A novel smart crop disease monitoring system utilizing the Internet of Things (IoT) has been developed, demonstrating substantial advancements in automated plant disease detection. The research presents the integration of the Henry Gas Chicken Swarm Optimization (HGCSO) algorithm with deep learning techniques to classify plant diseases more effectively within agricultural settings.
Recent advancements have proven the IoT's potential to facilitate proactive measures by farmers. The system collects images of plant leaves through IoT nodes deployed across crop fields, allowing for remote monitoring and timely disease diagnosis. By automally diagnosing diseases using the developed Caviar Henry Gas Chicken Swarm Optimization (CHGCSO), the model justifies the significance of optimized routing and classification methods amid growing food security concerns globally.
Utilizing the median filtering technique for image preprocessing, the system efficiently extracts important visual features from rice leaves. Key feature extraction methods include the Histogram of Oriented Gradient (HoG), statistical attributes, Spider Local Image Features (SLIF), and Local Ternary Patterns (LTP). With these features fed to the Deep Residual Network (DRN) model, the research achieved commendable accuracy rates of 94.3%, sensitivity of 93.3%, specificity of 92%, and F1-score of 93%. These results outstrip previous models, underscoring the promise of precise disease monitoring systems based on IoT.
This innovative model is not just about data collection; it’s about making sense of the data to empower farmers. By minimizing the time required for manual inspections and leveraging deep learning's predictive capabilities, the IoT framework can guide agricultural practices with unmatched accuracy and efficiency. Farmers can be alerted early about diseases, enabling timely intervention and minimizing potential crop losses.
Prior models faced challenges with image quality and computational complexity, but the current system addresses these through optimized routing and deep learning. The innovative approach of combining the CAViaR model with HGCSO exhibits the potential to improve previous classification methodologies significantly.
This breakthrough is particularly relevant as the agricultural sector increasingly adopts smart technologies to face challenges such as climate change, pest infestations, and resource management. The future of this research suggests exploring its application beyond rice crops to other agricultural plants, ensuring broader utility across various environments. Upcoming research aims to test this model under different conditions and configurations to validate its reliability and accuracy more comprehensively.
It is evident from the study's findings published by Saini et al. (2025) in Scientific Reports, the integration of IoT and advanced computational methods can redefine crop monitoring strategies, propelling sustainable agricultural practices forward.