Answers are grounded in this paper and cite their sources.
Crop leaf disease prediction is plagued by insufficient generalization over different crop species, the lack of differentiation between similar disease symptoms, and variable environmental conditions affecting image quality. Poorly labelled datasets, model over-fitting, and real-time deployment issues all affect the accuracy and reliability of detecting illnesses in agriculture applications. With the PlantVillage dataset, the Random Graph Diffusion Dual Channel Temporal Convolutional Network with Synergistic Fibroblast Optimization (RGD-DCTCNet-SFO) is employed to resolve these challenges in crop leaf disease detection. The pre-processing by the Blind DE-blurring based Light Weight Wiener Filter (BDE-LWWF) is the first step in the process, which enhances image quality by reducing noise and blurring. Return-Aligned Decision Transformer (RADT) provides accurate boundary definition to enable segmentation by identifying regions using deviation analysis. After features have been obtained, the Random Graph Diffusion Dual Channel Temporal Convolutional Network (RGD-DCTCNet) is utilized for effective crop leaf disease classification. Synergistic Fibroblast Optimization (SFO), which boosts the accuracy of classification and minimizes errors, performs optimization on the search process of ill regions for further improved performance. The RGD-DCTCNet-SFO algorithm surpasses existing methods, recording 99.9% efficiency and 99.8% sensitivity, based on experimental outcomes of a Python-based study. The approach provides a robust and reliable solution for agricultural analysis by significantly improving the accuracy of crop leaf disease diagnosis.
No takes yet. Share an insight, caveat, or question.
Betha et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: