Publication

Abstract : Agriculture plays a vital role in sustaining human life, providing the food and resources needed for survival. With the advent of digitalization, technology has permeated every aspect of our lives, including agriculture. The potato, a versatile tuber, is a widely enjoyed and commonly consumed staple in diets across the globe. To ensure a prosperous potato production, establishing a robust food security system becomes imperative, given its high nutritional value in terms of vitamins and minerals. Nonetheless, it is crucial to address diseases affecting potatoes including early blight and late blight. The manual assessment of these leaf diseases is labor-intensive and inconvenient. In the present investigation, we have implemented a system that utilizes a combination of machine learning and deep learning methodologies to classify two distinct types of diseases in potato plants, relying on the assessment of their leaf conditions. We utilized a range of models including support vector machine, naive bayes, K-Nearest Neighbor, Decision Tree, Random Forest, convolutional neural network, Sequential2, and VGG 16 to establish a highly accurate classification system. Notably, our experiment yielded an impressive accuracy of 95.36% within the initial 10 epochs of VGG 16 training, affirming the viability of the deep neural network approach. This research has made a substantial contribution to the agricultural sector and has provided valuable insights for farmers to effectively classify Potato leaf Disease, yielding optimal results.