Image Processing - How to making the Background black leaving barcode as white

I'm new to Matlab. I wondering how to make the background black and leaving the barcode white. I read on some noise elimaination technique that this step can detect the barcode. I have tried and error but cannot make it fully. Anybody has any ideal?

Noise Elimination Techniques

1. Clear any group/s of pixels that are touching the edge/border of the image

2. Clear any group/s of pixels less than 100 pixels^2 in area for image sizes > [300 300] or 50 pixels^2 for those smaller.

3. Clear any group/s of pixels that are not straight (since eccentricity of a straight line is 1).

code:

rgb  = imread('bar1.jpg');
rgb = imresize(rgb,0.33);
Igray= rgb2gray(rgb);
Ibw = im2bw(Igray, graythresh(Igray));
Ibw = ~Ibw
Iarea = bwareaopen(Ibw,10);
imshow(Iarea))

Original Image <http://www.mediafire.com/?o1hwqz954l85usi>

Answer 1

[Image Analyst]

1. imclearborder()

2. bwareaopen()

3. use regionprops to get coordinates of each bar, then use polyfit to see how well it fits to a line.

[Kim]

How to write regionprops code? Really hard to understand

[Image Analyst]

See my BlobsDemo for a full demo.

http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862

[Kim]

I have try out regionprops but the small group of pixel is there.

And I found out this could solve the problem.

This option uses the property area, so it will output the area of each of the groups of pixels left in the image both barcode and noise. Generally by this stage there are just small groups of noise left that weren't eliminated by the 100 threshold or maybe a really large group. Hence it is possibly to calculate the median area and this would be of one of the bars (assuming mainly bars left in the image). From this you can set boundaries for the area so lower boundary median/4 and upper boundary median * 4. We have chosen 4 since the greatest ratio of narrow to wide is 1:3. Hence 4 times the median and 1/4 of the median should definitely include the bars of the barcode (we have tested this and it is a fact). Advantage of this option is that it does not matter if noise has made it through to this stage or not and it is easy to calculate ( a lot more efficient than calculating the distance!).

[Image Analyst]

Sounds like you didn't use bwareaopen like I suggested to get rid of small noise blobs. You could also check the circularity = perimeter^2/(4*pi*area) to discriminate round vs. elongated blobs, or compare the ratio of MajorAxisLength/MinorAxisLength or Eccentricity.

[Kim]

I did use bwareaopen() however there are still some small pixel there and I cannot increase anymore too. If not the whole image will turn BLACK!!

Therefore I will require it to determine the barcode before I can do any further remove of the small pixels.

Thanks for the details, working on that currently! :)

[Kim]

Can I check with you how to compare the ratio of MajorAxisLength/MinorAxisLength or Eccentricity

[Kim]

Thank You!! :)

Answer 2

[Image Analyst]

Kim, your whole idea was flawed from the start. Now that I can see your image I can tell that you first want to do color segmentation on it before you do thresholding. See this demo. Once you have the bar code alone, then you can use Hough or some other ad hoc method to find the lines and their thicknesses.

workspace;  % Make sure the workspace panel is showing.
fontSize = 18;

% Read in a standard MATLAB gray scale demo image.
folder = 'D:\Temporary stuff';
baseFileName = 'bar1.jpg';
% Get the full filename, with path prepended.
fullFileName = fullfile(folder, baseFileName);
if ~exist(fullFileName, 'file')
  % Didn't find it there.  Check the search path for it.
  fullFileName = baseFileName; % No path this time.
  if ~exist(fullFileName, 'file')
    % Still didn't find it.  Alert user.
    errorMessage = sprintf('Error: %s does not exist.', fullFileName);
    uiwait(warndlg(errorMessage));
    return;
  end
end
rgbImage = imread(fullFileName);
% Get the dimensions of the image.  numberOfColorBands should be = 3.
[rows columns numberOfColorBands] = size(rgbImage);
% Display the original color image.
subplot(2, 2, 1);
imshow(rgbImage, []);
title('Original Color Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]); % Maximize figure.
set(gcf,'name','Demo by ImageAnalyst','numbertitle','off') 

% Extract the individual red, green, and blue color channels.
redChannel = rgbImage(:, :, 1);
greenChannel = rgbImage(:, :, 2);
blueChannel = rgbImage(:, :, 3);

% Find white objects.
thresholdValue = 128;
binaryImage = redChannel > 128 & greenChannel > thresholdValue & blueChannel > thresholdValue;
subplot(2, 2, 2);
imshow(binaryImage);
title('Initial Binary Image', 'FontSize', fontSize);

% Fill in bar code background.
binaryImage = imfill(binaryImage, 'holes');
% Remove regions touching border.
binaryImage = imclearborder(binaryImage);
% Get rid of regions less than 200000 pixels in area.
binaryImage = bwareaopen(binaryImage, 200000);
% Display the cleaned up image.
subplot(2, 2, 3);
imshow(binaryImage);
title('Final Binary Image', 'FontSize', fontSize);

% Find left and right columns.
% Find top and bottom rows.
verticalProfile = any(binaryImage, 2);
topRow = find(verticalProfile, 1, 'first');
bottomRow = find(verticalProfile, 1, 'last');
% Find left and right columns.
horizontalProfile = any(binaryImage, 1);
leftColumn = find(horizontalProfile, 1, 'first');
rightColumn = find(horizontalProfile, 1, 'last');
% Mask out the bars
barOnly = greenChannel .* uint8(binaryImage);
barOnly(~binaryImage) = 255;
% Crop out the image
barOnly = barOnly(topRow:bottomRow, leftColumn:rightColumn);
subplot(2, 2, 4);
imshow(barOnly);
title('Cropped, Masked Bar-Only Image', 'FontSize', fontSize);

msgbox('Now use hough, houghlines, or other methods to find the lines and their thicknesses.');

[Kim]

Actually my intention is to make the whole image black except the barcode white!

But whenever i use bwareopen() there are still unwanted pixel away from the barcode cause it is the same size as the barcode.

Therefore I trying to detect the area of the barcode first so that I can use imfill() to fill up the unwanted pixel outside the barcode.

my code

% Read Image

rgb = imread('bar1.jpg');

% Resize Image

rgb = imresize(rgb,0.33);

% Convert from RGB to Gray

Igray = rgb2gray(rgb);

% Threshold to minimize the intraclass variance

Ibw = im2bw(Igray, graythresh(Igray));

% Inverse Colour

Ibw = ~Ibw;

% Clear Border

Iarea = imclearborder(Ibw);

% Remove Small object

Iarea = bwareaopen(Iarea,50);

imshow(Iarea);

That's is what I trying to do! Hope this may give you a background info what my intention! :)

[Image Analyst]

This code is not very robust. You'll see that you have a variety of binary images in my code and you can get even more very easily. In my code you don't have a bunch of other blobs. It gets the bar code alone. Did you even run it?

[Kim]

Your coding really works with products that has a white border around the barcode!!

However I trying to put it to other product barcode which does not has a white border around the product, it won't works!!

So I was thinking of my previous method might work??

Here are some of the image I'm trying to do!!

<http://www.mediafire.com/?ccbrti18s4is95c>

Thank you for your respond! :)

[Kim]

http://www.mediafire.com/?ccbrti18s4is95c

[Kim]

Thank You!! :)

[vidyasagar S D]

can u please share the image used for it . the link is broken and not working

[Image Analyst]

After 8 years, no one has it anymore. If you have a problem with your own image, please post it and your code in a new question.