Pears classification by identifying internal defects based on X-ray images and neural networks

doi:10.1007/s40436-024-00512-1

Advances in Manufacturing ›› 2025, Vol. 13 ›› Issue (3): 552-561.doi: 10.1007/s40436-024-00512-1

• • 上一篇

Pears classification by identifying internal defects based on X-ray images and neural networks

Ning Wang¹, Sai-Kun Yu¹, Zheng-Pan Qi¹, Xiang-Yan Ding¹, Xiao Wu², Ning Hu¹

1. School of Mechanical Engineering, Hebei University of Technology, Tianjin, 300401, People's Republic of China;
2. Department of Mechanical Engineering, The University of Michigan, Dearborn, MI, 48128, USA

收稿日期:2023-10-10 修回日期:2024-01-29 发布日期:2025-09-19
通讯作者: Zheng-Pan Qi,E-mail:zhengpan_qi@hebut.edu.cn E-mail:zhengpan_qi@hebut.edu.cn
作者简介:Ning Wang is a master graduated from School of Mechanical Engineering, Hebei University of Technology, Tianjin, China. His major is mechanical engineering and his research field is intelligent non-destructive testing of fruit based on deep learning algorithms. His contributions include X-ray non-destructive testing of pears, image data analysis and processing, convolutional neural network developing and training, and manuscript writing.
Sai-Kun Yu is a master graduated from School of Mechanical Engineering, Hebei University of Technology, Tianjin, China. His major is mechanical engineering and his research field is nondestructive testing method and its application in fruit and metals. His contributions are X-ray non-destructive testing of pears and manuscript revising.
Zheng-Pan Qi is a lecturer at School of Mechanical Engineering, Hebei University of Technology, Tianjin, China. He works on artificial intelligence for mechanics, non-destructive testing and manufacturing, constitutive relations of solids and structures and their applications including analysis and prediction of failure modes, fatigue behavior of materials and so on. He got his bachelor degree on mathematics and applied mathematics at Beihang University and doctoral degree on engineering mechanics at Chongqing University. He studied and worked at University of Michigan-Dearborn and Ford Motor Company (US) as a joint doctoral student and intern, respectively. His contributions are methodology development, checking and qualification of the investigation, and manuscript revising.
Xiang-Yan Ding is a lecturer at School of Mechanical Engineering, Hebei University of Technology, Tianjin, China. She works on non-destructive testing of materials and structures, such as fruit and metals, especially based on nonlinear ultrasound and X rays. Her contributions are helping to conduct the nondestructive testing and manuscript revising.
Xiao Wu is a master graduated from Department of Mechanical Engineering, The University of Michigan, Dearborn, Michigan 48128, USA. His major is damage, fracture and fatigue behavior of various materials and structures. His contribution is manuscript revising, especially in grammar of and manner of English writing.
Ning Hu is a professor at School of Mechanical Engineering, Hebei University of Technology, Tianjin, China. His research fields include the finite element method and other numerical techniques in solid mechanics and thermal analysis, non-linear dynamics and active control of metallic and composite structures, analysis of strength and stability of metallic and composite structures, structural optimum design, fracture mechanics and micro-mechanics of composites, smart structures in structural health monitoring and active control system, inverse problems, and nanocomposites. His contributions are guidance in theory and feasibility assessment.
基金资助:
The research is fanatically supported by the Youth Program of National Natural Science Foundation of China (Grant No. 12102120), the Youth Program of Natural Science Foundation of Hebei Province (Grant No. A2021202019), the Key Project of University Research Program of Hebei Province (Grant No. ZD2021029), and the Key Program of Science and Technology Project of Tianjin (Grant No. 22JCZDJC00070).

Pears classification by identifying internal defects based on X-ray images and neural networks

Ning Wang¹, Sai-Kun Yu¹, Zheng-Pan Qi¹, Xiang-Yan Ding¹, Xiao Wu², Ning Hu¹

1. School of Mechanical Engineering, Hebei University of Technology, Tianjin, 300401, People's Republic of China;
2. Department of Mechanical Engineering, The University of Michigan, Dearborn, MI, 48128, USA

Received:2023-10-10 Revised:2024-01-29 Published:2025-09-19
Supported by:
The research is fanatically supported by the Youth Program of National Natural Science Foundation of China (Grant No. 12102120), the Youth Program of Natural Science Foundation of Hebei Province (Grant No. A2021202019), the Key Project of University Research Program of Hebei Province (Grant No. ZD2021029), and the Key Program of Science and Technology Project of Tianjin (Grant No. 22JCZDJC00070).

