Quan, *GOOD NEWS* Just now I tried your product on *anothe*r old machine Pentium-IV-HT - 512Ram -*win xP2* (machine under disposal) Downloaded both vietocr(jav) and Viet.net on the said old machine.. Successfully tested on *A4 size* kannada image file and *output was fine within matter of 2 or 5 seconds.* No error message. No memory problem. It is really strange!! and new experience for me! Whereas In New purchased Dell (Optiplex 330)- which has 3GB Ram - *win xp-with sp3* were giving trouble. I could not fathom the reason why it happens?
In Nutshell, it appears that all depends on configuration of machine or operating system installed??. It is hereby confirmed that your software has nothing to do with RAM capacity nor memory problem JavaOCR and Net ocr works wonderfully without any hitch and even faster than expected With Choicest Best Wishes, - Sriranga(76yrsold) On Tue, Jun 23, 2009 at 10:39 AM, 74yrs old <[email protected]> wrote: Quan, Thankful to you for the clarification. with Best Wishes, sriranga(76yrsold) ---------- Forwarded message ---------- From: 74yrs old <[email protected]> Date: Tue, Jun 23, 2009 at 11:32 AM Subject: Re: Fwd: vietocr - supported Kannada - problem. To: Quan Nguyen <[email protected]>, Rao <[email protected]> Quan, *GOOD NEWS* Just now I tried your product on *anothe*r old machine Pentium-IV-HT - 512Ram -*win xP2* (machine under disposal) Downloaded both vietocr(jav) and Viet.net on the said old machine.. Successfully tested on *A4 size* kannada image file and *output was fine within matter of 2 or 5 seconds.* No error message. No memory problem. It is really strange!! and new experience for me! Whereas In New purchased Dell (Optiplex 330)- which has 3GB Ram - *win xp-with sp3* was giving trouble. In Nutshell, all depends on configuration of machine or operating system installed??. It is hereby confirmed that your software has nothing to do with RAM capacity nor memory problem JavaOCR and Net ocr works wonderfully without any hitch and even faster than expected With Choicest Best Wishes, - Sriranga(76yrsold) On Tue, Jun 23, 2009 at 10:39 AM, 74yrs old <[email protected]> wrote: > Quan, > Thankful to you for the clarification. > with Best Wishes, > sriranga(76yrsold) > > On Tue, Jun 23, 2009 at 8:28 AM, Quan Nguyen <[email protected]> wrote: > >> I meant the size of the file in megabytes. >> >> >> 2009/6/22 74yrs old <[email protected]> >> >>> I could not follow "if the image is too large" - whether size of the page >>> of the image file or >>> every font image should be within 10 point size? Generally all images of >>> all formats will have or set to 300 dpi with help of irfanview software >>> before OCRed. >>> With regards, >>> -Sriranga(76yrsold). >>> >>> >>> On Tue, Jun 23, 2009 at 6:35 AM, Quan Nguyen <[email protected]>wrote: >>> >>>> The ocr.bat gives the Java virtual machine more memory to process or >>>> manipulate images, such as rotate or zoom in/out. Still, if the image is >>>> too >>>> large, it will pose a problem for any system. >>>> >>>> 2009/6/22 74yrs old <[email protected]> >>>> >>>> Quan, >>>>> Kindly parawise my comments are furnished below: >>>>> With regards, >>>>> -sriranga(76yrsold) >>>>> >>>>> On Mon, Jun 22, 2009 at 2:47 AM, Quan Nguyen <[email protected]>wrote: >>>>> >>>>>> I included "kan" in the re-released version. >>>>>> >>>>>> Nothing is wrong in vietocr.jpg; apparently, font Tunga cannot display >>>>>> English phrase "The quick brown fox jumps over the lazy dog. >>>>> >>>>> > You are wrong. In fact Tunga font will not display English phrase >>>>> but only Kannada script. >>>>> In case if the said font unable to display due to unforeseen problem >>>>> it will show as square boxes in place of font. This is first time I am >>>>> facing this situation only in your java product >>>>> >>>>> >>>>>> >>>>>> I looked at your .zip file from mediafire. When I unzipped it, many >>>>>> files were placed in wrong locations. I don't know how you zipped it in >>>>>> the >>>>>> first place and whether something was wrong with your setup. Anyway, >>>>>> after >>>>>> re-arranging the files, I was able to execute the program and >>>>>> successfully >>>>>> recognize your image. >>>>>> >>>>> >>>>> > Kan.datafiles and xml added to original