Image Classification of Ornamental Plants Leaf using Machine Learning Algorithms

Ornamental Plant image recognition and distinguishing have become a versatile focus in botanical taxonomy and computer science. A decorative plant must be recognized by one or more features, such as size, leaf shape, blossom colour, odour, etc., and these traits must be linked to a name, either a standard or scientific name. It is imperative and hard to know about the Variety of ornamental plants to take care of it and grow around the vicinity, i.e., difficult for the common person. Only an expert or a person with prior knowledge can identify it. The identification of ornamental plants by common people is a significant problem. Ornamental plants can be identified using the Machine Learning (ML) approach through their leaf's images. The authors have collected the 25 categories (total RGB image 5608 of the ornamental plant's leaf. And deployed several machine learning algorithms like Multi-Layer Perceptron (MLP), Logistic Regression (LR), Naive Bayesian (NB) Support Vector Machine (SVM), Decision Tree (DT), and k-nearest neighbour(KNN)). The comparative analysis shows that MLP has performed the best with 93.90 % accuracy among other ML algorithms. And confusion matric of MLP shows offers the class-wise performance of the algorithm.