CODEWITHSUNDEEP
node.js
FunnyCheese
FunnyCheese

Reputation: 341

remove noise from image text nodejs

how can i remove background noise and leave only the text? Example image:

enter image description here

My code:

var Tesseract = require('tesseract.js');
var Jimp = require("jimp");


Jimp.read("12.png").then(function (image) {
    image
        .color([
          { apply: 'brighten', params: [20] }
        ])
        .contrast(1)
        .greyscale()
        .write("img-opt.jpg");
})
.then(function() {
  Tesseract.recognize('img-opt.jpg', {
      tessedit_char_whitelist: 'AN%D%P'
  })
    .progress(function(message){console.log(message)})
    .catch(function(err){console.error(err)})
    .then(function(result){console.log(result.text)})
})
.catch(function (err) {
    console.error(err);
});

And i got this result:

enter image description here

So, how can i can remove background noise? Thanks!

Upvotes: 3

Views: 3745

Answers (1)

mihai
mihai

Reputation: 38543

I got better results by removing the brigthen transformation. Also desaturate seems to work better than grayscale.

image
    .color([{apply: 'desaturate', params: [90]}])
    .contrast(1)
    .write("img-opt.jpg");

result

You just need a little bit of trial and error.

Upvotes: 2

Related Questions