{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,12]],"date-time":"2025-10-12T04:31:46Z","timestamp":1760243506435,"version":"build-2065373602"},"reference-count":16,"publisher":"MDPI AG","issue":"5","license":[{"start":{"date-parts":[[2013,5,13]],"date-time":"2013-05-13T00:00:00Z","timestamp":1368403200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/3.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Remote Sensing"],"abstract":"<jats:p>This paper proposes a multi-level max-margin discriminative analysis (M3DA) framework, which takes both coarse and fine semantics into consideration, for the annotation of high-resolution satellite images. In order to generate more discriminative topic-level features, the M3DA uses the maximum entropy discrimination latent Dirichlet Allocation (MedLDA) model. Moreover, for improving the spatial coherence of visual words neglected by M3DA, conditional random field (CRF) is employed to optimize the soft label field composed of multiple label posteriors. The framework of M3DA enables one to combine word-level features (generated by support vector machines) and topic-level features (generated by MedLDA) via the bag-of-words representation. The experimental results on high-resolution satellite images have demonstrated that, using the proposed method can not only obtain suitable semantic interpretation, but also improve the annotation performance by taking into account the multi-level semantics and the contextual information.<\/jats:p>","DOI":"10.3390\/rs5052275","type":"journal-article","created":{"date-parts":[[2013,5,13]],"date-time":"2013-05-13T12:17:01Z","timestamp":1368447421000},"page":"2275-2291","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":25,"title":["Tile-Level Annotation of Satellite Images Using Multi-Level Max-Margin Discriminative Random Field"],"prefix":"10.3390","volume":"5","author":[{"given":"Fan","family":"Hu","sequence":"first","affiliation":[{"name":"Signal Processing Laboratory, School of Electronic Information, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3263-8768","authenticated-orcid":false,"given":"Wen","family":"Yang","sequence":"additional","affiliation":[{"name":"Signal Processing Laboratory, School of Electronic Information, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiayu","family":"Chen","sequence":"additional","affiliation":[{"name":"Signal Processing Laboratory, School of Electronic Information, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hong","family":"Sun","sequence":"additional","affiliation":[{"name":"Signal Processing Laboratory, School of Electronic Information, Wuhan University, Wuhan 430072, China"},{"name":"TSI Department, TELECOM ParisTech, F-75013 Paris, France"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2013,5,13]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"522","DOI":"10.1109\/LGRS.2010.2090034","article-title":"An object-oriented semantic clustering algorithm for high-resolution remote sensing images using the aspect model","volume":"8","author":"Yi","year":"2011","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"177","DOI":"10.1023\/A:1007617005950","article-title":"Unsupervised learning by probabilistic latent semantic analysis","volume":"42","author":"Hofmann","year":"2001","journal-title":"Mach. Learn."},{"key":"ref_3","first-page":"993","article-title":"Latent dirichlet allocation","volume":"3","author":"Blei","year":"2003","journal-title":"J. Mach. Learn. Res."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"523","DOI":"10.1016\/j.imavis.2008.04.022","article-title":"Latent mixture vocabularies for object categorization and segmentation","volume":"27","author":"Larlus","year":"2009","journal-title":"Image Vis. Comput."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"28","DOI":"10.1109\/LGRS.2009.2023536","article-title":"Semantic annotation of satellite images using latent dirichlet allocation","volume":"7","author":"Lienou","year":"2010","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1109\/LGRS.2012.2194770","article-title":"Unsupervised satellite image classification using markov field topic model","volume":"10","author":"Xu","year":"2013","journal-title":"IEEE Geosci. Remote Sens. Lett."},{"key":"ref_7","unstructured":"Wang, X., and Grimson, E. (2007, January 3\u20138). Spatial Latent Dirichlet Allocation. Vancouver, BC, Canada."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Verbeek, J., and Triggs, B. (2007, January 18\u201323). Region Classification with Markov Field Aspect Models. Minneapolis, MN, USA.","DOI":"10.1109\/CVPR.2007.383098"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"4232","DOI":"10.1109\/TIP.2012.2199127","article-title":"SAR-based terrain classification using weakly supervised hierarchical Markov aspect models","volume":"21","author":"Yang","year":"2012","journal-title":"IEEE Trans. Image Process."},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Zhu, J., Ahmed, A., and Xing, E.P. (2009, January 14\u201318). MedLDA: Maximum Margin Supervised Topic Models for Regression and Classification. Montreal, QC, Canada.","DOI":"10.1145\/1553374.1553535"},{"key":"ref_11","unstructured":"Blei, D.M., and McAuliffe, J.D. (2007, January 3\u20138). Supervised Topic Models. Vancouver, BC, Canada."},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Zhao, B., Li, F., and Xing, E. (2010, January 5\u201311). Image Segmentation with Topic Random Field. Crete, Greece.","DOI":"10.1007\/978-3-642-15555-0_57"},{"key":"ref_13","unstructured":"Lafferty, J., McCallum, A., and Pereira, F.C.N. (July, January 28). Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data. Williamstown, MA, USA."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1007\/s11263-011-0437-z","article-title":"Fast approximate energy minimization with label costs","volume":"96","author":"Delong","year":"2012","journal-title":"Int. J. Comput. Vis."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"147","DOI":"10.1109\/TPAMI.2004.1262177","article-title":"What energy functions can be minimized via graph cuts?","volume":"26","author":"Kolmogorov","year":"2004","journal-title":"IEEE Trans. Patt. Anal. Mach. Int"},{"key":"ref_16","unstructured":"Wang, Y., and Mori, G. (September, January 29). Max-Margin Latent Dirichlet Allocation for Image Classification and Annotation. Dundee, UK."}],"container-title":["Remote Sensing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2072-4292\/5\/5\/2275\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T21:46:41Z","timestamp":1760219201000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2072-4292\/5\/5\/2275"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2013,5,13]]},"references-count":16,"journal-issue":{"issue":"5","published-online":{"date-parts":[[2013,5]]}},"alternative-id":["rs5052275"],"URL":"https:\/\/doi.org\/10.3390\/rs5052275","relation":{},"ISSN":["2072-4292"],"issn-type":[{"type":"electronic","value":"2072-4292"}],"subject":[],"published":{"date-parts":[[2013,5,13]]}}}