Automated Deep Learning with Wavelet Neural Network based Rice Plant Classification

In the agricultural sector, a disease that occurs in plants is primarily responsible for the decrease in production and results in massive financial loss. Rice is considered the crucial food crop in Asian nations and is affected by distinct types of diseases. Due to the arrival of deep learning (DL) and computer vision methods, rice plant ailments will be identified and diminish the problem of the agriculturalists to save the crops. Currently, computer aided diagnosis (CAD) methods become accessible to observe pests and crop diseases with the help of plant images. An automatic rice disease prognosis technique could present details regarding preventing and controlling rice ailments for decreasing the monetary loss, decline the insecticide residue, and increase the number and quality of yields. In order to attain this method, authors are advised to advance effective image processing methods for noticing plant diseases. Therefore, this paper presents an Automated Deep Learning with Wavelet Neural Network Based Rice Plant Classification model named ADLWNN model. The proposed ADLWNN model focuses on the effectual recognition and categorization of rice plant images. The proposed ADLWNN model primarily exploits convolutional neural network (CNN) model to extract features from the input rice plant images. Moreover, manta ray optimization algorithm (MRFO) algorithm is applied as a hyperparameter optimizer. Besides, the WNN model is employed for the robust recognition and categorization of rice plant images. The simulation analysis of the ADLWNN model is tested using a set of rice plant images and the results indicated as 98.17% better outcomes for the ADLWNN model over other techniques.