Blur

This function block softens input images by applying a selectable blur filter. Use it to reduce noise, smooth textures, or prepare images for downstream analysis.

📥 Inputs

Image Any Accepts any image (color or grayscale) to be blurred.

📤 Outputs

Image Any Returns the blurred version of the input image.

🕹️ Controls

Blur Type Choose the blur algorithm. Available options include AVERAGE_BLUR, GAUSSIAN_BLUR, and MEDIAN_BLUR.

Kernel Size Adjust how strong the blur will be. Larger values produce stronger smoothing. (Median blur uses a single size value; other types use a rectangular kernel.)

⚙️ Running mechanism

When the block receives an image on the Image Any input it applies the selected blur method with the chosen kernel size and outputs the processed image on Image Any. Changing the controls updates how subsequent images are processed. Larger kernel sizes increase smoothing but reduce fine detail.

🎨 Features

• Simple selection between common blur methods for flexible smoothing.

• Adjustable blur strength via an intuitive slider.

• Works with both color and grayscale images.

• Fast, suitable for real-time preview and preprocessing pipelines.

📝 Usage instructions

1. Connect an image source to Image Any.

2. Select the desired Blur Type.

3. Move the Kernel Size slider to set smoothing strength.

4. Use the resulting Image Any output as input to viewers or analysis blocks.

💡 Tips and Tricks

• Use Image Resize before this block to speed up processing when full resolution is not needed.

• Use Blur as a preprocessing step before Image Threshold or Image Adaptive Threshold to reduce noise and get cleaner binary results.

• Apply Blur before Edge Filter to reduce false edge responses from high-frequency noise.

• For pattern or feature matching, try Blur together with Find Object to reduce background noise; adjust kernel size carefully so salient features remain detectable.

• Preview results with Show Image to quickly iterate on control settings.

• When analyzing intensity along lines, apply Blur before Histogram On Line to smooth spurious peaks.

🛠️ Troubleshooting

• If the output looks unchanged, increase the Kernel Size to see stronger smoothing.

• If important details disappear, reduce the kernel size or switch to a milder blur type (for example, try GAUSSIAN_BLUR).

• For salt-and-pepper noise, prefer MEDIAN_BLUR as it is more effective at removing isolated pixel noise.