This worksheet continuous with the computation of artificial images and covers the application of spatial filters which alter the pixel values of one band based on the pixels' neighborhood.
After completing this worksheet you should know how to apply spatial filters.
As always, please add these entries to your today's learning log in the beginning of your Rmd file you will use to push to your GitHub classroom.
As today's special, please complete the following assignment:
When it comes to filtering, one has to decide the type of filter and the area which will be included into the computation. While we will use a set of Harlick texture extraction filters which are made available through the R package glcm, we will still have to define the extension of the filter window.
For the following tasks, please choose one of the filters mean, variance, homogeneity, contrast, dissimilarity, entropy, second moment or correlation which are all provided by the glcm package.
Please write an R script which computes the filter for a set of different window extents. Visualize your data and decide which resolution work best. Save the best window size as well as one smaller and one larger size as a GeoTIFF to your hard disk.
Please write an Rmd file with html output which reads the above tiff files. Visualize the data and explain how your filter works with not more than two sentences. Add a maximum of three more sentences to explain why you think that your chosen window size works best in comparison with the two other images (i.e. one with a smaller, one with a larger resolution).
Knitr your Rmd file and include your two R scripts in your repository, update (i.e. commit) it and publish (i.e. push) it to the GitHub classroom. Make sure that the created html file is also part of your git repository.