An Enhanced Region-Based Model for Segmentation Images with Intensity Inhomogeneity
Segmentation of images with intensity inhomogeneity is always challenging due to low resolution, blurred boundaries and poor illumination. Although existing image segmentation methods were widely used, there exists some shortcomings in segmenting intensity inhomogeneous images, such as not considering the spatial interrelation within the central pixel and its neighborhood. Therefore, this paper proposes an enhanced region-based active contour model for segmenting images with intensity inhomogeneity. In this model, a range-based adaptive bilateral filter is utilized to preserve edge structures and resist the noise of the image. Then an effective energy functional is constructed into the level set framework. With the permission of keeping the original shape of the image, a regularization term is utilized to refrain from the process of re-initialization and speed up the curve evolution. In the end, some experiments on synthetic and real images and contrast with the classic segmentation models are executed. The proposed model is more accuracy than other classic models.
KeywordsImage segmentation Region-based model Level set
We would like to thank the editors and the reviewers for their pertinent comments. This work was supported in part by the National Natural Science Foundation of China (Grant No. 61502356) and the Hubei Provincial Department of Education Science and Technology Research Project (No. B2016590).
- 5.Qi, C.R., Su, H., Mo, K., et al.: PointNet: deep learning on point sets for 3D classification and segmentation. Proc. Comput. Vis. Pattern Recognit. (CVPR) 1(2), 4 (2017)Google Scholar
- 6.Prasoon, A., Petersen, K., Igel, C., Lauze, F., Dam, E., Nielsen, M.: Deep feature learning for knee cartilage segmentation using a triplanar convolutional neural network. In: Mori, K., Sakuma, I., Sato, Y., Barillot, C., Navab, N. (eds.) MICCAI 2013. LNCS, vol. 8150, pp. 246–253. Springer, Heidelberg (2013). https://doi.org/10.1007/978-3-642-40763-5_31CrossRefGoogle Scholar
- 16.Haiping, Yu., Huali, Z.: Regularized level set method by incorporating local statistical information and global similarity compatibility for image segmentation. In: Huang, D.-S., Bevilacqua, V., Prashan, P. (eds.) ICIC 2015. LNCS, vol. 9225, pp. 388–399. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-22180-9_38CrossRefGoogle Scholar
- 31.Gonde, A.B., Murala, S., Vipparthi, S.K., Maheshwari, R., Balasubramanian, R.: 3D local transform patterns: a new feature descriptor for image retrieval. In: Raman, B., Kumar, S., Roy, P.P., Sen, D. (eds.) Proceedings of International Conference on Computer Vision and Image Processing. AISC, vol. 460, pp. 495–507. Springer, Singapore (2017). https://doi.org/10.1007/978-981-10-2107-7_45CrossRefGoogle Scholar