Abstract
Segmenting meaningful visual structures from an image is a fundamental and most-addressed problem in image analysis algorithms. However, among factors such as diverse visual patterns, noise, complex backgrounds, and similar textures present in foreground and background, image segmentation still stands as a challenging research problem. In this article, the proposed method employs an unsupervised method that addresses image segmentation as subspace clustering of image feature vectors. Initially, an image is partitioned into a set of homogeneous regions called superpixels, from which Local Spectral Histogram features are computed. Subsequently, a feature data matrix is created whereupon subspace clustering methodology is applied. A single-stage optimization model is formulated with enhanced segmentation capabilities by the combined action of l½ and l2 norm minimization. Robustness of l½ regularization toward both the noise and overestimation of sparsity provides simultaneous noise robustness and better subspace selection, respectively. While l2 norm facilitates grouping effect. Hence, the designed optimization model ensures an improved sparse solution and a sparse representation matrix with an accurate block diagonal structure, which thereby favours getting properly segmented images. Then, experimental results of the proposed method are compared with the state-of-art algorithms. Results demonstrate the improved performance of our method over the state-of-art algorithms.
- [1] . 1996. A survey on evaluation methods for image segmentation. Pattern Recogn. 29, 8 (1996), 1335–1346.Google Scholar
Cross Ref
- [2] . 2002. Digital Image Processing (3rd Edition). Prentice-Hall, Inc..Google Scholar
- [3] . 2019. Understanding deep learning techniques for image segmentation. ACM Comput. Surv. 52, 4, Article 73 (
Aug. 2019), 35 pages.DOI: DOI: http://dx.doi.org/10.1145/3329784 Google ScholarCross Ref
- [4] . 2017. Image segmentation by correlation adaptive weighted regression. Neurocomputing 267 (2017), 426–435. Google Scholar
Digital Library
- [5] . 2015. Superpixel segmentation using linear spectral clustering. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 1356–1363.Google Scholar
- [6] . 2018. Image segmentation using sparse subset selection. In Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV’18). IEEE, 1470–1479.Google Scholar
Cross Ref
- [7] . 2018. Superpixel-based fast fuzzy C-means clustering for color image segmentation. IEEE Trans. Fuzzy Syst. 27, 9 (2018), 1753–1766.Google Scholar
Cross Ref
- [8] . 2020. Unsupervised learning of image segmentation based on differentiable feature clustering. IEEE Trans. Image Process. 29 (2020), 8055–8068.
DOI: DOI: http://dx.doi.org/10.1109/TIP.2020.3011269Google ScholarDigital Library
- [9] . 2012. Segmentation using superpixels: A bipartite graph partitioning approach. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 789–796. Google Scholar
Digital Library
- [10] . 2017. Subspace clustering guided unsupervised feature selection. Pattern Recogn. 66 (2017), 364–374. Google Scholar
Digital Library
- [11] . 2020. An unsupervised image segmentation method combining graph clustering and high-level feature representation. Neurocomputing 409 (2020), 83–92.Google Scholar
Cross Ref
- [12] . 2020. Unsupervised fuzzy model-based image segmentation. Signal Process. 171 (2020), 107483.Google Scholar
Digital Library
- [13] . 2017. A review of clustering techniques and developments. Neurocomputing 267 (2017), 664–681. Google Scholar
Digital Library
- [14] . 2020. Semisupervised approach to surrogate-assisted multiobjective kernel intuitionistic fuzzy clustering algorithm for color image segmentation. IEEE Trans. Fuzzy Syst. 28, 6 (2020), 1023–1034.
