|
|
|
|
|
|
| Comparative Study of Hyperspectral Preprocessing Methods and Multiple Models in Classification and Discrimination |
| JU Lei1, YU Jie1, WU Yan-miao2, LI Li2, LU Tian3, DING Ya-ping2, SHU Ru-xin1* |
1. Technology Center of Shanghai Tobacco Group Co., Ltd., Shanghai 200082, China
2. Department of Chemistry, Shanghai University, Shanghai 200444, China
3. Shanghai Shuzhiwei Information Technology Co., Ltd., Shanghai 200444, China
|
|
|
|
|
Abstract Identifying different parts of Solanaceae plants is crucial for their product formulation design and quality control. Hyperspectral technology, which can quickly and non-destructively acquire rich information, has become a widely used tool in plant research and monitoring. As important economic crops, Solanaceae plants have great research potential when combined with hyperspectral technology. This study employs hyperspectral technology to classify different parts of Solanaceae plant leaves after initial roasting. Firstly, hyperspectral sampling was conducted on 293 powder samples from different parts of Solanaceae plants using the Field Spec 3 spectroradiometer. Subsequently, data preprocessing was performed using S-G smoothing and first-order and second-order derivatives to enhance information and remove noise. To minimize redundant features, partial least squares (PLS) were then used for data dimensionality reduction. Finally, based on the dimensionality-reduced data, six machine learning classification models-support vector machine (SVM), logistic regression, K-nearest neighbors (KNN), decision tree, random forest, and gradient boosting decision tree—were used for modeling and analysis. The results showed that for the classification task, the SVM model performed best after first-order derivative processing, achieving an accuracy of 100.0% on the training set and 84.7% on the test set. After grid parameter optimization, the optimal parameters were determined: no restriction on maximum depth, a minimum sample split of 4, and 200 estimators. The accuracy of five-fold cross-validation after parameter optimization was 88.1%, with the training set accuracy at 100% and the test set accuracy at 86.4%. The study results indicate that preprocessing methods combined with dimensionality reduction can enhance data information, enabling classification models to capture the characteristics of Solanaceae plant samples better. This study is of great significance for the rapid, accurate, and non-destructive differentiation of parts of Solanaceae plants.
|
|
Received: 2024-06-18
Accepted: 2024-07-19
|
|
|
|
Corresponding Authors:
SHU Ru-xin
E-mail: shurx@sh.tobacco.com.cn
|
|
[1] MEI Ji-fan, GUO Wen-meng, LI Zhi-hui，et al(梅吉帆, 郭文孟, 李智慧, 等). Acta Tabacaria Sinica(中国烟草学报)，2024, 30(3): 51.
[2] LAI Jia-zheng，LI Bei-bei, CHENG Xiang，et al(赖佳政, 李贝贝, 程 翔, 等). Smart Agriculture[智慧农业(中英文)], 2023, 5(2): 68.
[3] FU Hu-yan, JIN Han-cheng, ZHANG Hong-liang，et al(付虎艳, 靳涵丞, 张洪亮, 等). Tobacco Science & Technology(烟草科技), 2015, 48(2): 21.
[4] CHEN Nan, FENG Hui-lin, YANG Yan-dong，et al(陈 楠, 冯慧琳, 杨艳东, 等). Journal of Agricultural Resources and Environment(农业资源与环境学报), 2021, 38(4): 570.
[5] LIU Hong-yun, WU Xue-mei, ZHANG Fu-gui，et al(刘红芸, 吴雪梅, 张富贵, 等). Computer and Digital Engineering(计算机与数字工程), 2022, 50(9): 2083.
[6] Guo T, Tan C, Li Q, et al. Journal of Ambient Intelligence and Humanized Computing, 2019, 10: 3239.
[7] Yu K, Fang S, Zhao Y. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2021, 245: 118917.
[8] ZHAO Chen, WANG Tao, GUO Wei-xiong，et al(赵 晨, 王 涛, 郭伟雄, 等). Journal of Hunan Agricultural University(Natural Sciences)[湖南农业大学学报(自然科学版)], 2023, 49(4): 405.
[9] YANG De-jian, ZHAO Liao-ying，HAO Xian-wei，et al(杨德建, 赵辽英, 郝贤伟, 等). Acta Tabacaria Sinica(中国烟草学报), 2023, 29(6): 23.
[10] LU Wen-cong, WU Yan-miao, LIU Tai-ang，et al(陆文聪, 吴炎淼, 刘太昂, 等). Journal of Henan Normal University(Natural Science Edition)[河南师范大学学报(自然科学版)], 2024, 52(4): 120.
[11] ZHAO Juan-juan, YE Shun, XU Ke, et al(赵娟娟, 叶 顺, 徐 可, 等). Journal of Henan Normal University(Natural Science Edition)[河南师范大学学报(自然科学版)], 2021, 49(1): 45.
[12] Deng W, Liu D, Guo F, et al. Agronomy, 2024, 14(4): 703.
[13] Wang H, Shao Y. Pattern Recognition, 2024, 146: 109987.
[14] Rahmatinejad Z, Dehghani T, Hoseini B, et al. Scientific Reports, 2024, 14(1): 3406.
[15] Sun Z, Wang G, Li P, et al. Expert Systems with Applications, 2024, 237: 121549.
[16] Hasan N, Ahmed N, Ali S M. Engineering Applications of Artificial Intelligence, 2024, 129: 107633.
