Research on Prediction Model of Soluble Solid Content in Plums Based on Near-Infrared Spectroscopy Data

doi:10.11924/j.issn.1000-6850.casb2024-0452

Abstract

Abstract:

To address the issues of complex data preprocessing, low prediction accuracy, and poor generalization capability of existing methods for predicting the soluble solid content (SSC) of plums, a one-dimensional convolutional neural network (1D-CNN) model specifically for plum varieties is proposed. Near-infrared (NIR) spectral data were constructed for 300 samples of Sanhua and Sanyue plums to predict their SSC. And the multi-layer structures such as input layer, one-dimensional convolution layer, pooling layer and full connection layer and output layer were designed to construct 1D-CNN model. Comparative experiments demonstrated that the model achieved a coefficient of determination (R²) of 0.980 and a root mean square error (RMSE) of 0.192, outperforming traditional machine learning methods such as support vector regression and random forest. As a lightweight model, it offers simplicity in the modeling process and strong generalization capability, making it suitable for practical applications.

Key words: near-infrared spectroscopy, soluble solids content, one-dimensional convolutional neural network, plum, deep learning

DI Biao, LIN Juan, LIU Xian. Research on Prediction Model of Soluble Solid Content in Plums Based on Near-Infrared Spectroscopy Data[J]. Chinese Agricultural Science Bulletin, 2024, 40(31): 133-138.

Figures/Tables 8

References 27

[1]	张仲雄, 刘昊灵, 魏子朝, 等. 苹果霉心病不同光谱检测方式对比研究[J]. 光谱学与光谱分析, 2024, 44(3):883-890.
[2]	毛欣然, 夏静静, 徐惟馨, 等. 手持式近红外光谱仪测定梨三种品质指标通用模型建模方法研究[J]. 光谱学与光谱分析, 2024, 44(2):406-412.
[3]	罗东杰, 王勐, 张小栓, 等. 基于Vis/NIR光谱传感的鲜食葡萄糖度检测系统[J]. 光谱学与光谱分析, 2023, 43(7):2146-2152.
[4]	宋丽华, 刘立峰, 陈焕英, 等. 花生籽仁蛋白质含量近红外光谱模型的建立[J]. 中国农学通报, 2011, 27(15):85-89.
[5]	李振, 侯名语, 崔顺立, 等. 花生籽仁黄酮含量近红外分析检测方法[J]. 光谱学与光谱分析, 2024, 44(4):1112-1116.
[6]	丁坤, 项安. 近红外高光谱成像技术结合偏最小二乘-判别分析所建模型快速鉴定核桃仁的品质[J]. 理化检验(化学分册), 2023, 59(7):844-848.
[7]	鹏瑶, 卞希慧, 王姿方, 等. 光谱预处理方法选择研究[J]. 光谱学与光谱分析, 2019, 39(9):2800-2806.
[8]	METLENKIN D A, PLATOV Y T, PLATOVA R A, et al. Non-destructive identification of defects and classification of Hass avocado fruits with the use of a hyperspectral image[J]. Agronomy research, 2022, 20(2):326-340.
[9]	原帅, 张娟, 刘美娟, 等. 基于主成分回归分析和偏最小二乘回归法的樱桃糖度检测的研究[J]. 分析科学学报, 2016, 32(4):553-556.
[10]	宋丽芳, 廖桂平, 陈敏, 等. 基于机器学习的油菜叶片水分含量高光谱估测[J]. 中国农业科技导报, 2024, 26(5):110-119.
[11]	CEVOLI C, IACCHERI E, FABBRI A, et al. Data fusion of FT-NIR spectroscopy and Vis/NIR hyperspectral imaging to predict quality parameters of yellow flesh "Jintao" kiwifruit[J]. Biosystems engineering, 2024,237:157-169.
[12]	郭阳, 郭俊先, 史勇, 等. CARS-SVM预测哈密瓜可溶性固形物含量[J]. 食品与机械, 2021, 37(6):81-85.
[13]	杨宝华, 高志伟, 齐麟, 等. 高光谱影像的鲜桃可溶性固形物含量预测模型[J]. 光谱学与光谱分析, 2021, 41(11):3559-3564.
[14]	吴爽, 李国建, 介邓飞. 基于深度学习的西瓜可见/近红外光谱可溶性固形物预测模型研究[J]. 食品与机械, 2020, 36(12):132-135.
[15]	孙海霞, 王润润, 张淑娟, 等. 基于可见/近红外光谱技术的玉露香梨可溶性固形物含量检测[J]. 农产品加工, 2023(14):46-48.
[16]	宋白玉, 张瑞鑫, 庄程翔, 等. 基于近红外光谱技术的草莓可溶性固形物与糖度的检测分析[J]. 科技与创新, 2023(7):9-12.
[17]	崔天宇, 卢中领, 薛琳, 等. 基于近红外反射光谱的番茄糖分快速检测模型研究[J]. 光谱学与光谱分析, 2023, 43(4):1218-1224.
[18]	ZHENG Y J, CAO Y C, YANG J, et al. Enhancing model robustness through different optimization methods and 1-D CNN to eliminate the variations in size and detection position for apple SSC determination[J]. Postharvest biology and technology, 2023, 205(1).
[19]	GUO Z M, ZHAI L X, ZOU Y, et al. Comparative study of Vis/NIR reflectance and transmittance method for on-line detection of strawberry SSC[J]. Computers and electronics in agriculture, 2024,218.
[20]	尚静, 孟庆龙, 张艳, 等. 紫外/可见光谱技术无损检测李子可溶性固形物含量[J]. 食品工业科技, 2020, 41(3):228-231.
[21]	SHEN L X, WANG H H, LIU Y, et al. Prediction of soluble solids content in green plum by using a sparse autoencoder[J]. Applied sciences, 2020, 10(11):3769.
[22]	ZHANG X, ZHOU C X, SUN Q, et al. Prediction of solid soluble content of green plum based on improved CatBoost[J]. Agriculture-basel, 2023, 13(6):1122-1122.
[23]	REDDY V K, SUHASINI A, BALARAM V V S S S. xACO_1D-CNN: Ant colony optimization based 1D-Convolution Neural Network based fruit grading and classification[J]. International journal of intelligent systems and applications in engineering, 2023, 11(1S):130-144.
[24]	PANDISELVAM R, PRITHVIRAJ V, MANIKANTAN M R, et al. Recent advancements in NIR spectroscopy for assessing the quality and safety of horticultural products: A comprehensive review[J]. Frontiers in nutrition, 2022,9:973457.
[25]	唐杰, 罗彦波, 李翔宇, 等. 基于近红外光谱数据的一维卷积神经网络模型研究[J]. 光谱学与光谱分析, 2024, 44(3):731-736.
[26]	李盛芳, 贾敏智, 董大明. 随机森林算法的水果糖分近红外光谱测量[J]. 光谱学与光谱分析, 2018, 38(6):1766-1771.
[27]	王菲菲, 刘彭, 孙凤伟, 等. 基于支持向量机和压力传感器的水果分类系统[J]. 食品与机械, 2023, 39(9):83-88.

