How do I find the bounding box of a fingerprint image

The patterns on the image are black while the background is white. I need to make the background which is white as minimal as possible. From research I think using regionprops will work but I don't understand regionprops. Thank you.

Answer 1

[Image Analyst]

Alternatively you could just sum the binary image horizontally and vertically, or use any(), and look for where it's first land last non-zero. Here's the demo for that method:

% Demo to find the bounding box of coins with any() and crop out the subimage.
clc;    % Clear the command window.
workspace;  % Make sure the workspace panel is showing.
format longg;
format compact;
fontSize = 20;

% Check that user has the Image Processing Toolbox installed.
hasIPT = license('test', 'image_toolbox');
if ~hasIPT
  % User does not have the toolbox installed.
  message = sprintf('Sorry, but you do not seem to have the Image Processing Toolbox.\nDo you want to try to continue anyway?');
  reply = questdlg(message, 'Toolbox missing', 'Yes', 'No', 'Yes');
  if strcmpi(reply, 'No')
    % User said No, so exit.
    return;
  end
end

% Read in a standard MATLAB gray scale demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
% Get the full filename, with path prepended.
baseFileName = 'eight.tif';
fullFileName = fullfile(folder, baseFileName);
% Check if file exists.
if ~exist(fullFileName, 'file')
  % File doesn't exist -- 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 in the search path folders.', fullFileName);
    uiwait(warndlg(errorMessage));
    return;
  end
end
grayImage = imread(fullFileName);
% Get the dimensions of the image.  
% numberOfColorBands should be = 1.
[rows, columns, numberOfColorBands] = size(grayImage);
% Display the original gray scale image.
subplot(2, 3, 1);
imshow(grayImage, []);
title('Original Grayscale Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% Give a name to the title bar.
set(gcf,'name','Demo by ImageAnalyst','numbertitle','off') 

% Let's compute and display the histogram.
[pixelCount, grayLevels] = imhist(grayImage);
subplot(2, 3, 2); 
bar(pixelCount);
grid on;
title('Histogram of original image', 'FontSize', fontSize);
xlim([0 grayLevels(end)]); % Scale x axis manually.

% Threshold the image to get a binary image.
binaryImage = grayImage < 175;
% Display the image.
subplot(2, 3, 3);
imshow(binaryImage, []);
title('Binary Image', 'FontSize', fontSize);

% Get vertical profile.
verticalProfile = any(binaryImage, 2);
subplot(2, 3, 4);
plot(verticalProfile, 'b-', 'LineWidth', 4);
grid on;
title('Vertical Profile', 'FontSize', fontSize);
xlabel('Row', 'FontSize', fontSize);

% Get Horizontal profile.
horizontalProfile = any(binaryImage, 1);
subplot(2, 3, 5);
plot(horizontalProfile, 'b-', 'LineWidth', 4);
grid on;
title('Horizontal Profile', 'FontSize', fontSize);
xlabel('Column', 'FontSize', fontSize);

% Find what index the bounding box is at
topRow = find(verticalProfile, 1, 'first')
bottomRow = find(verticalProfile, 1, 'last')
leftCol = find(horizontalProfile, 1, 'first')
rightCol = find(horizontalProfile, 1, 'last')

% Plot box around original
xBox = [leftCol, rightCol, rightCol, leftCol, leftCol]
yBox = [topRow, topRow, bottomRow, bottomRow, topRow]
subplot(2, 3, 1);
hold on;
plot(xBox, yBox, 'r', 'LineWidth', 2);

% Crop it out of the original:
croppedImage = grayImage(topRow:bottomRow, leftCol:rightCol);
% Display the image.
subplot(2, 3, 6);
imshow(croppedImage, []);
title('Cropped Image', 'FontSize', fontSize);

[Tola]

The code is running but it is not cropping the image. Thank you.

[Image Analyst]

I just copied and pasted and verified that it DOES indeed crop. You'll notice the bottom right image has edges just touching the edges of the coins while the original one has a lot of gray background surrounding the coins.

