Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A novel method for identifying rice seed purity based on hybrid machine learning algorithms

Phan Thi-Thu-Hong, Vo Quoc-Trinh, Nguyen Huu-Du

TL;DR

This work tackles rice seed purity identification by integrating deep CNN feature extraction with traditional ML classifiers in a hybrid framework. By extracting features from multiple blocks of VGG16 and ResNet-50 and combining them with LR or SVM, the method significantly outperforms both pure DL and hand-crafted feature-based ML approaches, achieving near-perfect accuracy on several varieties. The study emphasizes a practical, binary-separation setup aligned with real-world seed-purity screening and demonstrates that intermediate CNN features can preserve discriminative information crucial for high-precision classification. Overall, the proposed hybrid pipeline offers a robust, automatic system design for identifying rice seed purity with substantial potential for seed industry deployment.

Abstract

In the grain industry, the identification of seed purity is a crucial task as it is an important factor in evaluating the quality of seeds. For rice seeds, this property allows for the reduction of unexpected influences of other varieties on rice yield, nutrient composition, and price. However, in practice, they are often mixed with seeds from others. This study proposes a novel method for automatically identifying the rice seed purity of a certain rice variety based on hybrid machine learning algorithms. The main idea is to use deep learning architectures for extracting important features from the raw data and then use machine learning algorithms for classification. Several experiments are conducted following a practical implementation to evaluate the performance of the proposed model. The obtained results show that the novel method improves significantly the performance of existing methods. Thus, it can be applied to design effective identification systems for rice seed purity.

A novel method for identifying rice seed purity based on hybrid machine learning algorithms

TL;DR

This work tackles rice seed purity identification by integrating deep CNN feature extraction with traditional ML classifiers in a hybrid framework. By extracting features from multiple blocks of VGG16 and ResNet-50 and combining them with LR or SVM, the method significantly outperforms both pure DL and hand-crafted feature-based ML approaches, achieving near-perfect accuracy on several varieties. The study emphasizes a practical, binary-separation setup aligned with real-world seed-purity screening and demonstrates that intermediate CNN features can preserve discriminative information crucial for high-precision classification. Overall, the proposed hybrid pipeline offers a robust, automatic system design for identifying rice seed purity with substantial potential for seed industry deployment.

Abstract

In the grain industry, the identification of seed purity is a crucial task as it is an important factor in evaluating the quality of seeds. For rice seeds, this property allows for the reduction of unexpected influences of other varieties on rice yield, nutrient composition, and price. However, in practice, they are often mixed with seeds from others. This study proposes a novel method for automatically identifying the rice seed purity of a certain rice variety based on hybrid machine learning algorithms. The main idea is to use deep learning architectures for extracting important features from the raw data and then use machine learning algorithms for classification. Several experiments are conducted following a practical implementation to evaluate the performance of the proposed model. The obtained results show that the novel method improves significantly the performance of existing methods. Thus, it can be applied to design effective identification systems for rice seed purity.
Paper Structure (18 sections, 2 equations, 5 figures, 3 tables)

This paper contains 18 sections, 2 equations, 5 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Related works
  3. Data collection
  4. The proposed method
  5. Deep learning methods for feature extraction
  6. VGG16
  7. ResNet-50
  8. Machine learning algorithms
  9. Decision Tree, DT
  10. Random Forest, RF
  11. Extra Trees, ET
  12. K-Nearest Neighbors, KNN
  13. Logistic Regression, LR
  14. Support Vector Machine
  15. Experiments and results
  16. ...and 3 more sections

Figures (5)

  • Figure 1: An example of the rice seed images from the dataset
  • Figure 2: The schema of the proposed method
  • Figure 3: The architecture of the VGG16 model: Convolutional layer with kernel of 3$\times$3 and max pooling of 2$\times$2
  • Figure 4: The architecture of the ResNet-50 model
  • Figure 5: The number of positive and negative samples in each folder of the dataset.