[go: up one dir, main page]
More Web Proxy on the site http://driver.im/ skip to main content
research-article

Unsupervised Multiresolution Segmentation for Images with Low Depth of Field

Published: 01 January 2001 Publication History

Abstract

Unsupervised segmentation of images with low depth of field (DOF) is highly useful in various applications including image enhancement for digital cameras, target recognition, image indexing for content-based retrieval, and 3D microscopic image analysis. This paper describes a novel multiresolution image segmentation algorithm for low DOF images. The algorithm is designed to separate a sharply focused object-of-interest from other foreground or background objects. The algorithm is fully automatic in that all parameters are image independent. A multiscale approach based on high frequency wavelet coefficients and their statistics is used to perform context-dependent classification of individual blocks of the image. Unlike other edge-based approaches, our algorithm does not rely on the process of connecting object boundaries. The algorithm has achieved high accuracy when tested on more than 100 low DOF images, many with inhomogeneous foreground or background distractions. Compared with the state of the art algorithms, this new algorithm provides better accuracy at higher speed.

References

[1]
A. Adams, The Camera. Boston: New York Graphic Soc., 1980.
[2]
Y. Boykov and D. Huttenlocher, “A New Baysian Framework for Object Recognition,” Proc. DARPA IU Workshop, 1998.
[3]
C. Carson S. Belongie H. Greenspan and J. Malik, “Blobworld: Image Segmentation Using Expectation-Maximization and Its Application to Image Querying,” submitted for publication.
[4]
T. Darell and K. Wohn, “Depth from Focus Using a Pyramid Architecture,” Pattern Recognition Letters, vol. 11, no. 12, pp. 787-96, Dec. 1990.
[5]
I. Daubechies, Ten Lectures on Wavelets. Philadelphia: SIAM, 1992.
[6]
C. Faloutsos R. Barber M. Flickner J. Hafner W. Niblack D. Petkovic and W. Equitz, “Efficient and Effective Querying by Image Content,” J. Intelligent Information Systems, vol. 3, pp. 231-262, 1994.
[7]
R.G. Gonzalez and R.E. Woods, Digital Image Processing. Reading, Mass.: Addison-Wesley, 1992.
[8]
A. Gupta and R. Jain, “Visual Information Retrieval,” Comm. ACM, vol. 40, pp. 69-79, 1997.
[9]
A. K. Jain R. C. Dubes, Algorithms for Clustering Data. Englewood Cliffs, N.J.: Prentice-Hall, 1988.
[10]
J. Li and R.M. Gray, “Context-Based Multiscale Classification of Images,” Proc. IEEE Int'l Conf. Image Processing, Oct. 1998
[11]
E. Krotkov, Active Computer Vision by Cooperative Focus and Stereo. New York: Springer-Verlag, 1989.
[12]
J. Li J.Z. Wang and G. Wiederhold, “IRM: Integrated Region Matching for ImageRetrieval,” Proc. 2000 ACM Multimedia Conf., Oct. 2000.
[13]
W.Y. Ma and B.S. Manjunath, “Edge Flow: A Framework of Boundary Detection and Image Segmentation,” Proc. Computer Vision and Pattern Recognition, pp. 744-749, 1997.
[14]
S.K. Nayar and Y. Nakagawa, “Shape from Focus,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 16, no. 8, pp. 824-831, Aug. 1994.
[15]
M. Noguchi and S.K. Nayar, “Microscopic Shape from Focus Using a Projected Illumination Pattern,” Math. and Computer Modelling, vol. 24, nos. 5/6, pp. 31-48, Sept. 1996.
[16]
J. Shi and J. Malik, “Normalized Cuts and Image Segmentation,” Proc. Computer Vision and Pattern Recognition, pp. 731-737, 1997.
[17]
Basic Photographic Materials and Processes, L. Stroebel, ed. Boston: Focal Press, 1990.
[18]
M. Subbarao T. Yuan and J.-K. Tyan, “Integration of Defocus and Focus Analysis with Stereo for 3D Shape Recovery,” Three-Dimensional Imaging and Laser-Based Systems for Metrology and Inspection III, pp. 14-15, Oct. 1997.
[19]
D.-M. Tsai and H.-J. Wang, “Segmenting Focused Objects in Complex Visual Images,” Pattern Recognition Letters, vol. 19, no. 10, pp. 929-940, Aug. 1998.
[20]
J.Z. Wang G. Wiederhold O. Firschein and X.W. Sha, “Content-Based Image Indexing and Searching Using Daubechies' Wavelets,” Int'l J. Digital Libraries, vol. 1, no. 4, pp. 311-328, 1998.
[21]
J.Z. Wang J. Li D. Chan and G. Wiederhold, “Semantics-Sensitive Retrieval for Digital Picture Libraries,” D-LIB Magazine, vol. 5, no. 11, Nov. 1999. http://www.dlib.org.
[22]
C. Yim and A.C. Bovik, “Multiresolution 3D Range Segmentation Using Focus Cues,” IEEE Trans. Image Processing, vol. 7, no. 9, pp. 1283-1299, 1998.

