Related Experiment Video
Updated: Dec 25, 2025
11:49
Cereal Crop Ear Counting in Field Conditions Using Zenithal RGB Images
Published on: February 2, 2019
9.7K
Rapid Recognition of Field-Grown Wheat Spikes Based on a Superpixel Segmentation Algorithm Using Digital Images
Changwei Tan1, Pengpeng Zhang1, Yongjiang Zhang2
1Jiangsu Key Laboratory of Crop Genetics and Physiology/Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops/Joint International Research Laboratory of Agriculture and Agri-Product Safety of the Ministry of Education of China, Yangzhou University, Yangzhou, China.
Frontiers in Plant Science
|March 27, 2020
Summary
Accurate wheat spike counting using superpixel segmentation and color analysis improves crop monitoring. This method is effective in field conditions but less reliable for poorly or unevenly grown wheat.
Area of Science:
- Agricultural Science
- Computer Vision
- Image Processing
Background:
- Wheat spike number is crucial for crop growth monitoring and yield prediction.
- Accurate estimation of wheat spike number is essential for precision agriculture.
Purpose of the Study:
- To develop and validate a rapid and accurate method for estimating wheat spike number using image processing.
- To evaluate the effectiveness of superpixel segmentation compared to traditional pixel-based methods.
Main Methods:
- Simple Linear Iterative Clustering (SLIC) for superpixel segmentation of wheat images.
- Extraction of characteristic color parameters (Super green value, normalized red green index) for classification.
- Morphological transformations and inflection point detection for wheat spike extraction and counting.
Main Results:
- Superpixel segmentation significantly improved wheat spike recognition clarity and morphological integrity compared to pixel-based methods.
- High nitrogen levels correlated with better wheat growth and the highest estimation accuracy (94.01%).
- Estimation accuracy decreased under no nitrogen (80.8%), indicating sensitivity to growth status and heterogeneity.
Conclusions:
- Superpixel segmentation combined with color features offers a rapid and accurate approach for field-based wheat spike estimation.
- The method is recommended for uniformly growing wheat but not for poor or highly heterogeneous conditions.
- Findings provide valuable insights for field-grown wheat yield estimation.
