Abstract
Hyperspectral (HS) images captured from Earth by satellite and aircraft have become increasingly important in several environmental and ecological contexts (e.g. agriculture and urban areas). In the present study we propose an iterative learning methodology for the change detection of HS scenes taken at different times in the same areas. It cascades clustering and classification through iterative learning, in order to separate salient regions, where a change occurs in the scene from the unchanged background. The iterative learning is evaluated in both the clustering and the classification steps. The experiments performed with the proposed methodology provide encouraging results, also compared to several recent state-of-the-art competitors.
Similar content being viewed by others
Notes
In principle, any other pixel-wise distance measure can be implemented for this analysis.
As clarified in Section 3.3.1, clustering is performed on pixels spanned on the feature D—distance computed pixel-wise between the spectral vectors of the bi-temporal scene, according to the theory reported in Section 3.2—and Y — saliency label pixel-wise decided at the completion of the previous iteration. In the first iteration clustering is performed on D (as Y will be computed only after the completion of the classification step), while in each subsequent iteration clustering is performed on D × Y. However, as the information on D is available from the first iteration, we are able to compute (8) at each iteration.
The auto-encoding technique may also be used to reduce the dimensionality of the fused spectral data, even if a recent research shows that no significant improvement can be actually achieved in HS imagery analysis by considering auto-encoding instead of principal components [6].
Correspondence and requests for materials should be addressed to Annalisa Appice (email: annalisa.appice@uniba.it).
We use the Matlab implementation of US-ELM available on https://www.ntu.edu.sg/home/egbhuang/elm_codes.html. We run US-ELM with the dimension of the embedding layer varying between 2, 3, 4 and 5. We consider the results achieved from the trial with the highest accuracy.
References
Adar S, Notesco G, Brook A, Livne I, Rojik P, Kopackova V, Zelenková K, Misurec J, Bourguignon A, Chevrel S, Ehrler C, Fisher C, Hanus J, Shkolnisky Y, Ben-Dor E (2011) Change detection over sokolov open-pit mining area, czech republic, using multi-temporal hymap data (2009-2010). Image Signal Process Remote Sens Xvii 8180:81800T. https://doi.org/10.1117/12.898748
Alberti M, Marzluff MJ, Shulenberger S, Bradley G, Ryan C, Zumbrunnen C (2003) Integrating humans into ecology: Opportunities and challenges for studying urban ecosystems. Bioscience 53(12):1169–1179. https://doi.org/10.1641/0006-3568(2003)053[1169:IHIEOA]2.0.CO;2.
Appice A, Di Mauro N, Lomuscio F, Malerba D (2019) Empowering change vector analysis with autoencoding in bi-temporal hyperspectral images. In: MACHine learning for EArth observation workshop (MACLEANECMLPKDD), vol 2466. CEUR Workshop Proceedings, pp 1–10
Appice A, Guccione P, Malerba D (2016) Transductive hyperspectral image classification: toward integrating spectral and relational features via an iterative ensemble system. Mach Learn 103 (3):343–375. https://doi.org/10.1007/s10994-016-5559-7
Appice A, Guccione P, Malerba D (2017) A novel spectral-spatial co-training algorithm for the transductive classification of hyperspectral imagery data. Pattern Recogn 63:229–245. https://doi.org/10.1016/j.patcog.2016.10.010
Appice A, Malerba D (2019) Segmentation-aided classification of hyperspectral data using spatial dependency of spectral bands. ISPRS J Photogramm Remote Sens 147:215–231
Boisgontier H, Noblet V, Heitz F, Rumbach L, Armspach JP (2009) Generalized likelihood ratio tests for change detection in diffusion tensor images. In: Proceedings of the Sixth IEEE International Conference on Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2009. IEEE Press, pp 811–814
Borgelt C, Kruse R (2006) Finding the number of fuzzy clusters by resampling. In: 2006 IEEE International conference on fuzzy systems, pp 48–54. https://doi.org/10.1109/FUZZY.2006.1681693
