Advertisement

A Fast Summarization Method for Smartphone Photos Using Human-Perception Based Color Model

  • Kwanghwi Kim
  • Sung-Hwan Kim
  • Hwan-Gue Cho
Part of the Communications in Computer and Information Science book series (CCIS, volume 263)

Abstract

With an increasing number of smartphone user, people can easily take a hundreds of daily photos with their smartphone. However the growth of taken photos cause problems that the user are hard to browse, search and manage them. In this paper, we describe our spatial clustering method to enhance photo management in smartphone with considering perceptual color distribution. We address how to group nearly identical photos(NIP) taking duplicate photos in order to get a better quality photo. To measure perceptual differences between two photos, we conduct the CIELAB color metric and the optimal matching by dominant colors and specific colors. Also, we try to investigate the key features of NIP such as the similarity threshold and the number of dominant colors and specific colors. The result of experiments shows that our method enable to classify NIP groups similar to manual operation result and the average accuracy is 0.95.

Keywords

photo clustering CIELAB color-base clustering nearly identical photos 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Prasad, B.G., Biswas, K.K., Gupta, S.K.: Region-based image retrieval using integrated color, shape, and location index. J. Comput. Vis. Image Underst. 94(1-3), 193–233 (2004)CrossRefGoogle Scholar
  2. 2.
    Zhang, X., Wandell, B.A.: A spatial extension of CIELAB for digital color-image reproduction. J. SID 5(1), 61–63 (1997)Google Scholar
  3. 3.
    Cooper, M., Foote, J., Girgensohn, A., Wilcox, L.: Temporal event clustering for digital photo collections. J. ACM Trans. Multimedia Comput. Commun. Appl. 1(3), 269–288 (2005)CrossRefGoogle Scholar
  4. 4.
    Ryu, D., Kim, K., Park, S., Cho, H.: A web-based photo management system for large photo collections with user-customizable quality assessment. In: Proc. of the ACM SAC, pp. 1229–1236 (2011)Google Scholar
  5. 5.
    Ryu, D.-S., Kim, K., Cho, H.-G.: An Intelligent Clustering Method for Highly Similar Digital Photos Using Pyramid Matching with Human Perceptual 25 Color Histogram. In: Kim, T.-h., Adeli, H., Robles, R.J., Balitanas, M. (eds.) ISA 2011. CCIS, vol. 200, pp. 359–366. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  6. 6.
    Ryu, D., Chung, W., Cho, H.: PHOTOLAND: a new image layout system using spatio-temporal information in digital photos. In: Proc. of the ACM SAC, pp. 1884–1891 (2010)Google Scholar
  7. 7.
    Platt, J.C., Czerwinski, M., Field, B.A.: PhotoTOC: automatic clustering for browsing personal photographs. In: Proc. of the ICICS-PCM, pp. 6-10 (2003)Google Scholar
  8. 8.
    Graham, A., Garcia-Molina, H., Paepcke, A., Winograd, T.: Time as essence for photo browsing through personal digital libraries. In: Proc. of the 2nd ACM/IEEE-CS Joint Conference on Digital Libraries, pp. 326–335 (2002)Google Scholar
  9. 9.
    Jang, C., Cho, H.: A clustering method for concurrent photos obtained from multiple cameras using max-flow network model. J. Multimedia Systems, 1–23 (2011)Google Scholar
  10. 10.
    Mota, J., Fonseca, M.J., Gonçalves, D., Jorge, J.A.: Agrafo: a visual interface for grouping and browsing digital photos. In: Proc. of the AVI, pp. 494–495 (2008)Google Scholar
  11. 11.
    Blazica, B., Vladusic, D., Mladenic, D.: ShoeBox: A Natural Way of Organizing Pictures According to User’s Affinities. In: Proc. of the HCII, pp. 519–524 (2011)Google Scholar
  12. 12.

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Kwanghwi Kim
    • 1
  • Sung-Hwan Kim
    • 1
  • Hwan-Gue Cho
    • 1
  1. 1.Dept. of Computer Science and EngineeringPusan National UniversityBusanKorea

Personalised recommendations