This paper provides a comprehensive survey of the technical achievements in the field of image retrieval, particularly content-based image retrieval, which has seen significant advancements in recent years. The survey covers over 100 papers on image feature representation and extraction, multidimensional indexing, and system design, the three fundamental aspects of content-based image retrieval. The authors identify open research issues and suggest promising future directions based on current technology and real-world application demands. 1. **Feature Extraction**: The paper discusses various visual features such as color, texture, and shape, and their representations and matching techniques. It highlights the importance of feature extraction in content-based image retrieval. 2. **High Dimensional Indexing**: To make content-based image retrieval scalable to large collections, efficient multidimensional indexing techniques are essential. The paper explores dimension reduction methods and various indexing algorithms, including bucketing, k-d trees, and R-trees. 3. **Image Retrieval Systems**: The paper reviews several commercial and research image retrieval systems, including QBIC, Virage, RetrievalWare, Photobook, VisualSEEk, WebSEEK, Netra, and MARS, highlighting their unique features and characteristics. 4. **Future Research Directions**: The paper identifies several open research issues and suggests future directions, including integrating human interaction, linking low-level visual features to high-level concepts, web-oriented systems, high-dimensional indexing, performance evaluation criteria, and understanding human perception of image content. The paper emphasizes the need for a well-balanced large-scale testbed to evaluate image retrieval systems and the importance of integrating human perception into image retrieval systems to improve performance.This paper provides a comprehensive survey of the technical achievements in the field of image retrieval, particularly content-based image retrieval, which has seen significant advancements in recent years. The survey covers over 100 papers on image feature representation and extraction, multidimensional indexing, and system design, the three fundamental aspects of content-based image retrieval. The authors identify open research issues and suggest promising future directions based on current technology and real-world application demands. 1. **Feature Extraction**: The paper discusses various visual features such as color, texture, and shape, and their representations and matching techniques. It highlights the importance of feature extraction in content-based image retrieval. 2. **High Dimensional Indexing**: To make content-based image retrieval scalable to large collections, efficient multidimensional indexing techniques are essential. The paper explores dimension reduction methods and various indexing algorithms, including bucketing, k-d trees, and R-trees. 3. **Image Retrieval Systems**: The paper reviews several commercial and research image retrieval systems, including QBIC, Virage, RetrievalWare, Photobook, VisualSEEk, WebSEEK, Netra, and MARS, highlighting their unique features and characteristics. 4. **Future Research Directions**: The paper identifies several open research issues and suggests future directions, including integrating human interaction, linking low-level visual features to high-level concepts, web-oriented systems, high-dimensional indexing, performance evaluation criteria, and understanding human perception of image content. The paper emphasizes the need for a well-balanced large-scale testbed to evaluate image retrieval systems and the importance of integrating human perception into image retrieval systems to improve performance.