Jean-Marc i have posted the pic. https://groups.google.com/g/tesseract-ocr/c/1pi2LR4ug_A/m/NM74kUO7DgAJ
On Monday, 16 November 2020 at 19:51:53 UTC+5:30 Lorenzo Blz wrote: > Use hough lines detector to detect the lines and draw a thick white line > over them. > > > https://opencv-python-tutroals.readthedocs.io/en/latest/py_tutorials/py_imgproc/py_houghlines/py_houghlines.html > > Or use MSER to detect blobs of the appropriate size. > > https://answers.opencv.org/question/19015/how-to-use-mser-in-python/ > > https://stackoverflow.com/questions/40078625/opencv-mser-detect-text-areas-python > > > Lorenzo > > > Il giorno lun 16 nov 2020 alle ore 14:30 agentbond009 <kshit...@gmail.com> > ha scritto: > >> Thanks Martin Läuter for the tutorial. But still i am gonna use >> tesseract for this bc i don't think my PC will handle tenserflow. >> >> On Sunday, 15 November 2020 at 20:56:35 UTC+5:30 Martin Läuter wrote: >> >>> Hi, >>> Someone did some (apparently) solid sudoku solving without tesseract, >>> see >>> https://blog.scottlogic.com/2020/01/03/webassembly-sudoku-solver.html. >>> And the explanation is not bad either. >>> Hope that helps, >>> Martin. >>> On Sunday, November 15, 2020 at 12:25:49 PM UTC+1 agentbond009 wrote: >>> >>>> [image: sudoku_output.jpg]yes i think the image is always the same >>>> size but i am not able to cut out the columns(still figuring out ). Like >>>> the image below is the result after thresholding. The problem is to reduce >>>> the noise(all depends on quality) like for example if the pic is of bad >>>> quality like this one, then i think the ocr will mess it up. >>>> there are two pics for comparision. >>>> On Friday, 6 November 2020 at 20:56:25 UTC+5:30 Jean-Marc Spaggiari >>>> wrote: >>>> >>>>> Is the image always the same size? Can you cut the columns to remove >>>>> them and keep just the numbers? >>>>> >>>>> Le ven. 6 nov. 2020 à 10:21, agentbond009 <kshit...@gmail.com> a >>>>> écrit : >>>>> >>>>>> sorry for the late reply >>>>>> I have done the thresholding part by using opencv and converted the >>>>>> whole image to gray >>>>>> and still stuck with recognising the digits. i was thinking to use >>>>>> clear_border from skimage. >>>>>> >>>>>> On Friday, 30 October 2020 at 19:57:19 UTC+5:30 Jean-Marc Spaggiari >>>>>> wrote: >>>>>> >>>>>>> Hi, >>>>>>> >>>>>>> Did you give it a try? >>>>>>> >>>>>>> Try to have the numbers in white in black in a withe background. Are >>>>>>> you able to get each cell separately? >>>>>>> >>>>>>> JM >>>>>>> >>>>>>> Le ven. 30 oct. 2020 à 10:22, agentbond009 <kshit...@gmail.com> a >>>>>>> écrit : >>>>>>> >>>>>>>> hello. I am working on a project, i was wondering if tesseract is >>>>>>>> an OCR can it read the no.s from a sudoku. >>>>>>>> thanks >>>>>>>> >>>>>>>> -- >>>>>>>> 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/4abe9a73-9543-4a5b-b29e-a917874378f8n%40googlegroups.com >>>>>>>> >>>>>>>> <https://groups.google.com/d/msgid/tesseract-ocr/4abe9a73-9543-4a5b-b29e-a917874378f8n%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-oc...@googlegroups.com. >>>>>> >>>>> To view this discussion on the web visit >>>>>> https://groups.google.com/d/msgid/tesseract-ocr/f2c178b6-5861-49be-8b23-3506f58c14edn%40googlegroups.com >>>>>> >>>>>> <https://groups.google.com/d/msgid/tesseract-ocr/f2c178b6-5861-49be-8b23-3506f58c14edn%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-oc...@googlegroups.com. >> > To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/c66465e3-3640-4daa-9ae4-52ae2dbc8451n%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/c66465e3-3640-4daa-9ae4-52ae2dbc8451n%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/d81d85e4-e079-4347-a18a-e8fc8553bcf4n%40googlegroups.com.
