Geek Answers Handbook

OCR Image Noise Reduction

Hello, I am having a problem segmenting the following image below. It is a colored symbol to be recognized. I use sharpening, smoothing and smoothing. After that, I split the image using fuzzy-c clustering (3-grade). But in the case of the letter E, the best that I get does not have drastic changes, smoothing and smoothing, only with threshold segmentation fcm. However, I should get a better result than this, where I could combine these two parts as a whole (and not just the top white part with the other half black).

How can I solve this problem in order to be more reliable and work with other images, for example, 5 in the picture? Result 5 - with sharpening, deflation and smoothing, on top of fcm clustering. How can I make it more connected maybe?

I would really appreciate any help that I could get, please oh, and I do it in Matlab ... so it would be nice to get any help from there, thanks!

This is letter E, i would like to get one element as a whole

Second pic is number 5, should be more smooth and connected, without any spaces between lines

EDIT:

My next code is: function [bw, level] = fcmthresh (IM, sw) if (nargin <1) error ('You must provide an image.'); elseif (nargin == 1) SW = 0; elseif (sw ~ = 0 & sw ~ = 1) error ('sw must be 0 or 1.'); end

data=reshape(IM,[],1); [center,member]=fcm(data,3); [center,cidx]=sort(center); member=member'; member=member(:,cidx); [maxmember,label]=max(member,[],2); if sw==0 level=(max(data(label==1))+min(data(label==2)))/2; else level=(max(data(label==2))+min(data(label==3)))/2; end bw=im2bw(IM,level); function img=wienerDeblur(im) ImgNoisyBlurry = im2double(im); PSF = fspecial('laplacian'); %LEN, THETA add parameters for 'motion' noise_var = 0.0001; %0.0001 estimated_nsr = noise_var / var(ImgNoisyBlurry(:)); wnr3 = deconvwnr(ImgNoisyBlurry, PSF, estimated_nsr); img = wnr3; end H = fspecial('unsharp'); im = imfilter(im,H,'replicate'); im = wienerDeblur(im); im = wienerSmoothing(im);

This is all the code, plus I use only fcmthres for the letter E, because it works best. I read about morphological image processing (expansion, erosion) to possibly do the trick.

Is there a better technique for image contrast and noise removal?

+4
source share
2 answers

You can try canny edge detection on the h-channel (try all the channels and compare) (rgb2hsv) and then morphological operations to close the curves and fill (fill) the curves of numbers and letters. I do not have access to Matlab now, but I have had success with a similar problem.

http://www.mathworks.se/help/techdoc/ref/rgb2hsv.html http://www.mathworks.se/help/toolbox/images/ref/edge.html http://www.mathworks.se/ help / toolbox / images / ref / imfill.html

+1
source

after you get it in black and white, you can use the decomposition of singular values ​​(http://en.wikipedia.org/wiki/Singular_value_decomposition), and you can probably compare the special values.

To remove noise, set smaller singular values ​​to 0.

0
source

More articles:


All Articles