Automatic Feature Extraction Techniques

for

Content-based Image Archival and Retrieval

Image database systems use many low level features such as color, texture and shapes for content-based image archival and retrieval. Several techniques have been used for the automatic extraction of such features. This report will overview the different image features that are used to perform image archival and retrieval, the advantages and limitations of using these features in content-based image queries. The report will also consider some of the techniques used for extracting these features like color histograms, wavelet decomposition, etc.

Reading List (partial)

• J.R. Smith and S.F. Chang "Quad-Tree Segmentation for Texture-Based Image Query" Proceedings. ACM 2nd Multimedia Conference, San Francisco, Oct. 1994.
  
• Charles E. Jacobs, A. Finkelstein and D.H. Salesin "Fast Multiresolution Image Querying" Proceedings of SIGGRAPH 95.
  
• S.F. Chang , J.R. Smith, J. Meng "Efficient Techniques for Feature-Based Image/Video Access and Manipulation" Proceedings, 33rd Annual Clinc on Library Applications of Data Processing.
  
• W.Dsu, Chua T.S., and Pung H.K. " An Integrated Color-Spatial Approach to Content-based Image Retrieval" ACM 3rd Multimedia Conference, San Francisco, 1995.
  
• J.R. Smith and S.F. Chang "Automated Binary Texture Features Sets for Image Retrieval." Proceedings, Int. Conf. On Acoust. Speech and Signal Processing. (ICASSP). May, 1996. IEEE.
  
• A.Vellaikal and C.C.Jay Kuo "Content-Based Retrieval Using Multiresolution Histogram Representation."