@article{RN511,
   author = {Hagner, Olle and Reese, Heather},
   title = {A method for calibrated maximum likelihood classification of forest types},
   journal = {Remote Sensing of Environment},
   volume = {110},
   number = {4},
   pages = {438-444},
   abstract = {A modification to the maximum likelihood algorithm was developed for classification of forest types in Sweden's part of the CORINE land cover mapping project. The new method, called the “calibrated maximum likelihood classification” involves an automated and iterative adjustment of prior weights until class frequency in the output corresponds to class frequency as calculated from objective (field-inventoried) estimates. This modification compensates for the maximum likelihood algorithm's tendency to over-represent dominant classes and under-represent less frequent ones. National forest inventory plot data measured from a five-year period are used to estimate relative frequency of class occurrence and to derive spectral signatures for each forest class. The classification method was implemented operationally within an automated production system which allowed rapid production of a country-wide forest type map from Landsat TM/ETM+ satellite data. The production system automated the retrieval and updating of forest inventory plots, a plot-to-image matching routine, illumination and haze correction of satellite imagery, and classification into forest classes using the calibrated maximum likelihood classification. This paper describes the details of the method and demonstrates the result of using an iterative adjustment of prior weights versus unadjusted prior weights. It shows that the calibrated maximum likelihood algorithm adjusts for the overclassification of classes that are well represented in the training data as well as for other classes, resulting in an output where class proportions are close to those as expected based on forest inventory data.},
   keywords = {Prior weights
Maximum likelihood classification
Landsat
Forest inventory
Plot location
Haze correction
Area estimates
CORINE},
   ISSN = {0034-4257},
   DOI = {10.1016/j.rse.2006.08.017},
   url = {https://doi.org/10.1016/j.rse.2006.08.017},
   year = {2007},
   type = {Journal Article}
}

{"_id":"Y6sfZKWhpPqi9o3jN","bibbaseid":"hagner-reese-amethodforcalibratedmaximumlikelihoodclassificationofforesttypes-2007","author_short":["Hagner, O.","Reese, H."],"bibdata":{"bibtype":"article","type":"Journal Article","author":[{"propositions":[],"lastnames":["Hagner"],"firstnames":["Olle"],"suffixes":[]},{"propositions":[],"lastnames":["Reese"],"firstnames":["Heather"],"suffixes":[]}],"title":"A method for calibrated maximum likelihood classification of forest types","journal":"Remote Sensing of Environment","volume":"110","number":"4","pages":"438-444","abstract":"A modification to the maximum likelihood algorithm was developed for classification of forest types in Sweden's part of the CORINE land cover mapping project. The new method, called the “calibrated maximum likelihood classification” involves an automated and iterative adjustment of prior weights until class frequency in the output corresponds to class frequency as calculated from objective (field-inventoried) estimates. This modification compensates for the maximum likelihood algorithm's tendency to over-represent dominant classes and under-represent less frequent ones. National forest inventory plot data measured from a five-year period are used to estimate relative frequency of class occurrence and to derive spectral signatures for each forest class. The classification method was implemented operationally within an automated production system which allowed rapid production of a country-wide forest type map from Landsat TM/ETM+ satellite data. The production system automated the retrieval and updating of forest inventory plots, a plot-to-image matching routine, illumination and haze correction of satellite imagery, and classification into forest classes using the calibrated maximum likelihood classification. This paper describes the details of the method and demonstrates the result of using an iterative adjustment of prior weights versus unadjusted prior weights. It shows that the calibrated maximum likelihood algorithm adjusts for the overclassification of classes that are well represented in the training data as well as for other classes, resulting in an output where class proportions are close to those as expected based on forest inventory data.","keywords":"Prior weights Maximum likelihood classification Landsat Forest inventory Plot location Haze correction Area estimates CORINE","issn":"0034-4257","doi":"10.1016/j.rse.2006.08.017","url":"https://doi.org/10.1016/j.rse.2006.08.017","year":"2007","bibtex":"@article{RN511,\r\n author = {Hagner, Olle and Reese, Heather},\r\n title = {A method for calibrated maximum likelihood classification of forest types},\r\n journal = {Remote Sensing of Environment},\r\n volume = {110},\r\n number = {4},\r\n pages = {438-444},\r\n abstract = {A modification to the maximum likelihood algorithm was developed for classification of forest types in Sweden's part of the CORINE land cover mapping project. The new method, called the “calibrated maximum likelihood classification” involves an automated and iterative adjustment of prior weights until class frequency in the output corresponds to class frequency as calculated from objective (field-inventoried) estimates. This modification compensates for the maximum likelihood algorithm's tendency to over-represent dominant classes and under-represent less frequent ones. National forest inventory plot data measured from a five-year period are used to estimate relative frequency of class occurrence and to derive spectral signatures for each forest class. The classification method was implemented operationally within an automated production system which allowed rapid production of a country-wide forest type map from Landsat TM/ETM+ satellite data. The production system automated the retrieval and updating of forest inventory plots, a plot-to-image matching routine, illumination and haze correction of satellite imagery, and classification into forest classes using the calibrated maximum likelihood classification. This paper describes the details of the method and demonstrates the result of using an iterative adjustment of prior weights versus unadjusted prior weights. It shows that the calibrated maximum likelihood algorithm adjusts for the overclassification of classes that are well represented in the training data as well as for other classes, resulting in an output where class proportions are close to those as expected based on forest inventory data.},\r\n keywords = {Prior weights\r\nMaximum likelihood classification\r\nLandsat\r\nForest inventory\r\nPlot location\r\nHaze correction\r\nArea estimates\r\nCORINE},\r\n ISSN = {0034-4257},\r\n DOI = {10.1016/j.rse.2006.08.017},\r\n url = {https://doi.org/10.1016/j.rse.2006.08.017},\r\n year = {2007},\r\n type = {Journal Article}\r\n}\r\n\r\n","author_short":["Hagner, O.","Reese, H."],"key":"RN511","id":"RN511","bibbaseid":"hagner-reese-amethodforcalibratedmaximumlikelihoodclassificationofforesttypes-2007","role":"author","urls":{"Paper":"https://doi.org/10.1016/j.rse.2006.08.017"},"keyword":["Prior weights Maximum likelihood classification Landsat Forest inventory Plot location Haze correction Area estimates CORINE"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://www.slu.se/globalassets/ew/org/centrb/rt/dokument/publikationslistor/rikskogstaxeringsprojektet_bibtex_ny.txt","dataSources":["LErWk8avdgETh2iR9","Cac5nog9ND5ndLYhY","xTmt4jq9swAwpBHJB","LXbacBrgTRDPkh9C2","dLPsL5XH9N5Pjush7"],"keywords":["prior weights maximum likelihood classification landsat forest inventory plot location haze correction area estimates corine"],"search_terms":["method","calibrated","maximum","likelihood","classification","forest","types","hagner","reese"],"title":"A method for calibrated maximum likelihood classification of forest types","year":2007}