readme.txt

rang doppler imaging and motion compensation中的源代码

Translation:

The following is a brief description of the different matlab functions that are needed to execute some of the simulations discussed in the book. More information is given by typing help and the function name using matlab.

It is assumed that the user has available the following matlab toolboxes:

Optimization Toolbox
Signal Processing Toolbox
Statistic Toolbox
Image Processing Toolbox


Example1.m

This  is an example program to generate a simulated point imagery with translational and rotational motion. Given
generated signature, The target motion is estimated by iterative method of motion compensation using weighted least squares method.
The motion estimates is found via optimization for fast global minimum search.

Example2.m

This  is an example program to generate a simulated point imagery with translational and rotational motion. The
generated signature is then motion compensated and its image is displayed in spatial domain.

Centerm.m

Converts the data to m by nn matrix

Cia.m

Finds a starting and ending points of a segment using complex image analysis method.

Ciaphase.m

Finds starting and ending points of a segment using the linearity of the phase for complex image analysis.

Ciaproc.m

Decide final starting and ending points of a segment based on the information supplied by the amplitude information using CIA
function and the phase information using CIAphase for the complex image analysis.

Cleaning3.m

Cleanes the data in the spatial domain to raise SNR. The noise level is estimated using Weibull
distribution and thrshold level is determined. Linear filtering is used to clean the data.

Cramer2.m

Calculates Cramer-Rao lower bound of the velocity and acceleration estimates. It provides the precision
achievable in the motion estimates so that it limites the number of iteration in motion estimation.

Dispdft.m

Display or save the image of a burst in spatial domain.

FFTshift2.m

Swap the upper half part of a matrix with lower half part of the matrix.

Filtindex.m

Find the center of the selected region in the spatial domain.

Fmine5.m

Compares actual focal quality indicator with theoretical focal quality indicator which is based on Cramer-Rao
lower bound. If actual focal quality indicator is near Cramer-Rao bound, then the iteration stops.

Gradient5.m

Finds the gradient of the phase in a burst or in a segment.

Kaiser2.m

Generates two dimensional m by n Kaiser window.

Loaddata.m

Loads the data. For actual data, the name of data must be 'd727un' or 'dc10un'. For simulated data, the name 
of the data must be 'simuldata' or 'simuldat'. dataopt are 'd727un', 'dc10un', 'simuldata' or 'simuldat'.
ordata is original data. windata is kaiser windowed data.

Makevec.m

Makes the data as a matrix of desired size by zero padding.

Minimize.m

Find the polynomial estimates where the focal quality indicator is minimum. In this algorithm, the weighted
least squares combined with complex analysis method is used as focal quality indicator.

Morph.m

Determines the kind of morphological operation.

Morpharea.m

Finds the maximum length of the region selected by morphological image processing.

Motioncomp.m

Motion compensation algorithm using given data and motion parameter estimates. Returns motion compensated matrix.

Noisesect.m

Select the region composed by noise in spatial domain.

Normalize.m

Normalize to make the sum of magnitude in a burst to be one in spatial domain.

Rdisplay.m

Displays the image in various ways.

Repnan.m

Converts any NaN data to 0 so that the matrix is defined.

Shiftzero.m

This function moves the radar image to the center of the imaging window.

Signalgen.m

Generates radar echo transfer function given radar parameters.

Timegen.m

Generates the time coordinates for the given radar parameters.

tranfreqm.m

if m and n are defined only, a matrix such as
[0 1 2 3 .... n-1]
[0 1 2 3 .... n-1]
[. . . . ....  . ]
[0 1 2 3 .... n-1]
will be made.

Weibmen.m

Estimation of Weibull parameters using the method of Menon.

Wls.m

Calculates polynomial estimates using weighted least squares fitting.