Assessing the Effectiveness of CNNs, Decision Tree, and Voting Classifier for Plant Identification: A Comparative Study

This study employs a comparative study of Convolutional Neural Networks (CNN), Decision Trees (DT), and Voting Classifiers (VC) to identify different plants based on leaf images. To evaluate the effectiveness of the CNN model, we compared DT and VC consisting of six other classifiers: [14]Support Vector Machine (SVM), Random Forests (RF), Gaussian Naive Bayes (GNB), K-Nearest Neighbors (KNN)[14], and two ensemble techniques, Bagging and Adaboost. These models were trained and tested using publicly available input data of leaf images from 12 classes of plant species. The output of this study revealed that the CNN model is the most accurate model for identifying the class of the plant based on leaf images, with a 94.44 % accuracy rate, followed by the VC model with 91.66% accuracy, and the DT model with a comparable level of accuracy in terms of performance with an accuracy of 79.16%.