Back to results
US 2007/0183677 A1Published application
Dynamic range compression of high dynamic range imagery
AssigneeTeledyne Licensing, LLC
InventorMario Aguilar
Life of this patent
Estimating the term…
Abstract
An improved apparatus and method adaptively compresses the dynamic range of an image from a sensor by rescaling the value of every pixel to fit within the dynamic range of a low dynamic range display while preserving local image contrast and global scene appearance.
Claims7
An adaptive dynamic range compression apparatus, comprising:
- (a) a high frequency boost circuit that receives an input image comprising a plurality of input pixels, each of the input pixels being represented by a pixel value having a first predetermined first number of bits, the high frequency boost circuit producing an output image comprising a plurality of output pixels as a function of:
- (i) a first region of the input image neighboring each input pixel, the first region having a first area,
- (ii) a second region of the input image neighboring each pixel, the second region having a second area larger than the first area, and
- (iii) a third region of the input image neighboring each pixel, the third region having a third area larger than the second area;
- and
- (b) a dynamic range mapping circuit that receives the output image produced by the high frequency boost circuit and maps the output image to a target image comprising a plurality of target pixels, each of the target pixels being represented by a pixel value having a predetermined second number of bits, the second number of bits being less than the first-number of bits.
3 more elements
A method of adaptively compressing the dynamic range of an image having a plurality of pixels, comprising the steps of:
- convolving each pixel of the image and a first region of the image around each pixel with a Gaussian operator to produce a processed pixel value;
- and
- adaptively normalizing the processed pixel value based on a mean value in a second region of the image around each pixel, the second region being larger than the first region, but smaller than the entire image.
Dependencies are inferred from the claim text.