Hello Can you help us with what steps needs to be taken with image
Fonts size are unequal Somewhere background colors are dark text are light What can we increase the accuracy in this ? On Tuesday 19 December 2023 at 21:45:18 UTC+5:30 Zdenko Podobny wrote: > Hello, > > For Tesseract you need to remove all non-text parts (graphics element). > IMO also the outline number would be problematic. > > It would be better to post the original image so people can play with > preprocessing... > See e.g. this discussion > https://groups.google.com/g/tesseract-ocr/c/YqW9XhbWC_8/m/75juLKoJDwAJ > (not sure if this is possible with javascript) > > Zdenko > > > ut 19. 12. 2023 o 17:08 Aftab <afta...@gmail.com> napísal(a): > >> Hey guys, >> I am very new to image processing & OCR. But after a lot of trial and >> error. >> I have reached to this point. I have a small image, cropped from larger >> input and the image is pre-processed to maximise the visibility of the >> number. >> >> It is able to detect 10000 at the top, but it is not able to detect the >> number on the bottom. >> Here is the processed image I am working with. >> .[image: processed image.jpeg] >> >> I am running this in Browser using the tesseract.js node module, and here >> is my code for the detection: Tried with default pageseg_mode, as well as >> various other modes. 11 worked best out of all. >> >> async function recognizeText(image) { >> const worker = await createWorker('eng'); >> await worker.setParameters({ >> tessedit_char_whitelist: '0123456789', >> tessedit_pageseg_mode: '11', >> }); >> const ret = await worker.recognize(image); >> console.log(ret.data.text); >> await worker.terminate(); >> } >> >> -- >> You received this message because you are subscribed to the Google Groups >> "tesseract-ocr" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to tesseract-oc...@googlegroups.com. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/7e3e4683-2c35-42d2-ba72-8df2773d15b9n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/7e3e4683-2c35-42d2-ba72-8df2773d15b9n%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to tesseract-ocr+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/ac9c0ac6-2eed-4761-844e-ed67868b56f0n%40googlegroups.com.
