Overview
Thresholding turns grayscale or measured pixel values into binary or labeled regions using one or more cutoff values.
FoundationSource checked
Thresholding
Separates pixels into groups based on whether they cross a chosen value.
Quick Facts
Compact metadata for implementation and discovery.
- Category
- Image Processing
- Difficulty
- Beginner
- Family
- Segmentation
- Tier
- Foundation
- Status
- Verified
- Review
- Source Checked
- Sources
- 1
- Formulas
- 1
- Relationships
- 3
- Code example
- Ready
- Visualization
- Ready
Core Concepts
intensity thresholdbinary outputpixel classificationsegmentation maskpixel operationregion separationsegmentationpixel neighborhood
Quick facts
Complexity
Runtime and memory behavior at a glance.
Best
O(w h)
Average
O(w h)
Worst
O(w h)
Space
O(w h)
When to Use
- Use it when pixel values separate cleanly around one or more cutoffs.
- Use it for simple masks, document cleanup, or preprocessing before component labeling.
- Use adaptive thresholding when lighting varies across the image.
When Not to Use
- Avoid a fixed global threshold when illumination varies strongly.
- Avoid it when texture, shape, or context is needed to separate objects.
- Avoid thresholding without choosing the color space or channel intentionally.
What It Does
- Compares each pixel with a threshold
- Creates a binary or segmented image
Use Cases
- Object masks
- Document cleanup
- Simple segmentation
Inputs
- Image
- Threshold value
Outputs
- Binary or labeled image
Key Formulas (1)
Threshold rule
output = value ≥ threshold ? high : low
output = value >= threshold ? high : lowClassifies each pixel based on a cutoff value.
Pseudocode
- 1.Choose a threshold value
- 2.For each pixel:
- 3.Compare the pixel value to the threshold
- 4.Assign it to the low or high group
- 5.Write the result to the output
Canonical Example
- Input
- pixel values = [20, 120, 240], threshold = 100
- Output
- [0, 1, 1]
- Explanation
- Values above the threshold are assigned to the foreground group.
Implementation Notes
- Choose the intensity channel or feature to threshold before processing.
- Apply the same comparison rule to every pixel.
- Test on small images where expected mask values are easy to inspect.
Common Pitfalls
- Choosing a threshold that works only on one lighting condition.
- Ignoring noise before thresholding.
- Treating alpha, color, and grayscale values inconsistently.
Interactive Mechanism Walkthrough
Spatial Primitive & Sweep-Line CanvasShow grayscale values becoming black or white based on a cutoff.
Step 1 of 5: Choose a threshold value
image-thresholdNotes
Thresholding is often a preprocessing step rather than a complete segmentation method.
Sources (1)
What each reference supports for this entry.
Digital Image Processing
BookImage-processing textbook support for Thresholding's pixel-neighborhood operations, filtering rules, and image-analysis role.
Explore Next
Good next jumps from the relationship graph, ranked for useful discovery.
Alternative toCanny Edge Detection
Image Processing · Core
Thresholding segments by intensity, while Canny detects edges from gradients.
Compare with thisRelationships
View full relationship pageAlternative toCanny Edge Detection
Thresholding segments by intensity, while Canny detects edges from gradients.
Prerequisite ofConnected Component Labeling
Connected component labeling often runs on thresholded binary images.
Related Algorithms
Tags
Image ProcessingSegmentationClassic