Histogram · read exposure and color distribution
Five channels, mean / std dev / median. Local processing, nothing uploaded.
Under- or overexposed, blown highlights or crushed blacks — a glance at the histogram beats squinting at the picture.
Open Window > Histogram. The horizontal axis is 256 levels from 0 to 255, dark on the left and bright on the right, and a tall bar means many pixels sit at that level. Five channels are available — Luminosity, R, G, B, and Color — where "Color" overlays the three curves so a cast shows up at once; and two sources, Entire Image and Active Layer, the latter being what you want while grading one layer. Four readouts sit underneath: Mean (overall brightness), Std Dev (contrast — the smaller, the flatter), Median, and Pixels. Hover a bar and it also shows that Level, its count, and its Percentile — the share of pixels darker than that point, which is exactly what you want when setting black and white points. One thing to be clear about: the statistics run on a downsampled preview. An image longer than 256 px on its long edge is scaled to 256 before counting, so Pixels reports the sampled count, not the document total; at or under 256 nothing is scaled and the readouts are exact. The shape of the distribution is unaffected, but do not read that pixel count as your document size. It all runs in your own browser; the images are never uploaded.
Key features
Five channels
Luminosity, R, G, B, and Color — the last overlays all three so a cast shows at once.
Whole image or one layer
Switch the source to Active Layer while you are grading a single layer.
Four readouts plus hover detail
Mean, Std Dev, Median, Pixels; hover a bar for level, count, and percentile.
Runs locally
The images are never uploaded. Close the tab and nothing is left.
How to use
- 1 Open the imageDrag it straight in.
- 2 Open the histogram panelWindow > Histogram.
- 3 Pick a channel and sourceLuminosity for exposure, Color for casts; switch the source when grading one layer.
- 4 Correct accordinglyBunched left is underexposed, bunched right is overexposed; open it up with levels or curves.
FAQ
- How do I read a histogram?
- The horizontal axis is levels 0 to 255, dark on the left, bright on the right; the vertical axis is how many pixels sit at that level. All bunched left means underexposed, all bunched right means overexposed, bunched into a narrow middle band means flat. A bar hitting the ceiling at either end usually means crushed blacks or blown highlights — that detail is gone, and pulling it back only yields gray.
- Why does the Pixels number not match my image?
- Because the statistics run on a downsampled preview: an image longer than 256 px on its long edge is scaled to 256 before counting, so that number is the sampled count, not the document total. At or under 256 nothing is scaled and the readouts are exact. The shape of the distribution is unaffected, so judging exposure from it is perfectly sound — just do not read that number as your document size.
- What do Mean and Std Dev tell me?
- Mean is overall brightness — around 128 is neutral, and clearly lower means the picture is dark overall. Std Dev is contrast: a small value means pixels are crowded into a narrow band of levels, which reads as flat and gray; a large value means the tones are spread out. Making a photo look "clear" usually amounts to pushing the std dev up, with levels or curves.
- Is it free? Are my images uploaded?
- Free, no sign-up, no usage limit. Nothing is uploaded either — it all runs in your own browser.
Try Histogram · read exposure and color distribution now
Runs fully in your browser · no install · no signup · images never uploaded