Hi, How do I resize a video to satisfy google net's size input layer ?

I am currently doing a project whereby I have to detect vehicles using googlenet but I need to resize the size of the video for that. Anyone is able to help?

[Adam]

doc imresize

will resize an image. Do you need to resize the actual video? It is images taken from a video that you feed into googlenet isn't it, not a video itself?

[Syafiqah Daud]

I need to resize the video. I would want googlenet to detect vehicles in a real time video thus i would need to resize the video. I have tried imresize but it still doesn't work. Any other suggestions? Thanks for your reply though!

[Guillaume]

I have tried imresize but it still doesn't work

imresize works. So if it still doesn't work it's because you've done something wrong. However, since you haven't shown use what you've done we can't tell you what you've done wrong.

As far as I know the googlenet interface does not take videos anyway, so you'd have to pass it each frame of the video.

[Syafiqah Daud]

Are you willing to clarify about "pass it each frame of the video." As I am new to MatLab. Here is my code. Thank you!

      clear

    % camera = webcam; % Connect to the camera
    net = googlenet;   % Load the neural net
    net.Layers

    while true   
        labelType ='' 
        trafficVid = VideoReader('C:\Users\syafiqah\Desktop\project\traffic.mp4')
        get(trafficVid)
        implay('C:\Users\syafiqah\Desktop\project\traffic.mp4');  
        sz = net.Layers(1).Inputsize;
        trafficVid = imresize(trafficVid,[sz(1),sz(2)]);
        label = classify(net, video);        % Classify the picture     
         if (label == 'ambulance')
             labelType = 'Vehicle Recognised'
         elseif (label == 'moped')
                 labelType = 'Vehicle Recognised'
         elseif (label == 'tank')
                 labelType = 'Vehicle Recognised'
         elseif (label == 'sports car')
                 labelType = 'Vehicle Recognised'
         elseif (label == 'moving van')
                 labelType = 'Vehicle Recognised'
         elseif (label == 'trailer truck')
                 labelType = 'Vehicle Recognised'          
         elseif (label == 'garbage truck')
               labelType = 'Vehicle Recognised'
         elseif (label == 'trailer truck')
               labelType = 'Vehicle Recognised'
         else 
         end          
         title(char(labelType)); % Show the label 

        darkCarValue = 50;
        darkCar = rgb2gray(read(trafficVid,71));
        noDarkCar = imextendedmax(darkCar, darkCarValue);
        imshow(darkCar)
        figure, imshow(noDarkCar)
        sedisk = strel('disk',2);
        noSmallStructures = imopen(noDarkCar, sedisk);
        imshow(noSmallStructures)
        nframes = trafficVid.NumberOfFrames;
        I = read(trafficVid, 1);
        taggedCars = zeros([size(I,1) size(I,2) 3 nframes], class(I));

    for k = 1 : nframes
        singleFrame = READFRAME(trafficVid, k);

        % Convert to grayscale to do morphological processing.
        I = rgb2gray(singleFrame);

        % Remove dark cars.
        noDarkCars = imextendedmax(I, darkCarValue); 

        % Remove lane markings and other non-disk shaped structures.
        noSmallStructures = imopen(noDarkCars, sedisk);

        % Remove small structures.
        noSmallStructures = bwareaopen(noSmallStructures, 150);

        % Get the area and centroid of each remaining object in the frame. The
        % object with the largest area is the light-colored car.  Create a copy
        % of the original frame and tag the car by changing the centroid pixel
        % value to red.
        taggedCars(:,:,:,k) = singleFrame;

        stats = regionprops(noSmallStructures, {'Centroid','Area'});
        if ~isempty([stats.Area])
            areaArray = [stats.Area];
            [junk,idx] = max(areaArray);
            c = stats(idx).Centroid;
            c = floor(fliplr(c));
            width = 2;
            row = c(1)-width:c(1)+width;
            col = c(2)-width:c(2)+width;
            taggedCars(row,col,1,k) = 255;
            taggedCars(row,col,2,k) = 0;
            taggedCars(row,col,3,k) = 0;
        end
    end
        frameRate = trafficVid.FrameRate;
        implay(taggedCars,frameRate);

        drawnow;   
     end

Answer 1

[Diwakar Ravichandran]

Hi Syafiqah,

This is a question that was already answered on MATLAB answers. This should give you some headway for your problem. This is the link to the answer.

https://www.mathworks.com/matlabcentral/answers/322563-how-to-resize-video

Hope this helps

Cheers!

Answer 2

[Guillaume]

Are you willing to clarify about "pass it each frame of the video."

Your question is puzzling since at first glance you're doing the right thing in the 2nd half of your code. Yet, the first half of your code is nonsense.

The classification should be done similarly to the 2nd half. Something similar to:

trafficVid = VideoReader('C:\Users\syafiqah\Desktop\project\traffic.mp4');
for k = 1:trafficVid.NumberOfFrames     %or use while trafficVid.HasFrame
   label = classify(net, imresize(trafficis.readFrame, net.Layers(1).Inputsize));  %get image, resize and pass to classifier
   %...
end