 2013; 86:249-258. doi: 10.1016/j.postharvbio.2013.07.005

[22] Baranowski P, Mazurek W, Wozniak J, Majewska U. Detection of early bruises in apples using hyperspectral data and thermal imaging. J Food Eng. 2012; 110(3):345-355. doi: 10.1016/j.jfoodeng.2011.12.038

[23] Huang W, Zhang B, Li J, Zhang C. Early Detection of Bruises on Apples Using Near-Infrared Hyperspectral Image. In: Piageng 2013: Image Processing and Photonics for Agricultural Engineering. vol. 8761. SPIE; 2013. p. 133-139. doi: 10.1016/j.postharvbio.2024.113282

[24] Yin S, Bi X, Niu Y, Gu X, Xiao Y. Hyperspectral classification for identifying decayed oranges infected by fungi. Emirates J Food Agric. 2017; 29(8):601-609. doi: 10.9755/ejfa.2017-05-1074

[25] Li J, Zhang R, Li J, et al. Detection of early decayed oranges based on multispectral principal component image combining both bi-dimensional empirical mode decomposition and watershed segmentation method. Postharvest Biol Technol. 2019;158:110986. doi: 10.1016/j.postharvbio.2019.110986

[26] Lee WH, Kim MS, Lee H, et al. Hyperspectral near-infrared imaging for the detection of physical damages of pear. J Food Eng. 2014; 130:1-7. doi: 10.1016/j.jfoodeng.2013.12.032

[27] Prashar A, Jones HG. Infra-red thermography as a high-throughput tool for field phenotyping. Agronomy. 2014; 4(3):397-417. doi: 10.3390/agronomy4030397

[28] Kuzy J, Jiang Y, Li C. Blueberry bruise detection by pulsed thermographic imaging. Postharvest Biol Technol. 2018; 136:166-177. doi: 10.1016/j.postharvbio.2017.10.011

[29] Doosti-Irani O, Golzarian MR, Aghkhani MH, Sadrnia H, Doosti-Irani M. Development of multiple regression model to estimate the apple’s bruise depth using thermal maps. Postharvest Biol Technol. 2016; 116:75-79. doi: 10.1016/j.postharvbio.2015.12.024

[30] Kim G, Kim GH, Park J, Kim DY, Cho BK. Application of infrared lock-in thermography for the quantitative evaluation of bruises on pears. Infrared Physics Technol. 2014; 63:133-139. doi: 10.1016/j.infrared.2013.12.015

[31] Varith J, Hyde GM, Baritelle AL, Fellman JK, Sattabongkot T. Non-contact bruise detection in apples by thermal imaging. Innov Food Sci Emerg Technol. 2003; 4(2):211-218. doi: 10.1016/S1466-8564(03)00021-3

[32] Kavdır I, Guyer DE. Comparison of artificial neural networks and statistical classifiers in apple sorting using textural features. Biosyst Eng. 2004; 89(3):331-344. doi: 10.1016/j.biosystemseng.2004.08.008

[33] Ding L, Dong D, Jiao L, Zheng W. Potential using of infrared thermal imaging to detect volatile compounds released from decayed grapes. PLoS One. 2017; 12(6):e0180649. doi: 10.1371/journal.pone.0180649

[34] Zeng X, Miao Y, Ubaid S, Gao X, Zhuang S. Detection and classification of bruises of pears based on thermal images. Postharvest Biol Technol. 2020;161:111090. doi: 10.1016/j.postharvbio.2019.111090

[35] Haq ZA, Jaffery ZA.Impact of Activation Functions and Number of Layers on the Classification of Fruits using CNN. In:2021 8^th International Conference on Computing for Sustainable Global Development (INDIACom); IEEE; 2021. p. 227-231.