Abstract: In one aspect, a method, includes producing a centralness metric that is highest at a center of detail from a set of pixels and having a negative slope to neighboring pixels from a center pixel at the center of detail, calculating a delta signal for the set of pixels using the produced centralness metric and pixel values of the set of pixels, generating a squeezed pixel value for each of the pixel values of the set of pixels, and outputting the generated squeezed pixel values.

Abstract: In one aspect, a method to reduce compression artefacts, includes generating a first set of weights based on pixel value differences, producing a soft binned histogram from values in a neighborhood of pixels, generating an adapted histogram from the soft binned histogram using interpolation, generating occurrence density weights from the adapted histogram, combining the first set of weights and the occurrence density weights, filtering pixel values a bilateral filter using the combined weights, mixing the filtered pixel values and original pixel values, and outputting the mix of filtered pixel values and original pixel values.

Abstract: A multi-resolution cache includes a first, second and third cache segments the first segment having a first resolution and the second and third segments having a second resolution, the second resolution less than the first resolution, the first and third cache segments communicatively coupled to an off-chip memory, the first and third cache segments configured to each receive a cache line of data having the first and second resolutions, a fourth and fifth cache segments having the second resolution, a first downscaler communicatively coupled to the first and fourth cache segments configured to reduce the resolution when a first resolution cached data is shifted from the first cache segment to the fourth cache segment, a first upscaler communicatively coupled to the all cache segments that have the second resolution, and is configured to increase the reduced resolution cached data to the first resolution and output it.

Abstract: A method of recursive filtering of image data includes receiving sequentially a plurality of original pixel values, multiplying an original pixel value of a current pixel of the plurality of pixel values by a dynamically varying recursion coefficient, adding recursively filtered pixel values from left and right neighbors of the current pixel, retrieved from a memory buffer holding filtered pixel data from a previous image line, multiplying the added recursively pixel values by 1 minus the dynamically varying recursion coefficient, adding the two multiplication results together to yield a filtered pixel value for the current pixel, writing the filtered pixel value for the current pixel back into the memory buffer, and displaying the filtered pixel value on a display.

Abstract: A multi-resolution cache includes a first, second and third cache segments the first segment having a first resolution and the second and third segments having a second resolution, the second resolution less than the first resolution, the first and third cache segments communicatively coupled to an off-chip memory, the first and third cache segments configured to each receive a cache line of data having the first and second resolutions, a fourth and fifth cache segments having the second resolution, a first downscaler communicatively coupled to the first and fourth cache segments configured to reduce the resolution when a first resolution cached data is shifted from the first cache segment to the fourth cache segment, a first upscaler communicatively coupled to the all cache segments that have the second resolution, and is configured to increase the reduced resolution cached data to the first resolution and output it.

Abstract: A multi-resolution cache includes a first, second and third cache segments the first segment having a first resolution and the second and third segments having a second resolution, the second resolution less than the first resolution, the first and third cache segments communicatively coupled to an off-chip memory, the first and third cache segments configured to each receive a cache line of data having the first and second resolutions, a fourth and fifth cache segments having the second resolution, a first downscaler communicatively coupled to the first and fourth cache segments configured to reduce the resolution when a first resolution cached data is shifted from the first cache segment to the fourth cache segment, a first upscaler communicatively coupled to the all cache segments that have the second resolution, and is configured to increase the reduced resolution cached data to the first resolution and output it.

Abstract: Estimating and/or compensating motion in a video image is provided wherein an image segment of the video image is selected, a search area is defined around the image segment, the defining being based on a range of possible motion vectors for the image segment, and image data related to said search area is retrieved. The search area is defined to have its center offset from a center of the image segment. The offset may be determined by a global motion parameter.

Abstract: The quality of recorded images with for example a high speed camera can be detoriated by light modulation which is a consequence of for example gas-discharge lamps running at a lower frequency with respect to the camera. A camera according to the invention provides light modulation removal by handling fields of the image different depending on the light modulation.

Abstract: A processing circuit for motion compensated de-interlacing of video signals, having a line memory 21, a de-interlacing circuit 22, a frame memory 24, and a cache memory 25, further includes a pixel mixer 29 interposed between the cache memory 25 and the de-interlacing circuit 22.

Abstract: In a method of detecting text in a video signal (VS), high horizontal frequencies in the video signal (VS) are detected (ED) to obtain horizontal edge detection results (refo, refpp1, refpp2), and horizontal edge detection results (refo) at a given position are correlated (C) with horizontal edge detection results (refpp1, refpp2) at vertically neighboring positions.

Abstract: Estimating and/or compensating motion in a video image is provided wherein an image segment (B12) of the video image is selected, a search area (S) is defined around the image segment, the defining being based on a range of possible motion vectors (V) for the image segment, and image data related to said search area is retrieved. The search area is defined to have its center offset from a center of the image segment. The offset may be determined by a global motion parameter.

Abstract: A processing circuit for motion compensated de-interlacing of video signals, the processing circuit comprising: a line memory 21, a de-interlacing circuit 22, a frame memory 24, and a cache memory 25, in which a pixel mixer 29 is interposed between the cache memory 25 and the de-interlacing circuit 22.

Abstract: In a method of detecting text in a video signal (VS), high horizontal frequencies in the video signal (VS) are detected (ED) to obtain horizontal edge detection results (refo, refpp1, refpp2), and horizontal edge detection results (refo) at a given position are correlated (C) with horizontal edge detection results (refpp1, refpp2) at vertically neighboring positions.

Abstract: The quality of recorded images with for example a high speed camera can be detonated by light modulation which is a consequence of for example gas-discharge lamps running at a lower frequency with respect to the camera.