DE19944301A1 - Method and arrangement for processing or inverse processing of at least part of an image, as well as computer program products and computer-readable storage medium - Google Patents

Abstract

According to the invention, at least one given coding value is subtracted from each coding information for treating at least one component of an image with pixels, whereby coding information is allocated to said pixels. The pixels containing the subtracted coding information are subjected to a transformation.

Description

The invention relates to methods and arrangements for processing processing or inverse processing of at least part of a Picture.

Such methods and such arrangements are known from [1].

From [1] is a method for coding a digitized Known image.

The digitized image has pixels that Codie information is assigned.

Under coding information is brightness information (Lumi nance information) or color information (chrominance information) to understand.

The pixels are divided into blocks of 8. 8 pixels or also 16. 16 pixels grouped and the image blocks are called a Discrete Cosine Transformation (DCT) Subjected to transformation coding.

Below is a transformation coding To understand spectral transformation, e.g. B. the DCT, the Dis krete sine transformation or also the discrete Fourier Transformation.

The result of the transformation are transformation coefficients which are run-length encoded (run- Length coding) and entropy coding, e.g. B. a Huff man coding.  

In the context of the known coding method, two un Different coding modes are used, which as INTER- Image coding mode or INTRA image coding mode become.

In the context of the INTER image coding mode, the differences zen of coding information of two in time following pictures DCT-transformed and encoded.

As part of the INTRA image coding mode, the entire Co dation information of an image DCT-transformed and co dated.

A disadvantage of the method in the INTRA image coding mode in particular the great need for bits for coding the Bil of.

As part of the decoding of the encoded picture, that is, the Re construction of the coded image, the transformi on coefficients of an inverse quantization and an in verse transformation coding of the individual transforms on coefficients to reconstruct individual image blocks structure.

The reconstructed image blocks are used using ei The motion compensation process reconstructs the image.

The invention is therefore based on the problem, method as above how to specify arrangements for editing an image with a coding of the transformation coefficients ten with a reduced compared to the known method Number of bits is enabled.

The problem is caused by the procedures, the arrangements as well computer program products and computer readable Storage media with the features according to the independent Pa claims resolved.  

In a method for processing at least a part of egg nes image with pixels, which encoding information is ordered, at least one predetermined coding value subtracted from each coding information. The pixels with the subtracted coding information one Undergone transformation.

In a method for inverse processing at least one Part of a transformed image with transformed image score that transformed subtracted coding information tion and at least one predetermined coding value net, the transformed pixels with the trans formed subtracted coding information of an inverter undergone transformation. The default coding value each subtracted inversely transformed coding formation added.

An arrangement for processing at least a part of an image with pixels to which coding information is assigned contains a processor which is set up in such a way that the following steps can be carried out:

• - At least one predetermined coding value is used by everyone Coding information subtracted, and
• - The pixels with the subtracted coding information are undergoing a transformation.

An arrangement for the inverse processing of at least a part of a transformed image with transformed pixels, to which transformed subtracted coding information and at least one predetermined coding value is assigned, has a processor which is set up in such a way that the following steps can be carried out:

• - The transformed pixels with the transformed sub traced coding information becomes an inverse Undergone transformation, and  
• - The coding value becomes inversely trans every subtracted formed coding information added.

A computer program product comprises a computer-readable storage medium on which a program is stored which, after it has been loaded into a memory of the computer, enables a computer to carry out the following steps for processing at least a part of an image with image points which are associated with coding information is:

• - At least one predetermined coding value is used by everyone Coding information subtracted, and
• - The pixels with the subtracted coding information are undergoing a transformation.

A program is stored on a computer-readable storage medium, which, after it has been loaded into a memory of the computer, enables a computer to carry out the following steps for processing at least a part of an image with pixels to which coding information is assigned:

• - At least one predetermined coding value is used by everyone Coding information subtracted, and
• - The pixels with the subtracted coding information are undergoing a transformation.

Another computer program product comprises a computer-readable storage medium on which a program is stored which, after it has been loaded into a memory of the computer, enables a computer to carry out the following steps for the inverse processing of at least a part of a transformed image with transformed pixels , which is assigned transformed subtracted coding information and at least one predetermined coding value:

• - The transformed pixels with the transformed sub traced coding information becomes an inverse Undergone transformation, and  
• - The coding value becomes inversely trans every subtracted formed coding information added.

