{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,3]],"date-time":"2026-04-03T15:19:58Z","timestamp":1775229598189,"version":"3.50.1"},"reference-count":51,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2019,4,10]],"date-time":"2019-04-10T00:00:00Z","timestamp":1554854400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Rio Tinto Centre for Mine Automation","award":["Unknown"],"award-info":[{"award-number":["Unknown"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>This paper proposes novel autoencoders for unsupervised feature-learning from hyperspectral data. Hyperspectral data typically have many dimensions and a significant amount of variability such that many data points are required to represent the distribution of the data. This poses challenges for higher-level algorithms which use the hyperspectral data (e.g., those that map the environment). Feature-learning mitigates this by projecting the data into a lower-dimensional space where the important information is either preserved or enhanced. In many applications, the amount of labelled hyperspectral data that can be acquired is limited. Hence, there is a need for feature-learning algorithms to be unsupervised. This work proposes unsupervised techniques that incorporate spectral measures from the remote-sensing literature into the objective functions of autoencoder feature learners. The proposed techniques are evaluated on the separability of their feature spaces as well as on their application as features for a clustering task, where they are compared against other unsupervised feature-learning approaches on several different datasets. The results show that autoencoders using spectral measures outperform those using the standard squared-error objective function for unsupervised hyperspectral feature-learning.<\/jats:p>","DOI":"10.3390\/rs11070864","type":"journal-article","created":{"date-parts":[[2019,4,10]],"date-time":"2019-04-10T11:25:08Z","timestamp":1554895508000},"page":"864","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":49,"title":["Unsupervised Feature-Learning for Hyperspectral Data with Autoencoders"],"prefix":"10.3390","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2230-0632","authenticated-orcid":false,"given":"Lloyd","family":"Windrim","sequence":"first","affiliation":[{"name":"Australian Centre for Field Robotics, University of Sydney, Sydney 2006, Australia"}]},{"given":"Rishi","family":"Ramakrishnan","sequence":"additional","affiliation":[{"name":"Baymatob Operations Pty Ltd, Leichhardt, Sydney 2040, Australia"}]},{"given":"Arman","family":"Melkumyan","sequence":"additional","affiliation":[{"name":"Australian Centre for Field Robotics, University of Sydney, Sydney 2006, Australia"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3201-4768","authenticated-orcid":false,"given":"Richard J.","family":"Murphy","sequence":"additional","affiliation":[{"name":"Australian Centre for Field Robotics, University of Sydney, Sydney 2006, Australia"}]},{"given":"Anna","family":"Chlingaryan","sequence":"additional","affiliation":[{"name":"Australian Centre for Field Robotics, University of Sydney, Sydney 2006, Australia"}]}],"member":"1968","published-online":{"date-parts":[[2019,4,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"3066","DOI":"10.1109\/TGRS.2011.2178419","article-title":"Evaluating Classification Techniques for Mapping Vertical Geology Using Field-Based Hyperspectral Sensors","volume":"50","author":"Murphy","year":"2012","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"363","DOI":"10.1016\/S0169-555X(03)00150-8","article-title":"High spatial resolution hyperspectral mapping of in-stream habitats, depths, and woody debris in mountain streams","volume":"55","author":"Marcus","year":"2003","journal-title":"Geomorphology"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Wendel, A., and Underwood, J. (2016, January 16\u201321). Self-Supervised Weed Detection in Vegetable Crops Using Ground Based Hyperspectral Imaging. Proceedings of the IEEE International Conference on Robotics and Automation, Stockholm, Sweden.","DOI":"10.1109\/ICRA.2016.7487717"},{"key":"ref_4","first-page":"79","article-title":"Hyperspectral Image Processing for Automatic Target Detection Applications","volume":"14","author":"Manolakis","year":"2003","journal-title":"Linc. Lab. J."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1380","DOI":"10.1109\/36.934071","article-title":"Unsupervised Target Detection in Hyperspectral","volume":"39","author":"Chiang","year":"2001","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"517","DOI":"10.1080\/22797254.2017.1367963","article-title":"A new land-cover match-based change detection for hyperspectral imagery","volume":"50","author":"Seydi","year":"2017","journal-title":"Eur. J. Remote Sens."},{"key":"ref_7","first-page":"1","article-title":"High-Dimensional Data Analysis: The Curses and Blessings of Dimensionality","volume":"1","author":"Donoho","year":"2000","journal-title":"AMS Math. Chall. Lect."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"55","DOI":"10.1109\/TIT.1968.1054102","article-title":"On the Mean Accuracy of Statistical Pattern Recognizers","volume":"14","author":"Hughes","year":"1968","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"792","DOI":"10.1109\/36.239901","article-title":"Analyzing High Dimensional Multispectral Data","volume":"31","author":"Lee","year":"1993","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"665","DOI":"10.1109\/TIP.2017.2761542","article-title":"A Physics-Based Deep Learning Approach to Shadow Invariant Representations of Hyperspectral Images","volume":"27","author":"Windrim","year":"2018","journal-title":"IEEE Trans. Image Process."