Abstract
[Objective] To achieve rapid, non -destructive, and high -precision detection of six nutritional components in tomatoes, including dry matter, soluble sugar, total acid, vitamin C, soluble solids, and lycopene. [Methods] A tomato nutritional component detection model based on near -infrared spectroscopy (NIRS ) combined with light gradient boosting machine (LightGBM ) optimized using the particle swarm optimization (PSO ) algorithm (PSO -LightGBM ) was proposed. Using the six nutritional components as detection indicators, near -infrared spectral data (scanning range:400~2 500 nm) were collected from 471 tomato samples. Seven preprocessing methods were compared, and the optimal combination of standard normal variate (SNV ) transformation and first derivative was selected to eliminate noise and baseline drift. The PSO algorithm was used to optimize the key hyperparameters of LightGBM (number of leaves, learning rate, tree depth, feature sampling ratio, and Bagging ratio ) to construct a synchronous multi -component prediction model. Meanwhile, partial least squares regression (PLSR ), unoptimized LightGBM, and grid search -optimized LightGBM (GS-LightGBM ) models were established as comparison models. [Results] The PSO -LightGBM model achieved the highest prediction accuracy for all six nutritional components. The coefficients of determination for the prediction set (R2p) for dry matter and soluble solids reached 0.987 and 0.973, respectively, with residual prediction |deviation (ERPDp) values of 6.732 and 5.115, respectively. The R2p values for total acid, soluble sugar, and vitamin C all exceeded 0.95, and their ERPDp values were all greater than 2.5. The R2p for lycopene increased to 0.956, representing a 12.4% increase compared with that of the unoptimized LightGBM model. The prediction -set root mean square error (ERMSEP ) of PSO -LightGBM was, on average, 34.7% lower than that of PLSR and 15.2% lower than that of GS -LightGBM. [Conclusion] This method addresses the insufficient generalization ability caused by the random setting of LightGBM parameters and enables synchronous and accurate detection of multiple nutritional components in tomatoes.
Publication Date
10-21-2026
First Page
74
Last Page
82
DOI
10.13652/j.spjx.1003.5788.2026.60065
Recommended Citation
Lichun, Shi; Songwei, Wang; and Zhizhong, Wang
(2026)
"Rapid detection of tomato nutritional components based on NIRS and PSO-LightGBM,"
Food and Machinery: Vol. 42:
Iss.
9, Article 9.
DOI: 10.13652/j.spjx.1003.5788.2026.60065
Available at:
https://www.ifoodmm.cn/journal/vol42/iss9/9
References
[1] 魏子朝,卢苗,雷文晔,等.一种融合叶绿素荧光技术与可见—近红外光谱的番茄幼苗热胁迫无损检测方法 [J].光谱学与光谱分析,2024,44(6):1 613-1 619.Wei Z C,Lu M,Lei W Y,et al.A nondestructive method combined chlorophyll fluorescence with visible-NIR spectroscopy for detecting the severity of heat stress on tomato seedlings [J].Spectroscopy and Spectral Analysis,2024,44(6):1 613-1 619.
[2] 吴至境,刘富强,李志刚,等.基于 DCGAN 数据增强的樱桃番茄可溶性固形物含量光谱检测方法 [J].食品科学,2025,46(2):214-221.Wu Z J,Liu F Q,Li Z G,et al.Spectroscopic method for detection of soluble solid content in cherry tomato using deep convolutional generative adversarial network-based data augmentation [J].Food Science,2025,46(2):214-221.
[3] 黄彦发,张瑶,朱彦哲,等.基于深度学习的樱桃番茄 γ-氨基丁酸含量无损检测研究 [J].分析测试学报,2025,44(12):2 554-2 563.Huang Y F,Zhang Y,Zhu Y Z,et al.A study on non-destructive detection of GABA content in cherry tomatoes based on deep learning [J].Journal of Instrumental Analysis,2025,44(12):2 554-2 563.
[4] 高燕飞,徐雪峰,黄余,等.基于物联网和机器学习的番茄内外品质在线检测 [J].食品与机械,2025,41(7):78-85.Gao Y F,Xu X F,Huang Y,et al.Online detection of tomato internal and external quality based on IoT and machine learning[J].Food & Machinery,2025,41(7):78-85.
[5] 高翔堃,董璇,刘超,等.近红外光谱联合机器学习测定樱桃番茄中的番茄红素 [J].量子电子学报,2025,42(3):313-323.Gao X K,Dong X,Liu C,et al.Determination of lycopene in cherry tomatoes using near infrared spectroscopy combined with machine learning [J].Chinese Journal of Quantum Electronics,2025,42(3):313-323.
[6] 田文强,王巧华,徐步云,等.基于近红外光谱的腌制期咸鸭蛋理化指标无损检测 [J].食品科学,2023,44(2):319-326.Tian W Q,Wang Q H,Xu B Y,et al.Non-destructive detection of physical and chemical indicators of salted duck eggs during salting using near-infrared spectroscopy [J].Food Science,2023,44(2):319-326.
[7] 王冬,栾云霞,王欣然,等.近红外光谱无损分析肉类品质的研究进展 [J].肉类研究,2024,38(5):61-70.Wang D,Luan Y X,Wang X R,et al.Progress in non-destructive analysis of meat quality by near-infrared spectroscopy [J].Meat Research,2024,38(5):61-70.
[8] 康明月,王成,孙鸿雁,等.基于改进的 WOA-LSSVM 樱桃番茄内部品质检测方法研究 [J].光谱学与光谱分析,2023,43(11):3 541-3 550.Kang M Y,Wang C,Sun H Y,et al.Research on internal quality detection method of cherry tomatoes based on improved WOA-LSSVM [J].Spectroscopy and Spectral Analysis,2023,43(11):3 541-3 550.
[9] 周宇坤,陈孝敬,谢忠好,等.一种近红外光谱数据预处理组合优化策略 [J].光谱学与光谱分析,2025,45(1):52-58.Zhou Y K,Chen X J,Xie Z H,et al.A combinatorial optimization strategy for near-infrared spectral data preprocessing[J].Spectroscopy and Spectral Analysis,2025,45(1):52-58.
[10] 左晓佳,再努热・吐尔孙.肉品新鲜度评价及保鲜技术研究进展 [J].肉类研究,2023,37(12):69-75.Zuo X J,Tuerxun Z.Research progress in meat freshness evaluation and preservation technologies [J].Meat Research,2023,37(12):69-75.
[11] 方瑶,谢天铧,郭渭,等.基于近红外光谱的金鲳鱼新鲜度快速检测技术 [J].江苏农业学报,2021,37(1):213-218.Fang Y,Xie T H,Guo W,et al.Rapid detection technology of pomfret freshness based on near infrared spectroscopy [J].Jiangsu Journal of Agricultural Sciences,2021,37(1):213-218.
[12] 陈晶,温雅君,张延国,等.利用近红外光谱技术快速分析番茄6项常规营养成分 [J].中国蔬菜,2024 (11):106-113.Chen J,Wen Y J,Zhang Y G,et al.Rapid analysis of six nutritional components of tomatoes using near-infrared spectroscopy [J].China Vegetables,2024 (11):106-113.
[13] 刘艳群,肖付刚,陈彩虹.基于随机森林回归模型的小麦粉灰分含量快速测定 [J].食品与机械,2024,40(9):79-83.Liu Y Q,Xiao F G,Chen C H.Study on rapid determination method of ash content in wheat flour based on stochastic forest regression model [J].Food & Machinery,2024,40(9):79-83.
[14] 刘瑜明,王巧华,陈远哲,等.猪肉理化指标的近红外光谱无损检测 [J].光谱学与光谱分析,2024,44(5):1 346-1 353.Liu Y M,Wang Q H,Chen Y Z,et al.Non-destructive near-infrared spectroscopy of physical and chemical indicator of pork meat [J].Spectroscopy and Spectral Analysis,2024,44(5):1 346-1 353.
[15] 任智磊,赵霄霄,冯景,等.基于近红外光谱结合网格搜索—随机森林—自适应提升算法无损检测牛肉新鲜度 [J].肉类研究,2025,39(11):1-8.Ren Z L,Zhao X X,Feng J,et al.Non-destructive detection of beef freshness using near infrared spectroscopy combined with grid search-random forest-adaptive boosting algorithm [J].Meat Research,2025,39(11):1-8.
[16] 姜小刚,朱明旺,姚金良,等.基于近红外光谱技术苹果尺寸差异对糖度模型适用性的影响 [J].华中农业大学学报,2024,43(1):242-248.Jiang X G,Zhu M W,Yao J L,et al.Effects of apple size on applicability of model for predicting content of sugar based on near infrared spectroscopy [J].Journal of Huazhong Agricultural University,2024,43(1):242-248.
[17] Guo M Q,Lin H,Wang K Q,et al.Data fusion of near-infrared and Raman spectroscopy:an innovative tool for non-destructive prediction of the TVB-N content of salmon samples[J].Food Research International,2024,189:114564.
[18] Lan C F,Song B W,Zhang L,et al.State prediction of hydroturbine based on WOA-RF-adaboost [J].Energy Reports,2022,8:13 129-13 137.
[19] Liu T,Li J L,Xia R,et al.Experimental analysis of coal gangue identification using visible-near infrared spectroscopy under different working conditions with local and global model[J].International Journal of Coal Preparation and Utilization,2024,44(8):1-22.
[20] Zhang Y,Zhang C Y,Duan X X,et al.Visual/near-infrared spectroscopy combined with linear discriminant analysis and machine learning for classification of apple damage [J].Food Science,2024,45(22):255-261.
[21] 韩东,孙晓荣,李建伟.基于粒子群优化的深度极限学习机建模方法 [J].控制工程,2022,29(4):721-727.Han D,Sun X R,Li J W.Deep extreme learning machine improved by particle swarm optimization algorithm [J].Control Engineering,2022,29(4):721-727.
