<![CDATA[The classification of footwear types, such as boots, sandals, and shoes, is a significant challenge in the development of image recognition systems powered by artificial intelligence. This study aims to compare the performance of two popular classification models, namely Convolutional Neural Network (CNN) and Support Vector Machine (SVM), in recognizing footwear types. The dataset used is the Footwear-Shoe vs Sandal vs Boot Image Dataset, consisting of 3000 images for each category with a resolution of 136x102 pixels in RGB format. The methodology includes training and testing both models using optimized parameters to measure accuracy, precision, and computational efficiency. The results show that CNN achieves an accuracy of 98%, while SVM reaches an accuracy of 96%. The findings indicate that CNN is more suitable for applications requiring high accuracy, while SVM is an effective alternative in resource-constrained scenarios. This study offers significant contributions to understanding model performance in image-based footwear classification using machine learning.]]>
Copyrights © 2025
