You were right, I just found that my RPi is throttling. It explains the slowing down. Now I'm checking if heatsink could help. So I expect that there is nothing to tune up in my loop. I will check if I can try some smaller model.
best regards Jan Dne středa 30. ledna 2019 11:15:19 UTC+1 Lorenzo Blz napsal(a): > > Did you check this? > > > https://www.raspberrypi.org/forums/viewtopic.php?f=63&t=147781&start=50#p972790 > > Il giorno mer 30 gen 2019 alle ore 08:09 Jan Pohanka <[email protected] > <javascript:>> ha scritto: > >> I have already done that but haven't found anything interesting. >> I tried to ask here if there are eg. any part of algorithms that can be >> disabled etc. The image is preprocessed, binarized and contain only 8 >> digits (and point). I was also a bit surprised that resizing image from >> 400px to 50px has given only subtle speed up. >> >> I will try the fast model today (if I find how to switch it), maybe it >> will help. >> >> here are my measured times >> ocr time: 0.980876922607 >> ocr time: 0.435426950455 >> ocr time: 0.76907491684 >> ocr time: 0.836761951447 >> ocr time: 0.871710062027 >> ocr time: 0.803520917892 >> ocr time: 0.371052026749 >> ocr time: 0.732284069061 >> ocr time: 0.745162010193 >> ocr time: 0.836426019669 >> ocr time: 0.740739107132 >> ocr time: 0.379159927368 >> ocr time: 0.798940181732 >> ocr time: 0.3972260952 >> ocr time: 0.739762067795 >> ocr time: 0.7757999897 >> ocr time: 0.772871017456 >> ocr time: 0.435608863831 >> ocr time: 0.770547866821 >> ocr time: 0.870738983154 >> ocr time: 0.37126493454 >> ocr time: 0.837875127792 >> ocr time: 0.811723947525 >> ocr time: 0.865257024765 >> ocr time: 0.79048204422 >> ocr time: 0.435704946518 >> ocr time: 0.763910055161 >> ocr time: 0.391008853912 >> ocr time: 0.396636009216 >> ocr time: 0.38174700737 >> ocr time: 0.809095144272 >> ocr time: 0.773195028305 >> ocr time: 0.427488088608 >> ocr time: 0.403608083725 >> ocr time: 0.806233167648 >> ocr time: 0.948635101318 >> ocr time: 0.900885105133 >> ocr time: 0.829130887985 >> ocr time: 0.932774782181 >> ocr time: 1.09788799286 >> ocr time: 0.520708799362 >> ocr time: 0.448786973953 >> ocr time: 0.560626983643 >> ocr time: 0.993177175522 >> ocr time: 0.48442697525 >> ocr time: 1.1292309761 >> ocr time: 1.04695606232 >> ocr time: 0.8810338974 >> ocr time: 1.10285806656 >> ocr time: 1.05213713646 >> ocr time: 1.22593903542 >> ocr time: 1.04618191719 >> ocr time: 1.11645102501 >> ocr time: 1.05435395241 >> ocr time: 1.15162396431 >> ocr time: 0.547721862793 >> ocr time: 0.607867956161 >> ocr time: 1.14074802399 >> ocr time: 1.1790971756 >> ocr time: 1.18815803528 >> ocr time: 0.58503985405 >> ocr time: 1.10898280144 >> ocr time: 1.22723913193 >> ocr time: 1.2178709507 >> ocr time: 1.28540086746 >> ocr time: 1.28237104416 >> ocr time: 1.56176805496 >> ocr time: 1.2859480381 >> ocr time: 1.2599170208 >> ocr time: 1.42588591576 >> ocr time: 1.51333785057 >> ocr time: 1.34276986122 >> ocr time: 1.34283900261 >> ocr time: 1.39351201057 >> ocr time: 1.61450195312 >> ocr time: 1.44723105431 >> ocr time: 1.63176107407 >> ocr time: 0.82429599762 >> ocr time: 1.08239603043 >> ocr time: 0.755813121796 >> ocr time: 1.63984704018 >> ocr time: 1.84553313255 >> ocr time: 0.958009958267 >> ocr time: 1.52479290962 >> ocr time: 0.919597864151 >> >> thanks >> Jan >> >> Dne středa 30. ledna 2019 7:57:07 UTC+1 zdenop napsal(a): >>> >>> search issue tracker for "speed"... >>> >>> Zdenko >>> >>> >>> st 30. 1. 2019 o 7:51 Jan Pohanka <[email protected]> napísal(a): >>> >>>> It is 4.0. I'm satisfied with recognition results, but I need to make >>>> it faster (at constant times below 1s)... >>>> >>>> Dne středa 30. ledna 2019 7:48:23 UTC+1 zdenop napsal(a): >>>>> >>>>> What is your tesseract version? >>>>> >>>>> Zdenko >>>>> >>>>> >>>>> ut 29. 1. 