DOI: DOI: http://dx.doi.org/10.1109/TFUZZ.2020.2973121Google ScholarCross Ref
- [15] . 2020. Hypergraph-regularized low-rank subspace clustering using superpixels for unsupervised spatial-spectral hyperspectral classification. IEEE Geosci. Remote Sens. Lett. 18, 5 (2020), 871–875.Google Scholar
- [16] . 2020. A unified tensor framework for clustering and simultaneous reconstruction of incomplete imaging data. ACM Trans. Multimedia Comput. Commun. Appl. 16, 3 (2020), 1–24.Google Scholar
Digital Library
- [17] . 2016. Multitask low-rank affinity graph for image segmentation and image annotation. ACM Trans. Intell. Syst. Technol. 7, 4, Article 65 (
Mar. 2016), 18 pages.DOI: DOI: http://dx.doi.org/10.1145/2856058 Google ScholarCross Ref
- [18] . 2016. Spectral–spatial sparse subspace clustering for hyperspectral remote sensing images. IEEE Trans. Geosci. Remote Sens. 54, 6 (2016), 3672–3684.Google Scholar
Cross Ref
- [19] . 2014. Enhancing low-rank subspace clustering by manifold regularization. IEEE Trans. Image Process. 23, 9 (2014), 4022–4030.Google Scholar
Cross Ref
- [20] . 2018. Laplacian-regularized low-rank subspace clustering for hyperspectral image band selection. IEEE Trans. Geosci. Remote Sens. 57, 3 (2018), 1723–1740.Google Scholar
- [21] . 2019. Learning manifold structures with subspace segmentations. IEEE Trans. Cybernet. 51, 4 (2019), 1981–1992.Google Scholar
Cross Ref
- [22] . 2018. On the applications of robust PCA in image and video processing. Proc. IEEE 106, 8 (2018), 1427–1457.Google Scholar
Cross Ref
- [23] . 2019. Deep closed-form subspace clustering. In Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops. 0–0.Google Scholar
Cross Ref
- [24] . 2017. Subspace segmentation by correlation adaptive regression. IEEE Trans. Circ. Syst. Video Technol. 28, 10 (2017), 2612–2621. Google Scholar
Digital Library
- [25] . 2019. Nonlinear subspace clustering via adaptive graph regularized autoencoder. IEEE Access 7 (2019), 74122–74133.Google Scholar
Cross Ref
- [26] . 2013. Sparse subspace clustering: Algorithm, theory, and applications. IEEE Trans. Pattern Anal. Mach. Intell. 35, 11 (2013), 2765–2781. Google Scholar
Digital Library
- [27] . 2012. Robust recovery of subspace structures by low-rank representation. IEEE Trans. Pattern Anal. Mach. Intell. 35, 1 (2012), 171–184. Google Scholar
Digital Library
- [28] . 2010. Robust subspace segmentation by low-rank representation. In Proceedings of the 27th International Conference on Machine Learning (ICML’10). 663–670. Google Scholar
Digital Library
- [29] . 2012. Robust and efficient subspace segmentation via least squares regression. In Proceedings of the European Conference on Computer Vision. Springer, 347–360. Google Scholar
Digital Library
- [30] . 2011. Multi-task low-rank affinity pursuit for image segmentation. In Proceedings of the International Conference on Computer Vision. IEEE, 2439–2446. Google Scholar
Digital Library
- [31] . 2013. Correlation adaptive subspace segmentation by trace lasso. In Proceedings of the IEEE International Conference on Computer Vision. 1345–1352. Google Scholar
Digital Library
- [32] . 2015. Laplacian regularized low-rank representation and its applications. IEEE Trans. Pattern Anal. Mach. Intell. 38, 3 (2015), 504–517. Google Scholar
Digital Library
- [33] . 2000. Normalized cuts and image segmentation. IEEE Trans. Pattern Anal. Mach. Intell. 22, 8 (2000), 888–905. Google Scholar
Digital Library
- [34] . 2013. Graph-regularized low-rank representation for destriping of hyperspectral images. IEEE Trans. Geosci. Remote Sens. 51, 7 (2013), 4009–4018.Google Scholar
Cross Ref
- [35] . 2010. Graph regularized sparse coding for image representation. IEEE Trans. Image Process. 20, 5 (2010), 1327–1336. Google Scholar
Digital Library
- [36] . 2018. Hyper-Laplacian regularized multilinear multiview self-representations for clustering and semisupervised learning. IEEE Trans. Cybernet. 50, 2 (2018), 572–586.Google Scholar
Cross Ref