Bruzzone L, Prieto DF (2000) Automatic analysis of the difference image for unsupervised change detection. IEEE Trans Geosci Remote Sens 38(2):1171–1182
Celik T (2010) Method for unsupervised change detection in satellite images. Electron Lett 46(9):624–626. https://doi.org/10.1049/el.2010.0808
Choi S, Cha S, Tappert C (2010) A survey of binary similarity and distance measures. J Syst Cybern Inf 8:43–48
Clifton C (2003) Change detection in overhead imagery using neural networks. Appl Intell 18(2):215–234. https://doi.org/10.1023/A:1021942526896
Deng JS, Wang K, Deng Y, Qi GJ (2008) Pca-based land-use change detection and analysis using multitemporal and multisensor satellite data. Int J Remote Sens 29(16):4823–4838
Ding S, Zhang N, Zhang J, Xu X, Shi Z (2017) Unsupervised extreme learning machine with representational features. Int J Mach Learn Cybern 8(2):587–595. https://doi.org/10.1007/s13042-015-0351-8
Du S, Chen S (2014) Salient object detection via random forest. Signal Processing Letters. IEEE 21:51–54. https://doi.org/10.1109/LSP.2013.2290547
Du S, Ikenaga T (2019) Iterative autoencoding and clustering for unsupervised feature representation. In: 2019 IEEE International symposium on circuits and systems (ISCAS), pp 1–5. https://doi.org/10.1109/ISCAS.2019.8702659
Du Y, Chang CI, Ren H, Chang CC, Jensen OJ, D’Amico F (2004) New hyperspectral discrimination measure for spectral characterization. Opt Eng - OPT ENG 43:1777–1786. https://doi.org/10.1117/1.1766301
Du Z, Jeong YS, Jeong MK, Kong SG (2012) Multidimensional local spatial autocorrelation measure for integrating spatial and spectral information in hyperspectral image band selection. Appl Intell 36(3):542–552. https://doi.org/10.1007/s10489-010-0274-8
ESA: Fundamentals of atmospheric correction (2015). http://seom.esa.int/landtraining2015/files/Day_2/D2T1b_LTC2015_Gastellu-Etchegorry.pdf
Garea AS, Ordóñez Á, Heras DB, Argüello F (2016) Hyperview: an open source desktop application for hyperspectral remote-sensing data processing. Int J Remote Sens 37(23):5533–5550. https://doi.org/10.1080/01431161.2016.1244363
Gewali UB, Monteiro ST, Saber E (2018) Machine learning based hyperspectral image analysis: a survey. CoRR arXiv:1802.08701
GmbH C Rtk x8 hyperspectral mapping (2019). https://cubert-gmbh.com/product/rtk-x8-hyperspectral-mapping/
Golub GH, Van Loan CF (2013) Matrix Computations, 4th edn. JHU Press
Gupta DM, Munshi MK (1985) Land use and forestry studies of himachal pradesh. Int J Remote Sens 6:535–539
Kwon H, Nasrabadi NM (2005) Kernel rx-algorithm: a nonlinear anomaly detector for hyperspectral imagery. IEEE Trans Geosci Remote Sens 43(2):388–397. https://doi.org/10.1109/TGRS.2004.841487
Helmy A, El-Taweel G (2010) Neural network change detection model for satellite images using textural and spectral characteristics. Amer J Eng Appl Sci 3(4):604–610
Hoye G, Fridman A (2013) The mixel camera — keystone-free hyperspectral images. In: 2013 5Th workshop on hyperspectral image and signal processing: Evolution in remote sensing (WHISPERS), pp 1–4. https://doi.org/10.1109/WHISPERS.2013.8080703
Huang G, Song S, Gupta JND, Wu C (2014) Semi-supervised and unsupervised extreme learning machines. IEEE Trans Cybern 44(12):2405–2417. https://doi.org/10.1109/TCYB.2014.2307349
Hussain M, Chen D, Cheng A, Wei H, Stanley D (2013) Change detection from remotely sensed images: From pixel-based to object-based approaches. ISPRS J Photogramm Remote Sens 80:91–106
Ilsever M, Unsalan C (2012) Two-dimensional change detection methods: remote sensing applications. Springer Science & Business Media
Im J, Jensen JR, Tullis JA (2008) Object-based change detection using correlation image analysis and image segmentation. Int J Remote Sens 29(2):399–423. https://doi.org/10.1080/01431160601075582
Khanday W (2016) Review of various change detection techniques for hyperspectral images. Asian J Technol Manag Res (AJTMR) 6(2):39–42
Kiana E, Homayouni S, Sharifi MA, Farid-Rohani M (2015) Unsupervised change detection in sar images using gaussian mixture models. ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XL-1/W5:407–410. https://doi.org/10.5194/isprsarchives-XL-1-W5-407-2015
