function liveVideoTemplate()
% liveVideoTemplate - Template for running a video stream into Matlab.
%
% This function can be altered for general-purpose use of streaming video
% into Matlab from an attached camera.
%
% Helpful functions - You may find the following functions helpful.
%
% % To get information about hardware devices, use:
% h = imaqhwinfo
%
% % Look at the installed adaptors from h (using the above command). You
% % can find hardware components (cameras) that Matlab knows about
% % using:
% h = imaqhwinfo('ADAPTOR')
% eg: h = imaqhwinfo('winvideo')
%
% % From here, look at h.DeviceIDs; this field will list all devices that
% % can be used with the associated adaptor (a dll file). To see
% % information about a particular device, look at:
% h.DeviceInfo(DeviceID)
% eg: h.DeviceInfo(1)
%
% % Of important note is the field SupportedFormats, ie:
% h.DeviceInfo(1).SupportedFormats
%
% % This field lists all the potential formats that the connected camera
% % can return. These formats typically look something like
% % 'FORMAT_ImgWdthxImgHght'. Eg 'MJPG_1280x720' will return RGB images
% % with a resolution of 1280x720. Common formats include H264, MJPG,
% % and YUY2.
% %
% % NOTE: as far as I know streaming H.264 format is NOT supported
% % (even if a camera lists the format and Matlab allows you to set up
% % the object, the returned images will not be converted to an rgb
% % format Matlab recognizes and your image will look like noise).
%
% % To connect a camera device to Matlab you create a videoinput object:
% vid = videoinput('ADAPTOR', 'DeviceID', 'FORMAT');
% eg: vid = videoinput('winvideo', 1, 'YUY2_640x480');
%
% % You can see the available properties of our video object with:
% get(vid)
%
% % Of note are the camera properties which can be found with:
% vid.Source
%
% % You can see if the camera object is working with:
% preview(vid)
%
% % To get a single frame from the camera, use the getsnapshot command:
% img = getsnapshot(vid);
% imshow(img)
%
% % If your camera returns YUY2 you can set the video object to return
% % rgb images with the following command:
% set(vid, 'ReturnedColorSpace', 'rgb');
%
% % To stream video you will want to put the video object into manual
% % mode and then 'start' it. Connecting to the camera takes some
% % overhead so we just want to connect once; the camera will start
% % streaming video (without Matlab saving it) and then when we request
% % a frame Matlab can just get the most recent image from the stream:
% triggerconfig(vid,'manual');
% start(vid)
%
% % Now you can get and display images as desired:
% tic
% while toc<10
% img = getsnapshot(vid);
% imagesc(img); axis off
% end
%
% % After you are done you will want to 'stop' the video object. This
% % will allow other programs to use the camera.
% stop(vid)
%
% % If you lose scope of your video object (i.e. a function you have
% % using the video object crashes) you don't need to close Matlab in
% % order to get control of your video object again. You can use
% % imaqfind to find it:
% vid = imaqfind;
%
% % Finally, to clear out a video object, use the following command:
% delete(vid)
% Delete any previous [video] objects (necessary if this program previously
% crashed):
delete(imaqfind);
% Set up video object. Note: to change to a different camera (or camera
% setup) change the following line:
vid = videoinput('winvideo',1,'YUY2_640x480');
% Set the video object to always return rgb images:
set(vid, 'ReturnedColorSpace', 'rgb');
% Put the video trigger into 'manual', this starts streaming the video
% without saving it. We can then request frames at will while only having
% to run the startup overhead this one time.
triggerconfig(vid,'manual');
start(vid)
% Initialize frame counter and fps variable
counter = 0;
fps = 0;
% Set the total runtime in seconds
runtime = 15;
% Initialize figure to display video
h = figure(1);
% Start the timer and start keeping track of the time at the beginning of
% every 10 frames
tic
timeTracker = toc;
% We run a while loop to get and display a frame from the camera. The while
% loop runs for <runtime> seconds.
while toc < runtime
% Compute the frame rate averaged over the last 10 frames
if counter==10
counter = 0;
fps = 10/(toc-timeTracker);
timeTracker = toc;
end
counter = counter + 1;
% Get a new frame from the camera
img = getsnapshot(vid);
% PUT IMAGE PROCESSING HERE
% You may put any type of image processing you want to do here.
% Display image
% Note: use imagesc() instead of imshow() (it's faster).
imagesc(img); axis off
title(['FPS: ' sprintf('%2.1f', fps)])
end
% Stop the video stream
stop(vid)
C76