Overview
Template matching slides a template over an image and scores each location with correlation or a distance measure.
CoreSource checked
Template Matching
Finds image regions that resemble a smaller template patch.
Quick Facts
Compact metadata for implementation and discovery.
- Category
- Image Processing
- Difficulty
- Intermediate
- Family
- Image Matching
- Tier
- Core
- Status
- Verified
- Review
- Source Checked
- Sources
- 1
- Formulas
- 1
- Relationships
- 3
- Code example
- Ready
- Visualization
- Ready
Core Concepts
sliding templatesimilarity scorecorrelation mapbest location
Quick facts
Complexity
Runtime and memory behavior at a glance.
Best
O(pixels * template)
Average
O(pixels * template)
Worst
O(pixels * template)
Space
O(score map)
When to Use
- Use when image enhancement needs sliding template.
- Use when the input matches the assumptions behind image Matching.
When Not to Use
- Avoid when image noise, lighting, or scale breaks the algorithm's assumptions.
- Avoid when a simpler algorithm gives the required result with less setup.
What It Does
- Finds image regions that resemble a smaller template patch
- Uses sliding template and similarity score to guide the work.
- Produces processed image.
Use Cases
- Image enhancement
- Feature extraction
- Segmentation and motion analysis
Inputs
- Image or image sequence
- Kernel, threshold, or model parameters
Outputs
- Processed image
- Mask
- Feature map or motion field
Key Formulas (1)
Normalized correlation
score = dot(patch, template) / (||patch|| · ||template||)
score = dot(patch, template) / (||patch|| * ||template||)Scores how closely a patch matches the template independent of scale.
Pseudocode
- 1.Normalize the image and choose border handling
- 2.Evaluate the local filter, statistic, model, or response for each relevant pixel or region
- 3.Threshold, rank, refine, or combine the responses
- 4.Return the mask, transformed image, feature set, or geometric estimate
Canonical Example
- Input
- a small grayscale image patch with a clear edge, corner, or region
- Output
- Processed image
- Explanation
- Template Matching applies sliding template to produce processed image.
Implementation Notes
- Handle borders and pixel coordinate conventions consistently.
- Keep the sliding template step separate enough to test independently.
- Keep coordinate conventions, border handling, and image normalization explicit in tests.
Common Pitfalls
- Applying operations in the wrong color space can distort the result.
- Test edge cases, not only clean examples.
- Thresholds tuned on one clean image can fail under different lighting, scale, blur, or noise.
Interactive Mechanism Walkthrough
Spatial Primitive & Sweep-Line CanvasShow the source image beside masks, filters, or motion vectors produced by the algorithm.
Step 1 of 4: Normalize the image and choose border handling
image-processing-mechanismNotes
Template Matching should be used with the assumptions and tradeoffs of image Matching in mind.
Sources (1)
What each reference supports for this entry.
Computer Vision: Algorithms and Applications
BookComputer-vision textbook support for Template Matching's feature, matching, flow, or multiscale image-analysis workflow.
Open sourceExplore Next
Good next jumps from the relationship graph, ranked for useful discovery.
Relationships
View full relationship pageUsesConvolution
Template matching is commonly implemented as correlation or convolution over image windows.
Related Algorithms
Tags
Image ProcessingPattern MatchingCorrelationSearch