Table of Contents for
QGIS: Becoming a GIS Power User

Version ebook / Retour

Cover image for bash Cookbook, 2nd Edition QGIS: Becoming a GIS Power User by Alexander Bruy Published by Packt Publishing, 2017
  1. Cover
  2. Table of Contents
  3. QGIS: Becoming a GIS Power User
  4. QGIS: Becoming a GIS Power User
  5. QGIS: Becoming a GIS Power User
  6. Credits
  7. Preface
  8. What you need for this learning path
  9. Who this learning path is for
  10. Reader feedback
  11. Customer support
  12. 1. Module 1
  13. 1. Getting Started with QGIS
  14. Running QGIS for the first time
  15. Introducing the QGIS user interface
  16. Finding help and reporting issues
  17. Summary
  18. 2. Viewing Spatial Data
  19. Dealing with coordinate reference systems
  20. Loading raster files
  21. Loading data from databases
  22. Loading data from OGC web services
  23. Styling raster layers
  24. Styling vector layers
  25. Loading background maps
  26. Dealing with project files
  27. Summary
  28. 3. Data Creation and Editing
  29. Working with feature selection tools
  30. Editing vector geometries
  31. Using measuring tools
  32. Editing attributes
  33. Reprojecting and converting vector and raster data
  34. Joining tabular data
  35. Using temporary scratch layers
  36. Checking for topological errors and fixing them
  37. Adding data to spatial databases
  38. Summary
  39. 4. Spatial Analysis
  40. Combining raster and vector data
  41. Vector and raster analysis with Processing
  42. Leveraging the power of spatial databases
  43. Summary
  44. 5. Creating Great Maps
  45. Labeling
  46. Designing print maps
  47. Presenting your maps online
  48. Summary
  49. 6. Extending QGIS with Python
  50. Getting to know the Python Console
  51. Creating custom geoprocessing scripts using Python
  52. Developing your first plugin
  53. Summary
  54. 2. Module 2
  55. 1. Exploring Places – from Concept to Interface
  56. Acquiring data for geospatial applications
  57. Visualizing GIS data
  58. The basemap
  59. Summary
  60. 2. Identifying the Best Places
  61. Raster analysis
  62. Publishing the results as a web application
  63. Summary
  64. 3. Discovering Physical Relationships
  65. Spatial join for a performant operational layer interaction
  66. The CartoDB platform
  67. Leaflet and an external API: CartoDB SQL
  68. Summary
  69. 4. Finding the Best Way to Get There
  70. OpenStreetMap data for topology
  71. Database importing and topological relationships
  72. Creating the travel time isochron polygons
  73. Generating the shortest paths for all students
  74. Web applications – creating safe corridors
  75. Summary
  76. 5. Demonstrating Change
  77. TopoJSON
  78. The D3 data visualization library
  79. Summary
  80. 6. Estimating Unknown Values
  81. Interpolated model values
  82. A dynamic web application – OpenLayers AJAX with Python and SpatiaLite
  83. Summary
  84. 7. Mapping for Enterprises and Communities
  85. The cartographic rendering of geospatial data – MBTiles and UTFGrid
  86. Interacting with Mapbox services
  87. Putting it all together
  88. Going further – local MBTiles hosting with TileStream
  89. Summary
  90. 3. Module 3
  91. 1. Data Input and Output
  92. Finding geospatial data on your computer
  93. Describing data sources
  94. Importing data from text files
  95. Importing KML/KMZ files
  96. Importing DXF/DWG files
  97. Opening a NetCDF file
  98. Saving a vector layer
  99. Saving a raster layer
  100. Reprojecting a layer
  101. Batch format conversion
  102. Batch reprojection
  103. Loading vector layers into SpatiaLite
  104. Loading vector layers into PostGIS
  105. 2. Data Management
  106. Joining layer data
  107. Cleaning up the attribute table
  108. Configuring relations
  109. Joining tables in databases
  110. Creating views in SpatiaLite
