MATLAB - Graphics

MATLAB - Advanced Graphics and Visualization

MATLAB offers a powerful and flexible graphics system for visualizing data. This chapter explores how to create common plot types and customize them to produce publication-quality figures.

• Creating bar charts for categorical data
• Visualizing 3D data with contour plots
• Generating 3D surface and mesh plots
• Modern techniques for creating high-quality graphics

Drawing Bar Charts

The bar function is ideal for displaying quantities associated with different categories. Let’s visualize the test scores of a group of students.

Example

This script will generate a bar chart of student scores and save it as a high-resolution image file.

% Data: Student scores
student_ids = 1:10;
scores = [75, 58, 90, 87, 50, 85, 92, 75, 60, 95];

% Create the figure
figure;

% Draw the bar chart with custom color
b = bar(student_ids, scores, 'FaceColor', [0.2, 0.6, 0.8]);

% Add labels and title for clarity
xlabel('Student ID');
ylabel('Score (%)');
title('First Semester Exam Scores');
grid on; % Add a grid for better readability
ylim([0 100]); % Set y-axis limits

% Save the figure to a file
% exportgraphics is the modern, recommended function for saving plots
exportgraphics(gcf, 'student_scores.png', 'Resolution', 300);

% To run this, save the code in a .m file and execute it.
% A plot window will appear, and a file named 'student_scores.png' will be saved.

The code will produce a window with a bar chart showing ten bars, each corresponding to a student’s score. The x-axis is labeled ‘Student ID’, the y-axis is ‘Score (%)’, and the chart has a title. The bars are colored blue and a grid is visible in the background. A PNG file of this chart is also created.

Drawing Contour Plots

A contour plot is a 2D representation of a 3D surface. It uses lines or filled areas to connect points of equal value (level). It’s like a topographical map for data.

Example

To create a contour plot, we first need to define a grid of (x, y) points using meshgrid, then evaluate a function z = f(x, y) at each point on the grid.

% Step 1: Create a grid of X and Y coordinates
[X, Y] = meshgrid(-2:0.2:2, -2:0.2:2);

% Step 2: Define the function Z = f(X, Y)
Z = X .* exp(-X.^2 - Y.^2);

% Step 3: Create the contour plot
figure;
[C, h] = contour(X, Y, Z, 10); % Create a contour plot with 10 levels

% Step 4: Add labels to the contour lines
clabel(C, h);

% Add title and axis labels
title('Contour Plot of Z = X*exp(-X^2 - Y^2)');
xlabel('X-axis');
ylabel('Y-axis');

% Save the figure
exportgraphics(gcf, 'contour_plot.png', 'Resolution', 300);

This script generates a 2D plot with a series of concentric, oval-shaped lines. Each line is labeled with a number representing its constant Z-value. This shows the ‘hills’ and ‘valleys’ of the 3D function on a flat plane.

Three-Dimensional Plots

For a true 3D view of a surface z = f(x, y), MATLAB provides functions like surf and mesh.

surf: Creates a 3D surface plot with solid colored faces. mesh: Creates a 3D wireframe mesh plot.

Example

% Reuse the grid and function from the contour example
[X, Y] = meshgrid(-2:0.2:2);
Z = X .* exp(-X.^2 - Y.^2);

% Create a new figure
figure;

% Use 'tiledlayout' to show multiple plots in one figure
tiledlayout(1, 2); % 1 row, 2 columns

% First tile: surf plot
nexttile;
surf(X, Y, Z);
shading interp; % Smooths the color gradient
colormap('winter'); % Apply a color scheme
title('surf() plot');
xlabel('x'); ylabel('y'); zlabel('z');

% Second tile: mesh plot
nexttile;
mesh(X, Y, Z);
title('mesh() plot');
xlabel('x'); ylabel('y'); zlabel('z');

% Save the combined figure
exportgraphics(gcf, '3d_plots.png', 'Resolution', 300);

This code produces a single figure window containing two plots side-by-side. The left plot (surf) shows a smooth, colored surface representing the function, with one peak and one valley. The right plot (mesh) shows the same surface but as a wireframe, allowing you to see through it. This demonstrates the modern tiledlayout function for arranging subplots.

Best Practices for Publication-Quality Graphics

• Always Label: Every plot needs a title, and every axis needs a label with units (xlabel, ylabel, zlabel).
• Use Handles: Get handles to plot objects to programmatically control their properties (e.g., ax = gca; ax.FontSize = 12;). This offers more control than simple commands.
• Choose Appropriate Colormaps: Use colormaps that are intuitive and accessible (e.g., perceptually uniform colormaps like ‘viridis’ or ‘parula’).
• tiledlayout over subplot: For new code, prefer tiledlayout for arranging multiple plots. It provides better spacing and more intuitive controls.
• Export with exportgraphics: Use exportgraphics instead of the older print command. It offers better control over resolution, file formats (PNG, JPEG, PDF, EPS), and background color.