Agricultural and Biological Research

A model effective on cotton crop classification using convolutional neural networks

M. Roshini, Ramu Vankudoth*, Venkata Madhu Bindu, Mani Raju Komma and T. Sunil

Cotton Crop Classification is a crucial task in precision agriculture, enabling farmers to monitor crop health, growth stages, and disease prevalence for optimized crop management. Convolutional Neural Networks (CNNs) have shown remarkable success in image recognition tasks, making them an ideal candidate for accurate and automated Cotton Crop Classification. In this study, we propose a CNN-based approach for Cotton Crop Classification using diverse datasets of cotton crop images collected from satellite, aerial, and ground-based sources. The CNN model is designed to automatically learn relevant features from raw pixel values, eliminating the need for manual feature engineering. Data augmentation techniques are employed to enhance the dataset's diversity and prevent overfitting. Transfer learning with pre-trained models on large image datasets is used to fine-tune the CNN model on the cotton crop dataset, ensuring improved generalization and faster convergence. The performance of the CNN model is evaluated using standard metrics such as accuracy, precision, recall, F1-score, and confusion matrix. The results demonstrate the effectiveness of the proposed CNN approach, achieving high accuracy in classifying cotton crop classes, including mature cotton, young cotton, and diseased cotton. The integration of CNN-based systems with drones or autonomous vehicles enables automated and real-time crop monitoring, paving the way for more efficient and data-driven precision agriculture practices. Overall, the application of CNNs in Cotton Crop Classification showcases the potential to revolutionize modern agriculture, promoting sustainable and optimized crop management.