  111. Creating views in PostGIS
  112. Creating spatial indexes
  113. Georeferencing rasters
  114. Georeferencing vector layers
  115. Creating raster overviews (pyramids)
  116. Building virtual rasters (catalogs)
  117. 3. Common Data Preprocessing Steps
  118. Converting points to lines to polygons and back – QGIS
  119. Converting points to lines to polygons and back – SpatiaLite
  120. Converting points to lines to polygons and back – PostGIS
  121. Cropping rasters
  122. Clipping vectors
  123. Extracting vectors
  124. Converting rasters to vectors
  125. Converting vectors to rasters
  126. Building DateTime strings
  127. Geotagging photos
  128. 4. Data Exploration
  129. Listing unique values in a column
  130. Exploring numeric value distribution in a column
  131. Exploring spatiotemporal vector data using Time Manager
  132. Creating animations using Time Manager
  133. Designing time-dependent styles
  134. Loading BaseMaps with the QuickMapServices plugin
  135. Loading BaseMaps with the OpenLayers plugin
  136. Viewing geotagged photos
  137. 5. Classic Vector Analysis
  138. Selecting optimum sites
  139. Dasymetric mapping
  140. Calculating regional statistics
  141. Estimating density heatmaps
  142. Estimating values based on samples
  143. 6. Network Analysis
  144. Creating a simple routing network
  145. Calculating the shortest paths using the Road graph plugin
  146. Routing with one-way streets in the Road graph plugin
  147. Calculating the shortest paths with the QGIS network analysis library
  148. Routing point sequences
  149. Automating multiple route computation using batch processing
  150. Matching points to the nearest line
  151. Creating a routing network for pgRouting
  152. Visualizing the pgRouting results in QGIS
  153. Using the pgRoutingLayer plugin for convenience
  154. Getting network data from the OSM
  155. 7. Raster Analysis I
  156. Using the raster calculator
  157. Preparing elevation data
  158. Calculating a slope
  159. Calculating a hillshade layer
  160. Analyzing hydrology
  161. Calculating a topographic index
  162. Automating analysis tasks using the graphical modeler
  163. 8. Raster Analysis II
  164. Calculating NDVI
  165. Handling null values
  166. Setting extents with masks
  167. Sampling a raster layer
  168. Visualizing multispectral layers
  169. Modifying and reclassifying values in raster layers
  170. Performing supervised classification of raster layers
  171. 9. QGIS and the Web
  172. Using web services
  173. Using WFS and WFS-T
  174. Searching CSW
  175. Using WMS and WMS Tiles
  176. Using WCS
  177. Using GDAL
  178. Serving web maps with the QGIS server
  179. Scale-dependent rendering
  180. Hooking up web clients
  181. Managing GeoServer from QGIS
  182. 10. Cartography Tips
  183. Using Rule Based Rendering
  184. Handling transparencies
  185. Understanding the feature and layer blending modes
  186. Saving and loading styles
  187. Configuring data-defined labels
  188. Creating custom SVG graphics
  189. Making pretty graticules in any projection
  190. Making useful graticules in printed maps
  191. Creating a map series using Atlas
  192. 11. Extending QGIS
  193. Defining custom projections
  194. Working near the dateline
  195. Working offline
  196. Using the QspatiaLite plugin
  197. Adding plugins with Python dependencies
  198. Using the Python console
  199. Writing Processing algorithms
  200. Writing QGIS plugins
  201. Using external tools
  202. 12. Up and Coming
  203. Preparing LiDAR data
  204. Opening File Geodatabases with the OpenFileGDB driver
  205. Using Geopackages
  206. The PostGIS Topology Editor plugin
  207. The Topology Checker plugin
  208. GRASS Topology tools
  209. Hunting for bugs
  210. Reporting bugs
  211. Bibliography
  212. Index

