How to Remove Speckles and Non-Text Items in Acrobat OCR
Question details
The user needs to remove image artifacts such as speckles and punch holes during OCR processing.
- Product
- Adobe Acrobat Pro DC
- Device & OS
- not provided
- Scenario
- Correcting recognized text after performing OCR on a scanned PDF document.
- Observed behavior
- The previous "Not Text" option in the Correct Recognized Text tool is no longer available in Acrobat Pro DC for marking image artifacts.
Ensure you have a backup copy of your original scanned PDF document before you begin manually deleting objects and modifying the OCR results.
Use the Edit Tool to Manually Delete Artifacts
Since the "Not Text" option is removed, you can use the standard Edit tool to select and delete speckles and punch holes before running the text correction.
Acrobat Pro DC updated its OCR workflow, removing the dedicated "Not Text" button. Instead, users are expected to treat the document as an editable file, manually deleting unwanted image blocks that the OCR engine picked up.
Open your scanned document in Acrobat Pro DC and click on the "Edit PDF" tool in the right-hand pane to initiate the OCR rescan.
Once the page rescan is complete, locate and select the "Edit" option situated right beside the "Add Text" button in the top toolbar.
Click directly on the unwanted image objects, such as speckles, punch hole marks, or stray lines, to highlight their bounding boxes.
Press the "Backspace" or "Delete" key on your keyboard to permanently remove the selected non-text artifacts.
After clearing the page of image artifacts, proceed to use the "Correct Recognized Text" tool to fix any remaining typos in your document.
Use WPS PDF to Easily Clean Up Scanned Documents
WPS Office provides a highly capable PDF editor with built-in OCR tools. You can quickly recognize text and manually delete unwanted image artifacts, speckles, or scan marks using a straightforward, intuitive interface.
- 1. Open your PDF: Launch WPS Office and open your scanned PDF document using the WPS PDF tool.
- 2. Run OCR: Navigate to the "Tools" tab and click "OCR" to convert the scanned image into an editable document.
- 3. Enter Edit Mode: Switch to "Edit" mode from the top toolbar to make all elements on the page selectable.
- 4. Delete Artifacts: Click on any unwanted speckles, punch holes, or stray marks and press the "Delete" key on your keyboard.
- 5. Save the Clean Document: Click "Save" to finalize your changes and keep a clean, searchable PDF file.
Frequently Asked Questions
Why is the "Not Text" option missing in Acrobat Pro DC?
Adobe modernized the OCR correction interface in Acrobat Pro DC. The older "Not Text" workflow was replaced by direct editing capabilities, meaning you now manually delete artifacts using the Edit PDF tool.
Can I prevent punch holes from showing up during scanning?
Yes, you can place a solid white sheet of paper behind the document when scanning. Adjusting your scanner's brightness and threshold settings can also help eliminate punch holes and speckles before the PDF is created.
Will deleting speckles mess up the recognized text?
No. As long as you carefully click on the bounding boxes of the speckles or image artifacts, the separate text blocks will remain intact and fully editable.
