First CHToolsFree web tools

FIRST CH TOOLS / Images / 101 REDACTION SEE-THROUGH CHECK

Is Your Black-Out See-Through? Check It and Re-Cover It

Drop one redacted image and see it next to a view with brightness and contrast pushed to the extreme. Drag a box over anything that shows through to cover it with opaque black, then save it as a PNG.

Sample: a fictional invoice. The bank details got one pass of a marker; the contact name is opaque black.

Original (as it looks)
Enhanced (extreme brightness and contrast)

Drag a box over anything that shows through to cover it with opaque black (or tap two corners). When the enhanced view turns flat black there, it is gone.

Visual check of the shading
Tick once the enhanced view shows no text or shading in the covered areas
Boxes covered again
0When you save, everything inside these boxes becomes opaque black
Data after the end of the image
Reading

Only what is inside the boxes you drew is reliably gone. Anything outside them stays as it is.

How to Use

  1. Add your redacted imageDrop a screenshot or document image (PNG or JPEG) you are about to send or post, or tap to choose it. A fictional invoice is loaded as a sample at first.
  2. Look for see-through areas and box themIn the enhanced view (right, or below on a phone), text or shading inside an area that should be black means it shows through. Drag a box over it and it becomes opaque black. Once the enhanced view is flat black there, tick “Visual check of the shading”. Raise “Strength” if it is hard to see.
  3. Press “Save as it is now”You get a PNG rewritten from the pixels. Data that followed the end of the original image and metadata (camera details and so on) are not carried over. The saved file is read back to confirm every box is opaque black.

About This Tool

Blacked out, yet readable once brightened

Some pens and markers in phone photo and screenshot editors let you change the opacity of the colour. If the black or marker is not 100% opaque, the area looks black but the text underneath stays in the pixels as a faint difference in shading. Anyone who raises brightness or contrast can widen that difference and read it. The enhanced view here stretches the dark end of the range by 2× to 64× depending on Strength, and turns everything brighter into white. Pure black (0) stays 0 however much it is multiplied, so properly covered areas look flat black and see-through ones show text or brush strokes.

Covering overwrites the pixels in the box with opaque black (RGB 0,0,0). The save is a PNG rewritten from the canvas pixels, so data appended after the original image, metadata such as EXIF and ICC profiles are not included. The enhanced view is only for checking; what you save is the original image with black boxes added.

What “data after the end of the image” means

For PNG the chunks are read in order up to the end marker (IEND); for JPEG the markers are read in order up to the end of the main image (EOI). Whatever bytes are left after that are counted. The end marker of a small thumbnail embedded in a JPEG’s EXIF is not treated as the end. Files whose structure cannot be read — cut off, with a CRC that does not match and so on — are shown as “Could not analyze”. Trailing data can contain part of the image before it was edited (bugs where cropped screenshots kept part of the original: CVE-2023-21036, CVE-2023-28303), or another image or video added by a camera. This page does not read it back. The formats follow the PNG specification (W3C) and the JPEG standard (ITU-T T.81).

What this tool does not do

No text recognition, no automatic detection or covering of see-through areas, no blur or pixelation, no PDF redaction, no recovery of images from trailing data and no export of the enhanced view. You make the final call by looking at the enhanced view. The image is handled only inside this browser and is never uploaded or stored. AI agents can check for data after the end of an image with the MCP tool kuronuri_tail (for AI agents).