TWI565308B - Distortion video compression coding system and method thereof - Google Patents

Description

Distortionless video compression coding system and method thereof

The invention relates to an encoding system and a method thereof, in particular to a distortionless video compression coding system and method thereof for video distortion-free compression to reduce data transmission amount and power consumption.

Please refer to the invention of the Republic of China Patent Publication No. 201417585 "Encoder, Decoder and Encoding Method, Decoding Method". The prior art disclosed is to provide a decoder and an encoding method. In one stage, if the material does not already exist in the format of the data block, it divides the data to be encoded into data blocks; optionally, the result of such segmentation of the data to be encoded is that the data blocks have different sizes from each other. Size, ie, non-uniform, depends on the nature of the content included in the data block, which is different from many known coding algorithms that uniformly divide the data into corresponding data blocks; in the second stage, the encoding The apparatus includes an analysis unit and a plurality of coding units that use various types of coding algorithms, wherein the coding algorithms are different from each other, and some coding algorithms are known algorithms, wherein the coding unit is used to selectively Processing each data block; it can be understood that some of the coding algorithms used in the coding unit are substantially similar to each other, but are flat in terms of time. The manner of the lines is executed; at least one of the coding units organizes the data existing in the data block into at least two levels, and generates one or more masks, the masks describing the data blocks Which data value belongs to which corresponding level, at least one of the coding units also calculates an aggregated value of the data sorted into each level, such as an average value, or an intermediate value, or the like; advantageously, the mask is Implemented as a spatial bitmap; the third stage consists of compressing the bitmap and compressing the aggregated values of each stage to generate encoded data from the encoder. Alternatively, different compression algorithms can be used in the third stage. Other data compression algorithms may alternatively or additionally be employed, such as multi-level data compression. Alternatively, the encoder can be used in combination with other encoders to obtain a hybrid encoding of the data to generate encoded data.

However, the decoder and the encoding method of the prior art described above do not disclose their use. Lossless compression decoding and/or encoding for image data processing, and to reduce the response time of external memory connected to the general video encoding system in information/file access and transmission, taking up too much memory bandwidth And power consumption and other issues.

Therefore, the present invention provides a distortion-free video compression coding system and method thereof, which can effectively reduce transmission between external storage devices under the condition of undistorted visibility to the human eye by means of effective video lossless compression. The data size, the response time is fast, the memory bandwidth is small, and the power consumption is small, which is a technical problem that various circles are currently trying to solve.

In view of the above disadvantages of the prior art, the main object of the present invention is to provide a distortion-free video compression coding system and method thereof, which can effectively reduce the size of data transmitted between external storage devices by means of effective video lossless compression. The reaction time is fast, the memory bandwidth is reduced, and the power consumption is small.

In order to achieve the foregoing and other objects, an object of the present invention is to provide a distortion-free video compression coding method, comprising the steps of: inputting an initial picture to a discrete cosine transform module to generate a discrete cosine transform information; The discrete cosine transform information is input to a quantization module to generate a quantization information; the quantization information is input to the quantization module to generate an inverse quantization information; and the inverse quantization information is input to the discrete cosine transform module to Generating an inverse discrete cosine transform information; inputting the inverse discrete cosine transform information to a motion compensation module to generate an action compensation information; and inputting the motion compensation information to a picture reconstruction module to generate a picture reconstruction information, And temporarily reserving the image reconstruction information in a picture buffer module; the motion compensation module extracts the picture reconstruction information from the picture buffer module to generate an action estimation information; and input the action estimation information Go to the discrete cosine transform module and repeat all the above actions; wherein the screen reconstruction module is a sampling difference An encoder comprising a sample difference prediction unit and a Columbus coding unit, wherein the sample difference prediction unit is configured to perform prediction sampling in a complex direction of the content of the motion compensation information, and then encode the content in the Columbus coding unit. That is, the motion compensation information is compressed by the data amount to generate the picture reconstruction information.

In order to achieve the foregoing and other objects, another object of the present invention is to provide a distortion-free video compression coding system, comprising: a quantization module for generating a quantitative information. And at least one of the inverse quantized information; the discrete cosine transform module is configured to generate at least one of a discrete cosine transform information and an inverse discrete cosine transform information, and the discrete cosine transform module is coupled to the Quantization module; a motion compensation module for generating at least one of motion compensation information and motion estimation information, and the motion compensation module is coupled to the discrete cosine transform module; a picture reconstruction module For generating a picture reconstruction information and connecting to the action compensation module; and a picture buffer module for temporarily storing the picture reconstruction information and connecting to the picture reconstruction module; wherein the discrete cosine The conversion module is configured to receive an initial picture; the picture reconstruction module is a sample difference value encoder, and includes a sample difference prediction unit and a Columbus coding unit, wherein the sample difference prediction unit is used to perform the action The content of the compensation information is subjected to predictive sampling in the complex direction, and then encoded by the Columbus coding unit, that is, the data amount is compressed by the motion compensation information. To produce the reconstructed picture information.

When the distortion-free video compression coding system of the present invention is used, the initial picture is input to the discrete cosine transform module, and the discrete cosine transform module generates the discrete cosine transform information, and inputs the discrete cosine transform information to the quantization mode. a group, wherein the quantization module generates the quantization information, and then input the quantization information to the quantization module to generate the inverse quantization information, and input the inverse quantization information to the discrete cosine transform module to generate The inverse discrete cosine transform information is further input to the motion compensation module to generate an action compensation information, and the motion compensation information is input to a picture reconstruction module to generate a picture. Reconstructing the information; and then temporarily storing the screen reconstruction information in a picture buffer module; the motion compensation module extracts the picture reconstruction information from the picture buffer module to generate an action estimation information; The motion estimation information is input to the discrete cosine transform module, and all the above actions are repeated.

The picture reconstruction module can perform file size compression and effective video lossless compression on the content of the motion compensation information by using the sample difference prediction unit and the Columbus coding unit, thereby effectively reducing the transmission to the screen. The data size between the buffer modules reduces the occupied bandwidth transmitted to the picture buffer module 80 by 30%; in addition, the picture can be predicted in a plurality of directions to speed up the processing. Therefore, the invention can achieve the purpose of fast reaction time, low memory bandwidth consumption and low power consumption.

1‧‧‧ initial screen

10‧‧‧Quantification unit

11‧‧‧Quantitative information

20‧‧‧Anti-quantization unit

21‧‧‧Anti-quantification information

30‧‧‧Discrete cosine conversion unit

31‧‧‧Discrete Cosine Transform Information

40‧‧‧anti-discrete cosine transform unit

41‧‧‧Inverse Discrete Cosine Transform Information

50‧‧‧Action Compensation Unit

51‧‧‧Action Compensation Information

60‧‧‧Action Estimation Unit

61‧‧‧Action estimation information

70‧‧‧ Screen Reconstruction Module

71‧‧‧ Screen reconstruction information

72‧‧‧Sampling difference prediction unit

73‧‧‧Columbus coding unit

80‧‧‧ Picture buffer module

90‧‧‧ Entropy coding module

91‧‧‧ bit flow

1 is a schematic diagram showing the architecture of the distortion-free video compression coding system of the present invention; FIG. 2 is a schematic diagram showing sampling of the distortion-free video compression coding system of the present invention; and FIG. 3 is a flowchart showing the distortion-free video compression coding system of the present invention. schematic diagram.

The embodiments of the present invention are described by way of specific examples, and those skilled in the art can readily appreciate the other advantages and advantages of the present invention. The invention may be embodied or applied in various other specific embodiments, and various modifications and changes may be made without departing from the spirit and scope of the invention.

It is to be understood that the structure, the proportions, the size and the like of the present invention are intended to be used in conjunction with the disclosure of the specification, and are not intended to limit the scope of the invention. The conditions are not technically meaningful, and any modification of the structure, change of the proportional relationship or adjustment of the size should be disclosed in the present invention without affecting the effects and achievable effects of the present invention. The technical content can be covered.

In the following, a distortionless video compression coding system will be described in accordance with an embodiment of the present invention.

Referring to FIG. 1 to FIG. 3, a distortion-free video compression coding system according to an embodiment of the present invention includes: a quantization module, a discrete cosine transform module, a motion compensation module, and a picture re-modeling. Group 70 and a picture buffer module 80.

The quantization module is configured to generate at least one of a quantized information 11 and an inverse quantized information 21. In an embodiment of the present invention, the quantization module includes a quantization (Q) unit 10 and an inverse quantization (IQ) unit 20; wherein the quantization unit 10 is configured to receive a discrete cosine transform information 31, And used to generate the quantized information 11; wherein the inverse quantization unit 20 is configured to receive the quantized information 11 and used to generate the inverse quantized information 21.

The discrete cosine transform module is configured to generate at least one of discrete cosine transform information 31 and an inverse discrete cosine transform information 41, and the discrete cosine transform module is coupled to the quantization module. In an embodiment of the present invention, the discrete cosine transform module includes a Discrete Cosne Transform (DCT) unit 30 and an Inverse Discrete Cosne Transform (IDCT) unit 40; wherein the discrete cosine transform unit 30 series for receiving an initial At least one of the picture 1 and the motion estimation information 61 is used to generate discrete cosine transform information 31. The inverse discrete cosine transform unit 40 is configured to receive the inverse quantization information 21 and generate inverse discrete cosine transform information 41. .

