Method for determining whether to intra code a video block

Claim 2

In a video encoder, a method for determining, for an incoming video frame, whether a present video block of the incoming video frame will be encoded or whether a difference between the present video block and a best match video block of a previously stored video frame will be encoded, the method comprising the steps of: a) for the present video block, dividing the present video block into a predetermined set of sub blocks; b) for each sub block of the predetermined set of sub blocks, calculating an average pixel value; c) for each sub block of the predetermined set of sub blocks, calculating a sub block deviation based on the average pixel value and individual pixel values of the sub block; d) for the present video block, calculating a present video block deviation based on the sub block deviation for each of the sub blocks in the predetermined set of sub blocks; e) determining an error deviation between the present video block and the best match video block; f) when a comparison of the present video block deviation to the error deviation is unfavorable, encoding the present video block; and g) when the comparison of the present video block deviation to the error deviation is favorable, encoding the difference between the present video block and the best match video block.

Claim 3

In the method of claim 1, step (a) further comprises defining the predetermined set of sub blocks as four sub blocks.

Claim 4

In the method of claim 1, step (c) further comprises using the sub block deviation for rate control.

Claim 5

In the method of claim 1, step (c) further comprises using the sub block deviation to control quantization.

Claim 6

In the method of claim 1, step (c) further comprises calculating the sub block deviation as a sum of absolute differences between the average pixel value and individual pixel values of the sub block.

Claim 7

In the method of claim 1, step (c) further comprises calculating the sub block deviation as a mean square error between the average pixel value and individual pixel values of the sub block.

Claim 8

In the method of claim 1, step (d) further comprises calculating the present video block deviation by summing the sub block deviations for each of the sub blocks in the predetermined set of sub blocks.

Claim 9

In the method of claim 1, step (f) further comprises defining unfavorable as a poor best match video block.

Claim 10

In the method of claim 1, step (g) further comprises defining favorable as a good best match video block.

Claim 13

A video compression system comprising: a motion estimator that receives a present video block and compares it to a plurality of previously stored video blocks to produce a best match video block; and an intra/non-intra determiner, operably connected to the motion estimator, that receives the present video block and the best match video block and that determines whether to encode the present video block or whether to encode a difference between the best match video block and the present video block, where the intra/non-intra determiner: a) divides the present video block into a predetermined set of sub blocks; b) calculates an average pixel value for each sub block of the predetermined set of sub blocks; c) calculates a sub block deviation based on the average pixel value and individual pixel values of the sub block, for each sub block of the predetermined set of sub blocks; d) calculates a present video block deviation based on the sub block deviation for each of the sub blocks in the predetermined set of sub blocks; e) determines an error deviation between the present video block and the best match video block; f) compares the present video block deviation to the error deviation to produce a comparison deviation; g) determines that the present video block is to be encoded when the comparison deviation is unfavorable; and h) determines that the difference between the present video block and the best match video block is to be encoded when the comparison deviation is favorable.

Claim 14

A video compression system comprising: a motion estimator that receives a present video block and compares it to a plurality of previously stored video blocks to produce a best match video block; and an intra/non-intra determiner, operably connected to the motion estimator, that receives the present video block and the best match video block and that determines whether to encode the present video block or whether to encode a difference between the best match video block and the present video block, where the intra/non-intra determiner: a) divides the present video block into a predetermined set of sub blocks; b) calculates an average pixel value for each sub block of the predetermined set of sub blocks; c) calculates a sub block deviation based on the average pixel value and individual pixel values of the sub block, for each sub block of the predetermined set of sub blocks; d) calculates a present video block deviation based on the sub block deviation for each of the sub blocks in the predetermined set of sub blocks; e) determines an error deviation between the present video block and the best match video block; f) compares the present video block deviation to the error deviation to produce a comparison deviation; g) determines that the present video block is to be encoded when the comparison deviation is unfavorable; and h) determines that the difference between the present video block and the best match video block is to be encoded when the comparison deviation is favorable.