Answer 2

[Image Analyst]

Here's the way to do it with bwconvhull() and regionprops():

% Demo to find the bounding box of coins with bwconvhull()
% and regionprops(), and crop out the subimage.
clc;    % Clear the command window.
workspace;  % Make sure the workspace panel is showing.
clearvars;
format longg;
format compact;
fontSize = 20;

% Check that user has the Image Processing Toolbox installed.
hasIPT = license('test', 'image_toolbox');
if ~hasIPT
  % User does not have the toolbox installed.
  message = sprintf('Sorry, but you do not seem to have the Image Processing Toolbox.\nDo you want to try to continue anyway?');
  reply = questdlg(message, 'Toolbox missing', 'Yes', 'No', 'Yes');
  if strcmpi(reply, 'No')
    % User said No, so exit.
    return;
  end
end

% Read in a standard MATLAB gray scale demo image.
folder = fullfile(matlabroot, '\toolbox\images\imdemos');
% Get the full filename, with path prepended.
baseFileName = 'eight.tif';
fullFileName = fullfile(folder, baseFileName);
% Check if file exists.
if ~exist(fullFileName, 'file')
  % File doesn't exist -- 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 in the search path folders.', fullFileName);
    uiwait(warndlg(errorMessage));
    return;
  end
end
grayImage = imread(fullFileName);
% Get the dimensions of the image.  
% numberOfColorBands should be = 1.
[rows, columns, numberOfColorBands] = size(grayImage);
% Display the original gray scale image.
subplot(2, 3, 1);
imshow(grayImage, []);
title('Original Grayscale Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% Give a name to the title bar.
set(gcf,'name','Demo by ImageAnalyst','numbertitle','off') 

% Let's compute and display the histogram.
[pixelCount, grayLevels] = imhist(grayImage);
subplot(2, 3, 2); 
bar(pixelCount);
grid on;
title('Histogram of original image', 'FontSize', fontSize);
xlim([0 grayLevels(end)]); % Scale x axis manually.

% Threshold the image to get a binary image.
binaryImage = grayImage < 125;
% Display the image.
subplot(2, 3, 3);
imshow(binaryImage, []);
title('Binary Image', 'FontSize', fontSize);

% Get the convex hull
binaryImage = bwconvhull(binaryImage, 'union');
subplot(2, 3, 4);
imshow(binaryImage, []);
title('Convex Hull Image', 'FontSize', fontSize);

% Call regionprops, asking for BoundingBox.
measurements = regionprops(binaryImage, 'BoundingBox');

% Find what index the bounding box is at
leftCol = measurements(1).BoundingBox(1)+0.5
topRow = measurements(1).BoundingBox(2)+0.5;
width = measurements(1).BoundingBox(3)
height = measurements(1).BoundingBox(4)
rightCol = leftCol + width;
bottomRow = topRow + height;

% Plot box around original
xBox = [leftCol, rightCol, rightCol, leftCol, leftCol]
yBox = [topRow, topRow, bottomRow, bottomRow, topRow]
subplot(2, 3, 1);
hold on;
plot(xBox, yBox, 'r', 'LineWidth', 2);

% Crop it out of the original:
croppedImage = grayImage(topRow:bottomRow, leftCol:rightCol);
% Display the image.
subplot(2, 3, 6);
imshow(croppedImage, []);
title('Cropped Image', 'FontSize', fontSize);

[Tola]

This code is giving errors as shown below: Error in regionprops>ComputeConvexHull (line 916) hullIdx = convhull(rr, cc);

Error in regionprops>ComputeConvexImage (line 412)
    [stats, statsAlreadyComputed] = ...

Error in regionprops (line 221)
            [stats, statsAlreadyComputed] = ...

Error in bwconvhull (line 70)
blob_props = regionprops(labeled_image,'BoundingBox','ConvexImage');

Error in convhull (line 8)
binaryImage = bwconvhull(BWord, 'union');

[Image Analyst]

That is not my code you're showing. I didn't use any variables with those names (BWord, blob_props, labeled_image, etc.). You ran someone else's code instead of mine. If you look at my code you'll see a line like this:

binaryImage = bwconvhull(binaryImage, 'union');

which is NOT what you had. So either you're running some completely different code, or you made a mistake in adapting my code. Also, bwconvhull() is fairly new - within the last 2 or 3 years or so, so make sure you have the latest version of MATLAB.

Also, you didn't call your script convhull.m did you? Because that is not a good idea since it would blow away the built-in convhull() function.

[Tola]

its the code i ran and those are the errors i encountered and I dont know how to call the script convhull.m .

[Image Analyst]

Sorry, but my answer stands. I just copied and pasted my code from above and ran it and it ran fine. I hope someone else can do that to verify what I'm saying. You didn't say what version of MATLAB you have. Please say

which -all bwconvhull

and tell me what it says.