JP3416034B2 - Coded signal processing device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coded signal processing device, for example, a DCT (Discrete Cosine Transfor).
m) It can be applied to an apparatus that processes an image signal encoded by an encoding method.

[0002]

[Prior art]

Literature: Data compression technology (edited by Nikkei Electronics) Today, much information tends to be compression-encoded in a form suitable for transmission and storage. Especially in the case of video signals, DC
The compression coding of information has been advanced by applying various high efficiency coding methods such as the T coding method. Along with this, the mainstream editing method is shifting from linear editing to non-linear editing.

FIG. 2 shows a functional configuration of a processing device used for nonlinearly editing a video signal. Note that the coded signals V1 and V2 input to this processing device are already D
It is CT encoded and is read from a storage medium such as a hard disk.

Usually, the processing device will use such a coded signal V1.
And V2 are input to the decoding circuits 1 and 2, they are inverse DCT encoded and decoded into the original video signal (luminance signal and color difference signal). The signal form of the decoded video signal is a component video signal.

Since the processing in the processing circuit 3 (image processing such as editing, conversion and level conversion) is usually performed in the form of a composite video signal, the processing circuit 3 of FIG. , It is designed to be processed after being converted into a composite video signal. The processed video signal is converted into a component video signal again and then output.

However, the video signal after such processing is
When it is stored in a storage medium such as a hard disk or needs to be transmitted via a network such as a digital line, an encoding circuit provided in its output stage in order to reduce the storage capacity and communication band. DC again at 4
It is T-coded.

[0007]

However, in this method, after processing (after image processing such as editing, conversion, and level conversion)
When the image of (1) is stored again in a storage medium or transmitted over a network, it is necessary to decode the entire image and then re-encode, so that there is a problem that the image quality deteriorates while the processing is repeated.

In particular, in the case of lossy processing such as DCT encoding processing, the deterioration of image quality due to the repetition of processing was remarkable. Further, even in the case of simply changing the signal level, color, etc. of the video signal, in the conventional method, since it is necessary to perform processing after decoding once, decoding and encoding are repeated many times, and the image quality is improved. It was a factor of deterioration.

[0009]

(A) In order to solve such a problem, in the first invention, one or more encoded signals lossy-encoded for each predetermined unit are input, and a predetermined process is performed. The coded signal processing device for applying and outputting is provided with the following means.

That is, (1) decoding means for decoding each one or a plurality of coded signals, processing means for subjecting the decoded signal to predetermined processing, and lossy coding of the processed signal again A first signal system having a coding means for outputting; (2) a second signal system for directly transmitting one or more coded signals without decoding; (3) first and second A re-encoded signal and an encoded signal are input from the signal system, and the correlation between both signals is determined by a predetermined unit, and comparing means, (4)
When a high correlation is recognized by the comparison means, the encoded signal before decoding output from the second signal system is selectively output, and when a low correlation is recognized by the comparison means, And a selecting means for selectively outputting the re-encoded signal output from the first signal system.

As described above, according to the present invention, the correlation between the re-encoded signal that has been subjected to the predetermined processing and the coded signal that has not been subjected to the predetermined processing and that has not been subjected to the decoding is determined for each predetermined unit. When a high correlation is found, the coded signal itself before decoding, which is the original signal, is selectively output. Therefore, when a predetermined process is performed only on a part of all coded signals, Quality degradation can be kept to a minimum.

(B) In the second invention, in the coded signal processing device for inputting one or a plurality of coded signals lossy-coded for each predetermined unit, performing a predetermined process, and outputting the processed signal, The following means should be provided.

That is, (1) DC level generating means for generating an arbitrary DC level, and (2) level varying means for changing only the DC component of the encoded signal by the DC level generated by the DC level generating means. Be prepared.

As described above, in the present invention, only the target DC component is changed without decoding the coded signal to be processed, so that the deterioration of the quality associated with the decoding and the re-encoding is reduced. It can be lost.

[0015]

DETAILED DESCRIPTION OF THE INVENTION

(A) First Embodiment (A-1) Configuration of First Embodiment Hereinafter, a case where the coded signal processing device according to the present invention is applied to a processing device which processes a signal after DCT coding, A description will be given with reference to the drawings.

FIG. 1 shows a functional block configuration of a processing apparatus according to the first embodiment. Note that, in FIG. 1, the same parts as those in FIG.
The display of the input means used by the operator to give various instructions to the processing device is omitted.