[17] Sun X, Fu J. Energy, 2024, 288: 129840.
[18] XU Yu-ting, SUN Han-ran, GAO Xun，et al(许毓婷, 孙浩然, 高 勋, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2021, 41(11): 3572.
[19] Wu Y, Shang Z, Lu T, et al. Journal of Alloys and Compounds, 2024, 971: 172664.
[20] Fuadah Y N, Pramudito M A, Lim K M. Bioengineering-Basel, 2023, 10(1): 45.
|
| [1] |
WU Chang-yu1, DAI Jing-jing1, 2*, SONG Yang1, 2, CHEN Wei1, 2, LIU Zhi-bo1, LIU Hong-cheng3, BAI Long-yang1. Study on Core Hyperspectral Imaging and Its Significance in Exploration: A Case Study of Bangpu Large Polymetallic Deposit in Tibet[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 38-44. |
| [2] |
WU Meng-hong1, 2, DOU Sen1, LIN Nan2, JIANG Ran-zhe3, CHEN Si2, LI Jia-xuan2, FU Jia-wei2, MEI Xian-jun2. Hyperspectral Estimation of Soil Organic Matter Based on FOD-sCARS and Machine Learning Algorithm[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 204-212. |
| [3] |
TANG Bin1, HE Yu-long1, TANG Huan2*, LONG Zou-rong1, WANG Jian-xu1*, TAN Bo-wen2, QIN Dan2, LUO Xi-ling1, ZHAO Ming-fu1. Attention Mechanism Based Hyperspectral Image Dimensionality Reduction for Mold Spot Recognition in Paper Artifacts[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 246-255. |
| [4] |
ZHANG Chao1, 2, 3, WU Xuan1, 3, YANG Ke-ming4*, QI Fan-yu1, 3, XIA Tian5. Exploration of Spectral Characteristics of Crop Leaves Under Cu2+ Pollution[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 264-271. |
| [5] |
XU Yang1, MAO Yi-lin1, LI He1, WANG Yu1, WANG Shuang-shuang2, QIAN Wen-jun1, DING Zhao-tang2*, FAN Kai1*. Multispectral and Hyperspectral Prediction Models of REC, SPAD and MDA in Overwintered Tea Plant[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 256-263. |
| [6] |
XU Zi-yang1, 2, JIANG Xin-hua1, 2*, ZHAI Cheng-jun3, MA Xue-lei1, 2, LI Jing1, 2. Non-Destructive Detection of Multi-Indicator Chilled Mutton Freshness Based on Improved Artificial Neural Network[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(01): 291-300. |
| [7] |
CHEN Cheng1, YAN Bing1, YIN Zuo-wei1*, CAO Wei-yu2, WANG Wen-jing1. Study on the Spectrum and Visualization of “Trapiche” Tourmaline Based on Hyperspectral Imaging Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3429-3434. |
| [8] |
WANG Jing1, MA Ling1, MA Si-yan1, MA Yan1, ZHANG Yi-yang1, WU Long-guo1, 2*. Nondestructive Detection of Catalase Activity in Melon Leaves By
Fluorescence Hyperspectral Imagery[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3455-3462. |
| [9] |
XU Jing-yu1, BAO Ni-sha1, 2*, LANG Jie-shuang3, LIU Shan-jun1, 2, MAO Ya-chun1, 2, HE Li-ming1, 2. A Hyperspectral Recognition Method for Camouflaged Targets Based on Background Dictionary Sparse Representation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3534-3542. |
| [10] |
GAO Yu1, 2, SUN Xue-jian1, 3*, LI Guang-hua3, 4, ZHANG Li-fu1, 3, QU Liang3, 4, ZHANG Dong-hui5. Hidden Handwriting Recognition of Calligraphy Artifact Based on
Hyperspectral Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3485-3493. |
| [11] |
CAO Wang1, MAO Ya-chun1*, WEN Jie1, DING Rui-bo1, XU Meng-yuan1, FU Yan-hua2. Study on Inversion Method of Anshan-Type Iron Ore Grade Based on
Hyperspectral Image[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3494-3503. |
| [12] |
WANG Sa1, 2, QU Liang1, 2*, ZHANG Li-fu3, GAO Yu3, LI Guang-hua1, 2, CHANG Jing-jing1, 2. Research on the Inverse Model of Paper Viscosity Based on Hyperspectral Data[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3524-3533. |
| [13] |
ZHANG Ai-wu1, 2, 3, LI Meng-nan1, 2, 3, SHI Jian-cong1, 2, 3, PANG Hai-yang1, 2, 3. Hyperspectral Inversion Method for Natural Grassland Canopy SPAD Value Based on Scaling Up of Green Coverage Rate[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3513-3523. |
| [14] |
HUANG Lin-feng1, JIANG Xue-song1, 2*, JIA Zhi-cheng1, ZHOU Hong-ping1, 2, ZHOU Lei1, RONG Zi-fan1. Deep Learning-Based Monitoring of Nutrient Content in Pear Trees[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(12): 3543-3552. |
| [15] |
LIU Qing-song1, DU Wen-jing1, LUO Bo2, LI Kai-ge1, DAN You-quan1*, XU Luo-peng1, YANG Xiu-feng2, TANG Shen-lan1. Near Infrared Hyperspectral Identification of Surface Damage on Aircraft Wings[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2024, 44(11): 3069-3074. |
|
|
|
|
