FYI - this was trained using eng.traineddata and finetuned with 7segment fonts.
ShreeDevi ____________________________________________________________ भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com On Wed, Mar 29, 2017 at 9:09 PM, ShreeDevi Kumar <shreesh...@gmail.com> wrote: > Hi, > > I have built a 4.0 traineddata using some seven segment display fonts. > Trained mostly on numbers 0-9, capital letters A-Z, : etc. > > It is uploaded as a zip file at https://github.com/ > Shreeshrii/tessdata4alpha/raw/master/ssd1.zip > > unzip to get ssd1.traineddata > > I have not tested it much. Seemed to work with the sample images provided > in this email. > > Since B and 8, O and 0, S, Z and 5 all look similar in this display, there > would be errors. > > Should this be trained just for numbers, may have a better accuracy then. > > I am doing another run of training to see if there are improvements. > > Meanwhile, those who actually want this should give it a try and provide > feedback. > > > ShreeDevi > ____________________________________________________________ > भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com > > On Fri, Mar 24, 2017 at 3:04 PM, <komalagaw...@gmail.com> wrote: > >> >> Hello, >> I am basically working in electronics field and new to C#.Currently I am >> working on one project (Image processing in C#) where i am using C#,where >> in one of the part i have to detect text or digits of 7 segment display >> image for that on google i found Tesseract solution. >> >> For experiment i have first try to convert normal text image in to text >> file and it is working fine for some of the basic images but it is not >> working with 7 segment display.so i came to know i required trained data >> file for 7 segment. >> >> For training 7 segment data i follow the steps which are shown in vidoe >> of below link:https://www.youtube.com/watch?v=i_1-hGsXxy8. >> But the output.txt file showing in that video is not generating in my >> case.Due to which after using trained 7 segment data file ,i am getting >> garbage value in text file.So for checking that i am getting proper trained >> file or not , i have follow the procedure which is shown on that video but >> it is giving an error like outpt.txt file not found.Is this happened >> because of missing otput.txt file or something else i am missing to do.I >> have follow all the steps which are shown in that video for training 7 >> segment data. >> >> Also i have installed jTessBoxEditorFX.jar, serak trainer & Tesseract-ocr >> v3.02.So at the end i am just stuck in the point where i don't know >> where i am going wrong,is my procedure is wrong or software installation is >> not proper because after installing tesseract there is red cross mark >> against tesseract. >> >> Please somebody help me to figure it out.If possible please provide me 7 >> segment trained data file and also the exact steps to trained 7 segment >> data as i have to trained some more files for various display icons and >> some specific messages.Its very urgent as my project is stuck and i am >> helpless because after trying so much solutions in image processing for 7 >> segment display detection like pixel count & image comparison in C#, i came >> up on tesseract solution. >> If you have any doubts on understanding my query please let me know. >> >> Please do the needful. >> >> >>> >>> >>> >>> >>> -- >> 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 post to this group, send email to tesseract-ocr@googlegroups.com. >> Visit this group at https://groups.google.com/group/tesseract-ocr. >> To view this discussion on the web visit https://groups.google.com/d/ms >> gid/tesseract-ocr/b7fc9a05-8d8d-4e68-ac02-2e71b0078557%40googlegroups.com >> <https://groups.google.com/d/msgid/tesseract-ocr/b7fc9a05-8d8d-4e68-ac02-2e71b0078557%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> For more options, visit https://groups.google.com/d/optout. >> > > -- 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 post to this group, send email to tesseract-ocr@googlegroups.com. Visit this group at https://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduU9wd%2BwPwr5%3Dg-uAySsO7zN79xgUjkMdLZbzA_sxCfD5Q%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