Cited By

View all
  • (2023)Shot Boundary Detection Using Color Clustering and Attention MechanismACM Transactions on Multimedia Computing, Communications, and Applications10.1145/359592319:6(1-23)Online publication date: 12-Jul-2023
  • (2020)Computational Approaches to Aesthetic Quality Assessment of Digital Photographs: State of the Art and Future Research DirectivesPattern Recognition and Image Analysis10.1134/S105466182004008230:4(593-606)Online publication date: 1-Oct-2020
  • (2019)Single image defocus estimation by modified gaussian functionTransactions on Emerging Telecommunications Technologies10.1002/ett.361130:6Online publication date: 18-Jun-2019
  • Show More Cited By

Recommendations

Reviews

Toshiro Kubota

This paper describes an algorithm for segmenting a low depth of field (DOF) image into two regions: foreground and background. The underlying assumption is that an input image consists of focused foreground and blurred background, and the foreground has enough details so that it can be distinguished from the blurred background with high frequency features. The authors claim that the wavelet coefficients of detail channels are good features for the purpose. The segmentation is done in a multi-resolution fashion. First, the image is partitioned into a set of sub-blocks and these sub-blocks are classified using K-mean algorithm. Variances of wavelet coefficients in the detail channels are used as features for each block. At sub-sequence stages, each block is sub-divided into four smaller blocks and blocks at the segmentation boundary are re-classified to refine the segmentation map. The overall algorithm is simple and fast. According to the paper, segmentation of a 768x512 image took less than two seconds on 300MHz Pentium. However, judging from their experimental results, the accuracy of the segmentation is not outstanding; the classification error is 16% with one of test images. Like other multi-resolution algorithms, blocking effects appear in many of their experiments. It would have been useful if the authors could provide discussion on when the low DOF assumption is valid and their algorithm is applicable.

Access critical reviews of Computing literature here

Become a reviewer for Computing Reviews.

Comments

Please enable JavaScript to view thecomments powered by Disqus.

Information & Contributors

Information

Published In

cover image IEEE Transactions on Pattern Analysis and Machine Intelligence
IEEE Transactions on Pattern Analysis and Machine Intelligence  Volume 23, Issue 1
January 2001
96 pages
ISSN:0162-8828
Issue’s Table of Contents

Publisher

IEEE Computer Society

United States

Publication History

Published: 01 January 2001

Author Tags

  1. Content-based image retrieval
  2. image region segmentation
  3. low depth-of-field
  4. multiresolution image analysis.
  5. wavelet

Qualifiers

  • Research-article

Contributors

Other Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

  • Downloads (Last 12 months)0
  • Downloads (Last 6 weeks)0
Reflects downloads up to 01 Jan 2025

Other Metrics

Citations

Cited By

View all
  • (2023)Shot Boundary Detection Using Color Clustering and Attention MechanismACM Transactions on Multimedia Computing, Communications, and Applications10.1145/359592319:6(1-23)Online publication date: 12-Jul-2023
  • (2020)Computational Approaches to Aesthetic Quality Assessment of Digital Photographs: State of the Art and Future Research DirectivesPattern Recognition and Image Analysis10.1134/S105466182004008230:4(593-606)Online publication date: 1-Oct-2020
  • (2019)Single image defocus estimation by modified gaussian functionTransactions on Emerging Telecommunications Technologies10.1002/ett.361130:6Online publication date: 18-Jun-2019
  • (2018)Focused-Region Segmentation for Refocusing Images from Light FieldsJournal of Signal Processing Systems10.1007/s11265-018-1379-290:8-9(1281-1293)Online publication date: 1-Sep-2018
  • (2016)A Digitalized Recomposition Technique Based on Photo Quality Evaluation CriteriaWireless Personal Communications: An International Journal10.1007/s11277-015-2977-y86:1(301-314)Online publication date: 1-Jan-2016
  • (2016)Photo quality enhancement by relocating subjectsCluster Computing10.1007/s10586-016-0547-z19:2(939-948)Online publication date: 1-Jun-2016
  • (2015)A survey of image compression methods for low depth-of-field images and image sequencesMultimedia Tools and Applications10.1007/s11042-014-2032-074:18(7943-7956)Online publication date: 1-Sep-2015
  • (2013)Region-of-interest extraction in low depth of field images using ensemble clustering and difference of Gaussian approachesPattern Recognition10.1016/j.patcog.2013.03.00646:10(2685-2699)Online publication date: 1-Oct-2013
  • (2011)Spot detection in images with noisy backgroundProceedings of the 16th international conference on Image analysis and processing: Part I10.5555/2042620.2042690(575-584)Online publication date: 14-Sep-2011
  • (2011)Multi-resolution segmentation of high-resolution remotely sensed imagery using marker-controlled watershed transformProceedings of the International Conference & Workshop on Emerging Trends in Technology10.1145/1980022.1980167(674-678)Online publication date: 25-Feb-2011
  • Show More Cited By

View Options

View options

Media

Figures

Other

Tables

Share

Share

Share this Publication link

Share on social media