Abstract
Segmentation-based anomaly detectors proceeds to the clustering of the hyperspectral image as a first step. However, most of the well-known clustering methods cluster anomalous pixels as a part of the background. This paper presents a new hyperspectral image clustering approach based on the betweenness centrality measure. The proposed approach starts by the construction of an adaptive spatial and spectral neighborhood for each pixel. This neighborhood is based on the selection of the nearest spectral and spatial neighbors in multiple windows around each pixel to allow well-suited representation of the image features. In the next step, this neighborhood is clustered based on the edge betweenness measure algorithm that splits the image into regions sharing similar features. This approach (1) allows the reduction of intercluster relationship, (2) favors intracluster relations, and (3) preserves small clusters that can hold anomalous pixels. Experimental results show that the proposed approach is efficient for clustering and overcomes the state of the art approaches.
Similar content being viewed by others
References
Cariou C, Chehdi K (2015) Unsupervised nearest neighbors clustering with application to hyperspectral images. IEEE J Sel Top Sign Process 9:1105–1116. https://doi.org/10.1109/JSTSP.2015.2413371
Dawood H, Dawood H, Guo P (2012) Combining the contrast information with LPQ for texture classification. 2012 6th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications DOI: https://doi.org/10.1109/SETIT.2012.6481945
Duran O, Petrou M (2007) A time-efficient method for anomaly detection in hyperspectral images. IEEE Trans Geosci Remote Sens 45:3894–3904. https://doi.org/10.1109/TGRS.2007.909205
Ghamisi P, Ali A-R, Couceiro MS, Benediktsson JA (2015) A novel evolutionary swarm fuzzy clustering approach for hyperspectral imagery. IEEE J Sel Top Appl Earth Observ Remote Sens 8:2447–2456. https://doi.org/10.1109/JSTARS.2015.2398835
Gillis N, Kuang D, Park H (2015) Hierarchical clustering of Hyperspectral images using rank-two nonnegative matrix factorization. IEEE Trans Geosci Remote Sens 53:2066–2078. https://doi.org/10.1109/TGRS.2014.2352857
Glenn T, Dranishnikov D, Gader P, Zare A (2013) Subpixel target detection in hyperspectral imagery using piece-wise convex spatial-spectral unmixing, possibilistic and fuzzy clustering, and co-registered LiDAR. IEEE Int Geosci Remote Sens Symp. https://doi.org/10.1109/IGARSS.2013.6721347
Hu J, Guo P (2012) Spatial local binary patterns for scene image classification. 2012 6th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications. DOI: https://doi.org/10.1109/SETIT.2012.6481936
Luxburg UV (2007) A tutorial on spectral clustering. Stat Comput 17:395–416. https://doi.org/10.1007/s11222-007-9033-z
Mercovich R.A, Albano J, Messinger D (2011) Techniques for the graph representation of spectral imagery. 3rd Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing (WHISPERS), Lisbon Doi: https://doi.org/10.1109/WHISPERS.2011.6080912
Newman MEJ, Girvan M (2003) Mixing patterns and community structure in networks. Stat Mech Complex Networks. https://doi.org/10.1007/978-3-540-44943-0_5
Niazmardi S, Homayouni S, Safari A (2013) An improved FCM algorithm based on the SVDD for unsupervised Hyperspectral data classification. IEEE J Sel Top Appl Earth Observ Remote Sens 6:831–839. https://doi.org/10.1109/JSTARS.2013.2244851
Schaeffer S-E (2007) Survey: graph clustering. Comput Sci Rev 1:27–64. https://doi.org/10.1016/j.cosrev.2007.05.001
Yuan Y, Lin J, Wang Q (2016) Dual-clustering-based hyperspectral band selection by contextual analysis. IEEE Trans Geosci Remote Sens 54:1431–1445. https://doi.org/10.1109/TGRS.2015.2480866
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Saheb Ettabaa, K., Ben Salem, M. & Bouhlel, M.S. Hyperspectral image betweenness centrality clustering based adaptive spatial and spectral neighborhood approach for anomaly detection. Arab J Geosci 10, 412 (2017). https://doi.org/10.1007/s12517-017-3196-5
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12517-017-3196-5