xepost / academicfigures

Matlab scripts to create readable, color and high contrast figures for publications

AcademicFigures

Matlab scripts to create readable, color and high contrast figures for publications.

Academic Figures provides the command afigure that behaves as figure but creates a readable figure to embed it in a text document. Academic Figures produces figures with these properties by default (they can be customized):

1. White background instead of gray or operating system defined.
2. Large font size for legend, axis labels, ticks and title.
3. Wider lines when plotting.
4. No blank space around the chart. Thus, the image can be saved to a file without any margin.
5. Color and gray colormaps that keep the high contrast even when printed as grayscale.
6. Chart box and grid enabled by default.
7. The dimensions and size ratio can be fixed.
8. It nicely fits plot and bar commands.

Usage

Its usage is very simple. Just type afigure before any of your plot or bar commands. You can check AcademicFigures_Example to show a complete example.

You can also call aplot or abar directly without calling afigure first. These have the same input and output parameters as plot and bar.

Customization

The configuration of the Academic Figures is given by a structure returned by aconfig:

config = aconfig;

This defines these variables:

• BackgroundColor (default: [1 1 1]): figure background color.
• FontSize (default: 20): title, axes, labels and text font size.
• LineWidth (default: 2): width of plot lines.
• DrawBox (defaut: true): show a full box around around the axes.
• Grid (default: true): show grid.
• RemoveMargins (default: true): remove blank space around axes.
• Width (default: []): figure width in pixels. Ignored if empty.
• Height (default: []): figure height in pixels. Ignored if empty.
• SizeRatio (default: 4/3): width by height ratio of window. Ignored if empty of it both Width and Height are given.
• Colormap (default: 'cmr'): colormap. It can be a Nx3 matrix with values in [0, 1] representing N RGB colors, or the name of a predefined map (these maps keep high contrast when converted into grayscale): 'thermal', 'cmr', 'dusk', 'hsv2', 'gray'.
• LineStyles (default: '-|-.|--'): styles of plot lines.

You can change any variable and create an afigure with the new configuration:

config = aconfig;
config.BackgroundColor = [1 0 0]; % red background
config.Colormap = 'dusk';
afigure(config);

For convenience, you can modify the variables inline:

afigure(aconfig('BackgroundColor', [1 0 0], 'Colormap', 'dusk'));

Examples

You can check AcademicFigures_Example to show a complete examples. Some short examples are given in the following.

Example 1

afigure;

% some data
p1 = [95, 95:-1:0];
r1 = [0, 80 + [1:10, 10:0.1:15, 15:0.05:16.7]];
p2 = [90, 90:-1:0];
r2 = [0, 60 + [1:10, 10:0.1:15, 15:0.05:16.45]];

hold all;
plot(r1, p1);
plot(r2, p2);
xlabel('Recall (%)');
ylabel('Precision (%)');
legend('Dataset 1', 'Dataset 2', 'Location', 'SouthWest');
title('Academic Figure');

This produces this plot:

In comparison, this is the output of the same code using the figure command:

Example 2

You can also create bar plots:

afigure;

data = [10 20 30; 5 10 15];
bar(data);
xlabel('Dataset');
ylabel('Execution time (ms)');
set(gca, 'XTickLabel', {'Method 1', 'Method 2'});
legend('Step 1', 'Step 2', 'Step 3', 'Location', 'NorthEast');
title('Academic Bar Figure');

This produces:

In comparison, this is the output of the same code using the figure command:

Example 3

This example shows how to customize a figure:

afigure(aconfig('Colormap', 'dusk', 'LineWidth', 5, ...
    'LineStyles', ':|--'));
    
data = repmat(12:-1:1, 5, 1);
plot(data);
xlabel('X Axis');
ylabel('Y Axis');
legend('1', '2', '3', '4', '5', '6', '7', '8', '9', '10', '11', '12', ...
  'Location', 'EastOutside');
title('Academic Figure with custom parameters');

This produces: