Flower Identification by Deep Learning Approach and Computer Vision

Abstract

This study employed deep learning methods like VGG19, Xception, CNN, DenseNet201, and InceptionV3 to identify flowers. After applying these models, a confusion matrix was applied to evaluate the performances of the techniques. At an astounding 94% accuracy, the CNN model outperformed. VGG19 and DenseNet201, which came in at 82%. Inception V3 performed worst with 23% of accuracy. The culmination of the study is the accurate measurement of unknown blooms in real time and analysis of the result based on accuracy for different types of algorithms. The program demonstrated the usefulness of cutting-edge deep learning algorithms and functions as an efficient tool for smooth and dependable flower detection.