Abstract

Pest detection systems have lately become more sophisticated in order to boost yield productivity and quality. Disease identification is critical for increasing crop yield, however some diseases are difficult to detect early on by farmers, and the crop suffers as a result. As a result, pest detection is critical in order to prevent plant deterioration and increase product quality. Using image processing algorithms such as the K-means algorithm, a pest identification system will be developed. It will classify the leaf image based on criteria such as colour, texture, and their combinations to train three support vector machine classifier models. The dataset will be a trained dataset made up of thousands of photos gathered from various villages and farms. The algorithm will use a trained dataset to determine which leaves are healthy and which are unhealthy. It will provide various photographs of the diseased leaf as well as the healthy leaf. The system will be simpler to use because it only requires an image of a plant as input and returns results such as healthy and unhealthy plants. As a result, farmers may readily use the system to increase crop quality.Keywords: plant disease, image processing, android, etc