As shown in FIG. 1, also in the case of the processing apparatus according to this embodiment, the coded signals V1 and V2 which have already been DCT coded are converted into the original video signal (including the still picture) by the inverse DCT code. Decoding circuits 1 and 2 used for decoding
And a processing circuit 3 for converting the video signal decoded in these circuits into a composite video signal for processing, and a component part of a coding circuit 4 for again DCT-coding and outputting the video signal processed by the processing circuit 3. Is the same as in the case of the conventional circuit.

The difference is that (1) the re-encoded signal V3 output from the encoding circuit 4 and the original signals (encoded signals) V1 and V
2 and block (m pixels x n lines (m and n are integers))
Comparing circuits 5 and 6 for comparison in units, (2) Based on the result of this comparison, the original signal (encoded signal) V1 and post-decoding processing (image processing such as editing, conversion, level conversion, etc.) are performed, and code is performed again A switching circuit (SW) 7 that selectively outputs one of the encoded re-encoded signals V3 in block units, and (3) the original signal (encoded signal) V2 and the decoded signal based on the comparison result. Switching circuit (SW) for selecting either one of the re-encoded signal V3 re-encoded through processing
8 and (4) a combining circuit 9 for combining the outputs of the switching circuits and outputting the combined output V4.

Here, the comparison circuits 5 and 6 obtain the correlations of the two input signals, and when the correlation value is larger than a predetermined threshold value (that is, when the correlations are high and almost coincide with each other). ), A selection signal is given to the switching circuits (SW) 7 and 8 to select an encoded signal before decoding, and when the correlation value is smaller than a predetermined threshold value (that is, the correlation is low and some change is recognized). In this case, a selection signal is given to the corresponding switching circuits (SW) 7 and 8 so as to select the re-encoded signal V3 which has been re-encoded through the processing after decoding.

(A-2) Content of image processing operation Next, an operation when the image processing apparatus having the above-mentioned configuration is used to perform arbitrary image processing on a video signal will be described.

Generally, when performing image processing on a video signal,
In some cases, the entire screen is targeted for processing, but in many cases, only a part of the entire screen is processed. That is, with respect to the image portion that is not the processing target, the processing circuit 3 does not add any processing to the encoding circuit 4.
It is often encoded again in. Here, the processing operation in such a case will be described.

First, a process for a block corresponding to a video portion which is not a processing target will be described. At this time,
In the comparison circuits 5 and 6, the original signal (coded signal) V1
And V2 and a re-encoded signal V3 after re-encoding are highly correlated. Incidentally, since the DCT encoding is a lossy process, the image quality of the re-encoded signal V3 is somewhat deteriorated.

Therefore, for the block for which such a high correlation is recognized, a selection signal is issued so as to select the coded signals V1 and V2 which have not been subjected to any processing (before decoding). As a result, the combining circuit 9 outputs the combined output V4, which is a combination of signals that are not deteriorated in principle.

Next, the processing for the block corresponding to the video portion to be processed will be described. It should be noted that this processing includes not only the case of performing the predetermined processing on both of the encoded signals V1 and V2 but also the case of performing the processing on either one of them.

At this time, in the comparison circuits 5 and 6,
A high correlation cannot be recognized between both the original signals (coded signals) V1 and V2 and the recoded signal V3 of the coding circuit 4. That is, since some processing is added, the correlation between both signals in the block is significantly reduced.

Therefore, for the block in which such a low correlation is recognized, a selection signal is issued so as to select the output V3 of the encoding circuit 4. As a result, the synthesis circuit 9 obtains a synthesis output V4 that synthesizes the changed signals.

Thus, the synthesis circuit 9 sends the synthesized output of the re-encoded signal V3 which has been decoded and encoded only for the block corresponding to the processed image, and the other blocks may be deteriorated. A composite output of the original signals (encoded signals) V1 and V2 without the output is output.

In the block coding method applying the orthogonal transform method such as the DCT coding method, since the coding is independently executed for each block, signals are switched in block units as in the present embodiment. But there is no problem.

The original signals (encoded signals) V1 and V2
Since the switching process can be automatically determined only by the correlation result with the re-encoded signal V3 of the encoding circuit 4, it is not necessary to separately specify a block to be switched using a special control signal such as a key signal.

The simplest method of comparison in the comparison circuits 5 and 6 is to use the difference amount between both inputs as a parameter. However, this is not only the difference value but also the distribution numerical value of the DCT coefficient as a parameter. With this, it becomes possible to select with higher accuracy.

Further, although the present embodiment has been described on the premise that the coding blocks are independent of each other, in the actual processing, the difference of the direct current (DC) component is coded between adjacent blocks. It is common. Therefore, in practice, if only the direct current (DC) component amount is decoded, independent processing can be performed.

In addition, the DCT encoding operation is usually performed by DC
The T calculation process and the subsequent quantization process are also included, but since this is not an absolute condition, the description is omitted.

(A-3) Effects of First Embodiment As described above, according to the processing apparatus of the present embodiment, DC
Even when a T-coded coded signal is processed, the re-coded signal that has been subjected to both decoding and coding is output only for the block to which the process is applied, so that some other processing may be performed. The original signal can be used as it is for the blocks not added, and the image quality deterioration before and after the processing can be minimized.

Further, the determination of the processed block is
By making the determination based on the correlation between the original signals (encoded signals) V1 and V2 and the re-encoded signal V3 of the encoding circuit 4, automatic determination of switching is possible.

(B) Second Embodiment Hereinafter, a case where the coded signal processing device according to the present invention is applied to a processing device for processing a signal after DCT coding will be described with reference to the drawings.

FIG. 3 shows a functional block configuration of a processing apparatus according to the second embodiment. This processing device has a dedicated function of variable processing of the signal level.

This processing apparatus includes a DC level generating circuit 10 for generating a direct current (DC) level and a level changing circuit 11
Composed of.

The DC level generating circuit 10 is a circuit for generating an arbitrary direct current (DC) level according to an instruction from an operator, and the direct current (DC) level generated for each block is supplied to the level converting circuit 11 each time. It is designed to output.

On the other hand, the level conversion circuit 11 supplies the direct current (DC) from the DC level generation circuit 10 to the direct current (DC) level portion of the block in which the level change is instructed in the encoded signal V5 after DCT encoding. It is a circuit that performs replacement (including addition and subtraction processing) at the level, and outputs the signal after replacement as an encoded signal V6.

Here, how the direct current (DC) component of the coded signal that has been orthogonally transformed (axis-transformed) as in the DCT coding method appears in the coded signal will be described. Figure 4
FIG. 4 is a diagram showing a state in which DC coefficients in a block of 8 pixels × 8 lines are quantized. In FIG. 4, a direct current (DC) component appears in the upper left corner. Then, at the time of actual encoding, as shown in FIG. 5, a difference in DCT coefficient (direct current (DC) component) appearing in the upper left corner of each block is obtained between adjacent blocks and encoded.

Therefore, by changing this difference value, it is possible to change the direct current (DC) level in block units. Further, by changing the DCT coefficient (direct current (DC) component) in the block at the head position of the frame composed of a plurality of blocks, it is possible to shift the amount of all direct current (DC) components in the frame.

Moreover, since no level of decoding processing or coding processing is required for the level conversion operation of the coded signal V5, there is no need to worry about deterioration of image quality.

Incidentally, in the case of the conventional apparatus, when such level conversion processing is performed, the DCT-coded coded signal is subjected to the inverse DCT-coded processing to be decoded into a component video signal, which is further composited. It is necessary to change the direct current (DC) level in the state of being converted into the video signal, and again to perform the DCT coding processing of the component video signal, which is unavoidable.

As described above, in the second embodiment, when the level of the coded signal V5 is changed, the direct current (DC) component remains as the coded signal without using any decoding and re-encoding processing. Is changed, it is possible to minimize the risk of image quality deterioration in the process of signal processing.

The most suitable application of the processing apparatus according to the present embodiment is a video processing apparatus (so-called video processor) for performing level adjustment, color adjustment, etc. of an image signal.
Is.

(C) Third Embodiment (C-1) Configuration of Third Embodiment Hereinafter, the coded signal processing device according to the present invention is applied to a processing device for processing a signal after DCT coding. The case will be described with reference to the drawings.

FIG. 6 shows a functional block configuration of a processing apparatus according to the third embodiment. Note that, in FIG. 6, the same parts as those in FIG. 1 are designated by the same reference numerals.

The processing apparatus according to the third embodiment is obtained by adding the function explained in the second embodiment to the processing apparatus according to the first embodiment. That is, in the case of the first embodiment, since the method of selecting the signal to be output based on the correlation between the coding blocks is adopted,
Even when only the direct current (DC) component is changed, it is considered that there is no correlation, and the re-encoded signal (decoded → encoded) is selected. Avoidance is aimed at.

Therefore, the difference in configuration is that a new function is added to the comparison circuits 5'and 6 ', and an addition for adding a direct current (DC) level to the original signals (coded signals) V1 and V2. There are only two points in which the circuits 12 and 13 are added, and the other parts are the same as the configuration of the first embodiment.

Here, the configurations of the comparison circuits 5'and 6'and the addition circuits 12 and 13 will be described.

Comparison circuits 5'and 6'according to this embodiment.
Also in the case of, the point of obtaining the correlation between the original signals (coded signals) V1 and V2 and the recoded signal V3 is the same as in the first embodiment, but the quantization distributions are compared. However, when only the direct current (DC) component of the DCT coefficient is different, the selection signal is output so as to select the original signals (encoded signals) V1 and V2, and the direct current is supplied to the adder circuits 12 and 13. The difference is that the difference amount of the (DC) component is output.

The adder circuits 12 and 13 directly add the original signals (encoded signals) V1 and V2 when the difference amounts of the direct current (DC) components are given from the comparison circuits 5'and 6 '. Is. That is, as described in the second embodiment, it is a circuit that adds direct current (DC) components to the original signals (coded signals) V1 and V2 without performing any decoding processing.

(C-2) Content of Image Processing Operation Next, the operation when the image processing apparatus is used to perform arbitrary image processing on the video signal will be described.

The other operation contents are the same as those in the first embodiment already described, except that the processing contents in the processing circuit 3 are the change of the direct current (DC) level.

That is, as a result of comparing the original signals (coded signals) V1 and V2 with the recoded signal V3 after recoding in the comparison circuits 5'and 6 ', if a high correlation is found, , Outputs a selection signal instructing selection of the original signals (encoded signals) V1 and V2, and, on the contrary, when a high correlation is not recognized, instructing selection of the re-encoded signal V3 after re-encoding The operation of outputting the selection signal is executed.

However, the original signal (coded signal) V1
, And V2 and the re-encoded signal V3 after re-encoding, no difference is recognized from the comparison of the quantization distributions of the DCT coefficients, only the direct current (DC) component. In this case, the comparison circuits 5'and 6'include the switching circuit (SW) 7
And 8 and outputs a selection signal instructing selection of original signals (encoded signals) V1 and V2, and also adds circuit 12
And 13, the difference amount of the direct current (DC) component is output.

As a result, the original signals (encoded signals) V1 and V2 after the direct current (DC) components have been added in the adder circuits 12 and 13 and which have not undergone a series of processes involving decoding and the like. Will be selected and output to the synthesis circuit 9. That is, it is possible to obtain a signal which is basically free from signal deterioration.

(C-3) Effects of Third Embodiment As described above, according to the processing apparatus of this embodiment, the first
In addition to the effect similar to that of the embodiment described above, when the content of the processing in the processing circuit 3 is only the change of the direct current (DC) level, a high correlation is recognized in the determination of the correlation in the comparison circuits 5 ′ and 6 ′. It is possible to select the original signals (coded signals) V1 and V2 that have not been subjected to processing such as decoding even without them, and it is possible to minimize the image quality deterioration of the combined output V4.

(D) Application Example FIG. 7 shows an application example of each of the above-described embodiments. Here, a case where a coded signal is non-linearly edited using an electronic computer will be described. In addition,
FIG. 7A shows a general processing flow when a coded signal is processed using a computer, and FIG.
Shows an application example of each embodiment. Further, this example shows the case of processing a signal of one system.

First, the flow of general processing will be described. The encoded signal read from the disk (including a storage medium such as a memory card) 21 is decoded into a component signal by the decoding board 22, and then the signal processing board 2
3 is input and processed. Usually, the signal processing board 23 has a video signal memory having a capacity of one field or more, and after the decoded signal is stored in the video signal memory, predetermined signal processing is performed.

The signal after the signal processing is stored in the video signal memory again and then output to the CPU control board 24.
The processed signal input to the CPU control board 24 is displayed on the display via the display board 25 and is used for confirmation by the operator. The operator's operation on the displayed image is given to the signal processing board 23 via the CPU control board 24, and the signal processing and the display are repeated until the operator's confirmation is completed. When the signal obtained by the confirmation of the operator is stored in the disk 21, the signal is supplied to the encoding board 26 and encoded again.

By the way, in the case of applying this embodiment, as shown in FIG. 7B, a signal instructing storage in the disk 21 is once inputted to the comparison circuit 27 and decoded in block units. Compared to previous signal. Then, with respect to the block whose change is recognized by the signal processing, the signal re-encoded by the encoding board 26 is sent to the disk 21 through the switching circuit 28, and the block whose change is not recognized is sent from the disk 21. The read signal is written in the disk 21 again as it is. However, when selecting the signal read from the disk 21,
The original signal is used as-is and does not need to be rewritten. In this case, it is sufficient to write only the re-encoded signal.

(E) Other Embodiments Finally, an embodiment in which each of the above-described embodiments can be adopted will be described.

In the above embodiment, the case where two types of video signals are combined has been described, but when one type of video signal is handled as in the application example, three or more types of video signals are also handled. It can also be applied in some cases. In any case,
Basically, the re-encoded signal is compared with the pre-decoded signal in block units, and either the pre-decoded signal or the re-encoded signal is output depending on the comparison result. You just have to choose.

Further, in the above-mentioned embodiment, the picture structure of the video signal to be processed is not described at all, but the present invention can be applied to the case where a frame is used as a unit and the field is used as a processing unit.

Further, in the above embodiment, DC
Although the case of processing a T-coded video signal has been described, the present invention is not limited to this and is also applied to the case of processing a video signal subjected to other orthogonal coding processing (for example, Hadamard coding processing). obtain.

Further, in the above embodiment, the case where the block-coded signal is processed is described, but the present invention can be applied to the case where the signal of a predetermined period continuous in time series is targeted.

Further, in the above-described embodiment, the case where the video signal is the processing target is described, but the present invention can be applied to the case where the lossy compression processed audio signal is the processing target.

[0069]

As described above, according to the first aspect of the invention, even when the predetermined processing is performed on the encoded signal, the re-encoded signal after the predetermined processing and the predetermined processing are performed. The correlation of the coded signal before decoding which has not been performed is obtained for each predetermined unit, and when a high correlation is recognized, the coded signal before decoding which is the original signal itself is selectively output. It is possible to minimize the deterioration of quality when a predetermined process is performed on only a part of all encoded signals.

Further, according to the second aspect of the invention, since only the target DC component is changed without decoding the coded signal to be processed, deterioration in quality associated with decoding and re-encoding is prevented. It can be lost.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an encoded signal processing device according to a first embodiment.

FIG. 2 is a block diagram showing a configuration of a conventional device.

FIG. 3 is a block diagram showing a configuration of a coded signal processing device according to a second embodiment.

FIG. 4 is a diagram showing a result of quantization of DCT coefficients.

FIG. 5 is a diagram showing a DC component coding example between adjacent blocks.

FIG. 6 is a block diagram showing a configuration of an encoded signal processing device according to a third embodiment.

FIG. 7 is a block diagram showing a configuration example when applied to a computer.

[Explanation of symbols]

1, 2 ... Decoding circuit, 3 ... Processing circuit, 4 ... Encoding circuit,
5, 5 ', 6, 6' ... comparison circuit, 7, 8 ... switching circuit, 9
... synthesis circuit, 10 ... DC level generation circuit, 11 ... level variable circuit, 12, 13 ... addition circuit.

Claims (3)

(57) [Claims] 1. A coded signal processing apparatus for inputting one or a plurality of coded signals irreversibly coded for each predetermined unit, performing a predetermined process and outputting the coded signal, wherein each of the one or a plurality of coded signals is processed. A first signal system having a decoding means for decoding, a processing means for subjecting the decoded signal to the predetermined processing, and an encoding means for irreversibly encoding and outputting the processed signal; A second signal system for transmitting one or more coded signals as they are without decoding, and a re-coded signal and a coded signal from the first and second signal systems, and the correlation between the two signals. When a high correlation is found between the comparison means for judging every predetermined unit and the judgment in the comparison means, the coded signal before decoding outputted from the second signal system is selectively outputted, and A low correlation is judged by the comparison means. If because obtained, coded signal processing apparatus characterized by comprising selecting means for selectively outputting the re-encoded signal output from said first signal system. 2. The coded signal processing device according to claim 1, wherein the comparison means analyzes the coded signal to determine whether the two signals are low even if a low correlation is found between the signals. When it is recognized that the difference is only the change of the DC component, the selecting means is instructed to select the coded signal output from the second signal system, and the second signal system is selected. A value corresponding to the difference of the direct current component is given to the adding means provided between the output terminal of and the input terminal of the selecting means,
A coded signal processing device, characterized in that a coded signal input to the selecting means is modified. 3. A DC level generator for generating an arbitrary DC level in a coded signal processing device for inputting one or a plurality of coded signals irreversibly coded for each predetermined unit, performing a predetermined process and outputting the coded signal. A coded signal processing device comprising: means and level changing means for changing only the DC component of the coded signal according to the DC level generated by the DC level generating means.