Integrating machine learning, deep learning, and image analysis for seed species classification

Abstract

Premise: The growing demand for wildflower seeds in ecological restoration requires reliable species identification, yet current market products often contain heterogeneous species. As seed identification is labor‐intensive and requires advanced botanical knowledge, we evaluated multiple segmentation and classification approaches to determine which combination performed better for identifying species and quantifying their frequencies under real‐life constraints.

Methods: We generated seed images from common wildflower species using a flatbed scanner. Artificial intelligence (AI) and non‐AI segmentation pipelines were compared. We evaluated over 20 classifiers on tabulated features and convolutional neural networks (CNNs) on single seed image inputs. Open‐set conditions were simulated using “mock seed mixes” with previously unseen taxa using decision thresholds.

Results: With tabulated data, the XGBoost and Multi‐layer Perceptron (MLP) models exceeded F1 > 0.96, while AutoGluon's CNN and ensemble models reached F1 > 0.97. ResNet‐50 achieved an F1 score over 0.99 on single‐seed images obtained with Cellpose segmentation. Random forest achieved the highest accuracy (0.92) with unseen species in open‐set classification.

Discussion: Deep learning segmentation substantially enhanced CNN accuracy, yet tabulated‐feature machine learning models remained competitive and efficient. Threshold‐based rejection enabled robust open‐set classification, with random forest outperforming deeper models in rejecting unseen seeds. No single method was universally optimal; the best strategy depends on whether the task involves closed‐set or open‐set classification and on trade‐offs between accuracy and computational cost.

A hybrid machine learning framework integrating persian NLP for occupational risk classification in steel manufacturing
Ensieh Sadat Alamshah, Maryam Soleimani-Alyar, Saber Moradi Hanifi +2
Scientific Reports2 Oct 2026Peer-reviewedCC BY-NC-ND 4.0Full textCrossref
Automated brain MRI image classification through machine learning and deep learning approaches
Kamarujjaman, Srirupa Das, Diyasha Majumdar +3
Scientific Reports4 Oct 2026Peer-reviewedCC BY 4.0Full textNeurologyCrossref
Rapid Species-Level Classification of Urinary Pathogens from Raw LC-MS/MS Signals Using Machine Learning.
Simon J Pelletier, Antoine Lacombe-Rastoll, Florence Roux-Dalvai +11
Analytical chemistry1 Oct 2026Peer-reviewedCC BY-NC-NDFull textGeneral medicineEurope PMC
Enhanced Feature Extraction and Integration with Advanced Machine Learning Algorithms for Sports-Injury CT-Image Classification
Wang Jun, Chu Huiqin, Rashid Abbasi +1
Current Medical Imaging22 Sep 2026Peer-reviewedCC BY 4.0Full textMedical imaging & radiologyPMC
Code-Free Automated Machine-Learning (AutoML) Classification of Curated MRI Images Labeled for Rare Neurological Diseases: An Image-Level Proof-of-Concept Study
Mark A Bachir, Neel Nawathey, Alexander Bachir +3
Cureus1 Oct 2026Peer-reviewedCC BYFull textMedical imaging & radiologyEurope PMCPMC

Cite

BibTeX RIS

BibTeX

@article{ashworth2026integrating,
  title = {{Integrating machine learning, deep learning, and image analysis for seed species classification}},
  author = {Jonathan Ashworth and Eloise Barrett and Rebonto Haque and Ben Wildey and Anthony Hall and Felix Shaw and Jose J. De Vega},
  journal = {Applications in Plant Sciences},
  year = {2026},
  month = jul,
  doi = {10.1002/aps3.70072},
  url = {https://doi.org/10.1002/aps3.70072},
  note = {CC BY 4.0},
}

RIS

TY  - JOUR
TI  - Integrating machine learning, deep learning, and image analysis for seed species classification
AU  - Jonathan Ashworth
AU  - Eloise Barrett
AU  - Rebonto Haque
AU  - Ben Wildey
AU  - Anthony Hall
AU  - Felix Shaw
AU  - Jose J. De Vega
PY  - 2026
DA  - 2026/07/23
JO  - Applications in Plant Sciences
DO  - 10.1002/aps3.70072
UR  - https://doi.org/10.1002/aps3.70072
AB  - Premise: The growing demand for wildflower seeds in ecological restoration requires reliable species identification, yet current market products often contain heterogeneous species. As seed identification is labor‐intensive and requires advanced botanical knowledge, we evaluated multiple segmentation and classification approaches to determine which combination performed better for identifying species and quantifying their frequencies under real‐life constraints.  Methods: We generated seed images from common wildflower species using a flatbed scanner. Artificial intelligence (AI) and non‐AI segmentation pipelines were compared. We evaluated over 20 classifiers on tabulated features and convolutional neural networks (CNNs) on single seed image inputs. Open‐set conditions were simulated using “mock seed mixes” with previously unseen taxa using decision thresholds.  Results: With tabulated data, the XGBoost and Multi‐layer Perceptron (MLP) models exceeded F1 > 0.96, while AutoGluon's CNN and ensemble models reached F1 > 0.97. ResNet‐50 achieved an F1 score over 0.99 on single‐seed images obtained with Cellpose segmentation. Random forest achieved the highest accuracy (0.92) with unseen species in open‐set classification.  Discussion: Deep learning segmentation substantially enhanced CNN accuracy, yet tabulated‐feature machine learning models remained competitive and efficient. Threshold‐based rejection enabled robust open‐set classification, with random forest outperforming deeper models in rejecting unseen seeds. No single method was universally optimal; the best strategy depends on whether the task involves closed‐set or open‐set classification and on trade‐offs between accuracy and computational cost.
SN  - 2168-0450
ER  -

Abstract

Related

Cite