Adaptive clipping prevention for picture sharpness enhancement

Abstract

A contrast control and clipping device is configured to determine a maximum range of enhancement that may be applied to each pixel for picture sharpness enhancement. This maximum range is used to limit the enhancement that is determined by the conventional convolution kernel and gain control modules. Preferably, the maximum range is independent of the determined convolution value, thereby minimizing the bandwidth requirements among the modules used for picture sharpness enhancement. The reduced bandwidth requirement allows for an efficient partitioning of tasks between hardware and software embodiments, and eases the overall system design task. In a multi-media application, the amount of data that is transferred between the processor and sub-components is substantially reduced, and the efficient partitioning facilitates parallel processing. The maximum range of enhancement is determined based on the pixel values in the current field, using a computationally simple algorithm, thereby minimizing anomalies caused by temporal dependencies among the modules that are used for picture sharpness enhancement. Additionally, the sharpness enhancement is dynamically adjusted based on overall picture-level characteristics, including the average enhancement range, overall noise measures, and so on.

Description

technical field [0001] The invention relates to the field of video processing, in particular to methods for enhancing video images. Background technique [0002] The "sharpness" of an image can be improved by enhancing the high frequency components of the image; that is, emphasizing the edges and other parts of an image and sharpening changes rather than gradual transitions. However, high frequency enhancement can cause some changes to be overemphasized, causing "spotting" and other unappealing visual effects in the image. [0003] figure 1 A block diagram of a conventional sharpness enhancement device 100 is shown. The sharpness enhancement device 100 includes a convolution kernel 150, which functions as a high-frequency filter to determine high-frequency components of an image. Generally, an image on a television screen refers to a frame. As used herein, the terms image and frame are used interchangeably and are meant to include all or a portion of a collection of data...

AI Technical Summary

Problems solved by technology

Often, optimal process partitioning cannot be achieved due to bandwidth constraints

For example, in a multimedia processing system, based on the appropriate implementation of each specific task, or to facilitate parallel processing, the preferred partitioning is usually not possible due to the bandwidth requirements it creates on the processor bus.

Images