The motion compensation module is configured to generate at least one of motion compensation information 51 and motion estimation information 61, and the motion compensation module is coupled to the discrete cosine transform module. In an embodiment of the present invention, the motion compensation module includes a Motion Compensation unit 50 and a Motion Estimation unit 60. The motion compensation unit 50 is configured to receive the inverse discrete cosine transform information. And used to generate motion compensation information 51; wherein the motion estimation unit 60 is configured to receive a picture reconstruction information 71 and used to generate motion estimation information 61.

The screen reconstruction module 70 is configured to generate a screen reconstruction information 71 and is connected to the motion compensation module. The screen reconstruction module 70 is a Sample of Difference Encoder (SODE), which includes a Sample of Difference Prediction unit 72 and a Golomb-Rice coding unit. 73. The sampling difference prediction unit 72 is configured to perform prediction sampling in the complex direction on the content of the motion compensation information 51. As shown in FIG. 2, the sampling difference prediction unit 72 simultaneously samples the upward and right prediction results, and finally the phase. The prediction direction of the picture is obtained by adding the minimum value; then, encoding is performed by the Columbus encoding unit 73, that is, the data amount is compressed by the motion compensation information 51 to generate the picture reconstruction information 71 with a small amount of data.

The picture buffer module 80 is used to temporarily store the picture reconstruction information 71 and is connected to the picture reconstruction module 70. In this embodiment, the picture buffer module 80 is an external storage memory.

In this embodiment, an entropy coding module 90 is included, and the entropy coding module 90 is configured to receive the quantized information 11 and generate a bit stream 91.

Referring to FIG. 3, when the embodiment of the present invention is used, the method includes the following steps: inputting an initial picture to a discrete cosine transform module to generate a discrete cosine transform information (step S01); Converting the information into a quantization module to generate a quantization information (step S02); inputting the quantization information to the quantization module to generate an inverse quantization information (step S03); inputting the inverse quantization information to the discrete a cosine transform module to generate an inverse discrete cosine transform information (step S04); input the inverse discrete cosine transform information to a motion compensation module to generate an action compensation information (step S05); To a picture reconstruction module to generate a picture reconstruction information and temporarily store the picture reconstruction information in a picture buffer module (step S06); the motion compensation module extracts the screen reconstruction information from the image buffer module to generate an action estimation information (step S07); and inputs the motion estimation information to the discrete cosine transform module. And all the above actions are repeated (step S08).

Therefore, the screen reconstruction module 70 can perform file size compression and effective video lossless compression on the content of the motion compensation information 51 by using the sampling difference prediction unit 72 and the Columbus encoding unit 73. The size of the data transmitted between the picture buffer modules 80 is reduced. In the embodiment of the present invention, the original file size can be effectively reduced to reduce the occupied bandwidth of the picture buffer module 80 by 30%, and The direction prediction screen speeds up the processing. Therefore, the invention can achieve the purpose of fast reaction time, low memory bandwidth consumption and low power consumption.

Although the embodiments have been described with reference to the embodiments of the present invention, it will be understood that Inside. In particular, various changes and modifications can be made in the components and/or arrangements of the subject combination. Alternative uses will be apparent to those skilled in the art, in addition to variations and modifications in the component parts and/or arrangements.

1‧‧‧ initial screen

10‧‧‧Quantification unit

11‧‧‧Quantitative information

20‧‧‧Anti-quantization unit

21‧‧‧Anti-quantification information

30‧‧‧Discrete cosine conversion unit

31‧‧‧Discrete Cosine Transform Information

40‧‧‧anti-discrete cosine transform unit

41‧‧‧Inverse Discrete Cosine Transform Information

50‧‧‧Action Compensation Unit

51‧‧‧Action Compensation Information

60‧‧‧Action Estimation Unit

61‧‧‧Action estimation information

70‧‧‧ Screen Reconstruction Module

71‧‧‧ Screen reconstruction information

72‧‧‧Sampling difference prediction unit

73‧‧‧Columbus coding unit

80‧‧‧ Picture buffer module

90‧‧‧ Entropy coding module

91‧‧‧ bit flow

Claims (10)

