How can I take a snapshot from USB camera and then analyze it?
Mostra commenti meno recenti
On my previous work, I was trying to take an image from a file and find the centroids of the images. However, I am having trouble taking the snapshot and then having MATLAB save the image so then I can analyze it for the centroids of the dots. Is there any way this can be done?
% clear, clc
% establish video input
cam=videoinput('gentl',1,'Mono8');
%take snapshot
preview(cam)
triggerconfig(cam, 'manual');
snapshot=getsnapshot(cam);
%Display image
im=imagesc(snapshot)
im=max(im,[],3);
% Segment blobs
bw=imclose(im<20,strel('disk',5)); binaryImage = imfill(im, 'holes');
% Cenroids of individual blobs
RP=regionprops(bw,'Centroid'); N=numel(RP); C=zeros(N,2);
for i=1:N,
C(i,:)=RP(i).Centroid;
end
% Centroid of all blobs
C_net=regionprops(double(bw),'Centroid');
C_net=C_net.Centroid;
figure imshow(bw) hold on plot(C(:,1),C(:,2),'or','MarkerSize',10,'LineWidth',2) plot(C_net(:,1),C_net(:,2),'*g','MarkerSize',10,'LineWidth',2)
%Calculating the distance formula
D=sqrt(((C(4,1)) - (C(2,1))).^2 + ((C(4,2)) - (C(2,2))).^2);
formatSpec = 'Distance of top row is %4.2f pixels\n'; fprintf(formatSpec,D)
D2=sqrt(((C(3,1)) - (C(1,1))).^2 + ((C(3,2)) - (C(1,2))).^2);
formatSpec = 'Distance of bottom row is %4.2f pixels\n'; fprintf(formatSpec,D2)
closePreview(cam);
Specifically the im=max function appears to not be working. My goal is to get MATLAB to be able to take the snapshot image and determine the centroids between them.
1 Commento
Walter Roberson
il 27 Giu 2018
Make sure you install the GENTL support package; https://www.mathworks.com/hardware-support/gentl.html
Risposte (2)
Florian Morsch
il 27 Giu 2018
Modificato: Florian Morsch
il 27 Giu 2018
First install the support package "MATLAB Support Package for USB Webcams" ( https://de.mathworks.com/matlabcentral/fileexchange/45182-matlab-support-package-for-usb-webcams )
Then you can do something like this:
cam = webcam(); % This chooses the first connected webcam. If you have multiple use webcam(2), webcam(3) etc.
Snapshot = snapshot(cam);
image(Snapshot);
imwrite(Snapshot, 'SomeName.png'); % To save it to another location you can do something like
'C:/User/MATLABPictures/SomeName.png')
This will take a snapshot and save it as .png file to your current folder. Or you can work with the snapshot, e.g.:
cam = webcam();
cam.Resolution = '800x600';
Snapshot = snapshot(cam);
SnapshotGray = rgb2gray(Snapshot);
2 Commenti
PamunkeBoy
il 27 Giu 2018
Walter Roberson
il 27 Giu 2018
The GENTL package is perhaps more appropriate than the USB Webcams package for the Allied Vision devices.
PamunkeBoy
il 27 Giu 2018
4 Commenti
Walter Roberson
il 27 Giu 2018
for image_number = 1 : 100 %to process 100 images
snapshot = getsnapshot(cam);
filename = sprintf('framein_%04d.bmp', image_number);
saveas(gcf, filename)
im = imread(filename)
...
end
However, it you would then be analyzing the figure frame and axes frame and so on. You already have just the image itself in snapshot . It would seem to make more sense to use
for image_number = 1 : 100 %to process 100 images
snapshot = getsnapshot(cam);
filename = sprintf('framein_%04d.bmp', image_number);
imwrite(snapshot, filename);
im = snapshot;
...
end
PamunkeBoy
il 27 Giu 2018
Walter Roberson
il 27 Giu 2018
image_number = 0;
while true
key = input('Return to continue, Q to quit: ', 's');
if strcmpi(key, 'q'); break; end
snapshot = getsnapshot(cam);
image_number = image_number + 1;
filename = sprintf('framein_%04d.bmp', image_number);
imwrite(snapshot, filename);
im = snapshot;
...
end
PamunkeBoy
il 27 Giu 2018
Categorie
Scopri di più su ROI-Based Processing in Centro assistenza e File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!