Analyze photo assets to generate graphical RGB color distribution histograms and contrast details.
An image histogram is a graphical representation of the tonal distribution in a photograph — showing how many pixels exist at each brightness level from pure black (0) to pure white (255). Photographers, image editors, and computer vision engineers use histograms as an objective diagnostic tool to assess exposure accuracy, identify clipping, and evaluate color balance without relying on potentially uncalibrated display monitors.
This browser-based histogram analyzer processes uploaded images using the Canvas API to compute per-pixel luminance and RGB channel distributions, rendering separate histograms for each color channel (red, green, blue) and overall luminance. No image data leaves your device — all computation happens locally using browser JavaScript.
Histogram shape reveals exposure quality. A well-exposed photograph typically shows a mountain-shaped distribution across the full tonal range (0-255), with most pixels in the mid-tones and gentle falloffs at both ends. Clipping occurs when values pile up at the extreme ends: a spike at 255 (right edge) indicates blown highlights — overexposed areas where detail is permanently lost. A spike at 0 (left edge) indicates blocked shadows — crushed blacks with lost detail.
Low-key images (intentionally dark: silhouettes, noir portraits) show histograms shifted left toward shadows; high-key images (intentionally bright: snow scenes, overexposed fashion) shift right. Neither is inherently wrong — histogram shape must be interpreted relative to the intended look. The histogram is a tool for detecting unwanted clipping, not for enforcing a particular aesthetic standard.
Separate RGB channel histograms diagnose color casts and white balance problems. Correctly white-balanced images show similar distribution shapes and peaks for all three channels. When one channel is systematically shifted higher than others, the image has a color cast: red-dominated histogram suggests warm/orange cast; blue-dominated suggests cool/tungsten or underwater color shift; green-dominated suggests fluorescent lighting or foliage-heavy scenes.
Channel clipping analysis is more nuanced than luminance clipping. A highlights region might clip the red channel while the blue channel retains detail — appearing as orange/red color shift in the blown highlights rather than pure white. Color-space aware HDR tone mapping must prevent channel clipping independently in each color component to preserve natural color rendering in extreme contrast situations.
A well-exposed histogram for most photography shows a smooth, roughly bell-shaped distribution of pixels across the full tonal range (0-255), with no steep spikes at either extreme. The exact ideal shape depends on the scene: high-contrast scenes produce bimodal histograms; foggy scenes produce narrow mid-tone histograms. The key warning signs are spikes touching the left (0) or right (255) edges, indicating clipping.
Clipping occurs when pixel values exceed the representable range (0-255), losing tonal detail. Right-edge clipping (255) means overexposed highlights appear pure white with no gradation — cloud texture, specular highlights, and bright fabrics become featureless. Left-edge clipping (0) means blocked shadows — dark areas lose shadow detail. Clipped areas cannot be recovered in post-processing because the original data is gone.
Each color channel records the response of one type of photoreceptor. Real scenes have unequal distributions of colored light — a blue sky scene will have a high blue channel, a red sunset will dominate the red channel. Channel imbalance creates color casts. Correct white balance aligns the channel distributions to reflect actual scene luminance ratios regardless of the illuminant color.
Standard 8-bit images represent 256 tonal levels per channel. HDR (High Dynamic Range) images use 16-bit or 32-bit floating point per channel, representing far greater dynamic range. HDR histograms show tonal distribution across thousands of discrete levels or continuous floating-point values. This tool currently analyzes standard 8-bit images; HDR analysis requires specialized tools that handle extended range values.
Adjust exposure until the histogram fills the tonal range without clipping. Use curves adjustments to reshape the histogram — pulling the midtone anchor up brightens mid-tones without affecting white point; pulling highlights down recovers gradation from near-clipped areas. Compare histograms before and after adjustments to verify corrections are data-driven rather than monitor-dependent.