Lemoine G, Giovalli M (2013) Geo-correction of high-resolution imagery using fast template matching on a gpu in emergency mapping contexts. MDPI Remote Sens 5(9):4488–4502
Li S, Song W, Fang L, Chen Y, Ghamisi P, Benediktsson J (2019) Deep learning for hyperspectral image classification: an overview. IEEE Trans Geosci Remote Sens PP:1–20. https://doi.org/10.1109/TGRS.2019.2907932
Liang J, Zhou J, Bai X, Qian Y (2013) Salient object detection in hyperspectral imagery. In: 2013 20Th IEEE international conference on image processing (ICIP). https://doi.org/10.1109/ICIP.2013.6738493
Lin W, Wu Z, Lin L, Wen A, Li J (2017) An ensemble random forest algorithm for insurance big data analysis. IEEE Access 5:16568–16575
Liu S, Bruzzone L, Bovolo F, Du P (2015) Hierarchical unsupervised change detection in multitemporal hyperspectral images. IEEE Trans Geosci Remote Sens 53(1):244–260. https://doi.org/10.1109/TGRS.2014.2321277
Lopez-Fandino J, Garea AS, Heras DB, Argu̇ello F (2018) Stacked autoencoders for multiclass change detection in hyperspectral images. In: 2018 IEEE International geoscience and remote sensing symposium, IGARSS 2018. IEEE, Valencia, pp 1906–1909
Lopez-Fandino J, Heras DB, Arguello F, Duro RJ (2017) Cuda multiclass change detection for remote sensing hyperspectral images using extended morphological profiles. In: 2017 9Th IEEE international conference on intelligent data acquisition and advanced computing systems: Technology and applications (IDAACS), vol 1, pp 404–409
Lopez-Fandino J, Heras DB, Argu̇ello F, Mura MD (2019) GPU Framework for change detection in multitemporal hyperspectral images. Int J Parallel Prog 47(2):272–292. https://doi.org/10.1007/s10766-017-0547-5
Louppe G (2015) Understanding random forests: From theory to practice. arXiv:1407.7502
Lu D, Mause P, Brondizio E, Moran E (2010) Change detection techniques. Int J Remote Sens 25:2365–2401
Marpu P, Gamba P, Benediktsson JA (2011) Hyperspectral change detection with ir-mad and initial change mask. In: 2011 3Rd workshop on hyperspectral image and signal processing: Evolution in remote sensing (WHISPERS), pp 1–4. https://doi.org/10.1109/WHISPERS.2011.6080961
Michel U, Thunig H, Ehlers M, Reinartz P (2012) Rapid change detection algorithm for disaster management. ISPRS - International Archives of the Photogrammetry. Remote Sens Spatial Inf Sci I-4:107–111
Ming H, Wenzhong S, Zhang H, Chang L (2014) Unsupervised change detection with expectation-maximization-based level set. IEEE Geosci Remote Sens Lett 11(1):210–214. https://doi.org/10.1109/LGRS.2013.2252879
Moore TS, Campbell JW, Feng H (2001) A fuzzy logic classification scheme for selecting and blending satellite ocean color algorithms. IEEE Trans Geosci Remote Sens 39(8):1764–1776. https://doi.org/10.1109/36.942555
Mouroulis P, Green R, Chrien TG (2000) Design of pushbroom imaging spectrometers for optimum recovery of spectroscopic and spatial information. Appl Opt 39(13):2210–2220
Najafi A, Hasanlou M, Akbari V (2017) Land cover changes detection in polarimetric sar data using algebra, similarity and distance based methods. ISPRS Int Arch Photogramm Remote Sens Spat Inf Sci XLII:195–200
Nousi P, Tefas A (2018) Self-supervised autoencoders for clustering and classification. Evolv Syst:1–14. https://doi.org/10.1007/s12530-018-9235-y
Otsu N (1972) A threshold selection method from gray-level histograms. IEEE Trans Geosc Remote Sens 9 (1):62–66
Pedregosa F, Varoquaux G, Gramfort A, Michel V, Thirion B, Grisel O, Blondel M, Prettenhofer P, Weiss R, Dubourg V, Vanderplas J, Passos A, Cournapeau D, Brucher M, Perrot M, Duchesnay E (2011) Scikit-learn: Machine learning in Python. J Mach Learn Res 12:2825–2830
Penglin Z, Zhiyong L, Dan Z, Jiangping C (2012) A shape similarity based change detection approach of multi-resolution remote sensing images. ISPRS Int Arch Photogramm Remote Sens Spat Inf Sci I-7:263–266
Peters G, Crespo F, Lingras P, Weber R (2013) Soft clustering – fuzzy and rough approaches and their extensions and derivatives. Int J Approx Reason 54(2):307–322