zip files and as such it >>>>> might have not placed in the relevant data folder/tessdata folder I >>>>> presume. >>>>> Even exe renamed as exe-1 to overcome the rejection by the gmail. >>>>> >>>>>> >>>>>> Please get Rao to help you. I can't spend more time on this. >>>>>> > Sorry for the inconvenience caused to you on account of peculiar >>>>>> situation - This is happening first time fro me. However in one way. I >>>>>> gained some experience on account of your product >>>>>> >>>>> >>>>> >>>>>> I just released VietOCR.NET 0.9.4. I verified it could recognize >>>>>> Kannada also. Check it out if you still have problem with the Java >>>>>> version. >>>>> >>>>> > mostly may be of problem of memory for pre-processing? In fact I have >>>>> got installed 2.98(3GB)RAM memory New model is Dell. Question is how to >>>>> solve the problem of memory inspite of more freespace availalble in Hdd .I >>>>> never faced such situation so far. >>>>> Since less than one month I had installed Java for hands on experience >>>>> - which is new to me. I am thinking whether if I uninstalled java whether >>>>> memory problem go away for Net product? I doubt because freeocrNet 2.4 >>>>> still >>>>> works fine. >>>>> However latest version freeocrnet 2.6 does not work the latest version >>>>> of tesseract.* Please don't feel hurt if I compared with FreeOCRNet* - >>>>> just to have idea for your research purpose. >>>>> I tried to replace with old version of tesseract.exe to test. >>>>> Surprisigly eventhough new your exe had renamed as *exe 1*, another >>>>> new your exe re-installed. So your netocr will not accept the exe file >>>>> other >>>>> than generated by you only!!. However I restored all. >>>>> >>>>>> >>>>>> >>>>>> Regards, >>>>>> Quan >>>>>> >>>>>> >>>>>> 2009/6/21 74yrs old <[email protected]> >>>>>> >>>>>>> Again I downloaded zip from your website. When checked data file - I >>>>>>> am surprised to note "kan" xml already included. Also add tess data of >>>>>>> Kannada. But still problem. >>>>>>> when selected "kannada" - 'error assessing files whereas English or >>>>>>> Viet all kannada output is garbage - not readable. settings.jpg also >>>>>>> attached. when clicked on tessdata - no details appeared. >>>>>>> With regards, >>>>>>> -sriranga >>>>>>> >>>>>>> >>>>>>> On Sun, Jun 21, 2009 at 10:32 PM, 74yrs old <[email protected] >>>>>>> > wrote: >>>>>>> >>>>>>>> I have uploaded the zip file vide >>>>>>>> http://www.mediafire.com/?zzwmrmzdyiy. This is first time I am >>>>>>>> doing. >>>>>>>> >>>>>>>> On Sun, Jun 21, 2009 at 9:57 PM, Quan Nguyen >>>>>>>> <[email protected]>wrote: >>>>>>>> >>>>>>>>> By looking at the attached picture, I can tell that you've set >>>>>>>>> VietOCR to a wrong Tesseract folder. If you point to the correct >>>>>>>>> folder, >>>>>>>>> "Kannada" would show in the Language combobox. >>>>>>>>> >>>>>>>>> Rao, since you have successfully used VietOCR to recognize Kannada, >>>>>>>>> can you help Sriranga? >>>>>>>>> >>>>>>>>> 2009/6/21 74yrs old <[email protected]> >>>>>>>>> >>>>>>>>> Rao, >>>>>>>>>> as suggested by you I tried to run in comd.exe same error message >>>>>>>>>> displayed. attached jpg file which is explantaory. >>>>>>>>>> >>>>>>>>>> Reg: spell checker of DangAmbig - I wanted to request Nguyeng to >>>>>>>>>> examine and can be added as additional tool for frontendOCR. I think >>>>>>>>>> programming can be done in Java? >>>>>>>>>> with blessings, >>>>>>>>>> -sriranga(76yrsold) >>>>>>>>>> >>>>>>>>>> On Sun, Jun 21, 2009 at 4:26 PM, MVS RAO < >>>>>>>>>> [email protected]> wrote: >>>>>>>>>> >>>>>>>>>>> नमस्कार् SriRanga Sir >>>>>>>>>>> >>>>>>>>>>> For VierOcr , I was wondering if you are using the jar file >>>>>>>>>>> directly . >>>>>>>>>>> actually we have to use the given batch command file i.e. ocr.bat >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> the batch progarm stays on at commandline until we end VietOCR . >>>>>>>>>>> if there is a prolonged delay in OCR Gui , that is if it freezes >>>>>>>>>>> then >>>>>>>>>>> we can terminate the command line batch program by using CTRL C >>>>>>>>>>> >>>>>>>>>>> And regarding the spellchecker , will be checking it soon . >>>>>>>>>>> >>>>>>>>>>> many thanks for your help Sir >>>>>>>>>>> with best regards - mvs >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> --- On *Sun, 21/6/09, 74yrs old <[email protected]>*wrote: >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> From: 74yrs old <[email protected]> >>>>>>>>>>> Subject: Fwd: vietocr - supported Kannada - problem. >>>>>>>>>>> To: "svaram" <[email protected]> >>>>>>>>>>> Date: Sunday, 21 June, 2009, 4:01 PM >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> ---------- Forwarded message ---------- >>>>>>>>>>> From: 74yrs old >>>>>>>>>>> <[email protected]<http://mc/[email protected]> >>>>>>>>>>> > >>>>>>>>>>> Date: Sun, Jun 21, 2009 at 12:13 PM >>>>>>>>>>> Subject: Re: vietocr - supported Kannada - problem. >>>>>>>>>>> To: >>>>>>>>>>> [email protected]<http://mc/[email protected]> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> Just download and tested by clicking on vietocr.jar - error >>>>>>>>>>> message "errors acessing files .." - vide jpg files attached for >>>>>>>>>>> your >>>>>>>>>>> information. . >>>>>>>>>>> English/viet also works fine. Mouse also working fine. >>>>>>>>>>> With Best Wishes. >>>>>>>>>>> -sriranga(76yrsold) >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> On Sun, Jun 21, 2009 at 9:19 AM, nguyenq >>>>>>>>>>> <[email protected]<http://mc/[email protected]> >>>>>>>>>>> > wrote: >>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> I've just released a new update with numerous improvements. The >>>>>>>>>>>> FontDialog now shows all system fonts. Please check it out. >>>>>>>>>>>> >>>>>>>>>>>> http://vietocr.sf.net >>>>>>>>>>>> >>>>>>>>>>>> On Jun 14, 8:48 pm, 74yrs old >>>>>>>>>>>> <[email protected]<http://mc/[email protected]>> >>>>>>>>>>>> wrote: >>>>>>>>>>>> > Yes If Font Dialog is improved - it will be useful for all >>>>>>>>>>>> users of other >>>>>>>>>>>> > lang as universal tool. >>>>>>>>>>>> > Unfortunately SoftifreeOCR (version 2.6 net) now does not work >>>>>>>>>>>> with latest >>>>>>>>>>>> > tesseract.exe(2.04). >>>>>>>>>>>> > Graphic dispay has to be improved. it shows gray color till >>>>>>>>>>>> opened the file; >>>>>>>>>>>> > I am using XPwith sp3 and also installed ubuntu9.04(not used >>>>>>>>>>>> much) >>>>>>>>>>>> > With Best wishes, >>>>>>>>>>>> > -sriranga(76yrsold) >>>>>>>>>>>> > >>>>>>>>>>>> > On Mon, Jun 15, 2009 at 1:37 AM, nguyenq >>>>>>>>>>>> > <[email protected]<http://mc/[email protected]>> >>>>>>>>>>>> wrote: >>>>>>>>>>>> > >>>>>>>>>>>> > > Sorry, the FontDialog was specifically designed to show only >>>>>>>>>>>> fonts >>>>>>>>>>>> > > that contain Vietnamese glyphs. I'll remove the filter to >>>>>>>>>>>> show all >>>>>>>>>>>> > > system fonts in the next release. >>>>>>>>>>>> > >>>>>>>>>>>> > > It's not surprising since VietOCR GUI design was "inspired" >>>>>>>>>>>> by >>>>>>>>>>>> > > FreeOCR. ;-) >>>>>>>>>>>> > >>>>>>>>>>>> > > By the way, what OS and version are you running it on? >>>>>>>>>>>> > >>>>>>>>>>>> > > On Jun 14, 8:45 am, 74yrsold >>>>>>>>>>>> > > <[email protected]<http://mc/[email protected]>> >>>>>>>>>>>> wrote: >>>>>>>>>>>> > > > Quan, >>>>>>>>>>>> > > > Today I downloaded vietocr 0.9.12 tried with English. It >>>>>>>>>>>> works well. >>>>>>>>>>>> > > > Then I added Kannada >>>>>>>>>>>> > > > tessdata to the folder folder "tessdata" of Vietocr when >>>>>>>>>>>> tried to ocr >>>>>>>>>>>> > > > - error message displayed as "errors acessing files ". >>>>>>>>>>>> under Format I >>>>>>>>>>>> > > > tried to select Kannada font but does not appear. Even I >>>>>>>>>>>> tried with >>>>>>>>>>>> > > > English fonts to select also failed. I am surprised to see >>>>>>>>>>>> appearance >>>>>>>>>>>> > > > are similar to SoftiFreeOCR !! >>>>>>>>>>>> > > > As a sample zip file is uploaded in this forum for >>>>>>>>>>>> testing at your >>>>>>>>>>>> > > > end. >>>>>>>>>>>> > > > With regards,. >>>>>>>>>>>> > > > -sriranga(76yrsold) >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> ------------------------------ >>>>>>>>>>> Cricket on your mind? Visit the ultimate cricket website. Enter >>>>>>>>>>> now!<http://in.rd.yahoo.com/tagline_cricket_1/*http://cricket.yahoo.com> >>>>>>>>>>> >>>>>>>>>> >>>>>>>>>> >>>>>>>>> >>>>>>>> >>>>>>> >>>>>> >>>>> >>>> >>> >> > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en -~----------~----~----~----~------~----~------~--~---
