Single Characther Cursive Processing

Question

yoel y 2021-2-7

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/738367-single-characther-cursive-processing

评论： yoel y 2021-2-7

A2_final.png

Hello,

I'm researching on handwritten letter recognition using matlab. I have several pre-processed image that i took before, and i want to build a program that can automatically find the area of a character like below

from this,

into this

I've been searching for answers, but i only find cursive related question, which disccused how to extreact letters from words in an image. If anyone knows the solution, please let me know, thank you so much.

p.s. : i've attached one preprocessed image

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

Walter Roberson 2021-2-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/738367-single-characther-cursive-processing#answer_616887

在 MATLAB Online 中打开

info_struct = regionprops(imbinarize(YourImage), 'Area');
areas = [info_struct.Area];

This would give a separate area for each separate section. Separate sections could be due to having multiple characters in one image, or could be do to (for example) the dot being separate from the main stroke of a character

If you are certain that there is only one character in the image and you want to know the combined areas, then

area = nnz(imbinarize(YourImage))

3 个评论
显示 1更早的评论隐藏 1更早的评论

Walter Roberson 2021-2-7

在 MATLAB Online 中打开

bw = imbinarize(YourImage);
info_struct = regionprops(bw, 'Area', 'Image');

info_struct(K).Area is now the area of the K'th section, and info_struct(K).Image is now a binary image, on for each pixel inside the region.

If you are sure you want to count all the regions together (for example count the dot of an i as part of the same image) then

bw = imbinarize(YourImage);
info_struct = regionprops(double(bw), 'Area', 'Image');

You can also instead

bw = imbinarize(YourImage);
area = nnz(bw);
maskv = any(bw,1);
lb = find(maskv,1,'first');
rb = find(maskv,1,'last');
maskh = any(bw,2);
tb = find(maskh,1,'first');
bb = find(maskh,1,'last');
Image = YourImage(tb:bb, lb:rb);