Skip to content
DraconiticResearch
Search Sources API
210,049 studies31 sourcesUpdated 13 min ago
Peer-reviewedCC BYThe Plant Phenome Journal

MaizeEar‐SAM: Zero‐shot maize ear phenotyping

  • Hossein Zaremehrjerdi1
  • Lisa Coffey2,3
  • Talukder Zaki Jubery4
  • Huyu Liu2
  • Jonathan Turkus5
  • Kyle Linders5
  • James C. Schnable5
  • Patrick S. Schnable2,3
  • Baskar Ganapathysubramanian1,6,3,4
6 affiliations
  1. 1Department of Electrical and Computer Engineering Iowa State University Ames Iowa USA
  2. 2Department of Agronomy Iowa State University Ames Iowa USA
  3. 3Plant Science Institute Iowa State University Ames Iowa USA
  4. 4Translational AI Research and Education Center Iowa State University Ames Iowa USA
  5. 5Department of Agronomy University of Nebraska‐Lincoln Lincoln Nebraska USA
  6. 6Department of Mechanical Engineering Iowa State University Ames Iowa USA
Open at sourceDOICite

Abstract

Quantifying the variation in yield component traits of maize ( Zea mays L.), which collectively determine the overall productivity of this globally significant crop, is critical to plant genetics research, plant breeding, and the development of improved agronomic practices. Grain yield per acre is a function of the number of plants per acre ears per plant number of kernels per ear average kernel weight. Number of kernels per ear is the product of the number of kernel rows per ear ( kernel‐row‐number ) and the number of kernels per row ( kernels‐per‐row ). Traditional manual approaches to quantifying kernels‐per‐row, which exhibits strong environmental sensitivity, are time‐consuming, limiting the scale of data collected for either trait. Recent automation efforts using image processing and deep learning face challenges of high annotation costs and limited generalizability across diverse germplasm. We address these limitations by exploring large vision models for zero‐shot, annotation‐free maize kernel segmentation. Utilizing an open source large vision model—Segment Anything Model—we segment individual kernels in images of maize ears and implement a graph‐based algorithm to calculate kernels‐per‐row. Our approach effectively identifies kernels‐per‐row across diverse maize ears, demonstrating the potential of zero‐shot foundation vision models coupled with image processing routines to enhance automation and reduce subjectivity in agronomic data collection. All our codes are open‐sourced to democratize these frugal phenotyping approaches.

Related

  1. ASTRA enables zero-shot super-resolution of spatial transcriptomics across capture geometries
    J. Ye, Q. Cui
    bioRxivPreprintCC BY-NC-NDFull textBioinformatics & computational biologybioRxiv
  2. Inherent limitations of zero-shot mutation prediction in protein engineering
    P. R. Woolley, A. Feller, A. O. Ellington +1
    bioRxivPreprintCC BY-NCFull textSystems & synthetic biologybioRxiv
  3. Physics-Grounded Simulation Enables Zero-Shot Molecular and Reaction Analysis in Scanning Tunneling Microscopy
    Yuanqin He, Zhiyu Wang, Yang Bao +2
    arXivPreprintCC BY 4.0Full textBioinformatics & computational biologyCondensed matterarXiv
  4. Towards Sparse Causal Features for Zero-shot Mutation Effect Prediction in a Protein Language Model
    S. Mohanty, M. Phutela, A. G. Green
    bioRxivPreprintCC BY-NCFull textBioinformatics & computational biologybioRxiv
  5. Zero-Shot Generalization of Prospective Planning in Macaques
    M. Seong Bak, B. Zhu, Y. Minai +3
    bioRxivPreprintCC BY-NC-NDFull textNeurosciencebioRxiv

Cite

BibTeXRIS
BibTeX
@article{zaremehrjerdi2026maizeearsam,
  title = {{MaizeEar‐SAM: Zero‐shot maize ear phenotyping}},
  author = {Hossein Zaremehrjerdi and Lisa Coffey and Talukder Zaki Jubery and Huyu Liu and Jonathan Turkus and Kyle Linders and James C. Schnable and Patrick S. Schnable and Baskar Ganapathysubramanian},
  journal = {The Plant Phenome Journal},
  year = {2026},
  month = sep,
  doi = {10.1002/ppj2.70102},
  url = {https://doi.org/10.1002/ppj2.70102},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - MaizeEar‐SAM: Zero‐shot maize ear phenotyping
AU  - Hossein Zaremehrjerdi
AU  - Lisa Coffey
AU  - Talukder Zaki Jubery
AU  - Huyu Liu
AU  - Jonathan Turkus
AU  - Kyle Linders
AU  - James C. Schnable
AU  - Patrick S. Schnable
AU  - Baskar Ganapathysubramanian
PY  - 2026
DA  - 2026/09/20
JO  - The Plant Phenome Journal
DO  - 10.1002/ppj2.70102
UR  - https://doi.org/10.1002/ppj2.70102
AB  - Quantifying the variation in yield component traits of maize ( Zea mays L.), which collectively determine the overall productivity of this globally significant crop, is critical to plant genetics research, plant breeding, and the development of improved agronomic practices. Grain yield per acre is a function of the number of plants per acre ears per plant number of kernels per ear average kernel weight. Number of kernels per ear is the product of the number of kernel rows per ear ( kernel‐row‐number ) and the number of kernels per row ( kernels‐per‐row ). Traditional manual approaches to quantifying kernels‐per‐row, which exhibits strong environmental sensitivity, are time‐consuming, limiting the scale of data collected for either trait. Recent automation efforts using image processing and deep learning face challenges of high annotation costs and limited generalizability across diverse germplasm. We address these limitations by exploring large vision models for zero‐shot, annotation‐free maize kernel segmentation. Utilizing an open source large vision model—Segment Anything Model—we segment individual kernels in images of maize ears and implement a graph‐based algorithm to calculate kernels‐per‐row. Our approach effectively identifies kernels‐per‐row across diverse maize ears, demonstrating the potential of zero‐shot foundation vision models coupled with image processing routines to enhance automation and reduce subjectivity in agronomic data collection. All our codes are open‐sourced to democratize these frugal phenotyping approaches.
SN  - 2578-2703
ER  -