ComfyUI Node: Image Median Filter

Image Median Filter:

The Image Median Filter node is designed to enhance your images by reducing noise and smoothing out textures while preserving edges. This node applies a median filter, which is particularly effective for removing salt-and-pepper noise and other types of random noise from images. By replacing each pixel's value with the median value of the intensities in its neighborhood, the filter helps in maintaining the sharpness of edges while eliminating unwanted noise. This makes it an essential tool for AI artists looking to clean up their images without losing important details.

Image Median Filter Input Parameters:

image

This parameter accepts the image you want to process. The image should be in a format compatible with the node, typically as a tensor.

diameter

This integer parameter defines the diameter of the pixel neighborhood used during the filtering process. A larger diameter will consider more pixels around each point, leading to a stronger smoothing effect. The value ranges from 0.1 to 255, with a default of 2.0. Adjusting this value allows you to control the extent of noise reduction and detail preservation.

sigma_color

This float parameter controls the filter's sensitivity to color differences. A higher value means the filter will consider a wider range of colors as similar, leading to more aggressive smoothing. The value ranges from -255.0 to 255.0, with a default of 10.0. Fine-tuning this parameter helps in balancing between noise reduction and color detail retention.

sigma_space

This float parameter determines the filter's sensitivity to spatial differences. A higher value means the filter will consider a larger spatial area around each pixel, resulting in more extensive smoothing. The value ranges from -255.0 to 255.0, with a default of 10.0. Adjusting this parameter allows you to control the spatial extent of the filtering effect.

Image Median Filter Output Parameters:

IMAGE

The output is the processed image with the median filter applied. This image will have reduced noise and smoother textures while preserving important edges and details. The output is typically in the same format as the input image, usually as a tensor.

Image Median Filter Usage Tips:

• For images with a lot of fine details, start with a smaller diameter and gradually increase it to find the right balance between noise reduction and detail preservation.
• Adjust the sigma_color and sigma_space parameters in tandem to fine-tune the filter's effect. Higher values will result in more aggressive smoothing, which can be useful for very noisy images.
• Use this filter as a preprocessing step before applying other image enhancement techniques to ensure a cleaner and more refined final result.

Image Median Filter Common Errors and Solutions:

ValueError: Invalid diameter value

• Explanation: This error occurs when the diameter value is set outside the allowed range.
• Solution: Ensure that the diameter value is between 0.1 and 255.

ValueError: Invalid sigma_color value

• Explanation: This error occurs when the sigma_color value is set outside the allowed range.
• Solution: Ensure that the sigma_color value is between -255.0 and 255.0.

ValueError: Invalid sigma_space value

• Explanation: This error occurs when the sigma_space value is set outside the allowed range.
• Solution: Ensure that the sigma_space value is between -255.0 and 255.0.

TypeError: Input image is not in the correct format

• Explanation: This error occurs when the input image is not provided in a compatible format.
• Solution: Ensure that the input image is in a tensor format compatible with the node.