{"id":378,"date":"2018-05-03T14:35:43","date_gmt":"2018-05-03T12:35:43","guid":{"rendered":"https:\/\/blogs.fu-berlin.de\/reseda\/?page_id=378"},"modified":"2018-09-27T15:00:56","modified_gmt":"2018-09-27T13:00:56","slug":"classification-in-r","status":"publish","type":"page","link":"https:\/\/blogs.fu-berlin.de\/reseda\/classification-in-r\/","title":{"rendered":"Classification in R"},"content":{"rendered":"

We want to utilize a Random Forest (RF) and a Support Vector Classifier (SVM) algorithm in order to classify the Berlin land cover in six elementary categories: bare soil<\/em>, water<\/em>, grassland<\/em>, forest<\/em>, urban low density<\/em>, and urban high density<\/em>. Therefore, we need an image dataset and a shapefile containing points or polygons to which the respective class is attributed.<\/p>\n

The workflow will be exemplified by a L8 scene (ID: LC08_L1TP_193023_20170602_20170615_01_T1), which you may already have acquired during the L8 Download Exercise<\/a>. You need to preprocess the scene as shown in chapter Preprocess<\/a>. In addition, we narrowed our research area to Berlin to keep the data small (as shown in chapter Visualize in R<\/a>).<\/p>\n

You can download both the preprocessed image and the shapefile for testing purposes here<\/a>.<\/p>\n

\n

Section in a Box<\/h1>\n

This section guides you through a complete classification process for satellite imagery. The resulting classification maps will be validated in the next chapter<\/a>.<\/p>\n

Sample in QGIS<\/strong><\/a>
\n– some basic considerations and tips for sampling
\n– collect training polygons in QGIS for supervised classification
\nPrepare Samples in R<\/strong><\/a>
\n– import training polygons into R
\n– use training polygons to extract raster information
\n– put everything together in a data frame
\nRF Classification<\/strong><\/a>
\n– train a RF model with “randomForest” package
\n– classify image data and export a classification image
\nSVM Classification<\/strong><\/a>
\n– train a SVM (C-Classification method) with “e1071” package
\n– classify image data and export a classification image\n<\/div>\n

<\/br><\/br><\/p>\n

\n
\n