JP2003299080A - Coder, coding method, program and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform bit supply rate control for feedback rate control by a difficulty average avgD in a former GOP (group of picture). <P>SOLUTION: When a picture under currently processing is judged to be the head of the GOP, the difficulty average avgD in the former GOP is obtained, and a value of max<SB>-</SB>sum<SB>-</SB>sup being the upper limit value of the sum of value sum<SB>-</SB>sup of values of supplement is obtained. In the case of avgD>;Ox2,000 and sum<SB>-</SB>sup<;max<SB>-</SB>sum<SB>-</SB>sup, supplement having a positive value is provided to a usable bit rate R. When it is judged to be not avgD>;Ox2,000 and sum<SB>-</SB>sup<;max<SB>-</SB>sum<SB>-</SB>sup and avgD;Ox1,000 and sum<SB>-</SB>sup>;min<SB>-</SB>sum<SB>-</SB>sup, the supplement having a negative value is added to the usable bit rate quantity R, otherwise the supplement=0. <P>COPYRIGHT: (C)2004,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a coding device, a coding method, a program, and a recording medium, and particularly to a coding device suitable for use in a bit rate control in feedback rate control. And an encoding method, a program, and a recording medium.

[0002]

2. Description of the Related Art In recent years, various compression coding methods have been proposed as a method of compressing video data and audio data to reduce the amount of information, of which MPEG2 is a typical one.
(MovingPicture Experts Group Phase 2) is available.

In such an image compression method, TM5 (Test Model
There is 5). In step 1 of TM5, the target bit to be given for each picture is calculated. When calculating the target bit, the GC for each picture type is used.
The bit amount R that can be assigned to the remaining pictures in the GOP (Group of Pictures) is proportionally distributed according to each ratio of (Global Complexity), and the bit amount to be assigned to each picture is calculated.

TM4 is an excellent method for making the amount of generated bits per GOP almost constant, but when performing fixed rate coding, it is not always necessary to make the amount of generated bits of GOP constant. . In fixed-rate encoding, it is necessary to prevent the occupied amount of a VBV (Video Buffering Verifier) buffer from overflowing or underflowing a specified value.

In TM5, since the amount of generated bits per GOP is almost constant, the VBV buffer does not overflow or underflow. However, in TM5, the buffer capacity cannot be effectively used when encoding is performed at a low bit rate. For example, TM5 in MP @ PL of MPEG
, The VBV buffer capacity is about 1.8 Mbi.
However, since the number of bits per picture that is pulled out from the buffer is small, it is about 1.8 Mb.
It cannot be used effectively.

As described above, regardless of the input pattern,
By assigning a certain amount of bits, coding distortion will occur remarkably for pictures with high coding difficulty, while pictures with low coding difficulty will have less coding distortion. , As a whole, it becomes an unstable image with a lot of unevenness.

In order to solve such a problem, a larger amount of bits is allocated to a picture having a high degree of coding difficulty in a range where the buffer does not underflow, while a picture having a low degree of coding difficulty is distributed. Needs to allocate a small amount of bits suitable for the picture within the range where the buffer does not overflow.

Therefore, the applicant of the present invention has filed Japanese Patent Application Laid-Open No. 10-754.
43, a video data compression apparatus and method for adjusting the generated bit amount according to the complexity of a picture for each part of the video data so as to improve the quality of the video after compression as a whole. Disclosure.

In TM5, the available bit amount R that can be assigned to the remaining pictures of the GOP is an important parameter in rate control. For example, G
In the first half of the OP, images with complicated patterns continued, so if a large amount of bits were allocated, GOP
In the latter half of the above, the bit amount R becomes extremely small or becomes a negative number.

On the other hand, the applicant of the present invention has disclosed in Japanese Patent Laid-Open No. 10-7.
The bit replenishment rate control disclosed in 5443 is
A bit amount is added to or subtracted from the usable bit amount R allocated to a plurality of pictures to be encoded, depending on the image difficulty level of the encoding target and the VBV buffer occupancy amount (hereinafter, added Bits that can be or are subtracted are called supplements).

[0011]

The bit replenishment rate control that has been proposed in the past has been proposed in the case where information such as the difficulty levels of a plurality of picture images to be encoded is already known, that is, the feedforward (prefetching the encoding information). This is applied to feed forward type rate control. For example, after the data of 15 GOP data is accumulated, the image encoding difficulty level is judged, so that a certain delay occurs in the information accumulation. It is something that ends up. However, due to recent low-delay requirements for encoder systems, rate control that does not cause such a delay is required. Therefore, bit replenishment rate control is applied to feedback (Feed Back) type rate control where prefetch information cannot be obtained. I have to do it.

The present invention has been made in view of such a situation, and it is possible to perform the bit replenishment rate control in the feedback type rate control.

