> For the complete documentation index, see [llms.txt](https://docs.augelab.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.augelab.com/function-blocks/blocks-reference/image-transformations/transformation-filters/image-skeletonize.md).

# Image Skeletonize

This function block extracts the skeletal structure of objects in a binary/grayscale image. It is useful for feature extraction, topology analysis, and producing thin representations of shapes for measurement or further processing.

## 📥 Inputs <a href="#inputs" id="inputs"></a>

* `Image Gray` Grayscale or binary image to be skeletonized.

## 📤 Outputs <a href="#outputs" id="outputs"></a>

* `Skeletonized` Full skeleton result as a binary image (thin, single-pixel-wide representation).
* `Skeletonized Lite` Thinned version using a standard thinning method.
* `Skeletonized Partially` Partially thinned image controlled by the iterations setting.

## 🕹️ Controls <a href="#controls" id="controls"></a>

* `Iterations` Slider to adjust maximum iteration for partial thinning. Higher values produce more thinning in the `Skeletonized Partially` output.

## 🎨 Features <a href="#features" id="features"></a>

* Produces three different skeleton-style outputs for flexible use in analysis and visualization.
* Works directly on binary or grayscale images—no coding required.
* `Skeletonized Partially` allows controlled thinning so you can balance detail vs. simplification.

## 📝 Usage Instructions <a href="#usage" id="usage"></a>

1. Prepare the image input and connect it to the `Image Gray` input socket.
2. If your image is not already binary (black & white), consider thresholding or preprocessing (see Tips and Tricks).
3. Adjust the `Iterations` slider to change how aggressive the partial thinning should be.
4. Use the outputs for visualization, shape analysis, or as inputs to other blocks.

## 📊 Evaluation <a href="#evaluation" id="evaluation"></a>

When run, the block produces:

* `Skeletonized` for a full skeleton result,
* `Skeletonized Lite` for a standard thinned result,
* `Skeletonized Partially` which respects the `Iterations` control for gradual thinning.

These outputs can be inspected visually or fed to downstream blocks for measurement or further processing.

## 💡 Tips and Tricks <a href="#tips-and-tricks" id="tips-and-tricks"></a>

* Clean binary input produces the best skeletons. Consider using `Image Threshold` or `Image Adaptive Threshold` before this block to create a clear foreground/background separation.
* Reduce noise prior to skeletonization with `Blur` or `Denoising` to avoid spurious branches.
* Use `Morphological Transformations` (opening/closing) to remove small artifacts or to close small gaps in shapes before skeletonizing.
* Crop to the region of interest using `Image ROI Select` or `Image ROI` so the block focuses on the area you care about.
* After skeletonization, use `Find Contour`, `Approximate Contour`, or `Measure Position Distance` to extract measurements or further analyze the thin structures.
* If your images are large and processing is slow, try `Image Resizer` to reduce size before skeletonizing.

## 🛠️ Troubleshooting <a href="#troubleshooting" id="troubleshooting"></a>

* No visible skeletons: Ensure the input is binary or has sufficient contrast. Try `Image Threshold` or increase contrast with `Contrast Optimization`.
* Too many small branches or noise: Apply `Morphological Transformations` or `Blur` before skeletonizing to remove small artifacts.
* Skeleton too thin or details lost: Lower the `Iterations` setting for a gentler partial thinning, or use the `Skeletonized Lite` output which preserves more structure.
* Processing is slow: Reduce input resolution with `Image Resizer` or preprocess to limit the area using `Image ROI Select`.