Fuzzy color quantization and its application to scene change detection

ABSTRACT

It is both impractical and unnecessary to use true color, i.e., 24-bit colors, for most multimedia processing tasks since the computational complexity is extremely high and adjacent colors in the color model do not contribute much to visual and computational differences. Color quantization can reduce these complexities by using less number of bits to represent the 24-bit true color space. In this paper, we investigate the fuzzy color quantization technique based on the fact that the colors located nearby should not be quantized very differently. We introduce this technique to color histogram construction and show that it achieves better performance than uniform color quantization in video scene change detection using different types of movies.

REFERENCES

• 1: Balasubramanian, R., Bouman, C. A. and Allebach, J. P. Sequential Scalar Quantization of Vectors: An Analysis. IEEE Transactions on Image Processing, 4(9), 1282--1295.
• 2: Chaddha, N., Tan, W. C. and Meng, T. H. V. Color quantization of images based on human vision perception. Proc. of IEEE International Conference on Acoustics, Speech, and Signal Processing, 1994, ICASSP-94, 5, 89--92.
• 3: J. Furlani , L. McMillan , L. Westover, Adaptive colormap selection algorithm for motion sequences, Proceedings of the second ACM international conference on Multimedia, p.341-ff., October 15-20, 1994, San Francisco, California, United States [doi>10.1145/192593.192695]
• 4: Gu, E., Wang, J., Xu, D. and Chen, C. Perceptually Based Approach to Color Quantization. Proc. of International Symposium on Multispectral Image Processing and Pattern Recognition, ISMIPPR'2001, 4552, 292--297.
• 5: Gunsel, B. and Tekalp, A. M. Content-based video abstraction. Proc. of International Conference on Image Processing, 1998, 3, 128--132.
• 6: Paul Heckbert, Color image quantization for frame buffer display, ACM SIGGRAPH Computer Graphics, v.16 n.3, p.297-307, July 1982
• 7: Lienhart, R. Comparison of automatic shot boundary detection algorithms. Proc. of SPIE Storage and Retrieval for Image and Video Databases VII, 3656, 290--301.
• 8: Linde, Y., Buzo, A. and Gray, R. M. An algorithm for vector quantizer design. IEEE Transactions on Communications, 28, 84--95.
• 9: Oh, J. H. and Hua, K. A. An efficient technique for summarizing videos using visual contents. Proc. of International Conference on Multimedia and Expo 2000, 2, 1167--1170.
• 10: Park, Y. W., Lee, E. J., Kim, K. M., Lee, C. S. and Ha, Y. H. Color Enhancement by New Quantization of Fast Clustering Method. Proc. of Imaging Science and Technology: Evolution and Promise, 275--81.
• 11: Evgeny Roytman , Craig Gotsman, Dynamic Color Quantization of Video Sequences, IEEE Transactions on Visualization and Computer Graphics, v.1 n.3, p.274-286, September 1995 [doi>10.1109/2945.466721]
• 12: Sharma, G. and Trussell, H. J. Digital Color Imaging. IEEE Transactions on Image Processing, 6(7), 901--932.
• 13: Velzo, L., Gomes, J. and Sobreiro, M. Color Image Quantization by Pairwise Clustering. Proc. of the Brazilian Symposium of Computer Graphics and Image Processing, SIBGRAPI 97, 203--210.
• 14: Veprek, P. and Bradley, A. B. An Improved Algorithm for Vector Quantizer Design. IEEE Signal Processing Letters, 7(9), 250--252.
• 15: Zhigang Xiang , Gregory Joy, Color Image Quantization By Agglomerative Clustering, IEEE Computer Graphics and Applications, v.14 n.3, p.44-48, May 1994 [doi>10.1109/38.279043]
• 16: Contextual Color Quantization Algorithm, Proceedings of the 11th International Conference on Image Analysis and Processing, p.596, September 26-28, 2001

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.8 Scene Analysis
          Subjects: Color

Additional Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.3 Deduction and Theorem Proving
          Subjects: Uncertainty, "fuzzy," and probabilistic reasoning
  I.4 IMAGE PROCESSING AND COMPUTER VISION
      I.4.1 Digitization and Image Capture
          Subjects: Quantization

General Terms:
Algorithms, Design, Experimentation, Performance, Theory

Keywords:
color histogram, color quantization, scene change detection, video shot