Performance Juxtapose of Plant Leaf Disease Detection using Adaptive Deep Convolutional Recurrent Neural Network (ADCRNN) in MATLAB Versus Python

Leaf diseases can cause several detriments in crops' overall yield and fertility. While analyzing the various diseases that affect plants is imperative, identifying the diseases using algorithmic techniques that render optimal performance is crucial. In agriculture, detecting multiple diseases in plants is difficult. However, the provision to automate disease identification through machine learning approaches through the various phases of preprocessing and segmentation is implemented in this study. This indagation delineates the leaf disease identification using the proposed hybrid algorithmic Adaptive Deep Convolutional Recurrent Neural Network (ADCRNN) in bifurcated platforms such as MATLAB and Python. The Radial basis function is incorporated to optimize the throughput in MATLAB before observing the juxtaposed results in both platforms. The purpose of utilizing ADCRNN is to effectively combine the advantages and parallelly overcome the challenges of Deep Convolutional Neural Network (DCNN), RNN, and adaptive techniques that unsheathe the pivotal features from the provided input. The ADCRNN simulation is carried out in MATLAB and Python, and the results are successfully obtained. In addition, after optimization, the proposed ADCRNN technique’s accuracy increased to 94.5% in performance evaluation in MATLAB.