Read JM, Lam NS (2002) Spatial methods for characterizing land cover and detecting land cover changes for the tropics. Int J Remote Sens 23:2457–2474
Reynolds D (2009) Gaussian mixture models. In: Li S. Z., Jain A (eds) Encyclopedia of biometrics. Springer, Boston, pp 659–663. https://doi.org/10.1007/978-0-387-73003-5_196
Rufin P, Frantz D, Ernst S, Rabe A, Griffiths P, Özdoǧan M, Hostert P (2019) Mapping cropping practices on a national scale using intra-annual landsat time series binning, vol 11. https://doi.org/10.3390/rs11030232
Sahoo P, Soltani S, Wong AC (1988) A survey of thresholding techniques. Comput Vis Graph Image Process 41(2):233–260
Seydi ST, Hasanlou M (2017) A new land-cover match-based change detection for hyperspectral imagery. Eur J Remote Sens 50(1):517–533. https://doi.org/10.1080/22797254.2017.1367963
Song A, Choi J, Han Y, Kim Y (2018) Change detection in hyperspectral images using recurrent 3d fully convolutional networks. Remote Sens 10:1827. https://doi.org/10.3390/rs10111827
Eismann MT, Meola J, Hardie RC (2008) Hyperspectral change detection in the presence of diurnal and seasonal variations. IEEE Trans Geosci Remote Sens 46:237–249. https://doi.org/10.1109/TGRS.2007.907973
Tarabalka Y, Benediktsson JA, Chanussot J (2009) Spectral–spatial classification of hyperspectral imagery based on partitional clustering techniques. IEEE Trans Geosci Remote Sens 47(8):2973–2987
Truslow E, Manolakis D, Pieper M, Cooley T, Brueggeman M (2014) Performance prediction of matched filter and adaptive cosine estimator hyperspectral target detectors. IEEE J Sel Top Appl Earth Obs Remote Sens 7(6):2337–2350
Verbeek JJ, Vlassis N, Kröse B (2003) Efficient greedy learning of gaussian mixture models. Neural Comput 15(2):469–485. https://doi.org/10.1162/089976603762553004
Wang J, Liu S, Zhang S (2015) A novel saliency-based object segmentation method for seriously degenerated images. In: 2015 IEEE International conference on information and automation, pp 1172–1177
Wu C, Du B, Zhang L (2013) A subspace-based change detection method for hyperspectral image. IEEE J Sel Top Appl Earth Obs Remote Sens 6(2):815–830. https://doi.org/10.1109/JSTARS.2013.2241396
Xia J, Ghamisi P, Yokoya N, Iwasaki A (2018) Random forest ensembles and extended multiextinction profiles for hyperspectral image classification. IEEE Trans Geosci Remote Sens 56(1):202–216
Yang Z, Mueller R (2007) Spatial-spectral cross-correlation for change detection : a case study for citrus coverage change detection. In: ASPRS 2007 Annual conference, vol 2, pp 767–777
Yavariabdi A, Kusetogullari H (2017) Change detection in multispectral landsat images using multi-objective evolutionary algorithm, vol PP. https://doi.org/10.1109/LGRS.2016.2645742
Yazdani H, Ortiz-Arroyo D, Choros K, Kwasnicka H (2016) Applying bounded fuzzy possibilistic method on critical objects. In: IEEE 17th international symposium on computational intelligence and informatics, CINTI 2016, pp 000271–000276
Zhang N, Ding S (2017) Unsupervised and semi-supervised extreme learning machine with wavelet kernel for high dimensional data. Memet Comput 9(2):129–139. https://doi.org/10.1007/s12293-016-0198-x
Acknowledgments
This work fulfills the research objectives of the PON ”Ricerca e Innovazione” 2014–2020 project RPASinAir ”Integrazione dei Sistemi Aeromobili a Pilotaggio Remoto nello spazio aereo non segregato per servizi” (ARS01_00820), funded by the Italian Ministry for Universities and Research (MIUR). The authors thank Lynn Rudd for her help in reading the manuscript, López Fandiño Javier for providing bi-temporal imagery data of Hermiston1, Santa Barbara and Patterson, Mahdi Hasanlou for providing bi-temporal imagery data of Hermiston2 and Francesco Lomuscio for running experiments with AICA.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Appice, A., Guccione, P., Acciaro, E. et al. Detecting salient regions in a bi-temporal hyperspectral scene by iterating clustering and classification. Appl Intell 50, 3179–3200 (2020). https://doi.org/10.1007/s10489-020-01701-8
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10489-020-01701-8