> 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/deconvolution.md).

# Deconvolution

This function block helps restore sharpness and detail in images—especially useful for images of fast-moving objects that appear blurred. Use the provided sliders to tune the restoration effect for your camera and scene.

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

`Image Any` (input socket)\
A color or grayscale image to be processed.

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

`Image Any` (output socket)\
The processed image with deconvolution applied.

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

`Angle`\
Adjusts the direction or orientation of the deconvolution effect. Useful when motion blur has a predominant direction.

`Diameter`\
Controls the size/strength of the correction area. Higher values affect larger blur kernels.

`Noise Reduction`\
Controls how aggressively noise is reduced while restoring detail. Increasing this reduces noise but may soften fine detail.

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

* Restores sharpness for motion-blurred images while allowing user control over direction and strength.
* Works with both grayscale and color images (color images are processed per channel).
* Simple sliders make it easy to experiment without technical knowledge.

## ⚙️ Running mechanism <a href="#how-it-works" id="how-it-works"></a>

When the block runs it accepts the incoming image, applies the deconvolution processing (taking into account the chosen `Angle`, `Diameter`, and `Noise Reduction` values) and outputs the corrected image. If a color image is provided, each color channel is handled so the final output keeps correct color information.

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

1. Connect a camera or image source to the `Image Any` input.
2. Start with moderate values: set `Angle` to match the motion direction (if known), set `Diameter` to a small/medium value, and set `Noise Reduction` low.
3. Inspect the result and adjust sliders until the image looks sharp without excessive artifacts.
4. Use the output image downstream (display, save, measurement, etc.).

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

* Combine with `Denoising` before this block to reduce sensor noise and avoid amplifying noise during deconvolution.
* If the input image is very large, use `Image Resize` to lower the resolution for faster experimentation, then re-run at full size for final results.
* Use `Image ROI` to crop to the area of interest before processing—this speeds up tuning and prevents over-processing irrelevant areas.
* If contrast is low, try `Contrast Optimization` prior to deconvolution to make details more recoverable.
* Use `Show Image` to preview results quickly, and `Image Logger` to save processed frames for later review.

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

* Output looks noisy or has artifacts: increase `Noise Reduction` or place a `Denoising` block before deconvolution.
* No visible improvement: try changing `Angle` to match motion direction and increase `Diameter` gradually.
* Processing is slow: reduce image size with `Image Resize` or limit processing to a region with `Image ROI`.
* Colors look off after processing: confirm you started with a color image and preview using `Show Image`; if necessary, adjust the pipeline order (e.g., apply `Contrast Optimization` before deconvolution).