In the realm of agriculture, where the importance of sustainable practices looms large, the timely detection of diseases in crop production emerges as a critical factor for increasing productivity. However, Due to the scarcity of technologies used in crops, the diagnosis of diseases and pests is largely supported by human inspection, generating errors caused by the subjectivity of individuals. Even if disease is detected, finding the optimal treatment among the thousands of fertilizers and agricultural medicines is a challenging task.
Our goal is to develop a plant disease detection system that utilizes a subfield of artificial intelligence known as Computer Vision to solve this issue. This system can predict plant disease by comparing the images of plants and which those in the database then recommend the best treatment.The system utilizes an algorithm to analyze and process the captured image to classify the plant disease as: The diagnostic system is composed of image acquisition, image preprocessing, segmentation, feature extraction, feature selection, and subsequent classification of disease. The system can recognize the condition by utilizing deep computer vision with convolutional neural networks (CNN) which is particularly well-suited for image recognition and processing tasks. By utilizing a smartphone camera or a captured image, recommendations can be made to the agricultural shop where they will be available. on the other hand, the application will help the farmers with the information needed to treat the detected diseases produced by the image processing model. This solution aims to provide a versatile, scalable, approach to plant disease.
Overall, the system promises to enhance the process of detecting plant diseases and pests, while also providing farmers with information to enhance crop productivity, which will lead to a decrease in financial losses.