A distortion-free video compression coding method includes the steps of: inputting an initial picture to a discrete cosine transform module to generate a discrete cosine transform information; and inputting the discrete cosine transform information to a quantization module to generate a quantization Information; inputting the quantized information to the quantization module to generate an inverse quantization information; inputting the inverse quantization information to the discrete cosine transform module to generate an inverse discrete cosine transform information; and converting the inverse discrete cosine transform information Inputting to a motion compensation module to generate an action compensation information; inputting the motion compensation information to a picture reconstruction module to generate a picture reconstruction information, and temporarily storing the picture reconstruction information in a picture buffer module; The motion compensation module captures the image reconstruction information from the image buffer module to generate an motion estimation information; and inputs the motion estimation information to the discrete cosine transform module, and repeats all the above actions; The picture reconstruction module is a sample difference encoder including a sample difference prediction unit and a Columbus code And the sampling difference prediction unit is configured to perform prediction sampling in the complex direction of the content of the motion compensation information, wherein the sampling difference prediction unit simultaneously samples the upward and right prediction results of the content of the motion compensation information. Finally, adding the minimum value of the upward and rightward prediction results to obtain the prediction direction of the picture reconstruction information; and then encoding by the Columbus coding unit, that is, compressing the data amount of the action compensation information to generate the The picture reconstruction information reduces the occupied bandwidth of the picture buffer module by 30%. The distortion-free video compression coding method according to claim 1, wherein the discrete cosine transform module comprises a discrete cosine transform unit and an inverse discrete cosine transform unit; wherein the discrete cosine transform unit is configured to receive the At least one of an initial picture and the motion estimation information is used to generate the discrete cosine transform information; wherein the inverse discrete cosine transform unit is configured to receive the inverse quantization information and generate the inverse discrete cosine transform information. The distortion-free video compression coding method of claim 1, wherein the quantization module comprises a quantization unit and an inverse quantization unit, wherein the quantization unit is configured to receive the discrete cosine conversion information and use Generating the quantized information; wherein the inverse quantization unit is configured to receive the quantized information and used to generate the inverse quantized information. The non-distortion video compression coding method of claim 1, wherein the motion compensation module comprises a motion compensation unit and an action estimation unit, wherein the motion compensation unit is configured to receive the inverse discrete cosine transform The information is used to generate the motion compensation information; wherein the motion estimation unit is configured to receive the image reconstruction information and generate the motion estimation information. The undistorted video compression coding method according to claim 1, further comprising an entropy coding module, wherein the entropy coding module is configured to receive the quantization information and generate a bit stream. A distortion-free video compression coding system includes: a quantization module for generating at least one of a quantized information and an inverse quantization information; and a discrete cosine transform module for generating a discrete cosine transform information And at least one of the inverse discrete cosine transform information, and the discrete cosine transform module is coupled to the quantization module; and the motion compensation module is configured to generate at least one of motion compensation information and an motion estimation information And the motion compensation module is coupled to the discrete cosine transform module; a picture reconstruction module is configured to generate a picture reconstruction information and is connected to the motion compensation module; and a picture buffer module The screen reconstruction device is configured to receive an initial image; the image reconstruction module is a sample difference encoder, which includes a a sampling difference prediction unit and a Columbus coding unit, wherein the sampling difference prediction unit is configured to perform prediction sampling in a complex direction on the content of the motion compensation information. , The sample difference prediction compensation unit based the operation information of the same content Sampling up and right prediction results, and finally adding the minimum value of the upward and right prediction results to obtain the prediction direction of the picture reconstruction information; then coding with the Columbus coding unit, that is, compensating the information for the action The amount of data is compressed to generate the picture reconstruction information, so that the occupied bandwidth transmitted to the picture buffer module is reduced by 30%. The distortion-free video compression coding system of claim 6, wherein the discrete cosine transform module comprises a discrete cosine transform unit and an inverse discrete cosine transform unit; wherein the discrete cosine transform unit is configured to receive At least one of the initial picture and the motion estimation information is used to generate the discrete cosine transform information; wherein the inverse discrete cosine transform unit is configured to receive the inverse quantization information and generate the inverse discrete cosine transform information . The distortion-free video compression coding system of claim 6, wherein the quantization module comprises a quantization unit and an inverse quantization unit, wherein the quantization unit is configured to receive the discrete cosine conversion information and use Generating the quantized information; wherein the inverse quantization unit is configured to receive the quantized information and used to generate the inverse quantized information. The non-distortion video compression coding system of claim 6, wherein the motion compensation module comprises an action compensation unit and an action estimation unit, wherein the motion compensation unit is configured to receive the inverse discrete cosine transform The information is used to generate the motion compensation information; wherein the motion estimation unit is configured to receive the image reconstruction information and generate the motion estimation information. The undistorted video compression coding system according to claim 6, wherein the entropy coding module is configured to receive the quantization information and generate a bit stream.