Georeferencing vector layers

For various reasons, sometimes you have a vector layer that lacks projection information. This is often the case with CAD layers that were created only in local coordinates. When it is possible, try to track down the original projection information. As a last resort, you can attempt to warp the vector layer to match a known reference layer with the recipe described here.

Getting ready

You can open two instances of QGIS (or use one as you'll just be zooming back and forth a lot). In one instance, load a reference layer, something in the projection that you want your data to be in. Activate Coordinate Capture Plugin from the Manage Plugins menu.

Note

In Windows, you need the osgeo4w shell for this recipe. If you don't have a start menu item, look for the OSGeo4W.bat launcher in your QGIS or OSGeo4w installation folder.

This example uses cad-lines-only.shp, which is the line layer extracted from the CSS-SITE-CIV.dxf file. This file is a CAD rendering of design plans for Academy St. in the town of Cary, Wake County, North Carolina.

How to do it…

  1. Create a list of GCP matches between your unknown layer (cad-lines-only.shp) and your reference layer (CarystreetsND83NC.shp).
  2. Here are some specific adjustments to help with cad-lines-only.shp referenced to CarystreetsND83NC.shp. These will make it easier to find matches between the two layers:
    1. Load cad-lines-only.shp, and adjust its style properties using a rule-based style. Use the "Layer" = 'C-ROAD-CNTR' rule, which will only show you street centerlines.
    2. In your other QGIS session, load CarystreetsND83NC.shp in order to find the matching area, open the attribute table, and apply the following select expression: "Street" LIKE '%N ACADEMY%' OR "Street" LIKE '%S ACADEMY%' OR "Street" LIKE '%CHATHAM%'. The filter here highlights the three main streets of the original project, which is at the intersection of Chatham and N/S Academy streets in the center of the town. This may also be useful to change the color of the selected features to make it easier to find. The traffic circles at either end of the project are good landmarks:
      How to do it…
    3. Find an easy-to-identify feature that matches in both layers (street intersections).
    4. Use the coordinate capture plugin to copy the x,y value for the point in both layers.
    5. Save the coordinates in a text editor while you work.
    6. Repeat this procedure until you have at least four pairs of points. Try to pick points spread out across the whole layer:
      How to do it…

    Note

    There is currently no graphical interface in QGIS for the next step, which uses the OGR library that comes with QGIS. Take the list of points and using the ogr2ogr command-line, you're going to apply the GCP to the unknown layer.

  3. Each set of coordinate pairs will look as follows:
    -gcp sourceX sourceY destinationX destinationY
  4. Open a terminal (Mac or Linux) or an OSGeo4w shell (Windows).
  5. Change to the directory where you have the data (Hint: cd /home/user/Qgis2Cookbook/):
    ogr2ogr -a_srs EPSG:3358 -gcp 2064886.09740 741552.90836 629378.595 226024.853 -gcp 2066610.97021 741674.39817 629903.420 226064.049 -gcp 2064904.46214 743055.63847 629384.784 226485.725 -gcp 2062863.85707 741337.65243 628762.587 225960.900 cad_lines_nd83nc.shp cad-lines-only.shp

    -a_srs is the proj code for your reference layer.

    The command pattern is ogr2ogr <options> <destination> <source>.

    Tip

    Other useful advanced options include -order <n> to indicate polynomial level (default is based on the number of GCPs) or -tps to use Thin-plate-spline instead of polynomial. For more options refer to http://www.gdal.org/ogr2ogr.html.

  6. Now, load your new cad_lines_nd83nc.shp file in the same project, as CarystreetsND83NC.shp. They should line up without the need to enable projection-on-the-fly:
    How to do it…

How it works…

Given the list of input coordinates and matching output coordinates, a math formula is derived to translate between the two sets. This formula is then applied to all the points in the original data. The result of this is a reprojected dataset from an unknown projection to a known projection.

Note

The original data is actually EPSG:102719, but we're pretending that we didn't have this piece of information to demonstrate this example.

There's more…

When picking a reference layer, try to pick something in the projection that you want to use for your maps and analysis. That way you only have to reproject once, as each additional transformation can add an error. There's also more than one way to go about accomplishing this task, including moving the data by hand.

In this particular, example the transformation is autoselected based on the number of GCP point pairs. 4-5 is the first order polynomial, 6-9 is the second order polynomial, and 10+ is the third order polynomial. Refer to the previous recipe in this chapter for more information.

A related topic is Affine transformations when you simply want to shift or rotate a vector layer by a known amount. The QgsAffine plugin is great if you already know the parameters, or roughly know how far you want to rotate and shift the vector layer, as it then just needs some math to get the parameters.

Note

Maybe by the time you read this, all of the difficult things here will be worked in a plugin. Keep an eye open, and try the experimental plugins Vector Bender, vectorgeoref, and Affine Transformations.

See also