312 results found
-
Post Scan review and correction
During the correction process utilize a dictionary to look up words which contain most characters as determined by scan and in the same left to right order. Present these as choices to be picked rather than only allowing change via direct editing. Permit editing after a choice is made however because the choice might only be close.
So often I have seen "print" scanned as "primt" for instance. Since most text is words in a dictionary the correction process can offer alternative spellings with some high degree of probability that one of them is correct. This could also be used…
3 votes -
OCR automation (cloud API or SDK)
As a preprocessing step for our document analysis technology, we need to OCR hundreds of document scans. Since we need not only the recognized text, but also font information and text position, your OCR technology as integrated into the Acrobat DC provides the best quality for our needs, but unfortunately, there seems to be absolutely no possibility to automate the OCR process? Your mobile app seems to access some cloud api, since there you can make a picture with your phone and get the OCRed PDF within seconds from the Adobe servers, but the API is not public (at least…
3 votes -
OCR Japanese (maybe other language, too)
This is a bug report.
In macOS,
'language' of application and OCR language have to be the same.
Otherwise it won't work.In Windows,
this bug doesn't exist.Details are discussed as following.
https://forums.adobe.com/thread/2353148
https://forums.adobe.com/thread/23531453 votes -
Recognition of Indian Language OCR
The idea is not new, but the gravity is certainly very heavy.
Acrobat being so popular, wonder how its OCR is so inferior that it cannot recognise any of the Indian Languages?
It's even more intriguing when many free softwares like tesseract and professional software like Google Vision can do with ease. These software do their job for years now.
Wonder what's holding Acrobat and Adobe back in catching up with them.
2 votes -
Bug report: Toolbar emtpy
If I select any tool in the most recent version, the sidebar turns black/white but is otherwise empty. OCR and editing of a document are therefore impossible.
2 votes -
ocr multi
While doing OCR in one document, i would like to read another open pdf. Instead the program is fully locked in OCR for that pdf, and can't do additional task's. Some kind of multithreads function should been enabled/created for this, thanks.
2 votes -
Impove OCR Accuracy
Please improve the accuracy of OCR. The screen capture is from "review recognized text."
2 votes -
Uninterruptable error dialogs on OCR on a set of files.
If you select and drag/drop a large number of files into the dialog and, for some reason those files are unreadable (eg the drive is no longer mounted) you get uninterruptable "<filename> could not be added to the list because it does not exist." errors, one after another. You cannot interrupt this, so if there are 100+ files you were adding you get 100 error messages one after another that you have to manually click OK on to go past.
Other, more mature, applications notice this behavior and give an option to abort the entire process if there are multiple…
2 votes -
Changing OCR language results in immediate crash
I have documents in both French and English. Changing the OCR language from English US to French results in an immediate crash, repeatedly. Note that changing to English UK does not.
I removed by Prefs file, and all starts working again....for a couple of days. I first scan French! Then English! But when I go back to French, an immediate crash.
2 votes -
OCR search and copy on Japanese text
Just tried this out on a scanned PDF of Japanese text with the latest Acrobat Pro (2023.008.20470) and the results were completely unusable. Searching for any text on the page had zero results (even simple hiragana). Copying and pasting text gave me wildly inaccurate kanji and other symbols. macOS Preview does a pretty good job of letting me search and copy text from the same document, even my own poor handwriting, and I don't even have to wait for a progress bar.
2 votes -
We can't mark something in OCR as "not text" anymore in the new UX/UI
I'm trying Acrobat Pro and it seems that the new UI that we can try doesn't allow me to correct OCR by marking something as "not text". With the old UI, you can simply erase the garbage text next to the picture, and press "OK" or whatever, then this part of the document stops being considered as text.
2 votes -
FIND/SEARCH does not work
When i use CTRL+F (Find), the software does not find any match in this PDF. In Google Chrome the FIND in this PDF works very well. Why does Acrobat Reader 2023.01.20143 not work with this PDF?
2 votes -
Multicore support.
Please add multithreading. I am writing this as I am WAITING on my Mac M2 Ultra with 128 GB of RAM to complete an OCR, that is apparently a still single threaded application.
2 votes -
Bug: Adobe Acrobat Pro closes/crashes during OCR
Adobe closes immediately after clicking Recognize Text in the Scan & OCR tool or Optimize Scanned Pages in the Optimize PDF tool. This occurs on the Windows 10 desktop version. I've tried checking for updates (already up to date) and repairing the installation.
2 votes -
Bug: selection of text to be corrected produced erroneous selection.
1) Read in the attached file "bug.pdf"
2) Push "Scan & OCR"
3) Push "Correct recognized text."The resulting screen capture for me is shown in the attached file "a.jpg." To create this weird selection, I had toggled on "Review recognized text" then double clicked on "Guide-" which is in the upper right corner of the red box (as currently seen). Notice the text that was recognized was "Cnicl?b". I had hope to replace that text, but making the selection area so large seems nonsensicle.
I'm running on Windows.
2 votes -
Bug: OCR: removal of erroneous space within word doesn't work correctly
1) Read in the attached file "bug2.pdf"
2) Push "Scan & OCR"
3) Push "Correct recognized text". You should see the views shown in 1.jpg and 2.jpg.
4) In the "Recognized as" text area, notice there is a space after "pach." Delete the space and push "Accept."
5) Click the "OK" button of the message dialog window.
6) Toggle on "Review recognized text"
7) Double click on "pach" in the document, notice that despite the space having been removed, it still recognizes "pach" as a separate word from "ypus." See 3.jpgIt should combine them together into the single word…
2 votes -
Bug report: Scan & OCR>Correct recognized text: corrected text length no longer matches underlying text
1) Open the attached file in Adobe Pro DC.
2) Click on "Scan & OCR"
3) Click on "Correct recognized text"
4) Toggle on "Review recognized text"
5) In the "Recognized as" text area, change "th~" to "those"
6) Push Apply
7) Select Menu>Export a PDF>text (plain) and export a file and look at it.Notice after the correction, "those" slightly overlaps the word to the right of it and no longer matches the length of the underlying text of the image. Before the correction, there was a space between "th~" and "of", but after correction and export to the…
2 votes -
The Future of Furniture Shopping: Exploring the Furniture Store Phenomenon
The future of furniture shopping is likely to undergo significant changes as technology continues to advance and consumer preferences evolve. Here are some potential trends and developments that could shape the furniture store phenomenon in the future:
Online Shopping Dominance: Online furniture shopping has been growing steadily in recent years, and this trend is expected to continue. With the convenience of browsing and purchasing furniture from the comfort of one's home, online platforms will likely dominate the market. Virtual reality (VR) and augmented reality (AR) technologies will enhance the online shopping experience by allowing customers to visualize furniture in their…
2 votes -
Croatia
Croatia has been recognized by World for more than 2 decades and you guys still didn't implement your softwer to recognize all alphabets from multiple countries just like Croatians ć, č, đ, š and ž. OCR failed miserably not recognizing any of these letters. It's time to do smth about it guys...
2 votes -
scan dialog
Request: Improve dialog for scanning settings
Pain Point: When scanning documents, I sometimes need to select single-sided scanning and other times 2-sided scanning. It is cumbersome to have to click on the "gear icon" for document type, e.g., color and then also have to click on a very small option for 1 or 2-sided scanning.
Desired result: Build in sets of radio buttons to quickly establish document type, color mode, e.g., color, gray or black and white, and 1 sided or 2 sided scanning.
2 votes
- Don't see your idea?