Detection of Different Levels of Damage in Gong Pears Based on
Reflectance/Absorbance/Kubelka-Munk Spectroscopy
LI Bin1, 2, LU Ying-jun2, SU Cheng-tao2, LIU Yan-de1, 2
1. School of Intelligent Electromechanical Equipment Innovation Research Institute, East China Jiaotong University, Nanchang 330013, China
2. School of Mechatronics and Vehicle Engineering, East China Jiaotong University, Nanchang 330013, China
Abstract：Gong pear is prone to mechanical damage during harvesting, transportation, and sales, accelerating fruit decay and reducing its quality. Different treatment measures were taken to quickly distinguish the different degrees of damage to the gong pear and reduce economic loss. In the past, hyperspectral technology was used to study the damage degree of fruits. Usually, only the reflectance spectrum was used for thestudy. In this study, the reflectance (R), absorbance (A), and Kubelka-Munk (K-M) spectra of Gong pears were obtained by hyperspectral technology and combined with three deep learning algorithms to distinguish healthy and different damage degrees of Gong pears. Firstly, 60 fresh and undamaged Gong pears were selected as healthy samples, and 60 samples of Ⅰ, Ⅱ, and Ⅲ damaged Gong pears were prepared by free fall collision device. The spectral data of these 240 Gong pear samples were collected by hyperspectral imaging system, and the acquired spectra were corrected in black and white to obtain the original spectra of reflectance (R), absorbance (A), and Kubelka-Munk (K-M) of Gong pear. Then, three kinds of original spectral data were preprocessed by Baseline calibration, De-Trending, moving average (MA-S), multiple-scattering correction (MSC), convolution smoothing (SG-S), and standard normal variable transformation (SNV). BP neural network (BP), Limit gradient lift (XGBoost), and random forest (RF) discriminant analysis models were established to distinguish different damage degrees of Gongli. According to the discrimination results of the model on the damage degree of Gong pear, the accuracy of the BP model based on reflectance, absorbance, and K-M spectrum is better, with the overall accuracy reaching 85% or more. Itwas found that the BP model established by the baseline reflectance spectrum after pretreatment showed a greater improvement than that established by the unpretreated reflectance spectrum. The accuracy of discrimination reached 93.33%. To improve the accuracy and operation efficiency of the BP model, competitive adaptive reweighting (CARS) and no-information variable elimination (UVE) methods were used to screen out the spectral information of characteristic bands for the 3 kinds of original spectra and Baseline pre-treated spectra, and the BP model was established with the filtered characteristic spectral data. The discrimination results show that the A-RAW-CARS-BP model has the best discrimination accuracy, and the overall accuracy reaches 96.66%. The results show that it is feasible to discriminate the damage degree of Gong pear by using three kinds of original spectra, which provides a theoretical basis for detecting different damage degrees of Gong pear by hyperspectral technology.
李 斌，卢英俊，苏成涛，刘燕德. 基于反射率、吸光度和Kubelka-Munk光谱的贡梨不同损伤程度检测[J]. 光谱学与光谱分析, 2024, 44(11): 3101-3108.
LI Bin, LU Ying-jun, SU Cheng-tao, LIU Yan-de. Detection of Different Levels of Damage in Gong Pears Based on
Reflectance/Absorbance/Kubelka-Munk Spectroscopy. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3101-3108.
[1] Liao S, Wu Y, Hu X, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy（分子和生物分子光谱学）, 2023, 296: 122668.
[2] Yang X, Chen J, Jia L, et al. Sensors, 2020, 20（7）: 1866.
[3] Santana T C, Silva R A B, Pandorfi H, et al. Brazilian Journal of Food Technology, 2021, 24: e2020198.
[4] Hou L Z, Sun X Y, Pan L, et al. Food Science and Technology, Campinas, 2021, 41(4): 1062.
[5] DUAN Jie-li, ZHANG Han-yao, FU Han，et al（段洁利, 张汉尧, 付 函，等）. Transactions of the Chinese Society of Agricultural Engineering（农业工程学报）, 2023, 39（7）: 176.
[6] Min D, Zhao J, Bodner G, et al. Food Control, 2023, 152(1-3): 109830.
[7] Kuswidiyanto L W, Kim D E, Fu T, et al. Agriculture, 2023, 13(12): 2215.
[8] Zhang B, Ou Y, Yu S, et al. Plant Methods, 2023, 19(1): 148.
[9] Zhang L, Guan Y, Wang N, et al. Scientific Reports, 2023, 13(1): 14286.
[10] HE Fu-xian, MENG Qing-hua, TANG Liu, et al（何馥娴, 蒙庆华, 唐 柳，等）. Journal of Fruit Science（果树学报）, 2021, 38(9): 1590.
[11] Hou J M, He Z H, Liu D Y, et al. Food Science and Technology, Campinas, 2023, 43: e001823.
[12] Raj R, Cosgun A, Kulic D. Agronomy, 2022, 12(2): 425.
[13] Jiang Q Y, Wu G S, Tian C F, et al. Infrared Physics and Technology, 2021, 118: 103898.
[14] Davur Y J, Kämper W, Khoshelham K, et al. Horticulturae, 2023, 9（5）: 599.
[15] LI Jia-zong, GU Yan, LIU Cui-ling, et al（李佳琮, 谷 晏, 刘翠玲，等）. Journal of Food Safety and Quality（食品安全质量检测学报）, 2023, 14(5): 60.
[16] YIN Hai, LI Bin, LIU Yan-de, et al（殷 海, 李 斌, 刘燕德，等）. Journal of Huazhong Agricultural University（华中农业大学学报）, 2023, 42(3): 220.
