Pre-processing for video noise reduction

A system for performing noise-reduction in video, the system comprising:

• a camera device comprising an image sensor configured to generate captured video data comprising a plurality of frames;
• a memory device configured to store a plurality of programmed instructions;
• a processor configured to execute the plurality of programmed instructions to implement:
• a noise estimator for computing a local noise level estimate for a current frame of the plurality of frames;
• a motion classifier for determining a motion classification of a plurality of regions in the current frame of the captured video data based on the local noise level estimate;
• and
• a combiner configured to perform noise reduction on the current frame based on the local noise level estimate and the motion classification,
• wherein the noise reduction comprises processing a first adaptive spatial filter and a first adaptive temporal filter, and
• wherein determining a motion classification comprises processing, by the motion classifier, a second adaptive spatial filter and a second adaptive temporal filter.

A non-transitory computer-readable medium comprising a plurality of programmed instructions, which, when executed by a processor in a computing device, is configured to perform noise reduction on frames of video data from an image capture device, the plurality of programmed instructions comprising:

• instructions to receive video data captured by an image capture device, the video data comprising a plurality of frames;
• instructions to calculate a local noise level estimate of a current frame of the plurality of frames;
• instructions to determine a motion classification of a plurality of regions in the current frame of the captured video data based on the local noise level estimate;
• and
• instructions to perform noise reduction on the current frame based on the local noise level estimation and the motion classification, wherein the instructions to perform noise reduction comprise instructions to process a first adaptive spatial filter and a first adaptive temporal filter, and wherein the instructions to determine the motion classification comprise instructions to process a second adaptive spatial filter and a second adaptive temporal filter.