Abstract: A method and device for image noise estimation and image capture apparatus are provided. In the device, a global noise estimation unit is configured to determine a plurality of current sample blocks of a current image frame and a plurality of previous sample blocks of a previous image frame, and calculate a block feature of each of the current sample blocks, and calculate a block sum-of-absolute-difference (SAD) between each of the current sample blocks and corresponding one of the previous sample blocks, and cluster the current sample blocks into a plurality of segments according to the block features, and respectively establish a plurality of noise models for the segments according to the block features and the block SAD. A local noise calculation unit is configured to calculate noise level information of a local image block of the current image frame according to a corresponding noise model of the noise models.

Abstract: A method for image noise estimation includes: determining a plurality of current sample blocks of a current image frame and a plurality of pervious sample blocks of a pervious image frame; calculating a block feature of each of the current sample blocks, and calculating a block sum-of-absolute-difference (SAD) between each of the current sample blocks and corresponding one of the pervious sample blocks; clustering the current sample blocks into a plurality of segments according to the block features; respectively establishing a plurality of noise models for the segments according to the block features and the block SAD; and calculating noise level information of a local image block of the current image frame according to a corresponding noise model of the noise models. A device for image noise estimation and an image capture apparatus are also provided.

Abstract: In a method for estimating image noise, plural sample blocks of an image are determined; a mean of at least one color component of each of the sample blocks and a standard deviation of at least one color component of each of the sample blocks are calculated; the sample blocks are distributed into plural segments according to the means of the sample blocks; a weighted average of the standard deviations of all of the sample blocks of each of the segments is calculated according to at least one threshold value that is determined according to the minimum standard deviation among the standard deviations of all of the sample blocks of the segment. The weighted average may be applied to noise reduction, edge detection, or motion detection of the image.