CIV1900 Matlab - Plotting & Coursework

Contents for Lecture 5
• Plotting capabilities in Matlab
• Digital images and their construction
• Image processing

Matlab offers a rich graphics
environment and even if you do not
feel comfortable programming in
Matlab you should make use of its
graphics capabilities as it will allow
you to produce much higher quality
graphs and figures than is possible
in Excel, for example.

Pressure on the rib and in its lee

The most basic function you can use is to
generate scatterplots.
Imagine that you have an array of data where the
first column contains the year (1985-2010) in
which a sample was taken and the second is
some property of that sample.
A basic piece of exploratory data analysis would
be to plot these data against one another.

The Matlab command is “plot”
i.e.
plot(Data(:,1),Data(:,2))
This defaults to producing a solid line in blue

plot(Data(:,1),Data(:,2))

We can make use of a shorthand for line symbols,
line colours and line types to vary this. Some
examples:
Colours Types Symbols
b = blue - = solid line o = circles
r = red : = dotted s = square
k = black -- = dashed d = diamond
g = green ^ = triangle

Plot(Data(:,1),Data(:,2),‟:ok‟)

Plot(Data(:,1),Data(:,2),‟--^g‟,‟linewidth‟,2)

We can improve the look of our plots by editing
the axis properties.

Click Here

Here I have edited the text size and the upper limit
of the x-axis.

Here I have defined the y-axis with a mathematical
expression (you can‟t do this in Excel)

Here I have converted the axes to log axes

figure
plot (Data(:,1),Data(:,2),':^m','linewidth',1.5)
hold on
plot (Data(:,1),Data(:,3),„--m','linewidth',1.5)

figure
subplot(1,3,1:2)
plot (Data(:,1),Data(:,2),':^m','linewidth',1.5)
subplot(1,3,3)
plot (Data(:,1),Data(:,3),„--m','linewidth',1.5)

A 3D line plot:
figure
plot3(Data(:,1),Data(:,2),Data(:,3),‟k‟,‟linewidth‟,2)

If we have an array of data we can plot it as a surface,
where the x and y axes are given by the size of the array
and the z-axis by the values in the array

If we have an array of data we can plot it
as a surface, where the x and y axes are
given by the size of the array and the z-
axis by the values in the array.
surf(CIV1900_images{4})

Images may also be displayed using the “image”
or “imagesc” commands.
Hopefully the previous example has highlighted
that an image is simply an array of digital
numbers.
This example was coloured but it was a “false
colour” image. There was only level to the array
and the colouring was applied in bands:
high values (peaks) in red; low values (troughs) in
blue.
It might as well be a grey-scale image and we can
make it so by changing the “colormap”.

surf(CIV1900_images{4})
colormap(„gray‟)

A “true colour” image (like a jpeg file) is
constructed rather differently. Instead of a single
layer to the array, it has layers for each colour:
Three layers for an RGB image
Four for a CMYK image
Each layer ranges from 0 to 255 and tells you how
red or green or blue the colour is at that point.
When they are all combined they give you the
colour at that location, which using a legend, you
can relate to properties of the image.

Colour maps in Matlab by default range from 0 to
63.
The “image” command displays an image literally
relating the values in the array to the colour map.
The “imagesc” command rescales the values in
the array so that they fit the colour map.
For example, the values in our image go from 0-
255 so using “image” we would expect about 75%
of the image to wash out as white

figure
image(CIV1900_images{4})
colormap('gray')

figure
imagesc(CIV1900_images{4})
colormap('gray')

figure
imagesc(CIV1900_images{4})
axis („image‟)
colormap('gray')

figure
for loop1=1:9
subplot(3,3,loop1)
imagesc(CIV1900_images{loop1*3+4})
axis image
colormap('gray')
end

Matlab provides a flexible tool for producing
graphics.
This includes images, which are simply arrays of
numbers (1 layer for grey-scale, more for true
colour).
Image processing is an important area of
engineering science:
Stress analysis on beams using photostress
systems;
Particle Imaging velocimetry;
Monitoring of land use change using remote
sensing;
Face or fingerprint recognition software, etc.

The Coursework

Sobel Filtering – Edge Detection

The Coursework

The convolution integral

The Coursework Filter

Image

The Coursework

Sobel X Sobel Y

CIV1900 Matlab - Plotting & Coursework

More Related Content

What's hot

Viewers also liked

Similar to CIV1900 Matlab - Plotting & Coursework

CIV1900 Matlab - Plotting & Coursework