[0013]

The encoding device of the present invention is
First detecting means for detecting the complexity of the uncompressed data, encoding means for compression-encoding the uncompressed data, and non-encoding detected in the past by the encoding means detected by the first detecting means. Based on the complexity of the compressed data, the second bit amount added to the usable first bit amount of the buffer capacity of the VBV buffer allocated to the uncompressed data to be encoded is calculated. A first calculating unit; a second calculating unit that calculates a target bit based on a value obtained by adding the second bit amount calculated by the first calculating unit to the first bit amount; Determining means for determining the quantization index for encoding by the encoding means based on the target bit calculated by the calculating means.

A third calculating means for calculating the sum of the second bit amounts calculated by the first calculating means may be further provided, and the first calculating means may include the third calculating means. It is possible to calculate the second bit amount such that the total of the second bit amounts calculated by the means does not become larger than the predetermined reference value.

The first detecting means may be made to calculate an index indicating the complexity of the non-compressed data to detect the complexity of the non-compressed data. It is possible to compare the threshold value and the size of the index, and if the index is larger, to calculate the second bit amount of a positive value, and to calculate the second threshold value and the index smaller than the first threshold value. And the index is smaller, the second bit amount having a negative value can be calculated, and the index is
If the numerical value is between the first threshold and the second threshold, the second
The bit amount of can be set to 0.

The uncompressed data may be composed of a plurality of GOPs, and the index calculated by the first detecting means is 1 previously encoded by the encoding means.
It can be used as an index showing the average difficulty level of GOP.

A second detecting means for detecting the occurrence of the scene change and a third detecting means for detecting the image difficulty levels of the same picture type before and after the scene change can be further provided. , The first calculation means,
When the scene change is not detected by the second detecting means, the second bit amount is determined based on the complexity of the uncompressed data previously encoded by the encoding means, which is detected by the first detecting means. When the scene change is detected by the second detecting means, the second bit amount can be calculated based on the detection result by the third detecting means.

In the first calculating means, when the scene change is detected by the second detecting means and the image difficulty level after the scene change is higher than that before the scene change, the second value having a positive value is used. It is possible to calculate the bit amount and to set the second bit amount to 0 at other times.

The uncompressed data may be composed of a plurality of GOPs, and the first calculating means calculates the second bit amount at a predetermined timing in the GOP encoding process. You can let the second
If the newest second bit amount calculated by the first calculating unit is a negative value and the image difficulty level in the GOP has changed by a predetermined value or more, The target bit can be calculated by setting the second bit amount of the newest negative value calculated by the calculating means to 0.

The encoding method of the present invention comprises a detection step of detecting the complexity of uncompressed data, an encoding step of compressing and encoding uncompressed data, and an encoding step detected by the processing of the detection step. Based on the complexity of the uncompressed data encoded in the past by the processing, the second bit added to the usable first bit amount of the VBV buffer allocated to the uncompressed data to be encoded. A first calculation step for calculating the bit amount of
Calculated by the process of the second calculation step, which calculates the target bit based on the value obtained by adding the second bit amount calculated by the process of And a determining step of determining a quantization index for encoding by the processing of the encoding step based on the generated target bits.

The program recorded on the recording medium of the present invention comprises a detection step of detecting the complexity of uncompressed data, an encoding step of compressing and encoding the uncompressed data,
Based on the complexity of the uncompressed data previously encoded by the processing of the encoding step, which is detected by the processing of the detection step, of the VBV buffers allocated to the uncompressed data to be encoded A first calculation step of calculating a second bit amount added to the possible first bit amount and a second bit amount calculated by the process of the first calculation step are set as a first bit amount. A second calculation step of calculating a target bit based on the added value, and a quantization index for encoding by the processing of the encoding step is determined based on the target bit calculated by the processing of the second calculation step. And a determining step to perform.

The program of the present invention comprises a detection step of detecting the complexity of non-compressed data, an encoding step of compressing and encoding the non-compressed data, and a processing of the encoding step detected by the processing of the detection step. A second bit added to the available first bit amount of the VBV buffer allocated to the uncompressed data to be encoded, based on the complexity of the previously encoded uncompressed data. A first calculation step for calculating the quantity;
Calculated by the process of the second calculation step, which calculates the target bit based on the value obtained by adding the second bit amount calculated by the process of And a determining step of determining a quantization index for encoding by the processing of the encoding step based on the generated target bits.

In the encoding device, the encoding method, and the program of the present invention, the complexity of non-compressed data is detected, the non-compressed data is compression-encoded, and the complexity of non-compressed data encoded in the past is detected. Of the VBV buffers allocated to the uncompressed data to be encoded, a second bit amount added to the usable first bit amount is calculated, and the calculated second bit is calculated. The target bit is calculated based on the value obtained by adding the amount to the first bit amount, and the quantization index is determined based on the calculated target bit.

[0024]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an encoder 1 to which the present invention is applied.

