Authors : Nitesh Bhagat; Vasant; Payal Chandrakar

Volume/Issue : Volume 10 - 2025, Issue 5 - May

---

Google Scholar : https://tinyurl.com/vtm5ju76

DOI : https://doi.org/10.38124/ijisrt/25may2243

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

---

Abstract : This study investigates the application of AI-powered image recognition systems utilizing Convolutional Neural Networks (CNNs) and transfer learning. Leveraging benchmark datasets (ImageNet, CIFAR-10, MNIST), we evaluate model accuracy, precision, recall, and F1-score. Our findings reveal that deep learning architectures, especially transfer learning models like ResNet50 and InceptionV3, achieve high accuracy in object classification. However, concerns about data bias and interpretability remain. This paper emphasizes ethical deployment and outlines pathways for improving fairness and robustness in image recognition systems.

References :

1. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep Learning. MIT Press. A foundational textbook that provides comprehensive coverage of the principles and applications of deep learning.
2. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90. This landmark paper introduced AlexNet, a deep CNN that significantly advanced image recognition performance.
3. He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR),  770–778. The paper introduces ResNet, a deep learning architecture that solved the vanishing gradient problem in very deep networks.
4. Simonyan, K., & Zisserman, A. (2014). Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556. VGGNet is another influential CNN architecture known for its simplicity and effectiveness.
5. Dosovitskiy, A., et al. (2020). An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929. Introduces Vision Transformers (ViT), a novel architecture applying transformers to image data.
6. Selvaraju, R. R., Cogswell, M., Das, A., et al. (2017). Grad-CAM: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE International Conference  on  Computer  Vision  (ICCV),  618–626. Discusses techniques for interpretability in deep learning models.
7. Zeiler, M. D., & Fergus, R. (2014). Visualizing and understanding convolutional networks. In European Conference on Computer Vision (ECCV), 818–833. Provides insights into the internal workings of CNNs, contributing to model interpretability.
8. Deng, J., et al. (2009). ImageNet: A large-scale hierarchical image database. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 248–255.
9. Introduces the ImageNet dataset, which has been fundamental to the progress of image recognition research.
10. Dutta, A., Gupta, A., & Zisserman, A. (2019). VGG Image Annotator (VIA). arXiv preprintarXiv:1904.10699.
11. A useful tool for manually annotating image datasets used in image recognition projects.