Related Experiment Video
Updated: Jun 3, 2025
08:27
Author Spotlight: Efficient Image Recognition Using Directional Gradient Histogram Technique and Support Vector Machines
Published on: January 5, 2024
973
Maize Kernel Broken Rate Prediction Using Machine Vision and Machine Learning Algorithms.
Chenlong Fan1, Wenjing Wang1, Tao Cui2
1College of Mechanical and Electronic Engineering, Nanjing Forestry University, Nanjing 210037, China.
Foods (Basel, Switzerland)
|January 8, 2025
Summary
This study introduces a machine vision and machine learning approach for rapid online detection of broken maize kernels. This method accurately predicts kernel damage rates, preventing fungal contamination during harvest.
Area of Science:
- Agricultural Engineering
- Computer Science
- Data Science
Background:
- Maize harvest requires rapid detection of broken kernels to prevent fungal damage.
- Existing methods for assessing kernel damage are often inefficient and subjective.
Purpose of the Study:
- To develop an accurate and objective online detection method for broken maize kernels.
- To guide maize harvest practices for minimal kernel damage and reduced fungal contamination.
Main Methods:
- Constructed a dataset of high-moisture maize kernel phenotypic features, extracting seven geometric and shape characteristics.
- Developed regression models for predicting broken and unbroken kernel weights using machine learning algorithms.
- Established classification models for kernel defect detection using machine learning.
Main Results:
- Light Gradient Boosting Machine (LGBM) and Random Forest (RF) algorithms showed high accuracy (r values of 0.985 and 0.910) for weight prediction.
- Support Vector Machine (SVM) achieved over 95% accuracy in classifying kernel defects.
- A strong linear relationship was confirmed between predicted and actual broken rates.
Conclusions:
- The proposed machine vision and machine learning method provides an accurate, objective, and efficient approach for online broken rate detection in maize.
- This technology can significantly improve maize harvest quality and reduce post-harvest losses.
