Technical Computing

Technical Computing Description Topics

• The Power of Technical Computing with MATLAB
• Data Analysis and Exploration
• Mathematical Algorithms to Build On
• Modeling and Simulation
• Visualization and Image Processing
• Programming and Application Development
• Distributed and Parallel Programming

Visualization and Image Processing

MATLAB^® provides immediate access to graphics features that allow you to visually represent and understand your data. Powerful object-oriented graphics let you plot the results of your analyses, incorporate graphics into your system models, quickly render complex 3-D objects, and create presentation output. MATLAB provides:

• A variety of 2-D and 3-D plot types required in engineering and science
• Interactive tools to annotate, analyze, and customize plots
• Fast, accurate visualization of 3-D surfaces
• Integrated support for large, complex, irregularly sampled data sets
• Application-specific visualization via toolboxes
• Tools for image analysis and display

Powerful 2-D and 3-D Plotting Capabilities

MATLAB and its companion toolboxes have the specialized graphics required for technical computing built right in. From 2-D line plots of raw data, to labeled contour plots, to isosurfaces, these tools allow you to visually model and understand complex systems.

This graphic displays a time series of stock prices with minimum, maximum and mean prices. As with any MATLAB graphic it can be customized, imported by other software (such as Microsoft Word or PowerPoint), and exported for presentation. Click on image to see enlarged view. (7 KB)

Interactive Tools for Annotating, Analyzing, and Customizing Plots

New point-and-click tools support easy editing and annotation of 2-D and 3-D plots. For 2-D plots, the Basic Statistics and Basic Fitting tools let you interactively calculate statistics on the underlying data and optionally overlay descriptive numeric and graphical fit results directly on the plot. Additional interactive tools let you customize any plot result.

Here a slice plane is used to visualize the speed profile of a flow pattern in a tank.

Fast, Accurate Visualization of 3-D Surfaces

By combining features such as lighting, shading, camera control, and transparency, you can transform complex data structures into realistic 3-D surfaces and volume displays, giving you multiple layers of information in a single graphical view.

Integrated Support for Large, Complex, and Irregularly Sampled Data

Optimized to handle vector and matrix data, MATLAB can manipulate, analyze, and display data and 3-D objects saved in a variety of user-definable data structures and standard file formats. Displaying multidimensional data in MATLAB is easy. Efficient algorithms for data manipulation, interpolation, gridding, and triangulation enable the fast display of irregularly sampled, real-world data, all without preprocessing.

Here MATLAB is used to visualize a quantum mirage. This image is created using data from a scanning tunneling microscope, which was used in this experiment to capture information about the quantum nature of electron behavior. Image by H. C. Manoharan, C. P. Lutz, & D. M. Eigler, Nature 403, 512-515 (2000).

Application-Specific Visualization Via Toolboxes

The toolboxes include numerous application-specific graph types, plotting features, and interactive interfaces to view and evaluate function results. For example, the Neural Network Toolbox includes custom graphics to aid in analyzing network performance and training and you can use the Statistics Toolbox GUIs for model fitting and display.

			The Mapping Toolbox makes it easy to import, manipulate, and display geographic data. Shown here are a visual interface used to import data and a display of global digital topography data.

Tools for Displaying and Analyzing Images

MATLAB reads, writes, and displays byte and floating-point image data. The Image Processing Toolbox builds on this foundation and on two MATLAB core strengths - the matrix structure and data visualization - that make it the natural environment for operating on and displaying images and signals. Operations for image processing include:

• Filtering and 2-D filter design
• Binary image morphological operations
• Noise reduction and image enhancement
• Image transforms
• Image analysis and statistics
• Block processing
• Geometric operations
• Colormap manipulation
• Region-of-interest processing
• Colorspace conversions
• 8-bit and 16-bit integer and double precision data support

Image enhancement is often a crucial step in object identification. To simplify bubble segmentation, mathematical algorithms from MATLAB and the Image Processing Toolbox subtracted the estimated background intensity of the original image (top), resulting in more uniform intensity (bottom). (Data: R. Tutwiler, S. Brown, and W. Moore of ARL.)

Displayed at left are results of an image compression performed by the Wavelet Toolbox's two-dimensional compression tool. For this image, it was possible to force about half of the detail coefficients (high frequency components) to zero and still reconstruct our image with a good degree of accuracy. Click on image to see enlarged view. (39 KB)