Abstract
In order to smooth the multi-scale texture with strong gradient while maintaining the weak structure which is challenging for the existing texture filtering methods, we put forward a novel algorithm based on multi-scale inherent variation features. Based on statistical analysis of various kinds of structure/texture pixels, six-dimensional discriminating features are found and first extracted from the multi-scale inherent variation curve, which demonstrate superiority in recognizing the structure pixels. Then, a preliminary structure prediction map can be obtained with a structure/texture classification model, which is generated by the cross-validation-based SVM training process. Next, we design a post-processing-based fine structure detection scheme to deal with the defects in the structure prediction map with three main steps successively, i.e., removing the mistaken texture pixels with thinning and outlier rejection, retrieving the missed structure pixels with breakpoints connection, and repositioning the structure pixels with structure correction. Finally, we propose a structure guided adaptive image smoothing method to smooth texture while preserving structure without halo effect. Experimental results show that our algorithm works better than the state-of-the-art methods in the preservation of weak structure as well as the suppression of texture with strong gradient or varying scales.
Similar content being viewed by others
References
Baek, J., Jacobs, D.: Accelerating spatially varying Gaussian filters. ACM Trans. Gr. 29(6), 81–95 (2010)
Boser, B., Guyon, I., Vapnik, V.: A training algorithm for optimal margin classifiers. In: Proceedings of the Fifth Annual Workshop on Computational Learning Theory, pp. 144–152 (1992)
Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines. ACM Trans. Intell. Syst. Technol. 2(3), 389–396 (2011)
Cho, H., Lee, H., Kang, H., Lee, S.: Bilateral texture filtering. ACM Trans. Gr. 33(4), 128 (2014)
Cortes, C., Vapnik, V.: Support vector network. Mach. Learn. 20(3), 273–297 (1995)
Dollár, P., Zitnick, C.: Fast edge detection using structured forests. IEEE Trans. Pattern Anal. Mach. Intell. 37(8), 1558–1570 (2015)
Farbman, Z., Fattal, R., Lischinski, D., Szeliski, R.: Edge-preserving decompositions for multi-scale tone and detail manipulation. ACM Trans. Gr. 27(3), 67 (2008)
Gastal, E., Oliveira, M.: Domain transform for edge-aware image and video processing. ACM Trans. Gr. 30(4), 1244–1259 (2011)
Gonzalez, R.C., Woods, R.E.: Digital Image Processing. Addison-Wesley Longman Publishing Co., Inc, Boston (2001)
Ham, B., Cho, M., Ponce, J.: Robust image filtering using joint static and dynamic guidance. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4823–4831 (2015)
Hua, M., Bie, X., Zhang, M., Wang, W.: Edge-aware gradient domain optimization framework for image filtering by local propagation. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2838–2845 (2014)
Karacan, L., Erdem, E., Erdem, A.: Structure-preserving image smoothing via region covariances. ACM Trans. Gr. 32(6), 176 (2013)
Li, X., Gu, Y., Hu, S., Martin, R.: Mixed-domain edge-aware image manipulation. IEEE Trans. Image Process. 22(5), 1915–1925 (2013)
Martin, D., Fowlkes, C., Malik, J.: Learning to detect natural image boundaries using local brightness, color, and texture cues. IEEE Trans. Pattern Anal. Mach. Intell. 26(5), 530–549 (2004)
Subr, K., Soler, C., Durand, F.: Edge-preserving multiscale image decomposition based on local extrema. ACM Trans. Gr. 28(5), 147 (2009)
Tomasi, C., Manduchi, R.: Bilateral filtering for gray and color images. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 839–846 (1998)
Wang, S., Chen, Y.T., Feng, H.J., Xu, Z.H., Li, Q.: TWIST-TV regularization based image deblurring method. Infrared Laser Eng. 43(6), 2000–2006 (2014)
Xie, S., Tu, Z.: Holistically-nested edge detection. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1395–1403 (2015)
Xu, L., Lu, C., Xu, Y., Jia, J.Y.: Image smoothing via \(L0\) gradient minimization. ACM Trans. Gr. 30(6), 174 (2011)
Xu, L., Ren, S., Yan, Q., Liao, R., Jia, J.: Deep edge-aware filters. In: Proceedings of the 32nd International Conference on Machine Learning (ICML), pp. 1669–1678 (2015)
Xu, L., Yan, Q., Xia, Y., Jia, J.J.: Structure extraction from texture via relative total variation. ACM Trans. Gr. 31(6), 139 (2012)
Yang, Q., Wang, S., Ahuja, N.: SVM for edge-preserving filtering. In: Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1775–1782 (2010)
Yu, L.H., Feng, Y.Q., Chen, W.F.: Adaptive regularization method based total variational de-noising algorithm. J. Image Gr. 14(10), 1950–1954 (2009)
Zhang, Q., Shen, X., Xu, L., Jia, J.Y.: Rolling guidance filter. In: Proceedings of the European Conference on Computer Vision (ECCV), pp. 815–830. Springer (2014)
Acknowledgements
This work is supported by the Zhejiang Provincial Natural Science Foundation of China under Grant No. LY14F020004, the National Natural Science Foundation of China under Grant Nos. 61003188, 61379075 and U1609215, the Talent Young Foundation of Zhejiang Gongshang University under Grant No. QZ13-9, the National Key Technology R&D Program under Grant No. 2014BAK14B01, the Zhejiang Provincial Commonweal Technology Applied Research Projects of China under Grant No. 2015C33071 and the Zhejiang Provincial Research Center of Intelligent Transportation Engineering and Technology under Grant No. 2015ERCITZJ-KF1.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Liu, C., Shao, H., Wu, M. et al. Multi-scale inherent variation features-based texture filtering. Vis Comput 33, 769–778 (2017). https://doi.org/10.1007/s00371-017-1380-y
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00371-017-1380-y