DOI QR코드

DOI QR Code

Medical Image Classification and Retrieval Using BoF Feature Histogram with Random Forest Classifier

Random Forest 분류기와 Bag-of-Feature 특징 히스토그램을 이용한 의료영상 자동 분류 및 검색

  • Received : 2012.12.05
  • Accepted : 2013.01.09
  • Published : 2013.04.30

Abstract

This paper presents novel OCS-LBP (Oriented Center Symmetric Local Binary Patterns) based on orientation of pixel gradient and image retrieval system based on BoF (Bag-of-Feature) and random forest classifier. Feature vectors extracted from training data are clustered into code book and each feature is transformed new BoF feature using code book. BoF features are applied to random forest for training and random forest having N classes is constructed by combining several decision trees. For testing, the same OCS-LBP feature is extracted from a query image and BoF is applied to trained random forest classifier. In contrast to conventional retrieval system, query image selects similar K-nearest neighbor (K-NN) classes after random forest is performed. Then, Top K similar images are retrieved from database images that are only labeled K-NN classes. Compared with other retrieval algorithms, the proposed method shows both fast processing time and improved retrieval performance.

본 논문에서는 의료영상의 특성을 반영하여 픽셀 그래디언트의 방향 값을 특징으로 하는 OCS-LBP (Oriented Center Symmetric Local Binary Patterns) 특징을 개발하고 BoF(Bag-of-Feature)와 Random Forest 분류기를 이용한 영상 검색 방법을 제안한다. 학습영상에서 추출된 특징 값은 code book 으로 군집화 되고, 각 영상들은 code book을 통해 의미 있는 새로운 차원인 BoF특징으로 변환된다. 이렇게 추출된 BoF특징은 Random Forest 분류기에 적용되고 학습된 분류기에 의해 유사한 특성을 갖는 N개의 클래스별로 분류되게 된다. 질의 영상이 입력되면 동일한 OCS-LBP특징이 추출되고 code book을 통해 BoF특징이 추출된다. 전통적인 내용기반 영상검색과는 다르게, 본 논문에서는 질의 영상에서 추출된 BoF특징이 학습된 Random Forest에 적용되어 가장 유사한 K-근접 이웃 (K-nearest neighbor) 클래스들을 선택하고 선택된 클래스들에 포함된 영상들에 대해서만 질의 영상과의 BoF 유사도 측정을 통해 최종 유사한 영상을 검색하게 된다. 실험결과에서 본 논문에서 제안하는 방법은 빠르고 우수한 검색 성능을 보여 주었다.

Keywords

References

  1. Mi-suk Seo, Byoung-Chul Ko, Yu-ik Son, Hee - Juhn Park, Jae-Yeal Nam, "The human visual system and the MPEG-7 visual descriptors using a region of interest based Medical Image Retrieval" Burns Society of Korea Journal, Vol.13, No.2, pp.120-130, 2007.
  2. Gi-hui Park, Byoung-Chul Ko, Jae-Yeal Nam, "Medical image automatic annotation generated code using an array of multi-class SVM and comments," Journal of the Korea Information Processing Society B, Vol.16, No.4, pp.281-288, 2009.
  3. E. Uwimana and M. E. Ruiz, "Integrating an automatic classification method into the medical image retrieval process," AMIA Annual Symposium 2008, pp.747-751, 2008.
  4. U. Avni, M. Sharon, and J. Goldberger, "X-ray Image Categorization and Retrieval using Patch-based Visual Words Representation," International Conference on biomedical imaging, pp.350-353, 2009.
  5. J. Wang, Y. Li, Y. Zhang, H. Xie, and C. Wang, "Boosted leraning of visual word weighting factors for bag-of-features based medical image retrieval", Sixth International Conference on Image and Graphics, pp.1035-1040, 2011.
  6. B.C. Ko, J.-H. Lee, and J.-Y. Nam, "Automatic medical image annotation and keyword-based image retrieval using relevance feedback," Journal of Digital Imaging, Vol.25, No.4, pp.454-465, 2012. https://doi.org/10.1007/s10278-011-9443-5
  7. D. G. Lowe, "Object recognition from local scale-invariant features," IEEE International Conference on Computer Vision, Vol.2, pp.1150-1157, 1999.
  8. T. Ojala, M. Pietikainen, and T. Maenpaa, "Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.24, pp.971-987, 2002. https://doi.org/10.1109/TPAMI.2002.1017623
  9. M. Heikkila and M. Pietikainen, "Description of Interest Regions with Local Binary Patterns," Pattern Recognition Vol.42, pp.425-436, 2009. https://doi.org/10.1016/j.patcog.2008.08.014
  10. Y. G. Jiang C. W. Ngo, and J. Yang, "Towards Optimal Bag-of-Features for Object Categorization and Semantic Video Retrieval," 6th ACM international conference on Image and video retrieval, pp.494-501, 2007.
  11. L. Breiman, "Random Forests," Machine Learning Vol.45, pp.5-32, 2001. https://doi.org/10.1023/A:1010933404324
  12. B. C. Ko, J-H. Lee, and J-Y Nam, "Automatic medical image annotation and keyword-based image retrieval using relevance feedback," Journal of Digital Imaging, Vol.25, No.4, pp.454-465, 2012. https://doi.org/10.1007/s10278-011-9443-5
  13. P. Bhattacharya and M. M. Rahman, "Image Representation and Retrieval Using Support Vector Machine and Fuzzy C-means Clustering Based Semantical Spaces," International Conference on Pattern Recognition, Vol.2, pp.1162-1168, 2006.

Cited by

  1. Human action recognition in still images using action poselets and a two-layer classification model vol.28, 2015, https://doi.org/10.1016/j.jvlc.2015.01.003