Abstract: A video processing system may determine a scene in an image and may adaptively adjust one or more of three dimensional color components of the image based on the determined scene. The three dimensional color components may comprise luma, hue and/or saturation. Statistics for the three dimensional color components may be collected and utilized to determine content and/or scene type within the image. Embedded textual information may also be utilized. Configuration parameters for processing the image may be adapted based on the statistics and/or the knowledge of the content and/or scene type. Hues, saturation levels and/or luma levels may be adjusted based on the statistics, scene type and/or content. The three dimensional color components may be dynamically adapted for at least a portion of the image. The statistics may comprise distributions of saturation over hue and/or luma, distributions of hue over saturation and/or luma, and/or distributions of luma.

Abstract: A video frame may be divided into plurality of sub-frames to reduce motion blur in sample-and-hold display. The plurality of sub-frames may preserve, in their totality, the luminance and coloring of the original input frames. An input frame that is Y?CrCb encoded may be converted to R?G?B? to enable luminance conversion onto the plurality of sub-frame while preserving the coloring information of said input frame. The first of said plurality of sub-frames may comprise energy and/or luminance encoded into the original frame with remaining energy and/or luminance encoded into remaining sub-frames. Determining luminance encoding of said plurality of sub-frames may be performed dynamically or may be determined based on programmable look-up tables. Frame conversion may compensate for nonlinearity in sample-and-hold displays that may be utilized to display the output sub-frames, wherein said nonlinearity may be caused by the gamma characteristics of said displays.

Abstract: A ranging device is disclosed that monitors for changes in a scene of interest. The ranging device includes a transmitter that sends pulses into the scene, a receiver that generates pulse samples in response to pulse returned from the scene, and a target discriminator that analyzes the pulse samples. The target discriminator detects targets entering the scene, targets leaving the scene, and targets moving in the scene through the use of at least one non-averaging statistical characteristic of the pulse samples. The target discriminator may generate short term and long term high order statistical samples in response to the pulse samples as well as detector samples consisting of the ratio of high order statistical samples. The detector samples are highly indicative of the presence or absence of a target in the scene.

Abstract: A processing technique suitable for use with a search radar system which minimizes the false detection of targets. Rather than rely on a predetermined target trajectory, such as a straight line trajectory in range time space, the processing technique in accordance with the present invention is not based on a target dynamics model. The system in accordance with the present invention relies on past detections and assigns a confidence level to detections which are in the wake of a previous detection. A wake function is used to determine the amount by which each detection recursively accrues confidence based on a limited record of past detections. A detection is declared when the confidence level exceeds a predetermined threshold. As detections are encountered in range time space, such detections are used to form wakes based on the wake function which are used to assign confidence levels for future detections.