Incremental Learning of the Model for Watershed-Based Image Segmentation
Many image analysis methods need a lot of parameters that have to be adjusted to the particular image in order to achieve the best results. Therefore, methods for parameter learning are required that can assist a system developer in building a model. This task is usually called meta-learning. We consider meta-learning for learning the image segmentation parameters so that the image segmenter can be applied to a wide range of images while achieving good image segmentation quality. The meta-learner is based on case-based reasoning. The cases in the case base are comprised of an image description and the solutions that are the associated parameters for the image segmenter. First, the image description is calculated from a new image. The image description is used to index the case base. The closest case is retrieved based on a similarity measure. Then the associated segmentation parameters are given to the image segmenter and the actual image is segmented. We explain the architecture of such a case-based reasoning image segmenter. The case-description as well as the similarity function are described. Finally, we give results on the image segmentation quality.
Keywordsimage segmentation case-based learning image similarity feature description
Unable to display preview. Download preview PDF.
- 1.Attig, A., Perner, P.: A study on the case image description for learning the model of the watershed segmentation. Transactions on Case-Based Reasoning 2(1), 41–54 (2009)Google Scholar
- 6.Jobin Christ, M.C., Parvathi, R.M.S.: Segmentation of medical image using K-means clustering and marker controlled watershed algorithm. European Journal of Scientific Research 71(2), 190–194 (2012)Google Scholar
- 9.Perner, P.: Case-based reasoning for image analysis and interpretation. In: Chen, C., Wang, P.S.P. (eds.) Handbook on Pattern Recognition and Computer Vision, 3rd edn., pp. 95–114. World Scientific Publisher (2005)Google Scholar
- 12.Prewitt, J.M.S.: Object enhancement and extraction. In: Lipkin, B.S., Rosenfeld, A. (eds.) Picture Processing and Psychopictorics, pp. 75–149. Academic Press, New York (1970)Google Scholar
- 17.Zanaty, E.A., Afifi, A.: A watershed approach for improving medical image segmentation. Computer Methods in Biomechanics and Biomedical Engineering (forthcoming paper, 2012)Google Scholar