[tesseract-ocr] Benchmarking on Android

Hi all, I've done a small writeup on the performance differences running Tesseract on Android, building with GCC vs. Clang and also a bit about 3.05.02 vs. 4.0.0. In case you're interested, have a read here: https://rhardih.io/2019/01/benchmark-numbers-for-tesseract-on-android/ Best

Re: [tesseract-ocr] Android app using Tesseract v4 for OCR

to give it a spin. What is the reason for asking by the way? Best regards, René Hansen On Sun, 31 Mar 2019 at 20:25 Greg Dunkel wrote: > Please post to list. I am not the only one who would be interested in such > an app. > > On Sun, Mar 31, 2019, 10:34 AM Soumik Ranjan Dasgu

Re: [tesseract-ocr] Android app using Tesseract v4 for OCR

are it with you. > > On Mon, Apr 1, 2019 at 1:52 AM René Hansen wrote: > >> Here you go: >> >> https://camverter.rhardih.io >> >> I'm out of promo codes at the moment, but feel free to sign up as a beta >> tester, if you're not a buyer and want to try it

Re: [tesseract-ocr] Android app using Tesseract v4 for OCR

he > webapp in the following thread : > https://groups.google.com/forum/#!topic/tesseract-ocr/ViPGSNeYuiw > > On Thu, Apr 4, 2019 at 1:43 AM René Hansen wrote: > >> I'm not sure I follow. Tesseract, as a library, is already cross >> platform, is it not? You just need to co

Re: [tesseract-ocr] Re: Tesseract trained data for only digits.

The default traineddata works fine for numbers. If you whitelist only digits, that's all you're going to get back: ->SetVariable("tessedit_char_whitelist", "0123456789") /René On Wed, 5 Jun 2019 at 05:05, youcef bou wrote: > Any solution ? same problem > > Le jeudi 8 mai 2014 09:58:51 UTC+2,

Re: [tesseract-ocr] Image to Text conversion not happening correctly using Tesseract

This might be helpful: https://github.com/tesseract-ocr/tesseract/wiki/ImproveQuality /René On Sat, 13 Jul 2019 at 13:41, Vineet Garg wrote: > Output of Tesseract is > > hello2 > Welcome To CrossRoads Cafe! > 1/22/2019 8:00 > > , a Sor > > ae ed > sehen ee ae ve sais lanai sea > > ti > > ee

Re: [tesseract-ocr] Config hints to improve recognition accuracy.

cify-traineddata-language-path-and-language-code-when-using-tesser > > Thanks > > On Friday, August 30, 2019, René Hansen wrote: > > A few config params wont do the trick. You need to preprocess the image. > Make sure you read this > https://github.com/tesseract-ocr/tesseract/wiki/I

Re: [tesseract-ocr] Getting started with tesseract-ocr in a web app.

Your question isn't hard to understand at all. A web app that does OCR using tesseract. Simple. It's figuring out *how* to explain it that's hard. E.g., in your first email in this thread, you write: "I downloaded the tesseract-master.zip file, upload it to my web server and unzipped it" This

Re: [tesseract-ocr] Getting started with tesseract-ocr in a web app.

Hey Clint, Theres a ton of different ways you could build a system such as that. It all comes down to your experience level with the different technologies involved. E.g. if you know how to use the command-line version of tesseract, what you could do, is write a web application, that handles

Re: [tesseract-ocr] Open source (BSD) MICR dataset for Tesseract v4 + evaluation app

Very cool. Thank you for open sourcing this! /René On Tue, 17 Sep 2019 at 07:38, Mamadou wrote: > Hello, > > We've open sourced (BSD 3-Clause License) our MICR dataset and > *.traineddata for Tesseract v4. > > This was developed as an internal R project and never went to production > as we

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

A bit late to the party here, but I've just pushed changes that update build configs for tesseract 4 in https://github.com/rhardih/bad. It now supports building 4.0.0 *and* 4.1.0. I've tested both versions on x86, armv7-a and arm64-v8a. All seems to be working just fine. I'm using the default