The image rearrangement unit 12 rearranges the input uncompressed video data in the order of encoding. The scan conversion / macroblocking unit 13 performs picture / field conversion, and performs, for example, 3: 2 pulldown processing when the uncompressed video data is video data of a movie. Intra AC
The calculation unit 14 is processed by the image rearrangement unit 12 and the scan conversion / macroblocking unit 13, and is compressed from an I-picture to an intra AC (intra A) picture.
Calculate C).

Since the I picture is compression-encoded without reference to other pictures, the ME residual, which will be described later, cannot be obtained. Therefore, the intra AC is used as a parameter instead of the ME residual in order to obtain the coding difficulty of the I picture. Intra AC is MP
It is a parameter defined as a sum of variance values of video data for each DCT block in the DCT processing unit in the EG method, which indicates the complexity of the video and correlates with the difficulty of the picture pattern of the video and the amount of data after compression. Have sex. That is, the intra AC is the total sum of absolute values in the screen obtained by subtracting the average value of the pixel values of each block from the pixel value of each pixel in DCT block units.
The intra AC is represented by the following equation (1).

[0028]

[Equation 1] ... (1)

Further, in the equation (1), the equation (2) is established.

[Equation 2] ... (2)

The intra AC calculator 14 outputs the calculated intra AC value to the difficulty calculator 32 of the rate controller 15.

The arithmetic processing unit 16 performs motion compensation on the supplied video data based on the motion compensation information supplied from the motion compensating unit 25, and outputs it to the DCT unit 18. The DCT unit 18 performs, for example, discrete cosine transform (DCT) processing on the video data input from the arithmetic processing unit 16 in units of macroblocks of 16 pixels × 16 pixels, and converts the time domain data to the frequency domain data. And outputs to the quantizer 19.

The quantizing unit 19 converts the frequency domain data input from the DCT unit 18 into the rate control unit 15.
Quantized by the quantization index Q supplied from the quantization index determination unit 35 of
It outputs to (Variable Length Code) unit 20 and inverse quantization unit 22.

The VLC section 20 performs variable length coding processing on the quantized data input from the quantizing section 19 based on a predetermined conversion table, and outputs the resulting variable length coded data to the buffer 21. To do.

The buffer 21 buffers the input coded data to form a coded bit stream.
Output sequentially.

The inverse quantization unit 22 inversely quantizes the quantized data input from the quantization unit 19 in the quantization step of the quantization executed by the quantization unit 19, and the inverse DCT unit 23 as the inverse quantized data. Output to.

The inverse DCT unit 23 performs inverse DCT processing on the inverse quantized data input from the inverse quantization unit 22,
It is output to the arithmetic processing unit 24.

The arithmetic processing unit 24 adds the output data of the motion compensation unit 25 and the output data of the inverse DCT unit 23,
It is output to the motion compensation unit 25. The motion detector 17
The motion vector is searched for a macroblock that minimizes the sum of absolute values or the sum of squared differences between the target macroblock of the picture to be compressed (input picture) and the referenced picture (reference picture). Is calculated and output to the motion compensation unit 25. The motion compensation unit 25 performs motion compensation processing on the output data of the arithmetic processing unit 24 based on the motion vector input from the motion detection unit 17, and outputs the data to the arithmetic processing units 24 and 16. To do.

The rate control unit 15 is composed of an ME residual calculation unit 31, a difficulty calculation unit 32, a genbit detection unit 33, a target bit determination unit 34, and a quantization index determination unit 35, and the target bit and the quantization index. To decide.

The ME residual calculation unit 31 calculates the ME residual, which is a parameter having a strong correlation with the image coding difficulty. By motion estimation, it is possible to obtain a motion vector that reduces the sum of absolute difference values from the reference frame to the input frame, but in that case, the error obtained by the sum of absolute difference values or the sum of squares. The power of the component is the ME residual. In the P picture and the B picture, the ME residual and the image coding difficulty are
It has an almost simple proportional relationship.

The difficulty level calculation unit 32 uses the ME residual calculation unit 31.
By the approximation with the ME residual input from Equation (3),
Also, the coding difficulty Dj of the P picture and the B picture is calculated using the equation (4).

[Equation 3] ... (3)

[Equation 4] ... (4)

Here, MEj is the j-th picture.
KE ME residual, a_P, A_B, B _P, B_BIs that
Is a slope and a correction value when approximated by a linear expression.

Further, the difficulty calculation section 32 similarly calculates the coding difficulty Dj of the I picture by approximation by the intra AC input from the intra AC calculation section 14, and outputs it to the target bit determination section 34.

Then, the difficulty level calculator 32 calculates the average difficulty level avgD for each GOP from the coding difficulty level Dj calculated for each picture.

The genbit detection unit 33 detects the generated bit amount genbit of the most recently encoded I picture from the encoded data buffered in the buffer 21, and outputs the value to the target bit determination unit 34. To do.

