Contrast enhancement of images
AssigneeXerox Corporation
InventorsChristopher R. Dance, Damian Arregui
Life of this patent
Estimating the term…
Abstract
A method and system for contrast enhancement of an input image that utilizes luminance values of pixels of the input image to derive transforms. The method down-samples and partitions an image into sub-images whereby transforms are generated for selected sub-images. The selected sub-image transforms are used to generate an output transform which is applied to the input image for local contrast enhancement of the image. Furthermore, a decision method as to whether or not the input image is to receive local contrast enhancement wherein darkness features of the input image are compared to threshold values and combined into an expression which results in the determination.
Claims11
A method for contrast enhancement of an input image, the method comprising:
- filtering a down-sampled image with a low-pass filter;
- partitioning the down-sampled image into a set of sub-images;
- deriving a sub-image transform for at least one of the sub-images mapping pre-filtered luminance values of the sub-image to post-filtered luminance values of the sub-image;
- selecting at least one of sub-image transforms for a pixel of interest in the input image;
- generating an output transform from the selected sub-image transforms;
- applying the output transform to the pixel of interest in the input image;
- and
- determining if the input image is to receive local contrast enhancement wherein the determination comprises:
- applying the low-pass filter to a low-resolution version of the input image resulting in enhanced luminance values for pixels in the low-resolution image;
- partitioning the pixels of the filtered image into at least a first and second category based on luminance values of the pixels;
- computing characteristic darkness features of the filtered image using the categories of luminance values of the pixels;
- and
- comparing the darkness features of the image to threshold values, wherein comparing the darkness features to threshold values comprises combining the comparisons into a Boolean expression which returns a true or false result as to whether or not the image is to receive local contrast enhancement;
- wherein computing features and comparing features includes computing redness features using red component values of the pixels of the image and comparing the redness features to threshold values wherein the comparisons are combined into the Boolean expression.
A method to determine if an input image is to receive local contrast enhancement, the method comprising:
- applying a bilateral filter to the input image resulting in enhanced luminance values for pixels in the image;
- partitioning the pixels of the filtered image into at least a first and second category based on the luminance values of the pixels;
- computing characteristic darkness features of the filtered image using the categories of luminance values of the pixels;
- and
- comparing the darkness features of the image to threshold values;
- wherein computing and comparing features includes computing redness features of the image using red component values of the pixels of the image and comparing the redness features of the image to threshold values and wherein computing darkness features comprises computing a ratio of number of pixels in one of the categories comparing the darkness features of the image to threshold values comprises combining the comparisons into a Boolean expression which returns a true or false result as to whether or not the input image is to receive local contrast enhancement.
A method to determine if an input image is to receive local contrast enhancement, the method comprising:
- applying a bilateral filter to the input image resulting in enhanced luminance values for pixels in the image;
- partitioning the pixels of the filtered image into at least a first and second category based on the luminance values of the pixels;
- computing characteristic darkness features of the filtered image using the categories of luminance values of the pixels;
- and
- comparing the darkness features of the image to threshold values;
- wherein computing features further comprises:
- computing a first darkness feature as a ratio of number of pixels in the first category of luminance values to the total number of pixels in the image;
- computing a second darkness feature as a ratio of number of pixels in the second category of luminance values to the total number of pixels in the image;
- computing a first redness feature as a median of the red component values of pixels in the first category of luminance values;
- computing a second redness feature as a median of the red component values of pixels in the second category of luminance values;
- and
- computing a third redness feature as a median of the red component values of pixels in a third category of luminance values.
Dependencies are inferred from the claim text.