AI-Based Medicinal Plant Detection System

Abstract

Medicinal plants are really important for our health both in ways of healing and in modern hospitals.. Finding these plants can be very hard and it takes a long time a lot of work and you need to know a lot about plants. To make this easier we made a computer system that uses intelligence to identify medicinal plants by looking at pictures of their leaves. This system uses computer methods like Convolutional Neural Networks to look at the shape, texture, color and patterns on the leaves and sort them correctly. We also use a method called YOLO to find leaves even in complicated places. To make this work we take pictures of leaves clean them up with computer tools like OpenCV and then teach the computer to recognize them. We use a way of teaching the computer called transfer learning, which helps it learn faster with fewer pictures. We made a website where people can upload pictures of leaves and get information about the plant and its healing properties. This system is really good, for doctors, farmers and people who study plants because it is very accurate and works fast. Even though it is very good we still have problems to solve like getting types of pictures and making it work in different environments. In general our system shows how computers can help make finding medicinal plants easier more accurate and faster which can help us have healthcare and protect the environment. Medicinal plants are very important. Our system can help people who work with medicinal plants every day.