How to do segmentation for cursive character images?

Question

ana ain 2016-7-21

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/296592-how-to-do-segmentation-for-cursive-character-images

评论： Image Analyst 2016-7-23

I've done segmentation process for isolated character image using the following code. But it just work for isolated character using vertical projection method.

Soumyadeep have explained, The graph at the letter location will start from 0 , reach the peak and then fall but it will never touch 0 as the letters in this case are connected. by looking at the graph we could define the threshold (if I'm not mistaken), which should be the maximum of the deepest points of the valleys we see in the graph. The problem is, I didn't know how to set the threshold. I've already try to modify several values, but it didn't gave me a suitable result.

% Preprocessing + Segmentation
% // Original Code of Vertical Projection for Segmentation by Soumyadeep Sinha //
% // Modified by Ana Ainul S. : anaainul@gmail.com, Last modified : 20/04/16 //
% Saving each  single segmented character as one file 
function [se] = segmentt (a)
myFolder = 'D:\1. Thesis FINISH!!!\Data set';
%%Binarization %%
level = graythresh (a);
b = im2bw (a, level);
%%Complement %
c = imcomplement (b);
i=padarray(c,[0 10]);
% Vertical Projecttion for Character Segmentation
verticalProjection = sum(i, 1);
set(gcf, 'Name', 'Segmentation Trial', 'NumberTitle', 'Off') 
subplot(2, 2, 1);imshow(i); 
subplot(2,2,3);
plot(verticalProjection, 'b-');
grid on;
t = verticalProjection;
t(t==0) = inf;
mayukh = min(t); % return smallest element of t
% 0 where there is background, 1 where there are letters
letterLocations = verticalProjection > mayukh;
% Find Rising and falling edges
d = diff(letterLocations);
startingColumns = find (d>0);
endingColumns = find(d<0);
% Extract each region
y=1;
for k = 1 : length(startingColumns)
  % Get sub image of just one character...
  subImage = i(:, startingColumns(k):endingColumns(k)); 
   % im = subImage;
   s = subImage;
   figure, imshow (s);
   imwrite( s, fullfile( myFolder, sprintf('data.%d.png', k ) ) );
end;
se = (s);