- [37] . 2012. \(L\_\){\(1/2\)} regularization: A thresholding representation theory and a fast solver. IEEE Trans. Neural Netw. Learn. Syst. 23, 7 (2012), 1013–1027.Google Scholar
Cross Ref
- [38] . 2014. \(L\_\){\(1/2\)} regularization: Convergence of iterative half thresholding algorithm. IEEE Trans. Signal Process. 62, 9 (2014), 2317–2329. Google Scholar
Digital Library
- [39] . 2013. Sparse kernel logistic regression based on L 1/2 regularization. Sci. China Info. Sci. 56, 4 (2013), 1–16.Google Scholar
Cross Ref
- [40] . 2011. Hyperspectral unmixing via \(L\_\){\(1/2\)} sparsity-constrained nonnegative matrix factorization. IEEE Trans. Geosci. Remote Sens. 49, 11 (2011), 4282–4297.Google Scholar
Cross Ref
- [41] . 2019. A three-way optimization technique for noise robust moving object detection using tensor low-rank approximation, l1/2, and TTV regularizations. IEEE Trans. Cybernet. 51, 2 (2019), 1004–1014.Google Scholar
Cross Ref
- [42] . 2018. Multi-atlas segmentation of MR tumor brain images using low-rank based image recovery. IEEE Trans. Med. Imag. 37, 10 (2018), 2224–2235.Google Scholar
Cross Ref
- [43] . 2002. Laplacian eigenmaps and spectral techniques for embedding and clustering. In Advances in Neural Information Processing Systems. MIT Press, 585–591. Google Scholar
Digital Library
- [44] . 2015. Sparse subspace clustering with missing entries. In Proceedings of the International Conference on Machine Learning. PMLR, 2463–2472. Google Scholar
Digital Library
- [45] . 2011. Trace Lasso: A trace norm regularization for correlated designs. In Advances in Neural Information Processing Systems. MIT Press, 2187–2195. Google Scholar
Digital Library
- [46] . 2011. Subspace clustering. IEEE Signal Processing Magazine 28, 2 (2011), 52–68.Google Scholar
Cross Ref
- [47] . 2015. Adaptive \(L_{\frac{1}{2}}\) Sparsity-constrained NMF with half-thresholding algorithm for hyperspectral unmixing. IEEE J. Select. Top. Appl. Earth Observ. Remote Sens. 8, 6 (2015), 2618–2631.Google Scholar
Cross Ref
- [48] . 2016. On some properties of the Laplacian matrix revealed by the RCM algorithm. Czech. Math. J. 66, 3 (2016), 603–620.Google Scholar
Cross Ref
- [49] . 2020. Graph regularized low-rank representation for submodule clustering. Pattern Recogn. 100 (2020), 107145.Google Scholar
Cross Ref
- [50] . 2020. Noise robust image clustering based on reweighted low rank tensor approximation and \(l_{\frac{1}{2}}\) regularization. Signal, Image Video Process. 15, 2 (2020), 341–349.Google Scholar
Cross Ref
- [51] . 2020. Sparse graph connectivity for image segmentation. ACM Trans. Knowl. Discov. Data 14, 4 (2020), 1–19. Google Scholar
Digital Library
- [52] . 2020. Image segmentation evaluation: a survey of methods. Artific. Intell. Rev. 53, 8 (2020), 5637–5674.Google Scholar
Digital Library
Index Terms
An l½ and Graph Regularized Subspace Clustering Method for Robust Image Segmentation
Recommendations
Low-rank representation with graph regularization for subspace clustering
In this paper, we propose a low-rank representation method that incorporates graph regularization for robust subspace clustering. We make the assumption that high-dimensional data can be approximated as the union of low-dimensional subspaces of unknown ...
Structured Sparse Subspace Clustering with Within-Cluster Grouping
Highlights- We define a concept of GEWC to group data from the same subspace together and design a new regularization term to enforce it.
AbstractMany high-dimensional data in computer vision essentially lie in multiple low-dimensional subspaces. Recently developed subspace clustering methods have shown good effectiveness in recovering the underlying low-dimensional subspace ...
Image Segmentation Based on Non-convex Low Rank Multiple Kernel Clustering
Artificial IntelligenceAbstractSubspace clustering has been widely used in image segmentation. These methods usually use superpixel segmentation to pre-segment image, while the superpixel segmentation method always divides image into superpixel blocks of similar shape and size, ...






Comments