摘要/Abstract

摘要： In order to increase the sales and profitability, it is essential to classify the pears according to the external morphology (including shape, color and luster) and internal defects that can be quantitatively detected by various approaches. However, the existing classification methods concentrate mainly on the external quality rather than the internal defects. Therefore, this investigation develops an efficient and accurate classification method that can identify the internal sclerosis and bruises by combining the X-ray non-destructive testing and the convolutional neural network. Initially, the relations between the characteristics of the internal defects, i.e., internal sclerosis and bruises, and the grayscale features of the X-ray images are analyzed to provide the experimental data and demonstrate the theoretical foundations. Then, the X-ray images are processed by resolution reduction, feature enhancement and gradient reconstruction to improve the training efficiency and classification precision. Finally, the 18-layer residual network (ResNet-18) is optimized and trained to identify the internal bruises and sclerosis and classify the pears based on the identification results. It is found that the overall accuracy can reach 96.67% for identifying the bruised and sclerotic pears. The proposed method could also be applied to other fruits for defects identification and quality classification.

The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-024-00512-1

关键词: Pears classification, Internal defects, X-ray images, Residual network

Abstract: In order to increase the sales and profitability, it is essential to classify the pears according to the external morphology (including shape, color and luster) and internal defects that can be quantitatively detected by various approaches. However, the existing classification methods concentrate mainly on the external quality rather than the internal defects. Therefore, this investigation develops an efficient and accurate classification method that can identify the internal sclerosis and bruises by combining the X-ray non-destructive testing and the convolutional neural network. Initially, the relations between the characteristics of the internal defects, i.e., internal sclerosis and bruises, and the grayscale features of the X-ray images are analyzed to provide the experimental data and demonstrate the theoretical foundations. Then, the X-ray images are processed by resolution reduction, feature enhancement and gradient reconstruction to improve the training efficiency and classification precision. Finally, the 18-layer residual network (ResNet-18) is optimized and trained to identify the internal bruises and sclerosis and classify the pears based on the identification results. It is found that the overall accuracy can reach 96.67% for identifying the bruised and sclerotic pears. The proposed method could also be applied to other fruits for defects identification and quality classification.

The full text can be downloaded at https://link.springer.com/article/10.1007/s40436-024-00512-1

Key words: Pears classification, Internal defects, X-ray images, Residual network

Ning Wang, Sai-Kun Yu, Zheng-Pan Qi, Xiang-Yan Ding, Xiao Wu, Ning Hu. Pears classification by identifying internal defects based on X-ray images and neural networks[J]. Advances in Manufacturing, 2025, 13(3): 552-561.

