The exponential growth of digital visual data necessitates efficient and accurate Content-Based Image Retrieval (CBIR) systems. Traditional systems relying on handcrafted features often fail to capture the complex semantic meaning of images, resulting in a “semantic gap.” This paper proposes an Artificial Intelligence-based image retrieval system leveraging Deep Convolutional Neural Networks (CNNs) for automated hierarchical feature extraction. By utilizing a pre-trained ResNet-50 architecture, the system extracts high-dimensional feature vectors that accurately represent the semantic context of images. We compare the proposed AI-driven model against traditional techniques (SIFT and HOG) using the standard Corel-10k benchmark dataset. Experimental results demonstrate that the AI-based system significantly outperforms traditional methods, achieving a Mean Average Precision (map) of 91.2%. This paper details the methodology, comparison techniques, and quantitative results, proving the scalability and superior accuracy of deep learning in modern image retrieval.
This paper successfully demonstrates the implementation and superiority of an Artificial Intelligence-based image retrieval system. By replacing traditional handcrafted features with deep hierarchical representations extracted via CNNs (ResNet-50), the system effectively bridges the semantic gap. Our quantitative analysis reveals that the AI-driven approach yields a Mean Average Precision of 91.2%, outperforming legacy systems by a margin of nearly 30% while also reducing query retrieval time. Future research will explore the integration of Vision Transformers (ViTs) to capture global image context and the implementation of Approximate Nearest Neighbor (ANN) search algorithms to handle billion-scale image databases.
References
1.LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436-444.
2.Smeulders, A. W., Worring, M., Santini, S., Gupta, A., & Jain, R. (2000). Content-based image retrieval at the end of the early years. IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(12), 1349-1380.
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 (pp. 770-778).
4.Babenko, A., Slesarev, A., Chigorin, A., & Lempitsky, V. (2014). Neural codes for image retrieval. In Computer Vision–ECCV 2014 (pp. 584-599). Springer International Publishing.
5.Gordo, A., Almazán, J., Revaud, J., & Larlus, D. (2016). Deep image retrieval: Learning global representations for image search. In European conference on computer vision (pp. 241-257). Springer, Cham.
Latif, A., Rasheed, A., Sajid, U., Jameel, A., Ali, N., Rassi, D., & … & Zafar, T. I. (2019). Content-based image retrieval and feature extraction: a comprehensive review. Mathematical Problems in Engineering, 2019.
How to Cite This Paper
Dr.S.Thilagavathi (2026). AN ARTIFICIAL INTELLIGENCE-BASED CONTENT IMAGE RETRIEVAL SYSTEM. International Journal of Computer Techniques, 13(2). ISSN: 2394-2231.
