Update: I tried the Google Vision API, and it is actually ridiculously good. It hit all the targets except two, and that's without a tag dictionary.
On Tue, Nov 14, 2023 at 8:26 AM Olav Storøy <olavstor...@gmail.com> wrote: > Interesting! I'd be worried it could remove important text features, but > maybe tune it to not remove lines shorter than x. I definitely need to look > at cv2. Until now I've sort of assumed it's best to make Tesseract do as > much of this process as possible... Thanks for your input > > On Mon, Nov 13, 2023 at 8:05 PM Art Rhyno <artrh...@uwindsor.ca> wrote: > >> With such clear diagrams, there might be value in having OpenCV remove >> the horizontal and vertical lines, and then identifying and merging the >> blobs that are left to get the regions for recognition. I tried this a bit >> with one of your examples, it would take more refinement but there might be >> a path to getting good bounding boxes at the image level. >> >> >> >> art >> >> >> >> *From:* tesseract-ocr@googlegroups.com <tesseract-ocr@googlegroups.com> *On >> Behalf Of *Tom Morris >> *Sent:* Monday, November 13, 2023 11:14 AM >> *To:* tesseract-ocr <tesseract-ocr@googlegroups.com> >> *Subject:* Re: [tesseract-ocr] Re: Tesseract on technical drawings >> >> >> >> >> >> On Monday, November 13, 2023 at 5:35:20 AM UTC-5 olavs...@gmail.com >> wrote: >> >> >> >> Yeah it seems page segmentation is the crucial issue. If the bounding >> boxes are good, the recognition is usually very good. >> >> >> >> I think I've sort of reached the limit on what I can do with base >> Tesseract. I think the next step would be custom training / fine-tuning. >> >> >> >> Tesseract's page layout analysis / segmentation isn't training based, so >> I don't think this is going to help you. If you wanted to recognize the C/L >> glyph, you could do fine tuning training for it, but it's not going to help >> you with the problem of finding rotated text and accurately determining >> bounding boxes for text of interest. >> >> >> >> It's been ages since I've done serious image processing, but I'd >> recommend looking at something like OpenCV's text detection: >> >> https://docs.opencv.org/4.8.0/d4/d43/tutorial_dnn_text_spotting.html >> >> >> >> Aspirationally, you can get some idea of what's possible by playing with >> Google's Cloud Vision API demo >> >> https://cloud.google.com/vision/docs/drag-and-drop >> >> >> >> It lets you just drag & drop an image and then inspect the results both >> visually and via the JSON that the API produces. >> >> >> >> Good luck! >> >> >> >> Tom >> >> -- >> 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/3a6e0271-db4b-4624-bada-51167dd6d744n%40googlegroups.com >> <https://groups.google.com/d/msgid/tesseract-ocr/3a6e0271-db4b-4624-bada-51167dd6d744n%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/YQBPR0101MB990290D0BD05A1D3F3A8BA40DCB3A%40YQBPR0101MB9902.CANPRD01.PROD.OUTLOOK.COM >> <https://groups.google.com/d/msgid/tesseract-ocr/YQBPR0101MB990290D0BD05A1D3F3A8BA40DCB3A%40YQBPR0101MB9902.CANPRD01.PROD.OUTLOOK.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/CADVG04rff%3DzAa8WxnXEuVm45C%3DX%3Dz4M0nAg8xOTc4zEa%2BbASpQ%40mail.gmail.com.
