本研究以 Swedish Leaf Dataset 與 Flavia Dataset 為研究對象，建立一套以資料驅動為核心之植物葉片表型分析架構，研究主要從葉片輪廓與葉脈兩個層面探討植物形態特徵，藉由建立可量化且具解釋性的結構化特徵表示，進一步辨識不同葉種之典型表型。
在 Swedish Leaf Dataset 部分，利用葉片輪廓之質心輪廓距離曲線作為分析基礎，透過非參數離散化、直方圖表徵以及階層式聚類方法，重建葉片輪廓之結構依存關係，並辨識具有代表性的局部輪廓特徵與類別專屬表型。進一步藉由樹狀結構分析與局部拓撲視覺化，呈現不同葉種之間的同質性與異質性關係。
在 Flavia Dataset 部分，由原始葉片影像出發，透過影像前處理、葉脈萃取與葉脈追蹤技術，建立半圓極座標系統以描述葉脈結構，於此座標系統下，進一步量測葉脈角度、葉脈長度、葉尖位置與葉谷位置等幾何特徵，並建構可供比較之葉脈表型結構。
本研究建立了結合葉片輪廓與葉脈資訊之植物表型分析流程，將原本高維度且複雜的葉片影像資料轉換為具有生物意義之結構化特徵。研究結果顯示，透過輪廓與葉脈特徵之依存關係分析，可有效建構葉種專屬表型集合，並提供植物分類、形態分析以及未來表型與基因型關聯研究之基礎。

This study develops a data-driven phenotyping framework for plant leaves using the Swedish Leaf Dataset and the Flavia Dataset. For the Swedish Leaf Dataset, leaf outlines represented by Centroid-Contour Distance curves are analyzed through nonparametric discretization and hierarchical clustering to identify characteristic contour features and species-specific phenotypes. For the Flavia Dataset, leaf vein structures are extracted from original leaf images through image preprocessing, vein extraction, and vein-tracing techniques. A half-circle polar coordinate system is further established to quantify geometric features, including vein angles, vein lengths, leaf-tip locations, and valley locations.
By integrating leaf-outline and leaf-vein information, the proposed framework transforms complex leaf-image data into interpretable structural features and constructs species-specific phenotype collections. The results provide a foundation for plant classification, morphological analysis, and future studies on phenotype-genotype associations.

Das Choudhury, S., Samal, A., and Awada, T. (2019). Leveraging image analysis for high-throughput plant phenotyping. Frontiers in plant science, 10:508.
Diaz, G. (2017). Contour recognition of complex leaf shapes. Plos one, 12(12):e0189427.
Fushing, H., Chou, E. P., and Chen, T.-L. (2023). Multiscale major factor selections for complex system data with structural dependency and heterogeneity. Physica A: Statistical Mechanics and its Applications, 630:129227.
Fushing, H., Kao, H.-W., and Chou, E. P. (2024). Topological risk-landscape in metric-free categorical database. IEEE Access, 12:66296-66318.
Gonzalez, R. and Woods, R. (2002). Digital image processing. 2nd ed. Prentice-Hall.
Jeon, W.-S. and Rhee, S.-Y. (2017). Plant leaf recognition using a convolution neural network. International Journal of Fuzzy Logic and Intelligent Systems, 17(1):26-34.
Kuo, Y.-F., Lin, T.-T., and Wang, C.-J. (2004). Geometric modeling of plant leaves aided with image processing techniques. Journal of Agricultural Machinery, 13(2):1-17.
Lee, S. H., Chan, C. S., Wilkin, P., and Remagnino, P. (2015). Deep-plant: Plant identification with convolutional neural networks. In 2015 IEEE international conference on image processing (ICIP), pages 452-456. IEEE.
Lin, S.-C., Lo, H.-C., Tung, C.-W., Liu, L.-Y., and Hsieh, F. (2026a). Computationally unraveling genetic factors underlying morphological patterns of wheat's root growth: A scientific data analysis approach.
Lin, S.-C., Lo, H.-C., Tung, C.-W., Liu, L.-Y., and Hsieh, F. (2026b). An information-theoretic framework for mapping categorical disease phenotypes.
Michels, D. L. and Sobottka, G. A. (2013). A mobile application for robust feature extraction and cultivar classification of leaves. Journal of Mobile Multimedia, 9(1-2):145-154.
Minervini, M., Fischbach, A., Scharr, H., and Tsaftaris, S. A. (2016). Finely-grained annotated datasets for image-based plant phenotyping. Pattern recognition letters, 81:80-89.
Murphy, K. M., Ludwig, E., Gutierrez, J., and Gehan, M. A. (2024). Deep learning in image-based plant phenotyping. Annual review of plant biology, 75(1):771-795.
Pachitariu, M., Rariden, M., and Stringer, C. (2025). Cellpose-sam: superhuman generalization for cellular segmentation. BioRxiv, 2025-04.
Shi, R., López-Malvar, A., Knoch, D., Tschiersch, H., Heuermann, M. C., Shaaf, S., Madur, D., Santiago, R., Balconi, C., Frascaroli, E., et al. (2025). Integrating high-throughput phenotyping and genome-wide association analyses to unravel mediterranean maize resilience to combined drought and high temperatures. Plant Stress, page 100954.
Söderkvist, O. J. O. (2001). Computer vision classification of leaves from swedish trees. Master's thesis, Linkoping University.
Victorino, J. and Gómez, F. (2019). Contour analysis for interpretable leaf shape category discovery. Plant methods, 15(1):112.
Wu, S. G., Bao, F. S., Xu, E. Y., Wang, Y.-X., Chang, Y.-F., and Xiang, Q.-L. (2007). A leaf recognition algorithm for plant classification using probabilistic neural network. In 2007 IEEE international symposium on signal processing and information technology, pages 11-16. IEEE.
Xiao, Q., Bai, X., Zhang, C., and He, Y. (2022). Advanced high-throughput plant phenotyping techniques for genome-wide association studies: A review. Journal of advanced research, 35:215-230.
Zheng, X. and Wang, X. (2010). Leaf vein extraction based on gray-scale morphology. International Journal of Image, Graphics and Signal Processing, 2(2):25.