The target bit determination unit 34 receives the coding difficulty Dj and gen input from the difficulty calculation unit 32.
Based on the generated bit amount genbit of the I picture input from the bit detection unit 33, the target bit of each picture of each picture type is calculated and the rate control is performed.

That is, the target bit determining section 34
Is the value of the supplement added to the usable bit amount R assigned to the plurality of pictures to be encoded, based on the difficulty of the past image that has been encoded by the process described later ( su
pplement is 0 for positive and negative values
May be).

Since the bit replenishment amount supplement can take both positive and negative values, the integrated value sum_supplement of the supplement
By determining the maximum value and the minimum value of (hereinafter referred to as sum_sup) according to the VBV buffer size to be used, it is possible to adjust the supplement within the range of the VBV constraint. The target bit determination unit 34 determines the value of the target bit based on the usable bit amount R + supplement and outputs it to the quantization index determination unit 35.

The quantization index determination unit 35 generates a quantization index Q based on the value of the target bit input from the target bit determination unit 34, and outputs it to the quantization unit 19.

Next, referring to the flowchart of FIG.
The bit replenishment rate control processing 1 for determining the supplement to be added to R based on the difficulty of the past image after the encoding will be described.

In step S1, the target bit determining section 34 determines whether or not the picture currently being processed is the head of the GOP. In step S1, GO
When it is determined that it is not the head of P, the process of step S1 is repeated until it is determined that it is the head of GOP.

When it is determined in step S1 that the GOP is at the beginning, in step S2, the target bit determination unit 34 determines that the previous G from the difficulty level calculation unit 32 is determined.
Get the average difficulty level avgD in OP.

In step S3, the target bit determination unit 34 adds the available bit amount R.
max, which is the upper limit of sum_sup, which is the sum of supplement values
Get the value of _sum_sup.

In step S4, the target bit determining section 34 determines that avgD> 0x2000 and sum_sup <max_sum_s.
It is determined whether it is up. Here, 0x2000 which is compared with the average difficulty level avgD is a predetermined threshold value,
It is a value that can be set to obtain the required image quality while considering the image quality.

When it is determined in step S4 that avgD> 0x2000 and sum_sup <max_sum_sup, the target bit determining unit 34 determines in step S5.
A positive value supplement is added to the usable bit amount R. That is, since the previous GOP has a certain degree of difficulty or more, the target bit determination unit 34 predicts that the degree of difficulty of the GOP to be encoded is similar to that of the previous GOP, and can use the target bit. A positive value supplement is added to the bit amount R.

When it is determined in step S4 that avgD> 0x2000 and sum_sup <max_sum_sup are not satisfied,
In step S6, the target bit determination unit 34
Determines whether avgD <0x1000 and sum_sup> min_sum_sup. Here, 0x1000 compared with the average difficulty level avgD is a predetermined threshold value, which is a value smaller than 0x2000 described above (a value indicating that the image difficulty level is low), and is requested while considering the image quality. It is a value that can be set in order to obtain high image quality.

In step S6, avgD <0x1000,
And if it is determined that sum_sup> min_sum_sup,
In step S7, the target bit determination unit 34
Is a negative value for the available bit amount R
Add. That is, the target bit determination unit 34
Since the previous GOP had a certain level of difficulty or less (that is, a simple image), the difficulty of the GOP to be encoded is predicted to be the same as that of the previous GOP and can be used. Negative value of suppleme for bit amount R
add nt.

In step S6, avgD <0x1000,
When it is determined that sum_sup> min_sum_sup is not satisfied, the target bit determination unit 34 sets supplement = 0 in step S8. That is, the target bit determination unit 34 sets su to the usable bit amount R.
Do not increase or decrease pplement.

After the processing of step S5, step S7, or step S8 is completed, the target bit determination unit 34 is used in the processing of step S5, step S7, or step S8 in step S9.
Using the value of supplement, sum_sup = sum_sup + supple
ment, the process returns to step S1 and the subsequent processes are repeated.

By the process described with reference to FIG. 2, the usable bit amount R is added or reduced based on the difficulty level of the past image which has been encoded.
The value of ment is determined. For example, in GOP units, R + su
The previous GO if pplemet (supplement is a positive value, a negative value, or 0) is determined
Based on the average value of the image difficulty level of P (intra AC, ME residual, etc.), it is predicted that the difficulty level of the GOP to be encoded will be the same as the difficulty level of the previous GOP, and the usable bit Depending on the amount R, depending on the difficulty
supplement is added. Thereby, the bit supplement rate control can be applied to the feedback type rate control.

Although the image difficulty has been described as being calculated using the intra AC or the ME residual, the image difficulty may be calculated using other parameters.

The method of calculating the specific value of the supplement may be, for example, the method disclosed in Japanese Patent Laid-Open No. 10-75443, or a method other than that may obtain the required image quality. You may make it use a value.

