A Survey of High-Dimensional Similarity Search and Deep Feature-Based Indexing in Content-Based Image Retrieval
Keywords:
Content-Based Image Retrieval (CBIR), Convolutional Neural Networks (CNN), Deep learning, Facebook AI Similarity Search (FAISS), High-dimensional features, Hierarchical Navigable Small World (HNSW), Image Retrieval, Principal Component Analysis (PCA)Abstract
Content-Based Image Retrieval (CBIR) has progressed rapidly with advances in deep Convolutional Neural Networks (CNNs), high-dimensional feature embedding, and scalable similarity search frameworks. Traditional CBIR approaches rely on handcrafted features such as colour histograms, SIFT, and GLCM, which cannot capture semantic-level visual information, leading to inaccurate retrieval. Modern systems integrate CNN-based semantic embeddings, dimensionality reduction techniques like Principal Component Analysis (PCA), and efficient Approximate Nearest Neighbour (ANN) indexing methods such as FAISS (Facebook AI Similarity Search) and HNSW (Hierarchical Navigable Small World) to improve retrieval accuracy and scalability. This survey reviews 10 recent studies (2020–2025), with a strong focus on Indian contributions, covering deep feature extraction, hybrid CBIR models, indexing frameworks, and optimisation techniques. The survey also highlights existing research gaps and proposes a generalized framework combining CNN (Convolutional Neural Network) feature extraction, PCA compression, and ANN-based indexing.
References
A. Mathews, S. N, and V. K. R, “Analysis of Content Based Image Retrieval using Deep Feature Extraction and Similarity Matching,” International Journal of Advanced Computer Science and Applications, vol. 13, no. 12, 2022, doi: https://doi.org/10.14569/ijacsa.2022.0131277
M. Zubair, M. A. Alim, M. B. Ahmad, M. M. Alam, and M. Mohd Su’ud, “B-RISiC: A Balanced-Region Indexing Approach for Efficient Reverse Image Search in Collages,” IEEE Access, vol. 12, pp. 156915–156928, Jan. 2024, doi: https://doi.org/10.1109/access.2024.3486161
V. Lowanshi and K. Rai, “A small comparison of various content-based image retrieval approaches,” https://www.scienceopen.com/hosted-document?doi=10.14293/S2199-1006.1.SOR-.PP6PWQI.v1, Feb. 2023, doi: https://doi.org/10.14293/s2199-1006.1.sor-.pp6pwqi.v1
K. Lowanshi, S. Shrivastava, and V. Richhariya, “An Efficient Approach for Content based Image Retrieval using SVM, KNN-GA as Multilayer Classifier,” International Journal of Computer Applications, vol. 107, no. 21, pp. 43–48, Dec. 2014, Accessed: Dec. 24, 2025. Available: https://www.ijcaonline.org/archives/volume107/number21/19144-0558/
R. Praveen Kumar, G. Gautham Kumar, A. R. Arun, V. Vetriselvi, and P. M. Mohamed Ihsan, “A comprehensive security framework for cloud-based remote sensing image storage and retrieval with adversarial attack resistance,” International Journal of Computers and Applications, vol. 47, no. 9, pp. 766–790, Aug. 2025, doi: https://doi.org/10.1080/1206212x.2025.2545912
N. S. Osman and M. R. Mustaffa, “A Review on Content-Based Image Retrieval Representation and Description for Fish,” In 2015 4th International Conference on Advanced Computer Science Applications and Technologies (ACSAT), vol. 7, pp. 39–44, Dec. 2015, doi: https://doi.org/10.1109/acsat.2015.15
M. Momeny, A. M. Latif, M. Agha Sarram, R. Sheikhpour, and Y. D. Zhang, “A noise robust convolutional neural network for image classification,” Results in Engineering, vol. 10, p. 100225, Jun. 2021, doi: https://doi.org/10.1016/j.rineng.2021.100225
V. Lowanshi and S. Shrivastava, “Two Tier Architecture for Content Based Image Retrieval Using Modified SVM and knn-GA,” 2014. Accessed: Dec. 24, 2025. Available: https://www.ijcseonline.org/pub_paper/10-IJCSE-00540.pdf
M. Kashif, G. Raja, and F. Shaukat, “An Efficient Content-Based Image Retrieval System for the Diagnosis of Lung Diseases,” Journal of Digital Imaging, vol. 33, no. 4, pp. 971–987, May 2020, doi: https://doi.org/10.1007/s10278-020-00338-w
P. Elumalaivasan, S. Suthir, and S. Ravikumar, “CBIR: Retrieval of similar images using median vector algorithm,” In 2013 International Conference on Green Computing, Communication and Conservation of Energy (ICGCE), pp. 1–5, Dec. 2013, doi: https://doi.org/10.1109/icgce.2013.6823389
