Integrated CNN-Random Forest Model for Accurate Rice Variety Classification

The rice variety classification model has impressive performance, highlighted in this summary. Using accuracy measures like precision and recall, it effectively distinguishes five rice varieties (Arborio, Basmati, Ipsala, Jasmine, Karacadag). Precision, measuring how accurate the model's positive predictions are, was consistently high, over 94%, with Arborio receiving the highest score (95.42%). Similarly, recall, which shows how well the model can find true positives, followed a similar pattern. The study of a dataset with 11,700 photos is shown in the Abstract. The model's remarkable 94.87% overall accuracy is the result of this examination. This shows that the model can accurately capture the characteristics that set each rice variety apart. The model's accuracy is impressive, with F1 scores consistently above 94.84%. These scores measure the balance between the model's ability to correctly identify events and minimize false positives. The weighted average measures overall performance, considering different class sizes. The model's precision, recall, and F1-Score all reach 94.87% in the weighted average, showcasing its reliability in handling varying class sizes and maintaining accuracy across the entire dataset. The model's overall performance was confirmed by very high average accuracy measures, with a consistent 94.87% accuracy across all rice varieties. The model's outstanding 98% accuracy rate highlights its precision in differentiating between five specific rice types: Arborio, Basmati, Ipsala, Jasmine, and Karacadag. Not only does this accuracy demonstrate the model's effectiveness in classification, but it also establishes it as a reliable tool for various applications in agriculture, such as management, trade, and quality control. The model's ability to detect subtle differences in rice varieties positions it as an essential tool for stakeholders in the agricultural industry, providing them with valuable information.