Image histogram — check exposure and clipped highlights

How to use Image Histogram

1. Choose an image.
2. Under Show, pick Brightness or Red, green, blue.
3. Press Analyse.
4. Read the chart and the figures: Mean brightness, Median, Darkest / brightest, Clipped shadows (pure black) and Clipped highlights (pure white). A message underneath says whether clipping is significant.

What it does and when to use it

A histogram is how photographers judge exposure without trusting a screen that may be too bright or too dim. Use it to check whether a photo is underexposed (everything bunched left), overexposed (bunched right, with clipped highlights), or flat (a narrow hill in the middle). It is also a good before-and-after check: run it, adjust the photo with Image Brightness or Image Gamma, then run it again on the result to make sure you did not push the highlights to pure white.

For questions about the file itself, such as pixel size and format, use Image Info. For the main colours rather than tones, use Dominant Colors.

How it works

The picture is decoded in a Web Worker on your device. Pictures larger than 2,000 px on the long side are scaled down first to keep it quick. For every pixel that is not fully transparent, the tool adds one to the counter for its red, green and blue values and for its brightness, computed with the Rec. 709 weights (0.2126 R + 0.7152 G + 0.0722 B). The chart groups the 256 levels into 64 bars of four levels each and shows each bar as a percentage of all counted pixels. Mean, median, darkest and brightest come from the brightness counts.

Worked examples

We built a 100 × 100 px test image and ran the tool’s own functions: 8,500 pixels of mid-grey (128), 1,000 pure black and 500 pure white.

Both ends are over 2%, and the message warns that more than 2% of pixels are pure white, so highlight detail is lost (the tool reports highlight clipping first).

A typical daylight landscape. If the chart shows a hill that touches neither edge and both clipping figures are near 0%, the message reads “No significant clipping at either end”.

Limits and tips

• Scaling a very large photo blends neighbouring pixels, so a few isolated clipped pixels may not be counted. Large clipped areas, such as a white sky, still show clearly.
• The chart reflects the file’s values, not how a particular screen or printer shows them.
• A histogram cannot tell you whether the photo is in focus.