How use Adobe Photoshop for Color Analysis

The following guide will cover how to do Quantitative Color Analysis either using Adobe Photoshop alone or with an external tool for deeper analysis.

Step 1: Open Your Image in Photoshop

Open your target image (e.g. artwork, design piece, or photo).
Ideally, start with a .jpg or .png file.

Step 2: Pixelate the Image for Simplification

This step reduces visual complexity by grouping similar pixels together. Think of it like turning your image into digital LEGO blocks. This step makes color regions more uniform and easier to analyze.

Select your image in the layers (bottom right) then at the top, go to Filter > Pixelate > Mosaic
Set the Cell Size to around 20 pixels
(You can experiment with 10–50 depending on how much detail you want to keep.) The more detail you retain, the more colors and smaller percentages you will have for analysis.

Step 3: Convert the Image to Indexed Color Mode

This step limits the number of colors in the image to a fixed, countable set. It also helps isolate and quantify individual color usage.

At the top go to: Image > Mode > Indexed Color

Now you'll see a pop-up with several options. Here’s what they mean:

Option	Suggested Setting	Explanation
Palette	`Local (Perceptual)` or `Local (Selective)`	This tells Photoshop how to pick colors. Either is fine for analysis. Perceptual prioritizes smooth gradients and human vision. Selective prioritizes distinct, saturated colors.
Colors	`32`	Limits image to 32 distinct colors (you can choose fewer like 16 or so, but 32 is good for clarity and control).
Transparency	`Unchecked`	When checked, Photoshop keeps any transparent areas in your image as transparent. When unchecked, transparent areas are filled with a solid color (usually white or black).
Forced	`None`	Prevents Photoshop from adding unnecessary “reserved” colors like pure black or white unless they're already in the image.
Dither	`None`	Disabling dithering keeps colors solid instead of mixing pixels to fake gradients, which would mess with analysis.

Once you hit OK, the image will look a bit posterized (using only a small number of different color tones), this is what we want. Nice job!

If desired, at this point, you can see the exact colors used.

Image > Mode > Color Table
Optional: Screenshot it or export colors manually by saving the table for reference.

(Option A): Quantify colors using Photoshop’s built-in tools

This step outlines how to quantify colors by using built-in tools like the Magic Wand and Histogram, and analyze the results. You’ll identify how many pixels each color uses, calculate what percentage of the image they take up, and organize that data for analysis.

Sub step 1:

First, we need to determine the total pixel count of the image. We can go back to the original file, or go to: Image > Image Size

2048 × 1365 = 2,796,720 total pixels

Write this number down as you’ll use it for percentage calculations later.

Sub step 2: Use the Magic Wand Tool to Select a Color

This lets you select all pixels of a specific color across the entire image.

A. Select the Magic Wand Tool

Shortcut: W
Or find it in the toolbar (may be under the Quick Selection tool)

B. Set Tool Options at the Top:

Tolerance: 0 → ensures only the exact color is selected
Contiguous: Unchecked → selects all instances of that color in the entire image, not just touching areas
Sample All Layers: Optional, turn ON only if working on merged image layers

C. Click on Any Color Block

Click one of the squares in your mosaic-style image and see the magic happen! 🪄

Sub step 3: Find Out How Many Pixels Were Selected

Use the Histogram

Go to: Window > Histogram
Make sure the histogram is set to "Entire Image" be sure click the refresh button to get an accurate count of the selection. (This issue is due to a photoshop quirk showing cached data)

You’ll see a pixel count for the selected color (ex: 151160 pixels)

Sub step 4: Calculate Color Percentage

Use this formula: Percentage = (Color Pixel Count / Total Pixels) × 100

(151,160 / 2,796,720) × 100 ≈ 5.4%

You’ll need to repeat at this for the other colors. I would advise creating a like below in a spreadsheet which you can use to visualize the data.

Color (Hex / RGB)	Pixels	% of Image
#e2e5e9	151,160	5.4%
...	...	...

You can use the eye dropper tool and then click the foreground square to see the selected color and it’s RGB / Hex value.

(Option B): Online Color Analyzer

This method mentions how to take your modified image and running it through a third party online tool to quickly analyze

Sub step 1:

Towards the end Step 3 above, once you have the posterized image in Photoshop, you will need to export it as a PNG to your computer.

