Experimental Analysis of Plant Disease Identification Methodology Based on Improved Classification and Deep Learning Principle

The agriculture industry is essential for providing excellent food and significantly contributes to the growth of economies and people. Plant diseases can result in substantial reductions in food output and diminish species diversity. The prompt identification of plant diseases by precise or automated detection methods can improve food production quality and save economic losses. In recent years, deep learning has significantly enhanced the identification accuracy of image categorization and object detection systems. In this paper, a novel method termed Learning Enabled Disease Detection Logic (LDDL) is introduced for the efficient identification of plant diseases, which is cross-validated against the traditional Convolutional Neural Network (CNN) to assess the efficacy of the proposed scheme. The report delineates the specific crop illnesses addressed, the models utilized, the data sources referenced, and the overall efficacy as measured by the performance criteria applied in each study for plant disease diagnosis. The findings demonstrate that the proposed model, LDDL, achieves superior accuracy compared to prevalent illness detection methods and identifies the primary parameters influencing the efficacy of deep learning-based tools. The research offers significant insights for researchers, practitioners, and industry professionals in plant disease detection by addressing challenges and limitations, presenting a thorough understanding of the current research landscape, elucidating the advantages and drawbacks of various methods, and suggesting potential solutions to facilitate their implementation.