Products > Software > libdtcam
libdtcam
This package is developed for easy integration of DTI interferometers into user software.
It contains C control and image capture library, Python and Matlab bindings.
All modules are available in source code.
C library
Open-source, platform independent, socket based image capture and control C library.
It has simple functions, and can be easy integrated into new or existing software.
Functions
perform_http_request
( addr, port, uri, buf, size, err_buf, err_size )
This function creates socket and sends HTTP request to web server, waits response and stores it into buffer.
Parameters:
addr - server IP address
port - server port
uri - resource name
buf - response from server
size - buffer size
err_buf - buffer for error message
err_size - maximal size of error message
Return values:
response size - if >= 0
error code - if < 0
process_http_response
( buf, size, type, err_buf, err_size )
Parse response from HTTP server.
Parameters:
buf - response from server
size - buffer size
type - content type
err_buf - buffer for error message
err_size - maximal size of error message
Return values:
content size - if >= 0
error code - if < 0
extract_pgm
( buf, size, width, height, bpp, err_buf, err_size )
Extract binary array form PGM image format.
Parameters:
buf - PGM image
size - buffer size
width - image width
height - image height
bpp - image digitization (bits per pixel)
err_buf - buffer for error message
err_size - maximal size of error message
Return values:
image size - if >= 0
error code - if < 0
Usage
Image capture example
#include "dtcam.h"
...
result = perform_http_request(
inet_addr("192.168.0.32"), 80, "/img.pgm",
buf, sizeof(buf),
err_buf, sizeof(err_buf)
);
if(result >= 0) {
result = process_http_response(
buf, result,
&type,
err_buf, sizeof(err_buf)
);
}
if(result >= 0) {
if(type == 3) {
result = extract_pgm(
buf, result,
&width, &height, &bpp,
err_buf, sizeof(err_buf)
);
else {
fprintf(stdout, "wrong content type\n");
exit(1);
}
if(result >= 0) {
/* process image here */
...
}
else {
fprintf(stdout, "%s\n", err_buf);
exit(1);
}
...
python-dtcam
This package is a Python binding to the C library (libdtcam). It provides cam class,
to control and capture images from DTI interferometers.
Class cam
cam
( self, ip, port )
Constructor.
Parameters:
ip - ip address, string
port - http server port, positive integer
Methods
get_frame
( self, uri )
This method gets image from server
Parameters:
uri - resource name, string (only pgm format is supported)
Return values:
frame - image, numpy uint16 array
Usage
Image capture example
Import dtcam
cam = dtcam.cam("192.168.0.32", 80)
img = cam.get_frame("/img.pgm")
print img
Compiling & Installing
C library (libdtcam) is required
Compile module
python setup.py build
Compile under Windows with MinGW
python setup.py build -c mingw32
Install module
python setup.py install
matlab-dtcam
The matlab-dtcam binding module provides direct access to DTI interferometers from Matlab.
This module is established through a MEX-file.
Since each Matlab MEX-file is allowed to have only one entry point,
dtcam implements control over single function using sub-functions.
Depends on sub-function it is required to use variable number of "left hand side" and "right hand side" arguments.
The first input argument determines sub-function.
Interface
[result1, result2, ... ] = dtcam
( subfunction,
[parameter1, parameter2, ...]
)
Sub-functions:
0 - set communication parameters
1 - get image
2 - get phase map
set communication parameters
dtcam( uint8(0), ip, port )
Right hand side:
ip - device ip adress, uint8 array of size 8
port - server port, uint16
Left hand side:
None
get image
dtcam( uint8(1), shift )
Right hand side:
shift - phase shift, int16, optional
Left hand side:
image - uint16 2D matrix
get phase map
dtcam( uint8(2), ref )
Right hand side:
ref - phase shift reference points, int16 array of size 5
Left hand side:
image - uint16 2D matrix
Usage
Set communication parameters, ip 192.168.0.32, port 80
dtcam(uint8(0), uint8([192,168,0,32]), uint16(80));
Get image
z = dtcam(uint8(1));
Get image with phase shift 50%
z = dtcam(uint8(1), int16(16384));
Get phase map and plot it
dtcam(uint8(0), uint8([192,168,0,32]), uint16(80)); % set device address
ref = int16([0., 0.212, 0.374, 0.538, 0.707] * 32768); % reference points
z = dtcam(uint8(2), ref); % get phase map
z = uint8(z / 256); % convert phase map to 8 bit data format
image(z) % plot image
colormap(jet(256)) % set palette
axis off % Remove axis ticks and numbers
axis image % Set aspect ratio to obtain square pixels
Compiling
C library (libdtcam) is required
Compiling under Windows
mex dtcam.c ..\libdtcam\libdtcam.a wsock32.lib -I..\libdtcam
Compiling under Linux
mex dtcam.c ../libdtcam/libdtcam.a -I../libdtcam
Compiling for octave under Linux
mkoctfile --mex dtcam.c ../libdtcam/libdtcam.a -I../libdtcam
Requirements
matlab-dtcam
Matlab (with mex compiler) or
Octave (free software, mostly compatible with Matlab).
python-dtcam
Python 2.6 programming language
Numpy 1.5.1 (fundamental package needed for scientific computing with Python)
Links
Matlab http://www.mathworks.com/products/matlab/
Octave http://www.gnu.org/software/octave/
Python http://www.python.org
Numpy http://numpy.scipy.org
