ComfyUI Node: CLAHE

CLAHE:

CLAHE, which stands for Contrast Limited Adaptive Histogram Equalization, is a powerful image processing technique designed to enhance the contrast of images. Unlike traditional histogram equalization, which applies a global transformation to the entire image, CLAHE works on small regions in the image, called tiles. This localized approach allows for better contrast enhancement in areas with varying lighting conditions, making it particularly useful for improving the visibility of details in both bright and dark regions of an image. CLAHE also includes a clip limit parameter to prevent over-amplification of noise, ensuring that the enhanced image remains visually pleasing and free from artifacts. This method is widely used in medical imaging, night vision, and other applications where enhancing local contrast is crucial.

CLAHE Input Parameters:

src

The src parameter represents the source image that you want to enhance using CLAHE. This input should be an image in a compatible format that the node can process.

clip_limit

The clip_limit parameter controls the contrast enhancement limit. It prevents the amplification of noise by clipping the histogram at a specified value. The default value is 2, and it can be adjusted in steps of 1. A higher clip limit increases contrast but may introduce noise, while a lower clip limit reduces the risk of noise but may result in less contrast enhancement.

tile_grid_x

The tile_grid_x parameter specifies the number of tiles in the horizontal direction. The default value is 8, with a minimum value of 2. This parameter determines the size of the local regions where histogram equalization is applied. Increasing the number of tiles can enhance local contrast but may also increase computational complexity.

tile_grid_y

The tile_grid_y parameter specifies the number of tiles in the vertical direction. Similar to tile_grid_x, the default value is 8, with a minimum value of 2. This parameter, along with tile_grid_x, defines the grid size for local histogram equalization, affecting the balance between local contrast enhancement and computational load.

CLAHE Output Parameters:

IMAGE

The output parameter IMAGE represents the enhanced image after applying CLAHE. This image will have improved local contrast, making details in both bright and dark areas more visible. The output is in the same format as the input image, ensuring compatibility with subsequent processing steps.

CLAHE Usage Tips:

• Adjust the clip_limit parameter to balance between contrast enhancement and noise suppression. Start with the default value and increase or decrease it based on the visual quality of the output.
• Experiment with different values for tile_grid_x and tile_grid_y to find the optimal grid size for your specific image. Smaller tiles can enhance local details but may increase processing time.
• Use CLAHE for images with varying lighting conditions or where local contrast enhancement is needed, such as medical images or night-time photographs.

CLAHE Common Errors and Solutions:

"Invalid image format"

• Explanation: The input image format is not compatible with the CLAHE node.
• Solution: Ensure that the input image is in a supported format, such as a grayscale or RGB image.

"Clip limit out of range"

• Explanation: The clip_limit parameter is set to a value outside the acceptable range.
• Solution: Adjust the clip_limit parameter to be within the valid range, typically starting from 1 and increasing as needed.

"Tile grid size too small"

• Explanation: The values for tile_grid_x or tile_grid_y are set too low, resulting in insufficient tiles for processing.
• Solution: Increase the values for tile_grid_x and tile_grid_y to at least 2 to ensure proper grid formation for local histogram equalization.

"Tile grid size too large"

• Explanation: The values for tile_grid_x or tile_grid_y are set too high, causing excessive computational load or memory issues.
• Solution: Reduce the values for tile_grid_x and tile_grid_y to a more manageable size, balancing between local contrast enhancement and computational efficiency.