Image-Based Plant Disease Detection and Classification Using Deep Convolution Neural Network

Abstract

Agriculture is a country’s economic backbone. Numerous studies have shown that plant diseases are challenging to manage because their populations vary with environmental factors. Plants can become infected with a wide range of diseases, including fungal, bacterial, and viral infections. It has been discovered that fungal-like creatures infect 85% of plants. In developing countries, farmers use a more labor-intensive and time-consuming method. Manual detection or observation with the naked eye is similarly unlikely to generate useful data. Many farmers have also been observed using pesticides to reduce the effects of disease without first recognizing the exact ailment. Farmers use pesticides in an unrestrained manner, which can have negative consequences on the plant as well as human health. As a result, through the use of several machine learning approaches, the machine learning model aids in the identification of plant diseases. In this, we used image processing methods and as well as convolutional neural networks (CNNs). This research’s accuracy is 94.56%. The acquired results demonstrated that the proposed solution is usable and can be used by the farmers to identify plant diseases effectively.