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1016\/j.isprsjprs.2013.10.012","article-title":"Assessing the performance of two unsupervised dimensionality reduction techniques on hyperspectral APEX data for high resolution urban land-cover mapping","volume":"87","author":"Demarchi","year":"2014","journal-title":"ISPRS J. Photogramm. Remote Sens."},{"key":"ref_12","unstructured":"Bishop, C.M. (2006). Pattern Recognition and Machine Learning, Springer."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"9104","DOI":"10.3390\/rs6099104","article-title":"Mapping Layers of Clay in a Vertical Geological Surface Using Hyperspectral Imagery: Variability in Parameters of SWIR Absorption Features under Different Conditions of Illumination","volume":"6","author":"Murphy","year":"2014","journal-title":"Remote Sens."},{"key":"ref_14","unstructured":"Heege, H.J. (2015). Precision in Crop Farming, Springer."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"503","DOI":"10.1109\/LGRS.2007.900751","article-title":"Modified Fisher\u2019s Linear Discriminant Analysis for Hyperspectral Imagery","volume":"4","author":"Du","year":"2007","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1139","DOI":"10.1109\/TGRS.2008.2008308","article-title":"Kernel Nonparametric Weighted Feature Extraction for Hyperspectral Image Classification","volume":"47","author":"Kuo","year":"2009","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"2094","DOI":"10.1109\/JSTARS.2014.2329330","article-title":"Deep Learning-Based Classification of Hyperspectral Data","volume":"7","author":"Chen","year":"2014","journal-title":"IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Cheriyadat, A., and Bruce, L. (2003). Why principal component analysis is not an appropriate feature extraction method for hyperspectral data. Geosci. Remote Sens., 3420\u20133422.","DOI":"10.1109\/IGARSS.2003.1294808"},{"key":"ref_19","first-page":"115","article-title":"Principal Component Analysis for Hyperspectral Image Classification","volume":"62","author":"Rodarmel","year":"2002","journal-title":"Surv. Land Inf. Sci."},{"key":"ref_20","first-page":"3136","article-title":"Unsupervised hyperspectral image analysis using independent component analysis","volume":"7","author":"Chiang","year":"2000","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"439","DOI":"10.1109\/TGRS.2013.2241444","article-title":"Unsupervised feature learning for aerial scene classification","volume":"52","author":"Cheriyadat","year":"2014","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Windrim, L., Ramakrishnan, R., Melkumyan, A., and Murphy, R.J. (2017, January 4\u20137). Hyperspectral CNN Classification with Limited Training Samples. Proceedings of the British Machine Vision Conference (BMVC), London, UK.","DOI":"10.5244\/C.31.4"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"481","DOI":"10.1080\/01431160310001618040","article-title":"Unsupervised classification of hyperspectral data: An ICA mixture model based approach","volume":"25","author":"Shah","year":"2004","journal-title":"Int. J. Remote Sens."},{"key":"ref_24","unstructured":"Filho, A.G.D.S., Frery, A.C., Ara\u00fajo, C.C.D., Alice, H., Cerqueira, J., Loureiro, J.A., Lima, M.E.D., Oliveira, M.D.G.S., and Horta, M.M. (2003, January 8\u201311). Hyperspectral images clustering on reconfigurable hardware using the k-means algorithm. Proceedings of the 16th Symposium on IEEE Integrated Circuits and Systems Design, Sao Paulo, Brazil."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"504","DOI":"10.1126\/science.1127647","article-title":"Reducing the dimensionality of data with neural networks","volume":"313","author":"Hinton","year":"2006","journal-title":"Science"},{"key":"ref_26","doi-asserted-by":"crossref","first-page":"233","DOI":"10.1002\/aic.690370209","article-title":"Nonlinear principal component analysis using autoassociative neural networks","volume":"37","author":"Kramer","year":"1991","journal-title":"AIChE J."},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"291","DOI":"10.1007\/BF00332918","article-title":"Auto-association by multilayer perceptrons and singular value decomposition","volume":"59","author":"Bourlard","year":"1988","journal-title":"Biol. Cybern."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Licciardi, G., Frate, F.D., and Duca, R. (2009, January 12\u201317). Feature reduction of hyperspectral data using autoassociative neural networks algorithms. Proceedings of the 2009 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), Cape Town, South Africa.","DOI":"10.1109\/IGARSS.2009.5416882"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"447","DOI":"10.1109\/LGRS.2011.2172185","article-title":"Linear Versus Nonlinear PCA for the Classification of Hyperspectral Data Based on the Extended Morphological Profiles","volume":"9","author":"Licciardi","year":"2012","journal-title":"Geosci. Remote Sens. Lett. IEEE"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"2438","DOI":"10.1109\/LGRS.2015.2482520","article-title":"Unsupervised spectral\u2013spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification","volume":"12","author":"Tao","year":"2015","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"391","DOI":"10.1109\/TGRS.2017.2748160","article-title":"Unsupervised spectral\u2013spatial feature learning via deep residual