Related Experiment Video
Updated: Jun 29, 2026
11:01
Automated Quantification and Analysis of Cell Counting Procedures Using ImageJ Plugins
Published on: November 17, 2016
47.5K
Counting touching wheat grains in images based on elliptical approximation.
D R Avzalov1, E G Komyshev2, D A Afonnikov3
1Institute of Cytology and Genetics of the Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia.
Vavilovskii Zhurnal Genetiki I Selektsii
|July 23, 2025
Summary
Accurately identifying touching wheat grains in images is crucial for yield estimation. This study introduces an improved algorithm using concave point detection and elliptical approximation for precise grain boundary determination.
Area of Science:
- Agricultural Science
- Computer Vision
- Image Analysis
Background:
- Cereal grain number, size, and shape are key yield indicators.
- Digital image analysis is standard for estimating these parameters.
- Segmentation challenges arise with touching or densely packed grains.
Purpose of the Study:
- To develop an algorithm for accurate wheat grain identification and boundary determination in images, especially when grains are touching.
- To improve upon existing segmentation methods for touching grains.
Main Methods:
- A modified concave point search algorithm is employed.
- Contour boundary pixels are assigned to individual grains using elliptical approximation.
- The proposed method is compared against a non-approximated algorithm and the watershed algorithm.
Main Results:
- The proposed algorithm demonstrates higher accuracy in identifying wheat grains and their boundaries compared to the baseline methods.
- Elliptical approximation effectively refines contour boundary pixel assignment.
Conclusions:
- The developed algorithm offers improved accuracy for segmenting touching wheat grains.
- Computational time increases significantly with the number of grains and complex contours.