Also, here, the description is made assuming that the average difficulty level avgD in the previous 1 GOP is used, but the difficulty level calculation section 32 replaces the average difficulty level avgD in 1 GOP with, for example, a plurality of GOPs or GOPs. The average of the degree of difficulty may be calculated for a part of the above, or, instead of the simple average of the degree of difficulty, a weighted sum or a weighted average may be calculated as necessary.

However, in the bit replenishment rate control processing 1 described with reference to FIG. 2, when the tendency of the image difficulty changes abruptly due to a scene change or the like, the image quality may be adversely affected. For example, GO
When R + supplement is determined in P units, when a scene change occurs from a simple image to a difficult image,
Since the image difficulty level of the previous GOP is low, the supplement has a negative value, and the GOP of the difficult image is encoded with a small usable bit amount R.

Therefore, when encoding a GOP in which a scene change has occurred, by comparing the picture difficulty levels of the I picture at the beginning of the GOP and the I picture of the immediately preceding GOP, the supp
You may make it decide lement. As a result, when a scene change occurs, it is not necessary to allocate a small bit amount to the GOP of a difficult image, and the image quality can be improved.

Next, referring to the flowchart of FIG.
In a scene change, the bit replenishment rate control that determines the value of the supplement added to the usable bit amount R by comparing the image difficulty of the same picture type in the past with the image difficulty of the current picture in which the scene change has occurred The process 2 will be described.

In step S21, the target bit determining section 34 determines whether or not the picture currently being processed is the head of the GOP. In step S21,
When it is determined that it is not the head of the GOP, the process of step S21 is repeated until it is determined that the head of the GOP is reached.

If it is determined in step S21 that the GOP is at the beginning, the target bit determining section 34
In step S22, the difficulty level calculation unit 32
The image coding difficulty Di of the I picture is acquired, and step S
23, added to the available bit amount R
max, which is the upper limit of sum_sup, which is the sum of supplement values
Get the value of _sum_sup.

In step S24, the target bit determining section 34 determines whether or not there is a scene change. The determination as to whether or not there is a scene change is made by, for example, M
The determination may be made based on the ME residual value calculated by the E residual calculation unit 31, or may be made by any other method.

If it is determined in step S24 that there is a scene change, the target bit determining section 34 determines in step S25 the image coding difficulty Di of the I picture and the image code of the I picture of the immediately preceding GOP. Compare the difficulty level prevDi, and further compare it with the current value of sum_sup,
The values of max_sum_sup acquired in step S23 are compared to determine whether Di> prevDi and sum_sup <max_sum_sup.

When it is determined in step S25 that Di> prevDi and sum_sup <max_sum_sup, the target bit determining unit 34 is determined in step S26.
Is a positive value for the available bit amount R
Add. That is, the target bit determination unit 34
Since it detects that the image after the scene change is more difficult than the image before the scene change,
A positive value supplement is added to the usable bit amount R.

If it is determined in step S25 that Di> prevDi and sum_sup <max_sum_sup are not satisfied,
In step S27, the target bit determination unit 34
Sets supplement = 0. That is, the target bit determination unit 34 uses the available bit amount R as a supplement.
Do not increase or decrease ment.

If it is determined in step S24 that it is not a scene change, in step S28
The target bit determination unit 34 acquires the average difficulty level avgD in the previous GOP from the difficulty level calculation unit 32.

Then, the target bit determining section 34
In steps S29 to S33, the same processing as the processing executed in steps S4 to S8 of FIG. 2 is executed.

That is, the target bit determining unit 34
At step S29, avgD> 0x2000 and sum_su
Determine if p <max_sum_sup, avgD> 0x2000
And it is determined that sum_sup <max_sum_sup,
In step S30, a positive value supplement is added to the usable bit amount R.

In step S29, avgD> 0x2000
When it is determined that sum_sup <max_sum_sup is not satisfied, the target bit determination unit 34 determines in step S31 whether avgD <0x1000 and sum_sup> min_sum_sup, and avgD <0x1000 and sum_sup> m.
If it is determined to be in_sum_sup, in step S32, a negative value supplem for the available bit amount R.
Add ent, avgD <0x1000, and sum_sup> min_sum_su
When it is determined that it is not p, the supplement = 0 is set in step S33.

After the processing of step S26, step S27, step S30, step S32, or step S33 is completed, in step S34, the target bit determination section 34 acquires I obtained in step S22.
The picture coding difficulty level Di of a picture is set to prevDi = Di in order to be used for the processing of the next GOP.

At step S35, the target bit determining section 34 performs step S26, step S27, step S30, step S32, or step S3.
Using the value of supplement used in the process of 3,
Sum_sup = sum_sup + supplement, the process returns to step S21, and the subsequent processes are repeated.

By the processing described with reference to FIG. 3, in the scene change, the degree of difficulty of the same picture type in the past (I picture here) and the degree of difficulty of the current picture in which the scene change occurred (I picture here) The supplement can be determined by comparing.

