Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis

Hejun Yang

doi:10.47297/taposatWSP2633-456917.20230402

PDF
导出
分享
收藏
专辑

Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis

4卷第2期

浏览量: 0 Downloads: 2

Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis
Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis
Faculty of Information Technology, Beijing University of Technology,Beijing,China,100124
2023年4卷第2期页码：109-117
纸质出版日期： 2023 ，
DOI：10.47297/taposatWSP2633-456917.20230402
Accepted：

Scan QR Code

(2023). Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis. 科技理论与实践（英文版）, 4(2), 109-117.

HEJUN YANG. (2023). Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis. Theory and Practice of Science and Technology, 4(2), 109-117..
(2023). Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis. 科技理论与实践（英文版）, 4(2), 109-117. DOI： 10.47297/taposatWSP2633-456917.20230402.

HEJUN YANG. (2023). Empirical Advancements in Convolutional Neural Networks for Image Recognition: A Comparative Analysis. Theory and Practice of Science and Technology, 4(2), 109-117.. DOI： 10.47297/taposatWSP2633-456917.20230402.

摘要

Abstract

This paper presents a comprehensive comparison of various Convolutional Neural Networks (CNN) architectures employed in image recognition tasks. Utilizing the CIFAR-10 and ImageNet datasets

several prominent CNN architectures

including AlexNet

VGGNet

GoogLeNet

and ResNet

were rigorously tested and evaluated. The aim was to gain insights into their performance variations

understanding their strengths and weaknesses in relation to the complexity of image recognition tasks. The findings from this comparative analysis affirm that the choice of CNN architecture should be dictated by task-specific requirements

taking into account factors like computational resources

complexity of images

and the desired trade-off between accuracy and efficiency. Furthermore

the paper sheds light on the potential areas of improvement in existing CNN architectures and offers actionable recommendations for researchers and practitioners in the field. The choice of a CNN architecture should be guided by a multifaceted evaluation encompassing computational resources

the complexity of the images

and the balance between accuracy and efficiency. Researchers and practitioners should weigh these factors carefully to make an informed decision tailored to the specific requirements of their tasks.

关键词

Keywords

Convolutional Neural NetworksImage recognitionComparative analysisCNN architecturesMachine learningDeep learning

references

Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009). ImageNet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition (pp. 248-55). IEEE.[2] Goodfellow, I. J., Shlens, J., & Szegedy, C. (2014). Explaining and harnessing adversarial examples. arXiv preprint, arXiv:1412.6572.[3] Gupta, A., Vedaldi, A., & Zisserman, A. (2021). Synthetic data for learning deep models: a survey. Pattern Recognition Letters, 140, 251-58.[4] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 770-78.[5] Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., ... & Adam, H. (2017). MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications. arXiv preprint, arXiv:1704.04861.[6] Iandola, F. N., Han, S., Moskewicz, M. W., Ashraf, K., Dally, W. J., & Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and< 0.5 MB model size. arXiv preprint, arXiv:1602.07360.[7] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 25, 1097-1105.[8] Krizhevsky, A., & Hinton, G. (2009). Learning multiple layers of features from tiny images. Technical report, University of Toronto.[9] Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint, arXiv:1409.1556.[10]Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., ... & Rabinovich, A. (2015). Going deeper with convolutions. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1-9.[11]Sze, V., Chen, Y. H., Yang, T. J., & Emer, J. S. (2017). Efficient processing of deep neural networks: A tutorial and survey. Proceedings of the IEEE, 105(12), 2295-2329.