how to get horizontal projection of histogram of an binary image ?and based on that how to segment of each line in the image? please suggest me

Question

ayushi 2016-5-18

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/284847-how-to-get-horizontal-projection-of-histogram-of-an-binary-image-and-based-on-that-how-to-segment-o

评论： ayushi 2016-5-21

采纳的回答： Alessandro Masullo

<<

>>

<<

>> this is my ZjSqKcW.jpg and on the bases of this image i want the horizontal projection of histogram like g.png and after the histogram i want the segmentation of each line.how to do please suggest

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Alessandro Masullo 2016-5-20

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/284847-how-to-get-horizontal-projection-of-histogram-of-an-binary-image-and-based-on-that-how-to-segment-o#answer_222846

在 MATLAB Online 中打开

qeuCMg[1].jpg

Given your image im, you can binarize it with im2bw:

bw = 1-im2bw(image);

The horizontal histogram (I guess) is the sum of the object in the horizontal direction, which is

h = sum(bw,2)

You can then plot it with:

   figure
   plot(sum(bw,2),1:size(bw,1))

Which gives you the image attached.

The segmentation could be performed with findpeaks.

5 个评论
显示 3更早的评论隐藏 3更早的评论

Image Analyst 2016-5-21

Threshold it and then use regionprops(labeledImage, 'PixelIdxList') to find where each line starts and stops (what row indexes), or what rows have text in them.

ayushi 2016-5-21

在 MATLAB Online 中打开

sir will you please explain it with the help of example i tried the following code but its not working properly and also i am not able to detect where is the error in this:

   %DESKEWING
   % Find the coordinates of all the pixels
  [y, x] = find(Iedge2);
  % Fit them to a line
  coefficients = polyfit(x, y, 1);
  % Get a line from the left to the right
  xEnds = [1, columns];
  yEnds = polyval(coefficients, xEnds);
  hold on;
  line(xEnds, yEnds, 'Color', 'r');
% Get the angle
angle = atand((yEnds(2) - yEnds(1)) / (xEnds(2) - xEnds(1)));
rotatedImage = imrotate(Iedge2, angle);
% Display the rotated image.
subplot(2, 2, 4);
imshow(rotatedImage, [],'Parent',handles.axes2);
title('Binary Image', 'FontSize', 10);
%regionprops to get all the BoundingBoxes form your lines
stats=regionprops(rotatedImage,'BoundingBox');
figure, imshow(rotatedImage)
BoundingBoxes=struct2cell(stats);
BoundingBoxes=cell2mat(BoundingBoxes'); % making it into an array
[~,ind]=sort(BoundingBoxes(:,2)); % sorting it to y
BoundingBoxes=BoundingBoxes(ind,:); % apply the sorted vector 
   lineNr=8;
  imshow(original(BoundingBoxes(2,lineNr):...
  BoundingBoxes(2,lineNr)+BoundingBoxes(4,lineNr),...
  BoundingBoxes(1,lineNr):BoundingBoxes(1,lineNr)+BoundingBoxes(3,lineNr)  ))