A program is stored on a further computer-readable storage medium, which enables a computer, after it has been loaded into a memory of the computer, to carry out the following steps for the inverse processing of at least part of a transformed image with transformed pixels, which transformed subtracted Co dation information and at least one predetermined coding value is assigned:

• - The transformed pixels with the transformed sub traced coding information becomes an inverse Undergone transformation, and
• - The coding value becomes inversely trans every subtracted formed coding information added.

An advantage of the invention can be seen in particular in that within the framework of coding an image in the INTRA Image coding mode compared to the known procedure a considerable saving in the required bits is achieved.

Preferred developments of the invention result from the dependent claims.

The further developments described below relate on the procedures, the orders and the computer pro gram products and computer readable storage media.

The invention can be used both in software and in hardware, for example using a special electrical Circuit.

In a further training the procedure is based on the whole Image applied.  

In a further embodiment, the predetermined code averaging value determined by averaging the Co dation information, the pixels of the part of the image assigned. In this way, it becomes a coding value which averages the average content of coding in formation per pixel in the image or part to be coded represents the image. By subtracting the mean of each coding information of a pixel and coding The average itself becomes another on average Reduction of the total for coding the image or part required bits reached.

The part of the picture can be made using a process be determined for object detection within the image, which is used as part of the image in which the determined object is located.

The portion of the image can also be viewed using a can detection method can be determined.

In a further preferred development, it is hen that

• - The image content of the part with various saved Reference object models is compared,
• - a predefined distribution for each reference object model of coding values is stored,
• - The reference object model is determined, the regarding a predetermined similarity criterion to the part of the Bil the best match,
• - the code according to the shape of the reference object model from the coding information of the part of the Image are subtracted.

In a further embodiment it is provided that

• - The determined reference object with regard to the shape and Size is subjected to mapping so that the referencing z object better with respect to a given similar  criterion corresponds to the part of the image as without this illustration,
• - The coding values of the reference object shown one Undergo interpolation.

The invention is preferred for coding and / or decoding digital image.

An embodiment of the invention is shown in the figures and will be explained in more detail below.

Show it

Fig. 1 is a sketch in which the basic sequence of the process according to the embodiment is shown;

Figure 2 shows an arrangement of two computers, a camera and a screen, which carried the encoding, transmission and decoding and displaying the image data.

Fig. 3 is a sketch of an apparatus for block-based Co dation of a digitized image;

FIG. 2 shows an arrangement with two computers 202 , 208 and a camera 201 .

A camera 201 is connected to a first computer 202 via a line 219 . The camera 201 transmits recorded images 204 to the first computer 202 . The first computer 202 has a first processor 203 which is connected to an image memory 205 via a bus 218 . With the first processor 203 of the first computer 202 , a method for image coding is carried out. Image data 206 encoded in this way is transmitted from the first computer 202 to a second computer 208 via a communication link 207 , preferably a line or a radio link. The second computer 208 contains a second processor 209 , which is connected to an image memory 211 via a bus 210 . A method for image decoding is carried out with the second processor 209 .

Both the first computer 202 and the second computer 208 each have a screen 212 or 213 on which the image data 204 are visualized, the visualization on the screen 212 of the first computer 202 usually taking place only for control purposes. Input units are provided for operating both the first computer 202 and the second computer 208 , preferably a keyboard 214 or 215 , and a computer mouse 216 or 217 .

The image data 204 , which are transmitted from the camera 201 via the line 219 to the first computer 202 , are data in the time domain, while the data 206 , which are transmitted from the first computer 202 to the second computer 208 via the communication link 207 , Image data in the spectral range are rich.

The decoded image data are displayed on the screen 213 .

In the INTRA picture coding mode, the whole to be coded Image subjected to DCT, not just differential information on two successive pictures.

At the beginning of the coding of the picture in the INTRA picture coding mode, a coding value CW of all coding information is formed over all picture elements of the picture according to the following rule (step 100 , cf. FIG. 1):

being with

• - i denotes a row index with which the row is inside half of the image in which the pixel is located,
• - j denotes a column index with which the column in is specified within the image in which the pixel lies,
• x _{i, j} denotes the coding information which is assigned to the pixel which is located at the position determined by the i th row and the j th column,
• N denotes a number of lines contained in the image becomes,
• M denotes a number of columns contained in the image becomes.

