JP3206576B2 - Moving image encoding method, moving image encoding device, and storage medium storing moving image encoding program - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving picture coding method, a moving picture coding apparatus, and a storage medium storing a moving picture coding program. The present invention relates to a moving picture coding method for compressing and coding moving picture data at a variable rate to record a moving picture, a moving picture coding apparatus, and a storage medium storing a moving picture coding program.

[0002]

2. Description of the Related Art Conventionally, various methods have been proposed as a moving picture coding method for compressing and coding moving picture data. Among them, a moving picture coding method for compressing and coding moving picture data at a variable rate has been proposed. In this case, since the total generated code amount cannot be known in advance, there is a possibility that the recording time of the recording medium whose recording time is limited cannot be maintained. In this case, in order to maintain the initial recording time, if the bit rate is suddenly changed when the remaining capacity of the recording medium becomes small, the subjective video reproduction of the decoded moving image data may become difficult. Quality is not constant. In this regard, it is conceivable to set the average compression ratio to a higher value so as not to exceed the recordable capacity of the recording medium. In this case, however, mosquito noise, which is high-frequency noise, or a unit of image processing The block noise, which is a block-like transparent noise, is conspicuous, and the image quality of the entire moving image is degraded. Therefore, as described below, various techniques for maintaining the initial recording time and making the image quality of the entire moving image uniform have been conventionally proposed.

FIG. 10 is a block diagram showing an example of an electrical configuration of a conventional moving picture coding apparatus disclosed in Japanese Patent Application Laid-Open No. H10-174008. Hereinafter, this is referred to as a first conventional example. The moving image encoding apparatus of this example includes an encoding unit 1,
It comprises a recording medium 2, a free space detecting means 3, a generated code amount detecting means 4, and a code amount controlling means 5.
The encoding means 1 converts the moving image data into, for example, MPEG
(Moving Picture Experts Group) method to compress and encode encoded data and output. The recording medium 2 is composed of an optical disk, a magneto-optical disk, a magnetic drum, a magnetic tape, or the like, and stores encoded data supplied from the encoding unit 1. The free space detection unit 3 detects the free space of the recording medium 2 and supplies the same to the code amount control unit 5. The generated code amount detecting means 4 detects the generated code amount of the encoded data supplied from the encoding means 1 and supplies it to the code amount control means 5. The code amount control means 5 obtains the number of segments (constant time) corresponding to the scheduled recording time supplied from the outside, divides the free space supplied from the free space detection means 3 by the number of segments, and divides the free space per unit segment. A target code amount as a code amount is obtained, and the encoding unit 1 is controlled so that the generated code amount supplied from the generated code amount detection unit 4 approaches the target code amount. According to the moving picture coding apparatus having the above configuration, as shown in FIG. 11, the generated code amount converges to a substantially constant value (indicated by ● in the figure) for each segment. Averaging is performed and image deterioration is dispersed.

Japanese Patent Application Laid-Open No. 8-331555 discloses that a moving image data is compression-encoded into a first encoded data at a fixed bit rate and a large amount of information is encoded together with encoded information corresponding to a compression rate at that time. After recording on the first recording medium, the first encoded data is read from the first recording medium together with the encoded information and decoded into moving image data, and the moving image data is decoded based on the encoded information. A technique of compressively encoding the second encoded data at a variable rate and recording it on a second recording medium having a relatively small capacity so that the subjective image quality of each screen or each scene becomes uniform throughout the entire moving image. It has been disclosed. Hereinafter, this is referred to as a second conventional example. According to the second conventional example, it is possible to compress and encode moving image data with the best image quality according to the recordable capacity of the recording medium without worrying that the recordable capacity of the recording medium will be exceeded.

Further, Japanese Patent Application Laid-Open No. 10-98683 discloses an average in which the image quality after reproduction of moving image data to be recorded is substantially constant by dividing the recordable capacity of a recording medium by a recording time desired by a user. A technique for calculating a bit rate, setting the average bit rate as an initial bit rate, and changing the bit rate based on the remaining capacity of the recording medium, the remaining time of the recording time, and the complexity of the recorded content has been disclosed. I have. Hereinafter, this is referred to as a third conventional example.