2019 o 20:40 Jan Pohanka <[email protected]> napísal(a): >>>>> >>>>>> Thanks for suggestions. You are right that I'm reffering to >>>>>> api.GetUTF8Text() >>>>>> call, it is my bottleneck. >>>>>> I was not aware that there is some fast and best models in tesseract >>>>>> 4.0, I will give it a try. So far I used just lang=eng or osd. >>>>>> For me it is suspicious that the calls are getting longer during the >>>>>> time. Or to be more precise, first 10-15 calls are up to 500ms and >>>>>> latter >>>>>> ones rise above 1s... >>>>>> SetSourceResolution outside of the loop gives no change unfortunately. >>>>>> >>>>>> BR >>>>>> Jan >>>>>> >>>>>> Dne úterý 29. ledna 2019 18:08:49 UTC+1 Lorenzo Blz napsal(a): >>>>>>> >>>>>>> >>>>>>> First double check if the Pi is not throttling due to overheating or >>>>>>> lack of USB power. This may cause the slowdown. >>>>>>> >>>>>>> Usually 30/50 px of text height is fine. IF the problem is >>>>>>> tesseract, try to use the fast model (or "normal" if using best). I >>>>>>> assume >>>>>>> you are using the 4.x release. >>>>>>> >>>>>>> Try tesseract -v to see if you are using all the available CPU >>>>>>> optimizations. >>>>>>> >>>>>>> Try to move the SetSourceResolution outside the loop and see if it >>>>>>> changes something (MAYBE it may invalidate some caches or something). >>>>>>> >>>>>>> The time you are referring to is one single api.GetUTF8Text() call, >>>>>>> correct? >>>>>>> >>>>>>> >>>>>>> Lorenzo >>>>>>> >>>>>>> >>>>>>> Il giorno mar 29 gen 2019 alle ore 17:48 Jan Pohanka < >>>>>>> [email protected]> ha scritto: >>>>>>> >>>>>>>> Hello, >>>>>>>> >>>>>>>> I'm making a simple device used to recognize numbers on pictures >>>>>>>> taken by a webcam. All is running on raspberry pi 3. >>>>>>>> Everything is like following simple loop (in python for simplicity, >>>>>>>> but using C++ api it is the same), images are preprocessed to black >>>>>>>> and >>>>>>>> white >>>>>>>> >>>>>>>> api = PyTessBaseAPI(psm=tesserocr.PSM.SINGLE_WORD) >>>>>>>> >>>>>>>> for im in images: >>>>>>>> api.SetImage(im) >>>>>>>> api.SetSourceResolution(70) >>>>>>>> ot = api.GetUTF8Text() >>>>>>>> >>>>>>>> api.End() >>>>>>>> >>>>>>>> >>>>>>>> My problem is that api.GetUTF8Text() call is quite slow and more >>>>>>>> over it is getting slower and slower over time. Is there any options >>>>>>>> how to >>>>>>>> make recognition faster? I have tried to resize the image to around >>>>>>>> 50x10px. The times starts on around 300ms but then goes up to above 1s >>>>>>>> which is too slow for me. I tried both legacy and LSTM algorithms, but >>>>>>>> they >>>>>>>> are similar. >>>>>>>> >>>>>>>> best regards >>>>>>>> Jan >>>>>>>> >>>>>>>> -- >>>>>>>> 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 [email protected]. >>>>>>>> To post to this group, send email to [email protected]. >>>>>>>> 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/a53b4b25-97e3-47dc-823a-cbb219225eed%40googlegroups.com >>>>>>>> >>>>>>>> <https://groups.google.com/d/msgid/tesseract-ocr/a53b4b25-97e3-47dc-823a-cbb219225eed%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 [email protected]. >>>>>> To post to this group, send email to [email protected]. >>>>>> 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/baa59c86-b002-4607-8dda-16835cd3ea73%40googlegroups.com >>>>>> >>>>>> <https://groups.google.com/d/msgid/tesseract-ocr/baa59c86-b002-4607-8dda-16835cd3ea73%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 [email protected]. >>>> To post to this group, send email to [email protected]. >>>> 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/dedb7fd8-d61e-42bb-a492-34beaa8b1514%40googlegroups.com >>>> >>>> <https://groups.google.com/d/msgid/tesseract-ocr/dedb7fd8-d61e-42bb-a492-34beaa8b1514%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 [email protected] <javascript:>. >> To post to this group, send email to [email protected] >> <javascript:>. >> 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/30831c1e-4bd9-4eae-8118-ad9244949b80%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/30831c1e-4bd9-4eae-8118-ad9244949b80%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 [email protected]. To post to this group, send email to [email protected]. 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/2502861e-00ca-434a-89a6-51646d963f3f%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
