Scanned handouts
Reduce a dull background
Try background cleanup while checking that lightly printed words remain.
A cleaner starting point
Enhance Image.
Clean up scanned pages and document photos while keeping the details you need.
Advertisements
Review the preview before processing. Your original image stays unchanged.
100%
Your image is ready.
Download the processed image, return to adjust it, or continue to Image OCR.
Advertisements
Images are uploaded to prepare the preview and process your selected frame.
IMAGE
Advertisements
How to enhance a document image
- 01
Open the source
Upload one supported image up to 50 MB and choose the frame to prepare.
- 02
Review the cleanup
Adjust Background cleanup and Sharpness. Open cleanup editor for more control.
- 03
Enhance and download
Choose a format, click Enhance Image, then inspect and download the new image.
Prepare a cleaner document image before OCR
Enhance Image combines background cleanup and sharpening with a shared editor for brightness, contrast, color mode and local cleanup. It is designed for existing scan detail, not for inventing missing words or enlarging a low-resolution image.
QUICK VIDEO TUTORIAL
How to clean up a document image before OCR
A cleaner image should still tell the same story.
English narration
Watch the video
Document cleanup in practice
Photographed pages
Balance an uneven source
Combine tonal adjustments carefully rather than forcing every pixel to white or black.
OCR preparation
Save a readable reference
Create a separate source for recognition and keep the original for comparison.
Features and controls
Tune background cleanup and sharpness
The initial settings apply modest cleanup and sharpening. Adjust either control from 0 to 100 while reviewing the preview.
Starting values and useful detail
Background cleanup starts at 20 and Sharpness at 15. These are starting values, not an assessment that your source needs them. Reduce either to zero when the change is unnecessary.
Sharpening can make edges more distinct, but it cannot recover missing characters or undo serious blur. Excessive sharpening can add outlines and emphasize noise.
Use presets and additional adjustments
Open cleanup editor provides Clean scan and Faint text presets plus brightness, contrast and color controls.
Color, grayscale and black-and-white decisions
Color keeps useful color distinctions. Grayscale removes hue while retaining tonal levels. Black and white uses a threshold to decide which pixels become dark or light.
Inspect colored notes, light stamps and thin type before accepting a preset. Apply changes returns the chosen cleanup settings to the main preview; Cancel leaves the current main settings unchanged.
Clear only unwanted marks
The cleanup editor includes an eraser and rectangle clearing for selected image areas, with Undo and Redo.
Manual cleanup and recovery
Use local clearing on obvious defects or unwanted background areas, not on text whose meaning you have not checked. These controls change the image pixels; they do not decide whether a mark is important.
Compare with the original and use Undo, Redo or Reset while working. Download the prepared result separately, so the untouched source remains available if you need to revisit a decision.
Choose one frame or page
Open a still image or choose a frame from a supported animated or multipage source. Only that selected frame is exported.
Supported files and frame scope
PNG, JPG, JPEG, TIF, TIFF, BMP, GIF, WebP and HEIC are supported inputs. Open one file up to 50 MB. Files may contain up to 100 frames or pages; image dimensions and decoding limits also apply.
Frame navigation changes the image being edited. Saving a result does not preserve a whole animation, its timing or all pages of a multipage file. Choose each required frame separately.
Choose PNG, JPEG or lossless WebP
PNG is the default. PNG and WebP can keep transparency; JPEG uses a white background for transparent pixels.
Output quality and source preservation
PNG and lossless WebP avoid lossy encoding. JPEG is saved at its highest quality setting, but it is still a lossy format; it is not a lossless copy of the source.
The output is a new image. The original file stays unchanged, but the export does not preserve all original metadata or every source-format feature. Use a separate source copy when those details matter.
Supported files & processing
- Input
- PNG, JPG, JPEG, TIF, TIFF, BMP, GIF, WEBP, HEIC
- Output
- PNG, JPEG, WebP
Open one supported image up to 50 MB, with no more than 100 frames or pages. Pixel dimensions and decoding limits also apply. Each run exports only the selected frame as a still image.
Your image uploads to our servers to prepare its preview and process the selected frame. These tools do not run OCR. Download the result you need and keep the original separately; see the Privacy Policy for file handling details.
Advertisements
FAQ
Before you start
Formats, selected frames and output quality explained.
Server processingDoes Enhance Image reconstruct missing or blurred words?
No. It adjusts existing image pixels through cleanup, sharpening and tonal controls. It does not generate replacement text, add real resolution or guarantee a correct OCR result.
Do I have to keep the initial cleanup settings?
No. Background cleanup starts at 20 and Sharpness at 15. Adjust either value, reduce it to zero or open the cleanup editor to choose another treatment.
Should I choose black and white for every scan?
No. Black and white may suit strong printed text, but it can remove pale writing or color distinctions. Compare Color, Grayscale and Black and white before deciding.
Which files and sizes can I use?
Open one PNG, JPG, JPEG, TIF, TIFF, BMP, GIF, WebP or HEIC image up to 50 MB. Files may contain up to 100 frames or pages, with additional pixel and decoding limits. Only the selected frame is saved as a still image.
Which download format should I choose?
PNG is the default and uses lossless encoding. Lossless WebP also preserves image detail and transparency. JPEG is saved at its highest quality setting but remains lossy, with white replacing transparent areas. The export is a new image, not an unchanged copy of the original file and its metadata.
Does this run OCR or change my original?
No OCR runs in this image tool. The file uploads for preview preparation and image processing, and the result is a separate image. You can continue to Image OCR when you need text; that tool has its own limits and verification.
