Image encoding method and encoder

Patent metadata

Field	Value
Publication number	US-10469855-B2
Application number	US-201715720627-A
Country	US
Kind code	B2
Filing date	Sep 29, 2017
Priority date	Mar 31, 2015
Publication date	Nov 5, 2019
Grant date	Nov 5, 2019

Abstract

An image encoding method includes determining N P-frames from a sequence of images, determining a source refreshing region in each one of the P-frames, obtaining reconstructed images corresponding to the source refreshing regions by performing a first encoding on the source refreshing regions, obtaining updated P-frames by updating the source refreshing regions with the reconstructed images, and performing a second encoding on the updated P-frames.

First claim

What is claimed is:

1. An image encoding method comprising:
determining N P-frames from a sequence of images, N being a positive integer;
determining a source refreshing region in each one of the P-frames;
obtaining reconstructed images corresponding to the source refreshing regions by performing a first encoding on the source refreshing regions;
obtaining updated P-frames by updating the source refreshing regions with the reconstructed images, wherein each updated P-frame includes a new refreshing region and a source region separated from the new refreshing region, the new refreshing region corresponding to the reconstructed image corresponding to the updated P-frame; and
performing a second encoding on the updated P-frames, the second encoding including: performing the second encoding on the new refreshing regions according to an I-frame encoding mode; and performing the second encoding on the source regions according to a P-frame encoding mode.

2. The method of claim 1 , wherein performing the second encoding on the source regions according to the P-frame encoding mode includes:
if the source regions includes an unrefreshed region, performing the second encoding on the unrefreshed region according to the P-frame encoding mode;
if the source region includes a refreshed region, performing the second encoding on the refreshed region according to the P-frame encoding mode; or
if the source region includes an unrefreshed region and a refreshed region, performing the second encoding on the unrefreshed region and the refreshed region, respectively, according to the P-frame encoding mode.

3. The method of claim 1 , further comprising:
performing the first encoding and the second encoding successively on the P-frames.

4. The method of claim 3 , wherein:
each one of the P-frames includes a plurality of macro blocks, and
performing the first encoding and the second encoding successively on the P-frames includes:
on basis of the macro blocks, successively refreshing the N P-frames in order for rows of macro blocks from top to bottom or for columns of macro blocks from left to right until refreshing of an I-frame of the sequence of images corresponding to the P-frames is completed, wherein when one of the P-frames is refreshed, a P-frame immediately preceding the one of the P-frames and a P-frame immediately following the one of the P-frames overlap at boundaries thereof or do not overlap with each other.

5. The method of claim 1 , wherein:
each one of the P-frames includes N regions,
an (N−m) th region of an m th P-frame is identical to an (N−m) th region of an (m−1) th P-frame, 2≤m≤N, and
the new refreshing region of the m th region is obtained by referring to new refreshing regions of a 1 st part to an (N−2) th part of the (m−1) th P-frame.

6. The method of claim 1 , wherein performing the first encoding on the source refreshing regions of the P-frames according to the P-frame encoding mode includes:
performing the first encoding on each of the source refreshing regions using a quantization parameter QP according to the P-frames, and
recording the QP using a first quantization FirstQ, the QP being obtained by adjusting a bit rate required in transmitting the P-frames.

7. An image encoding method comprising:
determining N P-frames from a sequence of images, N being a positive integer, wherein the N P-frames are consecutive N P-frames including P 1 , P 2 , . . . , PN;
determining a source refreshing region in each one of the P-frames;
obtaining reconstructed images corresponding to the source refreshing regions by performing a first encoding on the source refreshing regions of the P-frames in an order of P 1 , P 2 , . . . , PN from left to right, including:
for each one of the P-frames, performing the first encoding on the corresponding source refreshing region to obtain the corresponding reconstructed image by referring to the new refreshing region of a previous P-frame on which the second encoding has been performed; and

obtaining updated P-frames by updating the source refreshing regions with the reconstructed images, including:
on basis of each one of a plurality of macro blocks included in the reconstructed images, assigning values to parameter values corresponding to each macro block in the source refreshing regions by taking parameter values corresponding to the reconstructed image as reference, so as to update the source refreshing regions with the reconstructed images and to obtain the new refreshing regions, thereby forming the updated P-frames; and

performing a second encoding on the updated P-frames in an order of P 1 , P 2 , . . . , PN from left to right.

8. An encoder comprising:
a processing circuit configured to determine N P-frames from a sequence of images, N being a positive integer, and determine a source refreshing region in each one of the P-frames; and
an encoding circuit configured to obtain reconstructed images corresponding to the source refreshing regions by performing a first encoding on the source refreshing regions,
wherein:
the processing circuit is further configured to obtain updated P-frames by updating the source refreshing regions with the reconstructed images, wherein each updated P-frame includes a new refreshing region and a source region separated from the new refreshing region, the new refreshing region corresponding to the reconstructed image corresponding to the updated P-frame; and
the encoding circuit is further configured to perform a second encoding on the updated P-frames, the second encoding including: performing the second encoding on the new refreshing regions according to an I-frame encoding mode; and performing the second encoding on the source regions according to a P-frame encoding mode.

9. The encoder of claim 8 , wherein the encoding circuit is further configured to:
if the source region includes an unrefreshed region, perform the second encoding on the unrefreshed region according to the P-frame encoding mode;
if the source region includes a refreshed region, perform the second encoding on the refreshed region according to the P-frame encoding mode; or
if the source region includes an unrefreshed region and a refreshed region, perform the second encoding on the unrefreshed region and the refreshed region, respectively, according to the P-frame encoding mode.

10. The encoder of claim 8 , wherein:
the N P-frames are consecutive N P-frames including P 1 , P 2 , . . . , PN, and the first encoding and the second encoding are performed on the P-frames in an encoding order of P 1 , P 2 , . . . , PN from left to right;
the encoding circuit is further configured to:
for each one of the P-frames, perform the first encoding on the corresponding source refreshing region to obtain the corresponding reconstructed image by referring to the new refreshing region of a previous P-frame on which the second encoding has been performed; and

the processing circuit is further configured to:
on basis of each one of a plurality of macro blocks included in the reconstructed images, assigning values to parameter values corresponding to each macro block in the source refreshing regions by taking parameter values corresponding to the reconstructed image as reference, so as to update the source refreshing regions with the reconstructed images and to obtain the new refreshing regions, thereby forming the updated P-frames.

11. The encoder of claim 8 , wherein the encoding circuit is further configured to perform the first encoding and the second encoding successively on the P-frames.

12. The encoder of claim 11 , wherein:
each one of the P-frames includes a plurality of macro blocks, and
the en

Assignees

• Sz Dji Technology Co Ltd

Inventors

• Yang Xiaohu
• Zhang Liangping
• You Zhongqian

Classifications

• H04N19/507

  using conditional replenishment · CPC title

• H04N19/174

  the region being a slice, e.g. a line of blocks or a group of blocks · CPC title

• H04N19/177

  the unit being a group of pictures [GOP] · CPC title

• H04N19/146

  Data rate or code amount at the encoder output · CPC title

• H04N19/107

  between spatial and temporal predictive coding, e.g. picture refresh · CPC title