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

Aug 3, 2026

Upload by Miles

In the grain industry, rice seed purity directly affects the quality evaluation of seeds and ultimately the rice yield, nutrient composition, and market price. Pure seeds must remain genetically true, yet real-world handling often mixes varieties during planting, production, packaging, and distribution. The study introduces an automatic identification approach using hybrid machine learning: deep learning for feature extraction from raw seed data and machine learning for classification, validated through practical experiments that show improved performance over existing methods.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages20

Reading Time50 min

AccessFree to download

Introduction

Background and importance of rice seed purity

Limitations of manual and traditional ML approaches

Proposed hybrid deep learning and machine learning method

Experimental setup and evaluation plan

arXiv :2406 .0758 1v 1 [ cs .CV] 9 Jun 2024 A novel method for identifying rice seed purity based on hybrid machine learning algorithms Thi-Thu-Hong Phana,∗, Quoc-Trinh Voa , Huu-Du Nguyenb a Artificial Intelligence Department, FPT University, Da Nang, 550000, Viet Namb School of Applied Mathematics and Informatics, Hanoi University of Science and Technology, Hanoi, Vietnam 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. Keywords: Rice seed purity identification, Feature extraction, Machine learning, VGG16, ResNet-50 1. Introduction Rice is one of the most important cereal grains for humans. It is a staple food full of nutrition and calories for over half of the humans in the world (Sohet al. [1]) . As the global population is increasing day by day, the role of rice is becoming more important and it is important to increase substantially the current level of rice production. An essential strategy for achieving high rice yield is to ensure the quality of rice seeds. A good seed has a high probability of producing a healthy, tolerant, and high-yielding rice plant. As a result, planting quality seeds is a necessary condition for a bountiful harvest, contributing to an increase in the quality of rice products. ∗ Corresponding author: [hongptt11@fe.edu.vn](hongptt11@fe.edu.vn) Email addresses: [trinhvq@fe.edu.vn](trinhvq@fe.edu.vn) (Quoc-Trinh Vo), [du.nguyenhuu@hust.edu.vn](du.nguyenhuu@hust.edu.vn)[ ](du.nguyenhuu@hust.edu.vn)(Huu-Du Nguyen) Preprint submitted to Heliyon June 13, 2024 There are several standards for the quality of rice seeds, including purity. The purity of seeds means that they must be genetically pure and free from all sorts of other seeds. This is an essential property of rice seeds to avoid unexpected influences of other varieties on their yield, nutrient composition, and price. However, purebred rice seeds inevitably get mixed up with others due to several stages of planting, producing, packaging, and circulating. The identification of rice seed purity is therefore a crucial task for the grain industry. Traditionally, this task is performed manually based on the experience of experts /technicians on the visual and physical characteristics of rice seeds. This method, however, may give inaccurate and inconsistent identification results due to subjective factors such as the different experiences of experts and the fatigue from concentrating on the same job for a long time. Recently, advanced technologies like computer vision and machine learning have been applied to handle the mission of identifying the purity of rice seeds. Phan et al. [2] used some machine learning (ML) algorithms including K-nearest neighbor (KNN), support vector machine (SVM), and random forest (RF) to identify rice seeds of a variety from a set of images of rice seeds mixed with the images of other types. The authors also suggested using feature extraction techniques of computer vision like Global Image Descriptor (GIST) and scale-invariant feature transform (SIFT) to extrac

FAQs

Why is rice seed purity important for grain production?

Rice seed purity helps prevent unexpected effects of other varieties on rice yield, nutrient composition, and price. It supports consistent quality in planting outcomes.

What problem does the proposed method address?

It targets the practical issue that rice seeds are often mixed with seeds from other varieties, making accurate purity identification necessary beyond manual expert inspection.

How does the proposed hybrid model perform rice seed purity identification?

It uses deep learning networks (VGG16 and ResNet-50) to extract discriminative features from rice seed image data, then applies multiple machine learning classifiers to categorize purity, with experiments demonstrating improved results.

Aug 3, 2026

Upload by Miles