In a further step (step 101 ), the coding value CW is subtracted from all coding information x _{i, j of} the pixels, with which subtracted coding information Δx _{i, j is} formed in accordance with the following rule:

Δx _{i, j} = x _{i, j} - CW ∀ i, j. (2)

The subtracted coding information Δx _{i, j} is then subjected to a discrete cosine transformation (step 102 ) and quantized in accordance with the known block-based image coding method. Quantized transformation coefficients 313 are formed by the quantization (cf. FIG. 3).

The quantized transform coefficients 313 are stored in a transform coefficient matrix.

The quantized transformation coefficients 313 are then read out in accordance with the Zig-Zag scan method

The transformation coefficients read out become one Run length coding and a Huffman  Coding as entropy coding in an entropy coding unit subjected.

The entropy-coded transformation coefficients and the coding value CW are transmitted as coded image data 317 via a channel, preferably a line or a radio link, to a decoder.

The selected quantized transform coefficients 313 are also fed to a reverse path.

An inverse quantization of the quantized spectral coefficients 313 takes place in the inverse quantization unit 315 . Spectral coefficients 318 obtained in this way are fed to an inverse transformation coding unit 319 (inverse discrete cosine transformation, IDCT). Reconstructed coding values (also differential coding values) 320 are supplied to an adder 321 in the differential image mode. The adder 321 also receives coding values of an image block which result from a temporally preceding image after motion compensation has already been carried out. Reconstructed image blocks 322 are formed with the adder 321 and stored in an image memory 323 .

Chrominance values 324 of the reconstructed image blocks 322 are fed from the image memory 323 to a motion compensation unit 325 . For brightness values 326 , an interpolation takes place in an interpolation unit 327 provided for this purpose. Using the interpolation, the number of brightness values contained in the respective image block is preferably quadrupled. All brightness values 328 are supplied to both the motion compensation unit 325 and the motion estimation unit 329 . The motion estimation unit 329 also receives the picture blocks of the macro block to be coded in each case (16 × 16 picture elements) via the connection 334 . In the motion estimation unit 329 , the motion estimation takes into account the interpolated brightness values ("motion estimation on a half-pixel basis").

The result of the motion estimation is a motion vector 330 , by means of which a local shift of the selected macroblock from the temporally preceding image to the macroblock 302 to be coded is expressed.

Both luminance information and chrominance based on the by the motion estimation unit 329, he mediated macroblock by the motion vector 330 ver inserted and sub tracted from the encoding values of the macroblock 302 (see data path 331).

The result of the motion estimation is thus the motion vector 330 with two motion vector components, a first motion vector component BV _x and a second motion vector component BV _y along the first direction x and the second direction y:

The coded image data 317 are received and stored in the decoder.

The decoder has the same structure as the back upward path of the encoder with the difference that at the beginning of the Decoding an inverse entropy coding is provided.

After entropy decoding has taken place, the quantized transformation coefficients 313 are present, which are written into the transformation coefficient matrix to be reconstructed in the order according to the Zig-Zag scan, ie are stored so that a correct reconstruction of the image can take place.

The one in the reconstructed transformation coefficient matrix stored transformation coefficients become inverse quantized and an inverse transformation coding un educated.

Using the motion vectors transmitted with the image using motion compensation for the one reconstructed individual image blocks.

The following are alternatives to those shown above Exemplary embodiments shown:

In an alternative embodiment it is provided that the subtracti only for at least part of the image on a coding value CW.

In a further alternative embodiment, an Ob detection process carried out. After performing the Object recognition method is an object recognized in the image been. The object detection method can have known edge ends Use the detection method to determine the edges of the object teln. Pixels lying within the determined object form the part of the image, possibly also with the image point ten that lie on the determined edge.

In this case, the average value is determined by Forming an average over the coding information of the pixels in the object.

In a further alternative embodiment there are several Reference object models saved. In the reference object mo dent is for each pixel within the reference object model has its own predefined coding value saved.

Using a reference object model will give it a typical look of such an object.  

The determined object O is compared with all stored reference object models and a similarity value AW is formed for each reference object model and the object O according to the following rule:

being with

• - x _{ir, jr} each denotes a coding value which is assigned to a pixel of the reference object model, where this pixel is located at the substantially same relative position as the pixel of the coding information considered, which is identified by x _{i, j}

referred to as.

Match the shape and / or size of the reference object model and the part of the picture do not match with each other, so subjected the reference object model to a mapping in such a way that shape and / or size are essentially the same voices.

After the mapping is complete, the coding information the pixels are interpolated so that a meaningful comparison to determine the subtracted coding information is guaranteed.

In such a case, the decoder, in which also the reference object models are stored, the specification of the used reference object model and possibly an image of the reference object model to the actually used reference object model, transferred.  

The following publications are cited in this document:
[1] ITU-T Draft Recommendation H.263, Video Coding for Low Bitrate Communication, September 1997

Claims (18)

1. Method for processing at least part of a picture with pixels to which coding information is assigned,

• - in which at least one predetermined coding value is subtracted from each of the coding information, and
• - In which the pixels with the subtracted coding information are subjected to a transformation.

2. The method according to claim 1, in which the method is applied to the entire image. 3. The method according to claim 1 or 2, in which the predetermined coding value is determined by Forming an average of the coding information Pixels of the part of the image are assigned. 4. The method according to any one of the preceding claims, in which the part of the image is determined by application a method for object detection within the image, which is used as part of the image in which the determined object is located. 5. The method according to any one of the preceding claims, in which the part of the image is determined by application an edge detection method. 6. The method according to any one of the preceding claims,

• in which the image content of the part is compared with different stored reference object models,
• - in which a predetermined distribution of coding values is stored for each reference object model,
• in which the reference object model is determined that best matches the part of the image with regard to a predefined similarity criterion,
• - In which, according to the shape of the reference object model, the coding values are subtracted from the coding information of the part of the image.

7. The method according to claim 6,

• in which the determined reference object is subjected to an image with regard to the shape and size, so that the reference object corresponds better to the part of the image with respect to a predetermined similarity criterion than without this image,
• - In which the coding values of the reference object shown are subjected to interpolation.

8. The method according to any one of the preceding claims, used to encode a digitized image. 9. Method for inverse processing of at least a part of a transformed image with transformed pixels, to which transformed subtracted coding information and at least one predetermined coding value are assigned,

• - In which the transformed pixels with the transformed subtracted coding information are subjected to an inverse transformation, and
• - in which the coding value of each subtracted inversely transformed coding information is added.

10. The method according to claim 9, used to decode the transformed image. 11. Arrangement for processing at least part of an image with pixels to which coding information is assigned,
with a processor that is set up in such a way that the following steps can be carried out:

• - At least one predetermined coding value is subtracted from each coding information, and
• - The pixels with the subtracted coding information are subjected to a transformation.

12. Arrangement according to claim 11, used to encode the transformed image. 13. Arrangement for the inverse processing of at least a part of a transformed image with transformed image points to which transformed subtracted coding information and at least one predetermined coding value is assigned,
with a processor that is set up in such a way that the following steps can be carried out:

• - The transformed pixels with the transformed subtracted coding information are subjected to an inverse transformation, and
• - The coding value is added to each subtracted inversely transformed coding information.

14. Arrangement according to claim 13, used for decoding a digitized picture. 15. A computer program product comprising a computer-readable storage medium on which a program is stored which enables a computer, after it has been loaded into a memory of the computer, to carry out the following steps for processing at least a part of an image with pixels, to which coding information is assigned:

• - At least one predetermined coding value is subtracted from each coding information, and
• - The pixels with the subtracted coding information are subjected to a transformation.

16. Computer-readable storage medium, on which a program is stored, which enables a computer, after it has been loaded into a memory of the computer, to carry out the following steps for processing at least a part of an image with pixels to which coding information is assigned:

• - At least one predetermined coding value is subtracted from each coding information, and
• - The pixels with the subtracted coding information are subjected to a transformation.

17. A computer program product comprising a computer-readable storage medium on which a program is stored which enables a computer, after it has been loaded into a memory of the computer, to carry out the following steps for the inverse processing of at least a part of a transformed image transformed pixels, to which transformed subtracted coding information and at least one predetermined coding value are assigned:

• - The transformed pixels with the transformed subtracted coding information are subjected to an inverse transformation, and
• - The coding value is added to each subtracted inversely transformed coding information.

18. Computer-readable storage medium on which a program is stored which enables a computer, after it has been loaded into a memory of the computer, to carry out the following steps for the inverse processing of at least one part of a transformed image with transformed pixels, the transformed subtracted Coding information and at least one predetermined coding value is assigned:

• - The transformed pixels with the transformed subtracted coding information are subjected to an inverse transformation, and
• - The coding value is added to each subtracted inversely transformed coding information.