Deep Learning Based Dry Weight Prediction of Rare Plant Persicaria chinensis (L.) H. Gross

Daeho Choi; Jungmok Kang; Yong-Woo Park

doi:10.7732/kjpr.2026.39.1.027

All Issue

2026 Vol.39, Issue 1 Preview Page

Research Article

Deep Learning Based Dry Weight Prediction of Rare Plant Persicaria chinensis (L.) H. Gross 딥러닝 기술을 활용한 희귀식물 덩굴모밀의 건중량 예측

1 February 2026. pp. 27-33

PDF XML

Abstract

This study developed image-based deep learning models for non-destructive prediction of leaf and stem dry weight of Persicaria chinensis and compared the performance of different convolutional neural network (CNN) architectures. P. chinensis is a rare species designated by the Korea Forest Service with high potential for medicinal and functional food applications, highlighting the need for efficient yield prediction methods. A baseline CNN and four transfer learning models (VGG16, ResNet50, DenseNet121, and MobileNetV2) were evaluated using leaf and stem images as input data. Ten percent of the dataset was reserved as an independent test set, while the remaining data were used for training and validation through K-fold cross-validation (K = 5). Data augmentation and early stopping were applied to improve model generalization. For leaf dry weight prediction, DenseNet121 achieved the highest test performance (R² = 0.89). For stem dry weight prediction, VGG16 achieved the highest coefficient of determination on the independent test set (R² = 0.91). Performance differences were attributed to variations in morphological and visual characteristics between leaves and stems. These results demonstrate the feasibility of image-based deep learning for non-destructive yield prediction of P. chinensis in precision agriculture and smart forestry.

Keywords

Color image

Convolutional neural network

Dry weight

Transfer learning

References

Chang, I.S. and G. Park. 2023. A study on the production forecasting of deep learning-based facility cultivation. J. Broadcast Eng. 28(4):448-456.

10.5909/JBE.2023.28.4.448

Cho, J.H., H.D. Jo, Y.E. Choi, S.H. Park, J.M. Yang, C.W. Jeong, D.H. Choi, E.S. Jung, Y.W. Park, and J.M. Kang. 2024a. A composition for alleviating immune hypersensitivity comprising Persicaria chinensis extract. Korea Patent. 10-2024-0124042 (in Korean).

Cho, Y., J.Y. Kim, and J. Ha. 2024b. Application of deep learning technology for phenotyping tissue specific length of sprout vegetables using YOLOv8. Korean J. Breed. Sci. 56(4):417-424.

10.9787/KJBS.2024.56.4.417

Choi, D., J. Kang, E.S. Jung, and Y.W. Park. 2025. Polyphenol profiling and comparison of useful component content according to extraction solvents and plant part of Persicaria chinensis. Korean J. Plant Res. 38(1):18-25.

Collins, C., D. Dennehy, K. Conboy, and P. Mikalef. 2021. Artificial intelligence in information systems research: a systematic literature review and research agenda. Int. J. Inf. Manag. 60:102383.

10.1016/j.ijinfomgt.2021.102383

Enquist, B., J. Brown, and G. West. 1998. Allometric scaling of plant energetics and population density. Nature 395:163-165.

10.1038/25977

Gu, C. and M. Lee. 2024. Deep transfer learning using real-world image features for medical image classification, with a case study on pneumonia x-ray images. Bioengineering 11(4):406.

10.3390/bioengineering1104040638671827PMC11048359

Hossen, M.J., K.S. Baek, E. Kim, W.S. Yang, D. Jeong, J.H. Kim, D.H. Kweon, D.H. Yoon, T.W. Kim, J.H. Kim, and J.Y. Cho. 2015. In vivo and in vitro anti-inflammatory activities of Persicaria chinensis methanolic extract targeting Src/Syk/NF-κB. J. Ethnopharmacol. 159:9-16.

10.1016/j.jep.2014.10.064

Korea National Arboretum. 2021. The national red list of vascular plants in Korea. Designpost, Pocheon, Gyeonggi, Korea (in Korean).

