Is there any way of extracting raw data from scanned plots of charts with X and Y scales defined on graph paper?

I want to use scanned image of a chart with plotted data. If I define the X/Y scaling and limits, is there a way to extract plotted data points from the scanned image using MATLAB without having the raw data available?

[John BG]

Hi Thomas

this is John BG <mailto:[e-mail] [e-mail]>

Indeed it is (possible).

Can you attach the image to your question so that readers can have a look?

Answer 1

[Walter Roberson]

Have a look at https://www.mathworks.com/matlabcentral/fileexchange/?utf8=%E2%9C%93&term=tag%3Adigitize

I went through and tagged a number of contributions for this purpose last year. There might also have been some newer ones added after I did the sweep.

[Thomas Gunther]

Thank you Walter. The link you provided offers many different options to meet my needs and I'm sure this will be very helpful in the future.

Answer 2

[Thomas Gunther]

I have attached a simple graph showing Tue Airspeed vs. Pressure Altitude and I want to be able to extract data from one of the Mach(M) Speed lines depicted.

Answer 3

[Jeff Miller]

This is not inside MATLAB, but it might be helpful anyway: WebPlotDigitizer

Answer 4

[James]

I know this is an old question, but for those who don't want to use the WebPlotDigitizer online resource, the following code does the same for a simple 2D X-Y plot (linear scale).

%% Get Image
cd('Path to directory containing the image')
image_data = imread('Path to image of figure');
imshow(image_data);


% Set the scale by selecting two points on the x-axis (small to large)
x_values = [min_x, max_x]; % Replace with the numerical values of the points you will select
disp('Click on point (min_x, 0.00)');
[x_pixel1, ~] = ginput(1); % Click on the pixel position for the minimum x-axis value
disp('Click on point (max_x, 0.00)');
[x_pixel2, ~] = ginput(1); % Click on the pixel position for the maximum x-axis value

% Now select two points on the y-axis (large to small this time)
y_values = [max_y, min_y]; % Replace with the numerical values of the points you will select
disp('Click on point (0.00, max_y)');
[~, y_pixel1] = ginput(1); % Click on the pixel position for the maximum y-axis value
disp('Click on point (0.00, min_y');
[~, y_pixel2] = ginput(1); % Click on the pixel position for the minimum y-axis value


%% Get Data

% Calculate scales for x and y axes
x_axis_scale = diff([x_values(1), x_values(2)]) / diff([x_pixel1, x_pixel2]);
y_axis_scale = diff([y_values(1), y_values(2)]) / diff([y_pixel2, y_pixel1]);

disp('Click on the curve to select points. Press Enter when finished.');
[x, y] = ginput; % Select points using mouse clicks

% Normalize the coordinates using the scale of the axes
x = (x - x_pixel1) * x_axis_scale + x_values(1);
y = (y_pixel1 - y) * y_axis_scale + y_values(1);