HDTV Overview (2)

HDTV Overview (2)

I. rec.709 HDTV color Encoding
Rec.709 defines the YCbCr color encoding. Unfortunately, unlike sdtv's range factor definition, YCbCr encoding is divided into two types. One is used for small image (sdtv) encoding, and the other is used for large image (HDTV) encoding.

Ii. Video Compression

Direct storage or transmission of YCbCr digital videos requires a large amount of data throughput-about 20 mb per second for sdtv and about 120 MB per second for HDTV. From the first generation of the photo studio digital recorder to today's ultra-high quality photo studio recorder, they all store uncompressed data.

(1) Data Compression
Data Compression reduces the number of spaces required to store text, numbers, binary, images, sounds, or other data. However, data compression also results in a large amount of computing required for compression and decompression.
In terms of definition, Data Compression should be lossless. After decompression, the data can be accurately restored, with the corresponding bit (or byte ).

Most important data compressionAlgorithmIncluding the RLE Algorithm and LZW algorithm.

(2) Image Compression
Image data usually has a strong correlation between portrait and landscape pixels. The RLE and LZW algorithms are suitable for storing two-layer (bilevel) or pseudo-color (pseudo-color) image data ., He usually usesProgramThe scanning line sequence and the correlation between the horizontal pixels are compressed. This is a moderate compression, perhaps 2: 1.
The image data algorithm can also compress the statistical results of the image data. Unlike any binary data, it is possible to improve the compression ratio. The ITU-T Fax standards use the horizontal and vertical correlation of image data to achieve a higher compression ratio of coin RLE or LZW.

The conversion technology is very effective for compressing continuous tone (gray or true color) image data. Discrete Cosine Transform DCT is suitable for compressing continuous tone image data.

(3) lossy compression
In general, lossless compression can only achieve moderate compression of consecutive tones. However, human sensory capabilities are limited, and some out-of-range data cannot be perceived. Therefore, the data can be discarded to achieve greater compression ratio. This is lossy compression.

Lossy compression is not suitable for dual-layer or pseudo-color images, but it is ideal for gray-level or true-color images.

(3.1) JPEG
JPEG is a Lossy Compression Method for images. Its Variant M-JPEG is dedicated to motion image sequences. The digital camera uses the M-JPEG algorithm.
MPEG is also a lossy compression algorithm standard for video sequence. MPEG-2 is used for the release of Digital TVs and DVDs.
Speaking of JPEG, it was also funny. At first, the JPEG Committee developed this standard to use it for color fax, but it was quickly used and deployed in desktop graphics and digital photos.
Generally, the JPEG compression tool converts 'G' B 'to y' cbcr, and then applies the secondary color sampling of to achieve 2:1 compression.
 
(3.2) motion-JPEG
Motion-JPEG simply compresses each domain or frame of a video sequence as a self-contained compressed image. Each domain or frame is "intra coded ). Since each image is individually compressed, The M-JPEG video sequence can be directly edited, and the disadvantage is the use of temporary coherence.
M-JPEG techniques and implementations increase the compression ratio from to around. The digital video data rate of 20 Mb/s can be compressed to 20 Mb/s, which is suitable for recording the consumption of electronic digital video tapes.
 
(3.3) MPEG
The Moving Picture Experts Group of ISO and IEC has developed the MPEG standard. At the beginning of MPEG, a self-contained image provides a base value, which forms an anchor picture ). Subsequent images can be encoded based on different pixels from the anchor image. This method is called interframe coding ).
Once the decoder receives the anchor image, it can provide basic predictions for subsequent images. This prediction can improve the prediction error after the encoder transmission.
Motion can change the position of the screen element. A fast moving element can move 10 pixels within one frame of time. In continuous motion, pixels at a certain position have different values in consecutive images. Therefore, motion can increase the number of incorrect prediction data.
Interpicture coding utilizes similar technologies. I will not go into details.
MPEG overcomes the issue of motion images and uses an encoder to calculate the motion vector ). The encoder replaces the pixel value of the anchor image by predicting motion. This process is called motion compensation ).