A program to help you rescue old data. Looks useful !
WebPlotDigitizer has been continuously developed since its creation in 2011 by Ankit Rohatgi at the University of Notre Dame (Indiana, USA). This software uses affine transformations to map pixel location in the image to data points on a figure based on the calibration points provided by the user. Various image processing algorithms included with this software can be used to extract large number of data points from uploaded figures with great precision.
In order to extract a spectra from an archival paper, the first step is to create an image that isolates the spectrum, including the horizontal and vertical axes. The presence of text and legends won’t interfere. The software will ask the user to select a few known points on the axes (2 points on the X-axis and 2 more points on the Y-axis for a 2D spectrum) and enter the corresponding values.
To acquire the data after plot axes calibration, several options are available. A tedious option is to manually select data points on the image (particularly helpful for low-resolution spectra with error bars). An alternative, faster method, is to use the automatic mode. In the automatic mode, the user can set up and execute an extraction algorithm that can differentiate between the data points and the image background and identify several data points in a short time. A background mode also allows the algorithms to include everything except the background color as potential data points, a very useful feature in the case of overlapping curves/spectra of different colors.
https://arxiv.org/abs/1708.02025
No comments:
Post a Comment
Due to a small but consistent influx of spam, comments will now be checked before publishing. Only egregious spam/illegal/racist crap will be disapproved, everything else will be published.