Detect Cars in Video using regionprops

Hi all,

I'm trying to write a code that detect cars from a video. Right now I'm trying to get the regionprops to work, it doesn't look like it's working at all since it's just giving random boxes and not connecting to the resulted video. Any help is greatly appreciated! Thank you.

clear all
close all
clc


%% Access Video
trafficVid= VideoReader('traffic.mj2');

gs = read(trafficVid,71);

implay('traffic.mj2'); %Explore video

%% Develop Algorithm
darkCarValue = 50;
darkCar = rgb2gray(gs);
noDarkCar = imextendedmax(darkCar, darkCarValue);

sedisk = strel('disk',2);
noiseFreeObject = imopen(noDarkCar, sedisk);
imshow(noiseFreeObject);
title('Object After Removing Noise');

%% Apply Algorithm
nframes = trafficVid.NumberOfFrames;
I = read(trafficVid, 1);
taggedCars = zeros([size(I,1) size(I,2) 3 nframes], class(I));

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

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

    noDarkCars = imextendedmax(I, darkCarValue);

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

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

    taggedCars(:,:,:,k) = singleFrame;

    stats = regionprops(noiseFreeObject, 'BoundingBox','Centroid');

   for object = 1:length(stats)
       bb = stats(object).BoundingBox;
       bc = stats(object).Centroid;
       rectangle('Position',bb,'EdgeColor','g','LineWidth',2);
   end
end

%% Results
frameRate = trafficVid.FrameRate;
implay(taggedCars,frameRate);

[KSSV]

You need to attach video so that we can check your code.

[Nguyen Nguyen]

Hi, thanks! The video is on Matlab library.

[prasanth s]

add following line at the end of first 'for' loop, so it displays the image and draw rectangle

figure(1),imshow(noDarkCars,[])

[Nguyen Nguyen]

@prasanth s this works!

I have an extra question if you have time. I want the code to detect all cars, not just black cars. But when I took out the lines that I think are used to exclude the black cars it doesn't work anymore. Do you know why?

clear all
close all
clc


%% Access Video
trafficVid= VideoReader('traffic.mj2');

gs = read(trafficVid,71);

implay('traffic.mj2'); %Explore video

%% Develop Algorithm
%darkCarValue = 50;
darkCar = rgb2gray(gs);
%noDarkCar = imextendedmax(darkCar, darkCarValue);

sedisk = strel('disk',2);
noiseFreeObject = imopen(darkCar, sedisk);
imshow(noiseFreeObject);
title('Object After Removing Noise');

%% Apply Algorithm
nframes = trafficVid.NumberOfFrames;
I = read(trafficVid, 1);
taggedCars = zeros([size(I,1) size(I,2) 3 nframes], class(I));

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

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

    % noDarkCars = imextendedmax(I, darkCarValue);

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

    % Remove small structures by removing all pixels less than 150 px.
    noiseFreeObject = bwareaopen(noiseFreeObject, 150);
    bw = bwlabel(noiseFreeObject);

    stats = regionprops(noiseFreeObject, 'BoundingBox','Centroid');

   for object = 1:length(stats)
       bb = stats(object).BoundingBox;
       bc = stats(object).Centroid;
       rectangle('Position',bb,'EdgeColor','g','LineWidth',2);
   end
   figure(1),imshow(noiseFreeObject,[])
end

[prasanth s]

following code includes black car detection using a theshold value 200

clear all
close all
clc


%% Access Video
trafficVid= VideoReader('traffic.mj2');

gs = read(trafficVid,71);

%% Develop Algorithm
darkCarValue = 50;
darkCar = rgb2gray(gs);
noDarkCar = imextendedmax(darkCar, darkCarValue);

sedisk = strel('disk',2);
noiseFreeObject = imopen(noDarkCar, sedisk);
imshow(noiseFreeObject);
title('Object After Removing Noise');

%% Apply Algorithm
nframes = trafficVid.NumberOfFrames;
I = read(trafficVid, 1);
taggedCars = zeros([size(I,1) size(I,2) 3 nframes], class(I));

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

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

    noDarkCars = imextendedmax(I, darkCarValue);

    % Remove lane markings and other non-disk shaped structures.
    noiseFreeObject = imopen(noDarkCars, sedisk);
    % Remove small structures.
    noiseFreeObject = bwareaopen(noiseFreeObject, 150);
    
    DarkCars = ~imextendedmax(I, 200);
    DarkCars = imopen(DarkCars, sedisk);
    DarkCars = imdilate(DarkCars, strel('disk',2));
    DarkCars = bwareaopen(DarkCars, 150);
    
    noiseFreeObject=noiseFreeObject|DarkCars;
    taggedCars(:,:,:,k) = singleFrame;

    stats = regionprops(noiseFreeObject, 'BoundingBox','Centroid');

   for object = 1:length(stats)
       bb = stats(object).BoundingBox;
       bc = stats(object).Centroid;
       rectangle('Position',bb,'EdgeColor','g','LineWidth',2);
   end
   
        figure(1),imshow(noiseFreeObject,[])
end

%% Results
% frameRate = trafficVid.FrameRate;
% implay(taggedCars,frameRate);

[Nguyen Nguyen]

This doesn't work :(