Working code of Optical Character Recognition (OCR)

[imsiddhant07]

Hello, If the issue is not assigned to anyone, may I work on it?

[Gaurav7888]

Hi , I opened this issue because i think its missing from this project if assigned then i want to work on this issue@adiXcodr

[imsiddhant07]

Yes true, but you should have provided a description while raising an issue and also asked about assigning it. Never the less let @adiXcodr decide :)

[Gaurav7888]

Ocr will take time and this is something 2 ppl can contribute together , thats what i think @siddhantjain07 .

[adiXcodr]

Currently, the OCR is made possible with the Tesseract.js framework, if you guys want, you can write a native alternative algorithm but it would need training data

[Gaurav7888]

@adiXcodr can we make ocr code using tensorflow. Js , that will be helpful for ur project ?

[adiXcodr]

Yes that was what I had in mind too. If you want to, you can proceed with it

On Tue, 8 Dec, 2020, 1:05 am Gaurav7888, notifications@github.com wrote:

@adiXcodr https://github.com/adiXcodr can we make ocr code using tensorflow. Js , that will be helpful for ur project ?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/adiXcodr/node-ocr/issues/7#issuecomment-740133226, or unsubscribe https://github.com/notifications/unsubscribe-auth/AKMGL4ZVPW4CCPPJ4KNQ7HLSTUU6JANCNFSM4UOGC6GQ .

[harshil15999]

@adiXcodr does the OCR module have to be in javascript?

[Gaurav7888]

@adiXcodr i will start working on it using tf.js ...

[adiXcodr]

@adiXcodr does the OCR module have to be in javascript?

Preferably. Doing it in some other language like python would need an extra API call. The computation needs to be done natively