As a result, for example, R + su is set for each GOP.
When pplement is decided, even if a scene change occurs from a simple image to a difficult image, G of the difficult image
It is possible to improve the image quality by avoiding encoding with a small R for OP, that is, assigning a small bit amount to a GOP of a difficult image.

However, in the bit replenishment rate control processing 1 and the bit replenishment rate control processing 2 described with reference to FIGS. 2 and 3, R + suppleme once determined.
Since the value of nt is not reviewed during encoding, the image quality may be adversely affected in the case where the difficulty tendency of a plurality of images being encoded changes slowly.

Bit replenishment rate control processing 1 and bit replenishment rate control processing 2 described with reference to FIGS. 2 and 3.
In the above, since R + supplement is determined for each GOP, for example, when the previous GOP is a simple image and no scene change has occurred, the supplement added to the usable bit amount R becomes a negative value. , The usable bit amount becomes smaller. If the GOP that started encoding in that state becomes slowly and difficult like an overlapped image, the bit amount becomes insufficient at the end of the GOP, and the image quality deteriorates.

In order to solve such a problem, when the supplement added to the usable bit amount R is a negative value, the image difficulty of the picture to be encoded and the immediately preceding picture of the same picture type. Compared with the difficulty level, if the difference is large to some extent, the process of restoring the amount corresponding to the supplement of the negative value added to the remaining R, in other words, to the subtracted supplement Introduce a process to restore the corresponding amount.

With reference to the flow chart of FIG. 4, the bit replenishment rate control processing 3 in which a mechanism for revising the supplement during the encoding is introduced to the bit replenishment rate control processing 1 described with reference to FIG. 2 will be described.

In step S51, the target bit determining section 34 determines whether or not the picture currently being processed is the head of the GOP. In step S51,
If it is determined that it is not the head of the GOP, step S5
In 2, the bit amount reviewing process described later with reference to FIG. 5 is executed, the process returns to step S51, and the subsequent processes are repeated.

If it is determined in step S51 that the GOP is at the beginning, steps S53 to S53
At 60, processing similar to that at steps S2 to S9 of FIG. 2 is executed.

Then, after the processing of step S60 is completed,
The process returns to step S51, and the subsequent processes are repeated.

Next, referring to the flowchart of FIG.
The bit amount review process executed in step S52 of FIG. 4 will be described.

In step S81, the target bit determination unit 34 is the value of the supplement of the previous process.
prev_sup, the difficulty value D (B, P) of the current picture (B picture or P picture), and the difficulty value prev of the previous picture of the same picture type
_D (B, P) is acquired, prev_sup <0, and D
It is determined whether or not (B, P)> 4 × prev_D (B, P).

In step S81, prev_sup <0,
When it is determined that D (B, P)> 4 × prev_D (B, P) is not satisfied, the value of the supplement of the immediately preceding process is not negative, or the image difficulty level of the encoding target picture is , The difference from the immediately preceding image difficulty level of the same picture type is not so large, the process proceeds to step S5 of FIG.
Return to 1.

Here, the difficulty level value D (B, P)
Is compared with 4 times the difficulty value prev_D (B, P). Here, the coefficient by which prev_D (B, P) is multiplied is a value of a property set while considering the image quality. Needless to say.

In step S81, prev_sup <0,
Further, when it is determined that D (B, P)> 4 × prev_D (B, P), the value of the supplement of the immediately preceding process is negative, and the image difficulty level of the encoding target picture, Since the difference from the image difficulty level of the immediately preceding picture type is large to some extent, the target bit determination unit 34 sets R = R-prev_sup in step S82. That is, this processing is synonymous with setting the value of the supplement calculated in the previous processing to 0 and executing the subsequent processing,
Since prev_sup <0 here, the value of R actually increases.

Then, in step S82, the target bit determining unit 34 determines in step S83 that su
m_sup = sum_sup-prev_sup, prev_sup = 0 in step S83, and the process is performed in step S of FIG.
Return to 51. Similarly in step S83, prev_sup <
Since it is 0, the value of sum_sup increases.

By the processing described with reference to the flow charts of FIGS. 4 and 5, the R + supplement once determined is reviewed in the middle of encoding. Therefore, for example, when the difficulty tendency of a plurality of images being encoded changes slowly. In the above, it is possible to prevent the image quality from being adversely affected.

Similarly, the bit replenishment rate control process 2 described with reference to FIG.
A process to review the pplement can be introduced.

With reference to the flowchart of FIG. 6, the bit replenishment rate control processing 4 in which a mechanism for revising the supplement during the encoding is introduced to the bit replenishment rate control processing 2 described with reference to FIG. 3 will be described.

In step S101, the target bit determining section 34 determines whether or not the currently processed picture is the head of the GOP. When it is determined in step S101 that it is not the head of the GOP, the bit amount reviewing process described with reference to FIG. 5 is executed in step S102, and the process returns to step S101.
The subsequent processing is repeated.

