Effective prevention and control of plant disease are made by early detection and identification of plant disease. To identify early disease detection and numerous prevailing methodologies were utilized. Some of the limitations were exhibited by these models so that the disease cannot be predicted accurately and on time. By using a Cauchy Logarithmic Linear Convolutional Neural Network Classification Algorithm, an efficient plant disease detection identification of hyperspectral image can classify the plant leaves into healthy leaves, diseased leaves are the ultimate goal of this work. Preprocessing the input leaf images is eventuated initially. The Gaussian filter eliminated the noises in the above process. The Adaptive Histogram Equalization elevated the image contrasts. Then, the Multi-threshold Watershed algorithm was utilized to segment the preprocessed images. The most optimal features are retrieved as of the segmented images. The Chaotic Bifurcation Woodpecker Mating Optimization algorithm is utilized to select the most effective features to abate the classification processes’ complexity. Cauchy Logarithmic Linear Convolutional Neural Network is a classifier where the chosen features are inputted to classify the plant leaves significantly. By utilizing certain quality parameters, the HI is analyzed.