Development of a Rice Plant Disease Classification Model in Big Data Environment

Area of Science:

• Agricultural Science
• Data Science
• Computer Vision

Background:

• Rice is a primary global food source, making crop health critical for food security.
• Agricultural losses due to plant diseases significantly impact food production and farmer livelihoods.

Purpose of the Study:

• To develop an efficient prediction model for identifying rice plant diseases.
• To minimize agricultural losses in paddy fields through early disease detection.

Main Methods:

• Utilized a big data framework to process large datasets of rice plant images.
• Applied feature extraction (color, shape, position, texture) and rough set theory-based feature selection.
• Implemented ensemble classification methods within a MapReduce framework for disease prediction.

Main Results:

• The proposed model demonstrated high efficiency in predicting rice plant diseases.
• Feature selection effectively reduced data dimensionality while retaining important disease indicators.

Conclusions:

• The developed prediction model is effective for early rice disease detection.
• Big data and machine learning approaches can significantly aid in agricultural loss mitigation.