评价指标	三华李		三月李
评价指标	训练集	测试集	训练集	测试集
R²	0.832	0.980	0.978	0.972
MSE	0.203	0.037	0.054	0.072
RMSE	0.451	0.192	0.233	0.269
MAE	0.361	0.128	0.174	0.158
MRE	2.978	1.061	2.203	2.146
CORR	0.917	0.990	0.988	0.986

评价指标	三华李		三月李
评价指标	训练集	测试集	训练集	测试集
R²	0.832	0.980	0.978	0.972
MSE	0.203	0.037	0.054	0.072
RMSE	0.451	0.192	0.233	0.269
MAE	0.361	0.128	0.174	0.158
MRE	2.978	1.061	2.203	2.146
CORR	0.917	0.990	0.988	0.986

评价指标	SVR		RF		1D-CNN
评价指标	训练集	测试集	训练集	测试集	训练集	测试集
R²	0.785	0.592	0.948	0.820	0.832	0.980
MSE	0.258	0.792	0.062	0.348	0.203	0.037
RMSE	0.508	0.890	0.249	0.590	0.451	0.192
MAE	0.288	0.507	0.125	0.344	0.361	0.128
MRE	2.441	3.827	1.02	2.667	2.978	1.061
CORR	0.891	0.786	0.979	0.933	0.917	0.990

评价指标	SVR		RF		1D-CNN
评价指标	训练集	测试集	训练集	测试集	训练集	测试集
R²	0.785	0.592	0.948	0.820	0.832	0.980
MSE	0.258	0.792	0.062	0.348	0.203	0.037
RMSE	0.508	0.890	0.249	0.590	0.451	0.192
MAE	0.288	0.507	0.125	0.344	0.361	0.128
MRE	2.441	3.827	1.02	2.667	2.978	1.061
CORR	0.891	0.786	0.979	0.933	0.917	0.990

评价指标	SVR		RF		CNN
评价指标	训练集	测试集	训练集	测试集	训练集	测试集
R²	0.716	0.763	0.963	0.882	0.978	0.972
MSE	0.695	0.620	0.090	0.308	0.054	0.072
RMSE	0.834	0.787	0.301	0.555	0.233	0.269
MAE	0.555	0.566	0.185	0.325	0.174	0.158
MRE	8.015	8.482	2.608	4.566	2.203	2.146
CORR	0.846	0.878	0.984	0.940	0.988	0.986