If it is determined in step S101 that it is the head of the GOP, the same processing as steps S22 to S35 in FIG. 3 is executed in steps S103 to S116.

After the processing of step S116 ends, the processing returns to step S101, and the subsequent processing is repeated.

By the processing described with reference to FIG. 6, it is possible to similarly introduce a mechanism for reconsidering the supplement during the encoding into the bit supply rate control processing 2 described with reference to FIG. It is possible to prevent the image quality from being adversely affected when the difficulty tendency of a plurality of images being encoded changes slowly.

The series of processes described above can be executed by hardware, but can also be executed by software. In this case, for example, the encoder 1
Is a personal computer 10 as shown in FIG.
It is composed of 1.

In FIG. 7, the CPU 111 uses the program stored in the ROM 112 or RA from the storage unit 118.
Various processes are executed according to the program loaded in M113. The RAM 113 also appropriately stores data necessary for the CPU 111 to execute various processes.

CPU 111, ROM 112, and RAM 113
Are mutually connected via a bus 114. An input / output interface 115 is also connected to the bus 114.

The input / output interface 115 includes an input unit 116 including a keyboard and a mouse, an output unit 117 including a display and a speaker, a storage unit 118 including a hard disk, a communication including a modem and a terminal adapter. The section 119 is connected. The communication unit 119 performs communication processing via a network including the Internet.

A drive 120 is also connected to the input / output interface 115 if necessary, and the magnetic disk 1
31, the optical disk 132, the magneto-optical disk 133, the semiconductor memory 134, or the like is appropriately mounted, and the computer program read from them is installed in the storage unit 118 as necessary.

When a series of processes is executed by software, a program that constitutes the software is installed in a computer in which dedicated hardware is installed, or various programs are installed to perform various functions. Is installed from a network or a recording medium into a general-purpose personal computer or the like capable of executing.

This recording medium, as shown in FIG.
A magnetic disk 131 (including a floppy disk) in which a program is stored, which is distributed in order to supply the program to a user separately from the apparatus body, and an optical disk 13.
2 (CD-ROM (Compact Disk-Read Only Memory),
DVD (including Digital Versatile Disk), magneto-optical disk 133 (including MD (Mini-Disk) (trademark),
Alternatively, in addition to being configured by a package medium including a semiconductor memory 134, a ROM 112 in which a program is stored and a hard disk included in a storage unit 118, which is supplied to a user in a state of being incorporated in the apparatus body in advance, is used. Composed.

In the present specification, the steps for writing the program stored in the recording medium are not limited to the processing performed in time series in the order of inclusion, but may be performed in parallel even if they are not necessarily performed in time series. Alternatively, it also includes processes that are individually executed.

[0108]

According to the present invention, image data can be encoded. Further, according to the present invention, it is possible to determine the supplement to be added to the usable bit amount R based on the degree of difficulty in the past image that has been encoded. Therefore, it is possible to apply the bit replenishment rate control to the feedback rate control. You can

Further, when encoding a GOP in which a scene change has occurred, the supplement added to the usable bit amount R is determined by comparing the image difficulty levels of the same picture type before and after the scene change. As a result, it is possible to improve the image quality without assigning a small bit amount to the GOP of a difficult image. Furthermore, since R + supplement once determined can be reviewed during encoding, it prevents the image quality from being adversely affected when the difficulty tendency of multiple images during encoding changes slowly. be able to.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an encoder to which the present invention has been applied.

FIG. 2 is a flowchart illustrating a bit replenishment rate control process 1.

FIG. 3 is a flowchart illustrating a bit replenishment rate control process 2.

FIG. 4 is a flowchart illustrating a bit supply rate control process 3;

FIG. 5 is a flowchart illustrating a bit review process.

FIG. 6 is a flowchart illustrating a bit replenishment rate control process 4.

FIG. 7 is a diagram illustrating a configuration of a personal computer.

[Explanation of symbols]

1 encoder, 12 image rearranging unit, 13 scanning conversion / macroblocking unit, 14 intra AC calculation unit, 15 rate control unit, 16 arithmetic processing unit, 17 motion detection unit, 18 DCT unit, 19
Quantization unit, 20 VLC unit, 21 buffer, 22
Inverse quantization section, 23 inverse DCT section, 24 arithmetic processing section, 25 motion compensation section, 31 ME residual calculation section,
32 difficulty calculation part, 33 genbit detection part, 34 target bit determination part, 35 quantization index determination part

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5C059 KK35 MA00 MA05 MA14 MA23                       MC11 MC38 ME01 NN01 PP05                       PP06 PP07 PP11 SS20 TA46                       TA60 TB03 TB04 TC02 TC03                       TC10 TC14 TC15 TC27 TC41                       TC42 TD02 TD03 TD04 TD06                       TD12 TD16 UA02 UA32

Claims (10)