Sub step 2: Go to the Color Summarizer Website

🔗 https://mk.bcgsc.ca/color-summarizer/

At the top, click the second tab ‘ANALYZE’

You’ll see a form labeled “Submit Image for Processing.”

Sub step 3: Upload Your Image & Set Options

A. Choose File

Select your .png file that was exported from Photoshop

B. Output Format

Select: html (easiest to view and copy)

D. Statistics

✅ Check: color clusters
✅ Check: histogram
Skip pixel unless you want raw pixel-level detail

E. Number of Color Clusters

(Match the number you set in Photoshop during Indexed Color conversion)

In this case where we choose 32 earlier in Photoshop, you’ll select: 20 (which is the maximum allowed on this website)

🧠 Why not 32?

In Photoshop, you may have reduced the image to 32 colors using Indexed Color Mode. It helps ensure your image is simplified and color-limited before analysis.

However, the Color Summarizer tool can only analyze up to 20 clusters (groups of similar colors), even if your image has more.

What happens behind the scenes?

The tool uses k-means clustering, which means it finds the 20 most dominant color regions in the image.
These clusters often represent 90–95% of all visible colors.
Less frequent colors (small details or near-duplicates) may be merged or ignored.

This is totally fine for most analysis, especially when your image is already reduced to 32 base colors. You’ll still get a reliable representation of the dominant colors and their proportions.

F. Delimiter

Leave on space

G. Precision

Choose: high (150 px) or vhigh (200 px)
(Higher = better accuracy, slower load)

H. Process Image

Wait a few seconds (can take up to 15–30s). The site will then generate a full color summary page.

Sub step 4: Understanding the Output

You’ll see a page including table like these. You can copy the data and paste it into a spreadsheet for further analysis. NOTE: Since we converted 32 colors into 20 earlier, some of the RGB/HEX values may have changed.

The table below in includes several points of data including:

RGB (Red, Green, Blue)

HSV (Hue, Saturation, Value)

LCH (Lightness, Chroma, Hue)

LAB (Lightness, Green–Red Axis, Blue–Yellow Axis)

📏 Color Model Range Reference Guide

🔴 RGB (Red, Green, Blue)

Each channel ranges from 0 to 255. Use this to assess how much red, green, or blue is in the image.

Value Range	Meaning
0–50	Very low — very dark or almost absent (e.g., black, shadows)
51–127	Moderate — some color, but not bright
128–200	Strong — clearly present, brighter areas
201–255	Very strong — very bright or vibrant

🌈 HSV (Hue, Saturation, Value)

H (Hue): 0–360 (represents color type)
S (Saturation): 0–100 (intensity of color)
V (Value): 0–100 (brightness)

Channel	Value Range	Meaning
H	0–360	Color wheel:
🔴 0 = Red
🟢 120 = Green
🔵 240 = Blue
🟠 60 = Yellow
🟣 300 = Magenta
S	0–30	Low saturation (grayish, dull)
	31–70	Moderate color intensity
	71–100	Vivid, punchy colors
V	0–30	Dark
	31–70	Normal brightness
	71–100	Very bright, close to white

🎨 LCH (Lightness, Chroma, Hue)

L: 0–100 (lightness)
C: 0–100+ (chroma = intensity)
H: 0–360 (hue angle)

Channel	Value Range	Meaning
L	0–30	Very dark
	31–70	Mid-tone lightness
	71–100	Very light
C	0–25	Dull, grayish
	26–50	Moderate vibrancy
	51+	Very saturated/vivid (rare in photos unless enhanced)
H	0–360	Same as HSV hue angle — tells you dominant color direction

🧪 LAB (Lightness, A, B)

L: 0–100 (same as LCH Lightness)
A: -128 (green) to +127 (red)
B: -128 (blue) to +127 (yellow)

Channel	Value Range	Meaning
L	0–30	Dark
	31–70	Normal lightness
	71–100	Bright
A	-128 to -20	Strong green tones
	-19 to +19	Neutral (no red or green bias)
	+20 to +127	Strong red tones
B	-128 to -20	Strong blue tones
	-19 to +19	Neutral (no blue or yellow bias)
	+20 to +127	Strong yellow tones

Think of each model like this:

RGB = how much of each basic light color is used
HSV = how the color looks to the eye (type, intensity, brightness)
LCH = more accurate perception of tone and vividness
LAB = mathematical color space good for comparing differences