Conv\u2013Deconv network for hyperspectral image classification","volume":"56","author":"Mou","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Windrim, L., Melkumyan, A., Murphy, R., Chlingaryan, A., and Nieto, J. (2016, January 25\u201328). Unsupervised Feature Learning for Illumination Robustness. Proceedings of the 2016 IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA.","DOI":"10.1109\/ICIP.2016.7533202"},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1927","DOI":"10.1109\/18.857802","article-title":"An Information-Theoretic Approach to Spectral Variability, Similarity, and Discrimination for Hyperspectral Image Analysis","volume":"46","author":"Chang","year":"2000","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"145","DOI":"10.1016\/0034-4257(93)90013-N","article-title":"The Spectral Image Processing System ( SIPS ) Interactive Visualization and Analysis of Imaging Spectrometer Data","volume":"44","author":"Kruse","year":"1993","journal-title":"Remote Sens. Environ."},{"key":"ref_35","unstructured":"Yuhas, R., Goetz, A.F.H., and Boardman, J.W. (1992, January 1\u20135). Descrimination among semi-arid landscape endmembers using the Spectral Angle Mapper (SAM) algorithm. Proceedings of the Summaries of the Third Annual JPL Airborne Geoscience Workshop, Pasadena, CA, USA."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"4162","DOI":"10.1109\/TGRS.2008.2001035","article-title":"Assessing the Influence of Reference Spectra on Synthetic SAM Classification Results","volume":"46","author":"Hecker","year":"2008","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_37","unstructured":"(2017, February 02). Hyperspectral and Colour Imaging. Available online: https:\/\/sites.google.com\/site\/hyperspectralcolorimaging\/dataset."},{"key":"ref_38","unstructured":"Ramakrishnan, R. (2016). Illumination Invariant Outdoor Perception. [Ph.D. Thesis, University of Sydney]."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Clark, R., Swayze, G., Wise, R., Live, K., Hoefen, T., Kokaly, R., and Sutley, S. (2007). USGS Digital Spectral Library splib06a: U.S. Geological Survey Data Series 231.","DOI":"10.3133\/ds231"},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"325","DOI":"10.1016\/S0038-092X(01)00054-8","article-title":"Parameterized Transmittance Model for Direct Beam and Circumsolar Spectral Irradiance","volume":"71","author":"Gueymard","year":"2001","journal-title":"Sol. Energy"},{"key":"ref_41","unstructured":"Theodoridis, S., and Koutroumbas, K. (1998). Recognition Pattern, Academic Press."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"76","DOI":"10.1021\/ci030295a","article-title":"Implementing the Fisher\u2019s Discriminant Ratio in a k -Means Clustering Algorithm for Feature Selection and Data Set Trimming","volume":"44","author":"Lin","year":"2004","journal-title":"J. Chem. Inf. Comput. Sci."},{"key":"ref_43","doi-asserted-by":"crossref","first-page":"8696","DOI":"10.1016\/j.eswa.2011.01.077","article-title":"A feature selection method based on improved fisher\u2019s discriminant ratio for text sentiment classification","volume":"38","author":"Wang","year":"2011","journal-title":"Expert Syst. Appl."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1007\/BF01908075","article-title":"Comparing partitions","volume":"2","author":"Hubert","year":"1985","journal-title":"J. Classif."},{"key":"ref_45","doi-asserted-by":"crossref","first-page":"846","DOI":"10.1080\/01621459.1971.10482356","article-title":"Objective Criteria for the Evaluation of Clustering Methods","volume":"66","author":"Rand","year":"1971","journal-title":"J. Am. Stat. Assoc."},{"key":"ref_46","doi-asserted-by":"crossref","first-page":"763","DOI":"10.1093\/bioinformatics\/17.9.763","article-title":"Principal component analysis for clustering gene expression data","volume":"17","author":"Yeung","year":"2001","journal-title":"Bioinformatics"},{"key":"ref_47","unstructured":"Van Rijsbergen, C.J. (1979). Information Retrieval, Dept. of Computer Science, University of Glasgow."},{"key":"ref_48","doi-asserted-by":"crossref","first-page":"503","DOI":"10.1007\/BF01589116","article-title":"On the limited memory BFGS method for large scale optimization","volume":"45","author":"Lui","year":"1989","journal-title":"Math. Program."},{"key":"ref_49","first-page":"153","article-title":"Greedy Layer-Wise Training of Deep Networks","volume":"19","author":"Bengio","year":"2007","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_50","doi-asserted-by":"crossref","first-page":"2798","DOI":"10.1109\/TGRS.2017.2783886","article-title":"Pretraining for Hyperspectral Convolutional Neural Network Classification","volume":"56","author":"Windrim","year":"2018","journal-title":"IEEE Trans. Geosci. Remote Sens."},{"key":"ref_51","unstructured":"Macqueen, J. (1965, January 27). Some Methods for Classification and Analysis of Multivariate Observations. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, CA, USA."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/7\/864\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T12:44:10Z","timestamp":1760186650000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/11\/7\/864"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2019,4,10]]},"references-count":51,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2019,4]]}},"alternative-id":["rs11070864"],"URL":"https:\/\/doi.org\/10.3390\/rs11070864","relation":{},"ISSN":["2072-4292"],"issn-type":[{"value":"2072-4292","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,4,10]]}}}