A Dual-Model Approach to Animal Image Classification: Logistic Regression and CNN

DOI

10.2991/978-94-6463-738-0_17How to use a DOI?

Keywords

Image Classification; Convolutional Neural Network (CNN); Logistic Regression; Animal-10 Dataset; Computer Vision; Binary Classification; Multi-class Classification; Deep Learning

Abstract

This study compares the presentation of Logistic Regression and CNN approaches for image classification using the Animal-10 dataset, which comprises ten distinct animal categories including horse, elephant, dog, butterfly, hen, cat, sheep, spider, cow, and squirrel. To classify photos as either dogs or non-dogs, it first used Logistic Regression for binary classification. CNN was then used for a more intricate multi-class classification. Preprocessing of the dataset included normalization, downsizing to 64x64 pixels, and grayscale conversion With an AUC score of just 0.57, the logistic regression model is weak at reliably identifying dog photos, and just 77.18% of occurrences are classified into the correct category. CNN, on the other hand, achieves the highest scores at 90.5% accuracy in multi-class classification to its several convolutional layers, max-pooling layers, and dense layers. he CNN was trained using a batch size of 32 over 10 epochs, with the Adam optimizer and categorical cross-entropy as the loss function. A test set consisting of 20% of the data was used to evaluate both models. Conclusion: CNNs are more suited for complicated image classification applications because they can automatically learn hierarchical features, even though logistic regression offers a simple solution for binary classification. According to the research, performance could be improved by using transfer learning strategies, data augmentation, and sophisticated structures.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - R. Rajesh Sharma
AU  - Jalaludin Khan
AU  - Akey Sungheetha
AU  - Sheila Mahapatra
AU  - G. S. Pradeep Ghantasala
PY  - 2025
DA  - 2025/06/22
TI  - A Dual-Model Approach to Animal Image Classification: Logistic Regression and CNN
BT  - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025)
PB  - Atlantis Press
SP  - 208
EP  - 217
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-738-0_17
DO  - 10.2991/978-94-6463-738-0_17
ID  - Sharma2025
ER  -