Quantization Method and Apparatus in Encoding/Decoding

Abstract

A method and apparatus is provided for encoding image data comprising receiving image data and an initial quantization matrix for encoding the image data; converting the initial quantization matrix into a modified scaling matrix, wherein the modified scaling matrix has a plurality of coefficients divided into different frequency bands; adjusting a value corresponding to at least one of the plurality of coefficients to control quality of an image to be encoded; generating a parameter set having a plurality of band parameters, each band parameter corresponding to a coefficient of the modified scaling matrix; writing the encoded image data and the parameter set having the plurality of band parameters into a bitstream; and transmitting the bitstream over the network to a decoding device. The present disclosure may improve quality of the encoded image efficiently.

What is claimed is: 
     
         1 . A method performed by an encoding device in a network for encoding image data, comprising:
 receiving image data and an initial quantization matrix for encoding the image data;   converting the initial quantization matrix into a modified scaling matrix, wherein the modified scaling matrix has a plurality of coefficients divided into different frequency bands;   adjusting a value corresponding to at least one of the plurality of coefficients to control quality of an image to be encoded by the image data;   encoding the image data by performing quantization on the image data using the modified scaling matrix;   generating a parameter set having a plurality of band parameters, each band parameter corresponding to a coefficient of the modified scaling matrix;   writing the encoded image data and the parameter set having the plurality of band parameters into a bitstream; and   transmitting the bitstream over the network to a decoding device.   
     
     
         2 . The method of  claim 1 , wherein the modified scaling matrix comprises at least one of the following items:
 a transformed coefficient matrix, a quantization matrix, a scaling transform matrix, and a transform matrix having bands correspondence with the transformed coefficient matrix and quantization matrix.   
     
     
         3 . The method of  claim 1 , further comprising:
 assigning a band distribution parameter corresponding to a distribution configuration of the coefficients in the modified scaling matrix, and performing the corresponding quantization operation in the process of encoding respectively using the modified scaling matrix and the band distribution parameter.   
     
     
         4 . The method of  claim 1 , wherein the modified scaling matrix comprises the coefficients of positive weighting and negative weighting, the positive weighting indicates that a frequency component is enhanced after being weighted with the coefficients, and the negative weighting indicates that a frequency component is reduced after being weighted with the weighting coefficient. 
     
     
         5 . The method of  claim 1 , further comprising:
 establishing the modified scaling matrix by computation by using scaling value coefficients.   
     
     
         6 . The method of  claim 1 , further comprising:
 establishing the modified scaling matrix by computation by using an arithmetic operation of addition, subtraction, multiplication, division, shifting, filtering or their combinations.   
     
     
         7 . The method of  claim 1 , further comprising:
 generating a set of parameters according to the modified scaling matrix, writing the set of parameters into a header of an encoding unit and the encoding unit comprises a sequence, a picture group, a picture, a slice or a macroblock.   
     
     
         8 . The method of  claim 1 , further comprising:
 each of the plurality of band parameters comprises a plurality of bits, wherein a number of the bits is constant.   
     
     
         9 . The method of  claim 1 , further comprising:
 each of the plurality of band parameters comprises a plurality of bits, wherein a number of the bits is variable.   
     
     
         10 . An apparatus for performing encoding on image data, comprising:
 a non-transitory memory storing executable instructions; and   a processor coupled to the memory that reads and executes the instructions to carry out operations of:   receiving image data and an initial quantization matrix for encoding the image data;   converting the initial quantization matrix into a modified scaling matrix, wherein the modified scaling matrix has a plurality of coefficients divided into different frequency bands;   adjusting a value corresponding to at least one of the plurality of coefficients to control quality of an image to be encoded by the image data;   encoding the image data by performing quantization on the image data using the modified scaling matrix;   generating a parameter set having a plurality of band parameters, each band parameter corresponding to a coefficient of the modified scaling matrix;   writing the encoded image data and the parameter set having the plurality of band parameters into a bitstream; and   transmitting the bitstream over the network to a decoding device.   
     
     
         11 . The apparatus of  claim 10 , wherein the processor is further configured to write the parameter set into a header of the bitstream for transmitting the image. 
     
     
         12 . The apparatus of  claim 10 , wherein the memory further stores a band distribution parameter corresponding to a distribution configuration of the scaling values in the modified scaling matrix, and the parameter set includes the band distribution parameter. 
     
     
         13 . The method of  claim 10 , further comprising:
 each of the plurality of band parameters comprises a plurality of bits, wherein a number of the bits is constant.   
     
     
         14 . The method of  claim 10 , further comprising:
 each of the plurality of band parameters comprises a plurality of bits, wherein a number of the bits is variable.