Skip to content

Digitizing the Plot

4 messages · Shubha Vishwanath Karanth, Philipp Pagel, Baptiste Auguie +1 more

#
Hi!
I am not aware of an R-package but there are several applications that
produce tables from graphs which you can import into R afterwards. A
few examples:

  g3data (LINUX, opensource)
  engauge (LINUX/Windows, opensource)
  Grab it! (Windows, commercial)
  Datathief (LINUX/Windows/Mac, shareware)
  UN-SCAN-IT (Windows/Mac, commercial)

Google digs up quite a few additional hits with "digitizing graph
data".

cu
	Philipp
#
The grImport package seems to provide such possibility for vector  
graphics,

http://www.stat.auckland.ac.nz/~paul/Talks/gddg.pdf

imageJ is another open-source option.

baptiste
On 16 Apr 2009, at 16:44, Shubha Vishwanath Karanth wrote:

            
_____________________________

Baptiste Augui?

School of Physics
University of Exeter
Stocker Road,
Exeter, Devon,
EX4 4QL, UK

Phone: +44 1392 264187

http://newton.ex.ac.uk/research/emag
#
I don't know of anything that does the automated process like the example software you link to, but here is one procedure that you can use to "do it by hand".

1. Brind the graph into R and plot it (as an image) using rimage or EBImage (or other) packages.

2. use the locator function to find the coordinates (under the current user coord system) of 2 points that you know the coordinates of (in the scale of the graph, desired coord system).

3. Use the updateusr function from the TeachingDemos package to convert the coordinates to match the axes (using the points from the previous step).

4. Use the locator function and click on the points/curve in the graph.  The result will be an approximation to the data of interest (you may want to round or do other clean-up, or just use as is).

Hope this helps,