Overview
Connected Component Labeling scans pixels and groups neighboring foreground pixels into distinct labeled components.
CoreSource checked
Connected Component Labeling
Assigns labels to contiguous foreground regions in a binary or segmented image.
Quick Facts
Compact metadata for implementation and discovery.
- Category
- Image Processing
- Difficulty
- Intermediate
- Family
- Image Segmentation
- Tier
- Core
- Status
- Verified
- Review
- Source Checked
- Sources
- 2
- Formulas
- 0
- Relationships
- 3
- Code example
- Ready
- Visualization
- Ready
Core Concepts
foreground maskpixel connectivitylabel equivalencecomponent id
Quick facts
Complexity
Runtime and memory behavior at a glance.
Best
O(pixels)
Average
O(pixels)
Worst
O(pixels)
Space
O(pixels)
When to Use
- Use when foreground regions need to be counted, measured, or filtered.
- Use when object counting is a central requirement.
When Not to Use
- Avoid when objects are not separable in the input mask.
- Avoid when lighting, noise, or scale changes dominate the image signal.
What It Does
- Scans foreground pixels
- Merges neighboring labels
- Outputs one label per connected region
Use Cases
- Object counting
- Segmentation cleanup
- Blob measurement
Inputs
- Binary or labeled image
- Connectivity rule
Outputs
- Component label image
- Component statistics
Pseudocode
- 1.Scan pixels in row order
- 2.Assign a new label or reuse neighboring labels
- 3.Record equivalent labels
- 4.Resolve equivalences
- 5.Replace provisional labels with final component labels
Canonical Example
- Input
- binary image has two separate white blobs
- Output
- labels = blob1 and blob2
- Explanation
- Each connected foreground region receives its own id.
Implementation Notes
- Use Union-Find to manage equivalent provisional labels in a two-pass implementation.
- Start with grayscale or single-channel tests before applying the method to full images.
- Keep a tiny example with expected output beside the implementation.
Common Pitfalls
- Changing connectivity changes whether diagonal pixels belong to the same region.
- Padding, threshold, and color-space choices can change the result dramatically.
- Not defining boundary, empty, or degenerate inputs before implementation.
Interactive Mechanism Walkthrough
Spatial Primitive & Sweep-Line CanvasShow separate foreground blobs receiving different colors or numeric labels.
Step 1 of 5: Scan pixels in row order
image-kernelNotes
Four-connectivity and eight-connectivity can produce different components.
Sources (2)
What each reference supports for this entry.
Digital Image Processing
BookImage-processing textbook support for connected-component labeling in binary images.
Sequential Operations in Digital Picture Processing
PaperReference for sequential connected-component labeling and equivalence handling.
Open sourceExplore Next
Good next jumps from the relationship graph, ranked for useful discovery.
Relationships
View full relationship pageUsesUnion-Find
Connected-component labeling can merge provisional labels efficiently with union-find.
Related Algorithms
Tags
Image ProcessingSegmentationGraph Connectivity