Histogram

What it does

Counts every pixel in the photo into 256 brightness levels — separately for the red, green and blue channels and for luminance (perceived brightness) — and draws them Lightroom-style: the three colour channels stacked with additive blending, so red over green reads yellow and all three read grey. Triangles in the top corners light up when shadows or highlights clip, coloured by the channel that's clipping. Hover the photo to read a pixel's RGB and mark it on the chart; hover the chart (or drag the slider) to light up every pixel in that brightness range on the photo. It shows the histogram; it doesn't adjust the photo.

What you'll need

• A photo (JPEG, PNG, WebP, etc.) from your device — the original export, not a screenshot of it.

Steps

1. Tap "Upload a photo" and choose an image.
2. Read the stacked RGB histogram — a channel reaching an edge before the others is colour clipping — and switch to Luminance or a single channel when you want one on its own.
3. Tap a lit triangle to paint the clipped pixels on the photo (blue for shadows, red for highlights); hover or drag across the chart to paint any other brightness range.
4. Hover or drag on the photo to read any pixel's R, G, B and luminance and see where it sits on the chart, then check the clipping table and the plain-English read.

Reading the result

Height is how many pixels sit at that brightness. A hump against the right edge with the right triangle lit means blown highlights; against the left edge, blocked shadows — tap the triangle to see exactly which pixels. A narrow hump in the middle is a flat, low-contrast image. In the RGB view, one channel reaching the edge before the others is colour clipping — a saturated sunset or red flower can blow its red channel while the luminance histogram still looks fine.

Tips

• The camera's default histogram is luminance-only — the stacked RGB view here catches single-channel clipping the camera didn't warn you about: a red hump on its own at the right edge is a blown red channel.
• The chart, the triangles and the overlays on the photo all use the same clipping windows (levels 0-4 and 250-255), so a lit triangle, a red band on the chart and red pixels on the photo are one fact shown three ways.
• There is no “correct” shape. A high-key portrait lives on the right, a night scene on the left; the histogram describes the light, it doesn't grade it.
• For a quick under/over-exposed verdict rather than the full chart, the Exposure Analyzer uses the same luminance channel and clipping windows.

What is a histogram?

A histogram is a bar chart of every pixel in a photo sorted by brightness: pure black at the left edge, pure white at the right, and bar height showing how many pixels sit at each of the 256 levels in between. A luminance histogram weights the three colour channels the way the eye does (green counts most, blue least) into one brightness value per pixel; an RGB histogram draws each channel separately, so you can see when a single colour reaches the edge before the others.

Clipping means pixels have piled up at the very edge — level 0 or 255 — because the scene was brighter or darker than the sensor could record at that exposure. Those pixels carry no detail: a blown sky is a flat white shape, a blocked shadow a flat black one, and no amount of editing can bring back what was never captured.

Why photographers use it

A screen is a poor exposure meter — a phone in sunlight, a camera LCD at night, a laptop with the brightness turned down — but the histogram is the same whatever you view it on. Photographers read it in the field to protect the highlights that matter (expose so the right edge is just short of clipping), to spot a flat file that needs contrast, and to compare two exposures of the same scene objectively. Reading the RGB version as well catches the colour clipping that ruins skin tones and sunsets while the luminance view sits innocently in the middle.

Luminance or RGB — which should I read?

Start with luminance for overall exposure: it's what the camera's default histogram shows and it answers “is this too dark or too bright?”. Switch to RGB whenever the scene has strong colour — sunsets, stage lighting, flowers, warm skin under tungsten — because a single channel can clip long before luminance does, and that shows up as smeared, detail-free colour. The individual red, green and blue views are for tracking down which channel is responsible.

Common mistakes

• Chasing a bell curve in the middle. a “perfect” centred hump is only right for an evenly lit mid-tone scene. A snow field belongs on the right, a night street on the left, a backlit silhouette at both ends. Judge the histogram against what the scene actually looked like, not against a textbook shape.
• Reading only the luminance channel. luminance is a weighted average, so a red channel at 255 can hide behind green and blue at 180. The result is a sunset or a red dress rendered as a flat, waxy blob while the luminance histogram reports no clipping at all.
• Treating every pixel at the edge as a mistake. specular highlights on chrome, the sun itself, a light bulb in frame, a genuinely black background — these are supposed to clip. The percentage tells you how much; the photo tells you whether it matters.
• Checking a screenshot or a re-shared copy. every re-encode — a screenshot, a messaging app, a social download — changes pixel values, often crushing shadows and compressing highlights. Read the histogram of the file you actually exported.
• Fixing a flat histogram by dragging the ends in editing without looking. a narrow hump means low contrast, but stretching it blindly to both edges creates the clipping you were trying to avoid. Add contrast until the ends approach the edges, then stop.

Example settings

FAQ

Is my photo uploaded anywhere?

No — the pixels are read from the file entirely in your browser. Nothing is sent to a server, and nothing is kept once you leave the page.

Why does the Luminance view look fine while the red channel says clipped?

Luminance is a weighted mix of the three channels, so one channel at maximum can be averaged down by the other two. The RGB view exists for exactly this — it's the most common way a saturated sunset or skin tone loses detail without an obvious warning.

How do I see which parts of the photo are blown out?

Tap the right-hand triangle above the histogram — every pixel in the highlight clipping window (levels 250-255) is painted red on the photo. The left triangle does the same for blocked shadows in blue. Hovering or dragging across any part of the chart paints that brightness range in magenta, so you can trace a hump on the histogram back to the part of the picture it comes from.

What does hovering the photo show?

The red, green, blue and luminance values of the pixel under the pointer (0-255 and as a percentage), with a vertical marker on the histogram at its luminance and a dot on each colour channel at that channel's value. On a phone, tap or drag on the photo instead — the reading stays put when you lift your finger.

What counts as clipping here?

Pixels within five levels of pure black (0-4) or pure white (251-255), reported as a percentage of the whole image — the same windows the Exposure Analyzer uses. Over 5% is called clipped, 0.5-5% is touching the edge, and under 0.5% is noise-level.

Why is the histogram slightly different from my camera's or Lightroom's?

Every app bins pixels a little differently — the camera reads its JPEG preview, Lightroom reads the RAW after its own profile, and this tool reads the exported file at up to 1000px on the long edge. The shape and the clipping story will agree; the exact bar heights won't.

Can this tool fix the clipping or adjust the photo?

No — it only shows the histogram and describes it. There are no curves or sliders. Use it to decide what to change in your editor, or whether to reshoot with a different exposure.