FLOWER RECOGNITION USING VGG16 AND K-NEAREST NEIGHBORS

Abstract

This study investigates the application of the VGG16 model in combination with the K-Nearest
Neighbors (KNN) algorithm for flower classification based on morphological features. The Flowers 102
dataset was utilized, accompanied by data preprocessing, model training, and performance evaluation
across different values of the parameter k. Experimental results confirm the model’s accuracy, highlighting
the feasibility and effectiveness of using KNN for addressing basic image classification tasks. Moreover,
this approach demonstrates potential for integration into STEM education and practical artificial
intelligence training in academic settings.