参考文献

[1] Pholpho T, Pathaveerat S, Sirisomboon P (2011) Classification of longan fruit bruising using visible spectroscopy. J Food Eng 104(1):169-172
[2] Opara UL, Pathare PB (2014) Bruise damage measurement and analysis of fresh horticultural produce-a review. Postharvest Biol Technol 91:9-24
[3] Doerflinger FC, Rickard BJ, Nock JF et al (2015) An economic analysis of harvest timing to manage the physiological storage disorder firm flesh browning in “empire” apples. Postharvest Biol Technol 107:1-8
[4] Zeng X, Miao Y, Ubaid S et al (2020) Detection and classification of bruises of pears based on thermal images. Postharvest Biol Technol 161:111090. https://doi.org/10.1016/j.postharvbio.2019.111090
[5] Baranowski P, Lipecki J, Mazurek W et al (2008) Detection of watercore in “gloster” apples using thermography. Postharvest Biol Technol 47(3):358-366
[6] Huang Y, Lu R, Chen K (2020) Detection of internal defect of apples by a multichannel VIS/NIR spectroscopic system. Postharvest Biol Techno 161:111065. https://doi.org/10.1016/j.postharvbio.2019.111065
[7] Han D, Tu R, Lu C et al (2006) Nondestructive detection of brown core in the chinese pear “yali” by transmission visible-NIR spectroscopy. Food Control 17:604-608
[8] Khatiwada BP, Subedi PP, Hayes C et al (2016) Assessment of internal flesh browning in intact apple using visible-short wave near infrared spectroscopy. Postharvest Biol Technol 120:103-111
[9] Mogollón R, Contreras C, Da Silva Neta ML et al (2020) Non-destructive prediction and detection of internal physiological disorders in “keitt” mango using a hand-held VIS-NIR spectrometer. Postharvest Biol Technol 167:111251. https://doi.org/10.1016/j.postharvbio.2020.111251
[10] Herremans E, Melado-Herreros A, Defraeye T et al (2014) Comparison of X-ray CT and MRI of watercore disorder of different apple cultivars. Postharvest Biol Technol 87:42-50
[11] Van Dael M, Verboven P, Zanella A et al (2019) Combination of shape and X-ray inspection for apple internal quality control: in silico analysis of the methodology based on X-ray computed tomography. Postharvest Biol Technol 148:218-227
[12] Van De Looverbosch T, Raeymaekers E, Verboven P et al (2021) Non-destructive internal disorder detection of conference pears by semantic segmentation of X-ray CT scans using deep learning. Expert Syst Appl 176:114925. https://doi.org/10.1016/j.eswa.2021.114925
[13] Muziri T, Theron KI, Cantre D et al (2016) Microstructure analysis and detection of mealiness in “forelle” pear (Pyrus communis L.) by means of X-ray computed tomography. Postharvest Biol Technol 120:145-156
[14] Magwaza LS, Opara UL (2014) Investigating non-destructive quantification and characterization of pomegranate fruit internal structure using X-ray computed tomography. Postharvest Biol Technol 95:1-6
[15] Shahin MA, Tollner EW, Mcclendon RW (2001) AE-automation and emerging technologies: artificial intelligence classifiers for sorting apples based on watercore. J Agric Eng Res 79:265-274
[16] Matsui T, Kamata T, Koseki S et al (2022) Development of automatic detection model for stem-end rots of “hass” avocado fruit using X-ray imaging and image processing. Postharvest Biol Technol 192:111996. https://doi.org/10.1016/j.postharvbio.2022.111996
[17] Kotwaliwale N, Weckler PR, Brusewitz GH et al (2007) Non-destructive quality determination of pecans using soft X-rays. Postharvest Biol Technol 45:372-380
[18] Haff RP, Slaughter DC, Sarig Y et al (2006) X-ray assessment of translucency in pineapple. J Food Process Preserv 30:527-533
[19] Li J, Rao X, Wang F et al (2013) Automatic detection of common surface defects on oranges using combined lighting transform and image ratio methods. Postharvest Biol Technol 82:59-69
[20] Zhang X, Zhu Y, Su Y et al (2021) Quantitative extraction and analysis of pear fruit spot phenotypes based on image recognition. Comput Electron Agric 190:106474. https://doi.org/10.1016/j.compag.2021.106474
[21] Wang B, Yin J, Liu J et al (2022) Extraction and classification of apple defects under uneven illumination based on machine vision. J Food Process Eng 45:13976. https://doi.org/10.1111/jfpe.13976
[22] Zhang Y, Dong Z, Chen X et al (2019) Image based fruit category classification by 13-layer deep convolutional neural network and data augmentation. Multimed Tools Appl 78:3613-3632
[23] Alhudhaif A, Polat K, Karaman O (2021) Determination of COVID-19 pneumonia based on generalized convolutional neural network model from chest X-ray images. Expert Syst Appl 180:115141. https://doi.org/10.1016/j.eswa.2021.115141
[24] Yasaka K, Akai H, Kunimatsu A et al (2018) Deep learning with convolutional neural network in radiology. Jpn J Radiol 36:257-272
[25] Khan E, Rehman MZU, Ahmed F et al (2022) Chest X-ray classification for the detection of COVID-19 using deep learning techniques. Sensors 22:1211. https://doi.org/10.3390/s22031211
[26] Lu Y, Lu R, Zhang Z (2021) Detection of subsurface bruising in fresh pickling cucumbers using structured-illumination reflectance imaging. Postharvest Biol Technol 180:111624. https://doi.org/10.1016/j.postharvbio.2021.111624
[27] Yu X, Lu H, Wu D (2018) Development of deep learning method for predicting firmness and soluble solid content of postharvest korla fragrant pear using VIS/NIR hyperspectral reflectance imaging. Postharvest Biol Technol 141:39-49
[28] Bobelyn E, Serban A, Nicu M, Lammertyn J, Nicolai BM, Saeys W (2010) Postharvest quality of apple predicted by NIR-spectroscopy: study of the effect of biological variability on spectra and model performance. Postharvest Biol Technol 55:133-143
[29] Nicolaï BM, Beullens K, Bobelyn E et al (2007) Nondestructive measurement of fruit and vegetable quality by means of NIR spectroscopy: a review. Postharvest Biol Technol 46:99-118
[30] Colnago LA, Andrade FD, Souza AA et al (2014) Why is inline NMR rarely used as industrial sensor? Challenges and opportunities. Chem Eng Technol 37:191-203
[31] Yang W, Zhang J, Wang H et al (2011) Peroxisome proliferator-activated receptor γ regulates angiotensin II-induced catalase downregulation in adventitial fibroblasts of rats. FEBS Lett 585:761-766
[32] Nguyen HD, Cai R, Zhao H et al (2022) Towards more efficient security inspection via deep learning: a task-driven X-ray image cropping scheme. Micromachines 13:565. https://doi.org/10.3390/mi13040565
[33] Van De Looverbosch T, Bhuiyan RMH, Verboven P et al (2020) Nondestructive internal quality inspection of pear fruit by X-ray CT using machine learning. Food Control 113:107170. https://doi.org/10.1016/j.foodcont.2020.107170
[34] Vélez Rivera N, Gómez-Sanchis J, Chanona-Pérez J et al (2014) Early detection of mechanical damage in mango using NIR hyperspectral images and machine learning. Biosyst Eng 122:91-98
[35] Wei H, Gu Y (2020) A machine learning method for the detection of brown core in the Chinese pear variety huangguan using a MOS-based e-nose. Sensors 20:4499. https://doi.org/10.3390/s20164499

Pears classification by identifying internal defects based on X-ray images and neural networks

Pears classification by identifying internal defects based on X-ray images and neural networks

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价