Image guides

Read an image histogram before comparing exposure

·

Transparency can change which pixels contribute to a histogram. Compare both settings on the same image before interpreting the shadows or highlights.

Image Histogram Viewer

Read the red distribution on the sample

Choose Load example and Analyze histogram. The status reports 7 counted pixels and 1 excluded. Choose Red in Chart channel and check that the peak is 0 and two pixels reach level 255.

Download CSV and check that the red count at intensity 0 is 3. The table records one row for each level from 0 through 255, including levels with no hits.

Include RGB where alpha is zero

Turn on Include fully transparent pixels and analyze again. The report now counts all 8 pixels. Download the CSV again and compare red intensity 18. Its count rises from zero to one.

The partially transparent second pixel on the bottom row already counted once. The toggle adds only the fully transparent pixel.

W3C PNG Third Edition, image layout and alpha

Read the linear-light luminance result

Choose Luminance. The input’s sRGB components are converted to linear values and weighted before binning. The white pixel reaches bin 255 while pure red reaches bin 54.

Use RGB when you need one component’s encoded level, and use luminance when you want the weighted distribution. Compare the counts before making an exposure edit.

W3C sRGB relative luminance definition

Common questions

What does Peak level show when bins tie?

The table uses the lowest intensity among the bins with the highest count.

Does the graph apply alpha as a fractional weight?

Each included pixel adds one count. The default excludes only pixels with alpha zero.

Sources

Put it into practice.

Related guides