Image recognition in depth: comparative study of CNN and Pre-trained VGG16 architecture for classification tasks

Nanxiang Zhou

doi:10.1117/12.3026829

15 March 2024 Image recognition in depth: comparative study of CNN and Pre-trained VGG16 architecture for classification tasks

Nanxiang Zhou

Author Affiliations +

Proceedings Volume 13075, Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023); 130751Y (2024) https://doi.org/10.1117/12.3026829
Event: Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023), 2023, Kunming, China

Abstract

Image recognition and classification have found extensive applications in the field of artificial intelligence. This study conducts a comparative study of conventional Convolutional Neural Network (CNN) and the Visual Geometry Group Network (VGG16) to explore the effectiveness of model classification. The aim is to comprehensively understand the interactions among different hierarchical image feature representations in scenery classification tasks, the models' performance in complex feature learning, and non-linear expression capabilities. The experiments conducted on the "Intel Image Classification" dataset demonstrate the superior accuracy of the pre-trained VGG16 model over the CNNs model, particularly in terms of feature learning and generalization capabilities. Moreover, this comparative analysis approach enhances the understanding of the characteristics and suitable scenarios of different network architectures. It guides the selection, design, and optimization of deep learning models for practical image classification applications. Therefore, the study makes a significant contribution to the field of scene image classification and offers practical implications. It establishes a foundation for future research directions that will emphasize innovation in models, integration of multimodal data, and improvement in robustness and interpretability, aiming to drive breakthroughs and applications of image recognition and classification technology across various fields.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Nanxiang Zhou "Image recognition in depth: comparative study of CNN and Pre-trained VGG16 architecture for classification tasks", Proc. SPIE 13075, Second International Conference on Physics, Photonics, and Optical Engineering (ICPPOE 2023), 130751Y (15 March 2024); https://doi.org/10.1117/12.3026829

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Image classification

Data modeling

Education and training

RGB color model

Performance modeling

Deep learning

Neurons

Show All Keywords

Keywords/Phrases

Search In:

Publication Years