Lai, S.M., D. Sudhahar, and K. Anandarajagopal. 2012. Evaluation of antibacterial and antifungal activities of Persicaria chinensis leaves. Int. J. Pharm. Sci. Res. 3(8):2825-2830.

Lăzăroiu, G., M. Andronie, M. Iatagan, M. Geamănu, R. Ștefănescu, and I. Dijmărescu. 2022. Deep learning-assisted smart process planning, robotic wireless sensor networks, and geospatial big data management algorithms in the internet of manufacturing things. ISPRS Int. J. Geo-Inf. 11:277.

10.3390/ijgi11050277

Lee, J.H., N.B.N. Han, J. L, L.B.T. An, G.J. Kwon, and J.Y. Kim. 2024. Deep learning-based weight and grade prediction algorithm from apple images. J. Korean Inst. Inf. Technol. 22(6):31-39.

10.14801/jkiit.2024.22.6.31

Li, G., M. Zhang, J. Li, F. Lv, and G. Tang. 2021. Efficient densely connected convolutional neural networks. Pattern Recognit. 109:107610.

10.1016/j.patcog.2020.107610

Luo, J., A. Wei, H. Zhu, F. Xie, G. Huang, P. Huang, W. Yang, L. Deng, S. Lin, and H. Qin. 2024. Optimization of extraction process and content determination of total flavonoids from Polygonum chinense L. produced in Guangxi. Hubei Agri. Sci. 63(1):164-168.

Park, W.P. and S. Heo. 2022. Development of machine learning models classifying nitrogen deficiency based on leaf chemical properties in shiranuhi (Citrus unshiu × C. sinensis). Korean J. Plant Res. 35(2):192-200.

Pound, M.P., J.A. Atkinson, A.J. Townsend, M.H. Wilson, M. Griffiths, A.S. Jackson, A. Bulat, G. Tzimiropoulos, D.M. Wells, E.H. Murchie, T.P. Pridmore, and A.P. French. 2017. Deep machine learning provides state-of-the-art performance in image-based plant phenotyping. Gigascience 6(10):1-10.

10.1093/gigascience/gix08329020747PMC5632296

Razzaq, K. and M. Shah. 2025. Machine learning and deep learning paradigms: from techniques to practical applications and research frontiers. Computers 14(3):93.

10.3390/computers14030093

Sadras, V.O. and R.A. Richards. 2014. Improvement of crop yield in dry environments: benchmarks, levels of organisation and the role of nitrogen. J. Exp. Bot. 65(8):1981-1995.

10.1093/jxb/eru061

Sandler, M., A. Howard, M. Zhu, A. Zhmoginov, and L.C. Chen. 2018. MobileNetV2: inverted residuals and linear bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4510-4520.

10.1109/CVPR.2018.00474

Sarker, I.H. 2021. Machine learning: algorithms, real-world applications and research directions. SN Comput. Sci. 2:160.

10.1007/s42979-021-00592-x33778771PMC7983091

Ye, F., Y. Su, H. Xiao, X. Zhao, and W. Min. 2018. Remote sensing image registration using convolutional neural network features. IEEE Geosci. remote Sens. Lett. 15(2):232-236.

10.1109/LGRS.2017.2781741

Zhang, L., Z. Xu, D. Xu, J. Ma, Y. Chen, and Z. Fu. 2020. Growth monitoring of greenhouse lettuce based on a convolutional neural network. Hortic. Res. 7:124.

10.1038/s41438-020-00345-632821407PMC7395764

Information

Publisher :The Plant Resources Society of Korea
Publisher(Ko) :한국자원식물학회
Journal Title :Korean Journal of Plant Resources
Journal Title(Ko) :한국자원식물학회지
Volume : 39
No :1
Pages :27-33
Received Date : 2025-12-23
Revised Date : 2026-01-09
Accepted Date : 2026-01-14
DOI :https://doi.org/10.7732/kjpr.2026.39.1.027

Korean Journal of Plant Resources ISSN:1226-3591(Print) 2287-8203(Online) 한국자원식물학회지

All Issue