IoT-based plant disease detection using enhanced Elman spike neural network with Capuchin search optimisation algorithm

In recent years, the internet of things (IoT) has gained attention for its transformative role in agriculture. A main challenge in agriculture is early identification of plant disease which is needed to prevent crop loss and ensure food preservative. Typical plant disease detection techniques are often time-consuming and labour-intensive, making it important to replace them with automated systems. Therefore, IoT-based plant disease detection using enhanced Elman spike neural network together with Capuchin search optimisation algorithm (IoT-PDD-OEESNN) is proposed in this paper for detecting potato plant. The input data is preprocessed using altered phase preserving dynamic range compression (APPDRC) filtering model for extracting the leaf region of the image and also eliminates the noise and blur image. The proposed IoT-PDD-OEESNN approach is implemented in Python using certain metrics. The IoT-PDD-OEESNN method attains better accuracy of 30.12%, 26.75% and lower computation time of 27.18%, 26.29%, and 29.56% when analysed with the existing methods.