[Claims] 1. An encoding device for encoding uncompressed data, comprising: first detecting means for detecting complexity of the uncompressed data; encoding means for compressing and encoding the uncompressed data; A buffer of a VBV buffer allocated to the uncompressed data that is to be encoded, based on the complexity of the uncompressed data that was previously encoded by the encoding unit and that was detected by the first detecting unit. Out of capacity
First calculating means for calculating a second bit amount added to the usable first bit amount; and the second bit amount calculated by the first calculating means for the first bit Second calculating means for calculating a target bit based on the value added to the quantity, and a quantization index for encoding by the encoding means based on the target bit calculated by the second calculating means. An encoding device comprising: determining means for determining. 2. A third calculating means for calculating the sum of the second bit amounts calculated by the first calculating means, wherein the first calculating means is provided by the third calculating means. The encoding device according to claim 1, wherein the second bit amount is calculated such that the sum of the calculated second bit amounts does not become larger than a predetermined reference value. 3. The first detecting means detects the complexity of the uncompressed data by calculating an index indicating the complexity of the uncompressed data, and the first calculating means comprises a first threshold value. And the size of the index are compared, and when the index is larger, the second bit amount having a positive value is calculated, and the second threshold smaller than the first threshold is compared with the index. If the index is smaller, the second bit amount having a negative value is calculated, and if the index is a numerical value between the first threshold and the second threshold, the second bit amount is calculated. The encoding device according to claim 1, wherein the bit amount of 2 is set to 0. 4. The uncompressed data is composed of a plurality of GOPs, and the index calculated by the first detecting means indicates an average difficulty level of 1GOP encoded in the past by the encoding means. The encoding device according to claim 3, wherein the encoding device is an index. 5. A second detection means for detecting the occurrence of a scene change, and a third detection means for detecting the image difficulty level of the same picture type before and after the scene change, the first detection means. When the scene change is not detected by the second detecting means, the calculating means of the complex of the uncompressed data previously encoded by the encoding means, detected by the first detecting means. Based on this, the second bit amount is calculated, and when the scene change is detected by the second detecting means, the second bit amount is calculated based on the detection result by the third detecting means. The encoding device according to claim 1, wherein 6. The first calculating means is positive if the scene change is detected by the second detecting means and the image difficulty level is higher after the scene change than before the scene change. 6. The encoding apparatus according to claim 5, wherein the second bit amount of the value is calculated, and the second bit amount is set to 0 in other cases. 7. The uncompressed data is composed of a plurality of GOPs, and the first calculating means calculates the second bit amount at a predetermined timing in the encoding process of the GOPs, The second calculation means is such that the newest second bit amount calculated by the first calculation means is a negative value,
Further, when the image difficulty level changes in the GOP by a predetermined value or more, the target bit is calculated by setting the second bit amount of the latest negative value calculated by the first calculating unit to 0. The encoding device according to claim 1, wherein 8. A coding method of a coding device for coding uncompressed data, comprising a detecting step of detecting complexity of the uncompressed data, and an encoding step of compressing and coding the uncompressed data. The VBV assigned to the uncompressed data to be encoded based on the complexity of the uncompressed data previously encoded by the encoding step, detected by the detecting step. A first calculation step of calculating a second bit amount to be added to the usable first bit amount of the buffer, and the second bit amount calculated by the process of the first calculation step. , A second calculation step of calculating a target bit based on a value added to the first bit amount, and the target calculated by the processing of the second calculation step. Ttobitto based on the encoding method characterized by including a determination step of determining a quantization index encoding in the process at the encoding step. 9. A program for an encoding device that encodes uncompressed data, the detecting step of detecting complexity of the uncompressed data, and the encoding step of compressing and encoding the uncompressed data. Based on the complexity of the uncompressed data previously encoded by the process of the encoding step, detected by the process of the detecting step, and assigned to the uncompressed data to be encoded from now on. A first calculation step of calculating a second bit amount to be added to the usable first bit amount of the VBV buffer, and the second bit amount calculated by the processing of the first calculation step. Based on a value added to the first bit amount, a second calculation step of calculating a target bit, and the target calculated by the processing of the second calculation step. Based on target bit, a recording medium from which a computer readable program is recorded, characterized in that it comprises a determining step of determining a quantization index encoding in the process at the encoding step. 10. A program executable by a computer that controls an encoding device that encodes uncompressed data, the detecting step of detecting the complexity of the uncompressed data, and the compression code of the uncompressed data. An encoding step to be encoded, and the uncompressed data to be encoded from now on the basis of the complexity of the uncompressed data previously encoded by the processing of the encoding step, which is detected by the processing of the detecting step. Of the VBV buffers allocated to the first bit amount that is added to the usable first bit amount; A second calculation step of calculating a target bit based on a value obtained by adding the second bit amount to the first bit amount; and the second calculation step. Based on the target bit calculated by the processing of flops, program characterized by comprising a determination step of determining a quantization index encoding in the process at the encoding step.