How to count black pixels in a region of an image that can only have 1 white neighbor pixel
조회 수: 2 (최근 30일)
이전 댓글 표시
I have a binary image (black background and white pixels). How do i count the number of blackpixels witch in it's neighbourhood of 5*7 (5 lines and 7 colums) can only have a white pixel.
I can try using bwhitmiss() but i would need 35 matrixs with one white pixel in each "matrix slot".
Can you tell me an easyest way to achieve this?
댓글 수: 0
채택된 답변
Ashish Uthama
2012년 4월 6일
You could try leveraging conv2.
Here is the idea:
in = [
0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 ]
k = [
1 1 1
1 1 1
1 1 1]
r = conv2(in,k,'same')
The result r will have 0 at all locations which DO NOT have 1's in the immediate 3x3 neighborhood.
댓글 수: 6
Ashish Uthama
2012년 4월 9일
João, conv2 pads with zeros around the edges, and it ought to work. convole these two images with [1 1 ; 1 1] to see whats going on and to understand the 'neighborhood' being used.
추가 답변 (1개)
Image Analyst
2012년 4월 6일
Here is the solution in case you want to compare yours to mine:
clc; % Clear the command window.
close all; % Close all figures (except those of imtool.)
workspace; % Make sure the workspace panel is showing.
fontSize = 20;
% Create an image of all zeros.
imageArray = false(50, 70);
% Get up to a tenth of the pixels as white dots.
numberOfSinglePixels = 100 %int32(rand * numel(imageArray) / 10)
scrambledCoords = randperm(numel(imageArray));
coordinatesToSet = scrambledCoords(1:numberOfSinglePixels)';
imageArray(coordinatesToSet) = true;
% Display the image
imshow(imageArray, 'InitialMagnification', 300);
title('Original Image', 'FontSize', fontSize);
% Enlarge figure to full screen.
set(gcf, 'units','normalized','outerposition',[0 0 1 1]);
% Count up the dots in a sliding 5x7 window.
windowRows = 5;
windowCols = 7;
sumImage = conv2(double(imageArray), ones(windowRows, windowCols), 'same');
% Extract only those pixels where the window is centered
% at dots in the input image.
sumImageMasked = sumImage .* double(imageArray);
% Count those places where the count is exactly 1.
numberOfDots = sum(sumImageMasked(:) == 1)
% Get the locations of those dots in a 5x7 black neighborhood.
[dotRows dotCols] = find(sumImageMasked == 1);
% Plot crosses on those dots.
hold on;
plot(dotCols, dotRows, 'r+', 'MarkerSize', 25);
caption = sprintf('Red crosses over the dots that are in an isolated %x by %d window',...
windowRows, windowCols);
title(caption, 'FontSize', fontSize);
message = sprintf('%d of the %d pixels are in a black neighborhood of %d by %d\n',...
numberOfDots, numberOfSinglePixels, windowRows, windowCols);
fprintf('%s\n', message);
msgbox(message);
댓글 수: 0
참고 항목
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!