Plant Leaves Classification System

Plant leaves could be classified using machine learning techniques. The plant leaves could be used to identify the different types of the plants. Lot of plant leaves classification systems are developed using machine learning techniques. Support Vector Machine, Random Forest, K Nearest Neighbour, and Convolutional Neural Network algorithms could be used to develop the plant leaves classification systems. Colour, Shape, Texture, and Vein details could be used as the features to classify the plant leaves. Though we have lot of plant leaves classification systems, the accuracy of the classification systems is low. We use the shape, and the colour of the plant leaves as features to classify the plant leaves. We have collected the length, width, and colour data from five different plants, which are Jack fruit, Lemon, Nochi, Guava, and Cinnamon. Nearly fifty samples are collected from each type of the plant. Our classification system uses machine learning algorithms to identify the plants. The system uses K Nearest Neighbour, Support Vector Machine, and Random Forest algorithms. The system could be used to classify plant leaves in the agricultural industry, and Ayurvedic medicine industry. The accuracy, precision, recall, and F1 score of the models are obtained. We have achieved the highest prediction accuracy of 99% in the Random Forest algorithm. The local pattern of the plant species could be found in the first layer of the convolutional neural network model.