Multi-Dimensional Dynamic Time Warping for Image Texture Similarity
Modern content-based image retrieval systems use different features to represent properties (e.g., color, shape, texture) of the visual content of an image. Retrieval is performed by example where a query image is given as input and an appropriate metric is used to find the best matches in the corresponding feature space. Both selecting the features and the distance metric continue to be active areas of research. In this paper, we propose a new approach, based on the recently proposed Multidimensional Dynamic Time Warping (MD-DTW) distance , for assessing the texture similarity of images with structured textures. The MD-DTW allows the detection and comparison of arbitrarily shifted patterns between multi-dimensional series, such as those found in structured textures. Chaos theory tools are used as a preprocessing step to uncover and characterize regularities in structured textures. The main advantage of the proposed approach is that explicit selection and extraction of texture features is not required (i.e., similarity comparisons are performed directly on the raw pixel data alone). The method proposed in this preliminary investigation is shown to be valid by proving that it creates a statistically significant image texture similarity measure.
KeywordsContent-Based Image Retrieval Texture Dynamic Time Warping Similarity Measure Distance Measure Chaos Theory
Unable to display preview. Download preview PDF.
- 1.ten Holt, G.A., Reinders, M.J.T., Hendriks, E.A.: Multi-Dimensional Dynamic Time Warping for Gesture Recognition. In: Annual Conference of the Advanced School for Computing and Imaging (2007)Google Scholar
- 2.Ono, A., Amano, M., Hakaridani, M., Satoh, T., Sakauchi, M.: A flexible content-based image retrieval system with combined scene description keywords. In: IEEE International Conference on Multimedia Computing and Systems, pp. 201–208 (1996)Google Scholar
- 3.Shen, H.T., Ooi, B.C., Tan, K.L.: Giving meanings to WWW images. In: ACM Multimedia, pp. 39–48 (2000)Google Scholar
- 8.Kruskall, J., Liverman, M.: The symmetric time warping problem: from continuous to discrete. In: Time Warps, String Edits and Macro-molecules: The Theory and Practice of Sequence Comparison, pp. 125–161. Addison-Wesley, Reading (1983)Google Scholar
- 9.Edmonds, A.N.: Time Series Prediction Using Supervised Learning and Tools from Chaos Theory. PhD Thesis, University of Luton (1996)Google Scholar
- 15.The Multiple-Dimensions Mutual Information Program (Matt Kennel), http://www-ncsl.postech.ac.kr/en/softwares/archives/mmi.tar.Z
- 16.Takens, F.: Detecting strange attractors in turbulence. In: Dynamical Systems and Turbulence, pp. 366–381. Springer, Heidelberg (1980)Google Scholar
- 17.Mañé, R.: On the dimension of the compact invariant sets of certain nonlinear maps. In: Dynamical Systems and Turbulence, pp. 230–242. Springer, Heidelberg (1980)Google Scholar
- 18.Medio, A., Gallo, G.: Chaotic Dynamics: Theory and Applications to Economics. Cambridge University Press, Cambridge (1993)Google Scholar
- 20.Picard, R., Graczyk, C., Mann, S., Wachman, J., Picard, L., Campbell, L.: MIT media lab: Vision texture database, http://vismod.media.mit.edu/vismod/imagery/VisionTexture/