Comparative Analysis of Deep Learning Algorithms for Image Classification

Image classification is one of the highly explored fields in artificial intelligence and computer vision, due to its widespread applications such as medical image analysis, autonomous cars, geographical classification of satellite photos, and facial recognition. Generally, conventional image classif...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:2024 2nd International Conference on Signal Processing, Communication, Power and Embedded System (SCOPES) S. 1 - 6
Hauptverfasser: Chowdary, Ginjupalli Pranay, K, Rajakumar, Narendra Yalla, Sri Satya, Yadav, Varthala Charith, Kasetty, Sai Bhargav
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 19.12.2024
Schlagworte:
Accuracy
Classification algorithms
Data Augmentation
Data models
Deep learning
Deep Learning Models
Deep Neural Networks
Feature extraction
Image classification
Manuals
Multi-Scale Feature Extraction
Signal processing
Signal processing algorithms
Solid modeling
UC Merced land use dataset
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Image classification is one of the highly explored fields in artificial intelligence and computer vision, due to its widespread applications such as medical image analysis, autonomous cars, geographical classification of satellite photos, and facial recognition. Generally, conventional image classification models depend upon manual feature extraction, but due to the exponential rise in data volume, manual adjustment is no longer feasible or efficient. This difficulty has led to the adoption of deep learning algorithms, which automate feature extraction and provide benefits such as high processing capacity, efficient feature filtering, and improved classification speed and accuracy. This research compares multiple deeplearning techniques for image classification. The study examines the results, offering a fundamental analysis and discussion of the findings where CNN proved to perform better when compared to other algorithms like GAN, RNN, and DBN with the highest accuracy of 92.8%.
DOI:10.1109/SCOPES64467.2024.10991323