14
2023
-
06
What are the video encodings?
Why video encoding?
Along with the increase in user demand for HD video, the amount of video data for video multimedia is also increasing. If uncompressed, these videos are difficult to be used for actual storage and transmission. Take 1080P 60FPS 1H video as an example, the size of the uncompressed video will reach 20G+.
The basic principle of video encoding:
Video image data has a strong correlation, that is to say, there is a large amount of redundant information. The redundant information can be categorized into spatial redundant information and time domain redundant information. Compression technology is to remove the redundant information in the data (remove the correlation between the data), compression technology includes intra-frame image data compression technology, inter-frame image data compression technology and entropy coding compression technology.
Removing redundant information in the time domain
Time domain redundant information can be removed by using inter-frame coding technique, which includes the following three parts:
- Motion Compensation
Motion compensation is to predict and compensate the current local image by the previous local image, which is an effective method to reduce the redundant information in the frame sequence.
- Motion Representation
Different regions of the image need to use different motion vectors to describe the motion information. Motion vectors are compressed by entropy coding.
- Motion Estimation
Motion estimation is a set of techniques for extracting motion information from a video sequence.
Note: Common compression standards use block-based motion estimation and motion compensation.
De-redundancy in the null domain
Inter-frame coding and entropy coding techniques are mainly used:
- Transform coding
Both intra-frame images and predicted differential signals have a high level of null-space redundancy information. Transform coding transforms the null domain signal to another orthogonal vector space so that its correlation decreases and data redundancy is reduced.
- Quantization Coding
After transform coding, a batch of transform coefficients are produced, which are quantized to bring the output of the encoder to a certain bit rate. This process leads to a reduction in accuracy.
- Entropy coding
Entropy coding is lossless coding. It takes the coefficients and motion information obtained after transformation and quantization and compresses them further.
What are the coding schemes:
There are many video coding schemes, it is not too much to describe them as a hundred, and the common audio-video coding has the following categories:
MPEG series
By the ISO (International Standards Organization) under the MPEG (Motion Picture Experts Group) development Video coding is mainly Mpeg1 (vcd), Mpeg2 (DVD), Mpeg4 (DVDRIP are used in its variants, such as: divx, xvid, etc.), Mpeg4 AVC; audio coding is mainly MPEG Audio Layer 1/2, MPEG Audio Layer 3 (mp3), MPEG-2 AAC, MPEG-4 AAC and so on. Note: DVD audio does not use Mpeg. MPEG is the Moving Picture Experts Group (MPEG).
Keywords: intelligent lighting
Zhongshan Zhongsheng Optoelectronic Technology Co., Ltd.
One of the eighth floors of Building E, No. 13-2 Huanzhen North Road, Henglan Town, Zhongshan City
Copyright©2022 Zhongshan Zhongsheng Optoelectronic Technology Co., Ltd. All rights reserved. | Powered by 300.cn Zhongshan | SEO