1. Forum
  2. Usenet
  3. COMP.OS.LINUX.ADVOCACY

  • Tesseract Is Junk. FOSS Has No OCR.

    From Farley Flud@21:1/5 to All on Tue Aug 26 19:01:46 2025
    XPost: comp.os.linux.misc

    Optical Character Recognition (OCR) is perhaps a critical necessity
    in today's digital world.

    Google has contributed an OCR software called Tesseract to
    to the FOSS community.

    I am sorry to have to say that, in my experience, Tesseract is junk.
    It can be very inaccurate and its accuracy is quite inconsistent.

    I have began several OCR software projects only to have to abandon
    each of them due to the poor performance of Tesseract.

    If you disagree, then please report, with full examples.

    But I doubt that anyone who seriously applies Tesseract will be
    satisfied with its results.


    --
    Gentoo: the only road to GNU/Linux perfection.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From vallor@21:1/5 to Farley Flud on Wed Aug 27 04:20:19 2025
    On Tue, 26 Aug 2025 19:01:46 +0000, Farley Flud <ff@linux.rocks> wrote in <pan$a6c73$f3d41061$d19e7b60$2f573631@linux.rocks>:

    Optical Character Recognition (OCR) is perhaps a critical necessity in today's digital world.

    Google has contributed an OCR software called Tesseract to to the FOSS community.

    I am sorry to have to say that, in my experience, Tesseract is junk.

    You know tesseract has been in development since 1985, right?

    You know it started at Hewlitt Packard, right?

    You know that it has _two_ engines, which can be combined for
    better results, right?

    I didn't think so.

    (FTR: I've had good success with tesseract.)

    --
    -v System76 Thelio Mega v1.1 x86_64 NVIDIA RTX 3090Ti 24G
    OS: Linux 6.16.3 D: Mint 22.1 DE: Xfce 4.18
    NVIDIA: 580.76.05 Mem: 258G
    "A cat is always on the wrong side of a door."

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)
  • From Farley Flud@21:1/5 to vallor on Wed Aug 27 12:29:29 2025
    XPost: comp.os.linux.misc

    On 27 Aug 2025 04:20:19 GMT, vallor wrote:


    (FTR: I've had good success with tesseract.)


    Pics or it didn't happen.

    Check out this image of the Gettysburg Address (or a part
    of it). It is Liberation Serif 12-point font:

    https://i.postimg.cc/rmTZhHYg/gburg.png

    Now what follows is the tesseract output using this command:

    tesseract --tessdata-dir /usr/share/tessdata gburg.png gburg --psm 6 --oem 1

    This uses the best engine with the best training data.

    Output:

    =================================================================================================================

    But, in a larger sense, we can not dedicate—we can not consecrate—we can not hallow—this ground.
    ‘The brave men, living and dead, who struggled here, have consecrated it, far above our poor power to
    add or detract. The world will little note, nor long remember what we say here, but it can never forget
    ‘what they did here. It is for us the living, rather, to be dedicated here to the unfinished work which they
    ‘who fought here have thus far so nobly advanced. It is rather for us to be here dedicated to the great
    task remaining before us—that from these honored dead we take increased devotion to that cause for
    ‘which they gave the last full measure of devotion—that we here highly resolve that these dead shall not
    have died in vain—that this nation, under God, shall have a new birth of freedom—and that.
    government of the people, by the people, for the people, shall not perish from the earth.

    =================================================================================================================

    Where in fuck did those "`" characters come from? It is actually
    a sequence: 0xE2, 0x80, 0x98, which is the Unicode "left single quotation mark."

    Why is there a period after "that" in the second-to-last line?

    I suppose that if I tried more samples I could get much worse results
    but this quickie demonstrates that tesseract is not really up to snuff.


    --
    Gentoo: the only road to GNU/Linux perfection.

    --- SoupGate-Win32 v1.05
    * Origin: fsxNet Usenet Gateway (21:1/5)