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

t; > This said, to setup and run ndk-build on any supported platform (Linux, > Windows, MacOS) is IMHO easier than to use docker. > > BR, > Alex > > On Tuesday, July 30, 2019 at 3:05:52 PM UTC+3, René Hansen wrote: >> >> A bit late to the party here, but I've just pu

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

epository. > > Zdenko > > > po 5. 8. 2019 o 8:35 René Hansen napísal(a): > >> Awesome! Thanks Zdenko. >> >> Would it be possible to tag c5a50b93ce as something like 4.1.1? >> >> That way I can target an official release and get rid of my own fork. >&

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

o.cpp.o - file is move to training part >> >> Zdenko >> >> >> št 1. 8. 2019 o 19:05 René Hansen napísal(a): >> >>> Thanks Alex. >>> >>> Cool Zdenko, >>> >>> I can't find any reference to the unittest sub-directory

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

Agreed. Maybe the real solution after all, is to drop the usage of glob, and go for a portable solution? This is how I got around it initially. Not the best code though: https://github.com/tesseract-ocr/tesseract/compare/4.1.0...rhardih:4.1.0-rhardih On Wed, 7 Aug 2019 at 10:56, 'Stefan Weil'

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

I can completely understand the reasons and need for the way tess-two does things. If I was working with Android Studio and Java/Kotlin, I would probably never have spend time on this. Last time I used tess-two it worked flawlessly. I am coming at this from the perspective of Qt projects however.

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

more. /René On Wed, 31 Jul 2019 at 19:32, Alex Cohn wrote: > On Wednesday, July 31, 2019 at 1:43:24 PM UTC+3, René Hansen wrote: >> >> Thanks Alex, I'll go and have a look. One would imagine that -D >> BUILD_TRAINING_TOOLS=OFF should be enough. >> > > Disablin

Re: [tesseract-ocr] Re: Tesseract 4.1.0 released

/René On Thu, 1 Aug 2019 at 15:16, Zdenko Podobny wrote: > Thanks. Attached patch should fix it (it does not solve unittest part > @Shree: are you able to fix unittest). Can you test it? > > Zdenko > > > št 1. 8. 2019 o 13:03 René Hansen napísal(a): > >> Go

Re: [tesseract-ocr] Why do i have to install libtesseract-dev? (sudo apt install libtesseract-dev)

Other projects can be built and linked against Tesseract. One example is OpenCV. In that case, you can install the -dev package, to avoid having to compile tesseract yourself, in order to get headers and linkable libraries. /René On Thu, 1 Aug 2019 at 17:57, Areo Twister wrote: > I already

Re: [tesseract-ocr] Use Tesseract dll with c project

It's *literally* one of the main items in list of wiki pages: https://github.com/tesseract-ocr/tesseract/wiki/APIExample Is it me, or is this a growing trend on this mailing list? E.g. people just firing off emails with completely open questions instead of doing a minimal effort to search or

Re: [tesseract-ocr] Please advise

Start by reading the available documentation: https://github.com/tesseract-ocr/tesseract/wiki If you feel some of it could be clearer or improved, feel free to request changes and/or provide the updates yourself. /René On Wed, 16 Oct 2019 at 12:44, myquest wrote: > Hi > > > I am newbie and

Re: [tesseract-ocr] Single Randomly placed Digits in a image

This might help you: https://www.pyimagesearch.com/2018/09/17/opencv-ocr-and-text-recognition-with-tesseract/ On Thu, 26 Dec 2019 at 08:41, Kundan Karna wrote: > First of all ,i am not much experienced in python and tesseract . > I want to extract the bounding box of texts in an image which has

Re: [tesseract-ocr] difficulty in compiling tesseract 4 file (libtesseract.so) for Android

Hey Nikko, Could you provide an MRE ? That would make it easier to troubleshoot. Alternatively you could also try to build and run the provided tesseract test apps on an android device, providing the proper path to the shared libs you