Japanese Patent Laid-Open Publication No. Hei 10-98683 describes that a pre-process such as a noise filter process is performed before a coding unit, and a filter rate is adjusted to control a bit rate. This is to reduce the generated code amount by reducing the information amount of the moving image data to be coded. As a technique for adjusting a filter parameter of a filter provided in a preceding stage of the encoding means to reduce a generated code amount, for example, Japanese Unexamined Patent Application Publication No.
No. 276 and JP-A-8-79766. First, the technique disclosed in Japanese Patent Application Laid-Open No. 6-225276 discloses a technique in which a pass band of a filter provided before a coding unit and a quantization step size of a quantization unit constituting the coding unit are encoded in the past. By changing the coefficient of the filter for the obtained moving image data or the quantization step size of the quantizing means, etc. This is a technique for preventing image quality deterioration. Hereinafter, this is referred to as a fourth conventional example.

The technique disclosed in Japanese Patent Application Laid-Open No. Hei 8-79766 discloses a technique in which moving image data is encoded a plurality of times, and the moving image data is encoded in the final encoding in accordance with the generated code amount. In addition to performing the filtering process, according to the generated code amount and the filter characteristics used in the filtering process,
This is a technique for determining a bit rate at the time of final encoding and encoding moving image data at a variable rate. Hereinafter, this is referred to as a fifth conventional example.

[0008]

The above-mentioned various conventional moving picture coding techniques have the following drawbacks and problems. First, in the first conventional example, as shown in FIG. 11, the encoding means 1 can be controlled so that the generated code amount approaches the target code amount. In the end of the scheduled recording time, if moving image data that changes more rapidly than at the beginning of the scheduled recording time is supplied, the target code amount cannot be maintained, If the compression ratio is not set to a high value, the remaining recordable capacity of the recording medium 2 will be exhausted before the scheduled recording time is reached. However, when the compression ratio is set to a high value, mosquito noise and block noise are conspicuous, and the image quality of the entire moving image is reduced. Also in the third conventional example, since the complexity of the recorded content is determined from the data that has been compressed and coded first, if the recorded content changes drastically between the first half and the second half, the first conventional example is used. The problem described above also occurs. In this regard, in the second conventional example, since the moving image data is once compressed and encoded into the first encoded data and recorded on the first recording medium, the total generated code amount can be grasped. Although the problem of the third conventional example does not occur, it is necessary to encode the moving image data twice,
The device becomes large, complicated and expensive.

In the fourth conventional example, the pass band of the filter and the quantization step size of the quantization means constituting the coding means are changed. For this purpose, a filter control circuit and a motion amount parameter measuring section are used. And they form a complicated loop with the encoding means, which has the disadvantage of complicating the device. Furthermore, in the fifth conventional example, since encoding processing is performed a plurality of times, processing time is required, and there is a problem that moving image data cannot be encoded in a short time.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has a simple and inexpensive configuration, and is capable of reducing the quality of moving image data that changes rapidly even in the latter half of the estimated recording time in a short time. It is an object of the present invention to provide a moving image encoding method, a moving image encoding device, and a storage medium storing a moving image encoding program that can encode a moving image.

[0011]

According to a first aspect of the present invention, there is provided a moving image encoding method for compressing and encoding moving image data at a variable rate and recording the compressed data on a recording medium. From the start of recording on the recording medium to the present, the total generated code amount which is the total of the code amounts generated as a result of the compression encoding of the moving image data, the recordable capacity of the recording medium, and the image quality set by the user When the difference value from the total generated code amount at the target average rate, which is the target bit rate for maintaining the image quality of the recording mode determined from the recording mode, exceeds at least one threshold value. Changes the compression ratio in the compression encoding stepwise according to the threshold value, and the change in the compression ratio
In both cases, a hysteresis characteristic is given to one threshold value.
Cormorant is characterized in that.

According to a second aspect of the present invention, there is provided the moving picture encoding method according to the first aspect, wherein the changing of the compression ratio includes changing a ratio of removing a predetermined component from moving image data before compression encoding. It is characterized by performing by doing.

According to a third aspect of the present invention, there is provided the moving picture coding method according to the second aspect, wherein the predetermined component includes noise,
It is characterized by a slight change between current moving image data and past moving image data.

According to a fourth aspect of the present invention, there is provided the moving picture encoding method according to the second or third aspect, wherein the predetermined component is a diagonal component of the moving image data.

According to a fifth aspect of the present invention, there is provided the moving picture coding method according to any one of the first to fourth aspects, wherein the change of the compression ratio is performed in accordance with the property of the moving picture data. It is characterized by:

[0016]

According to a sixth aspect of the present invention, there is provided a moving image encoding apparatus, comprising: encoding means for compressing and encoding moving image data at a variable rate and recording the same on a recording medium; Generating code amount detecting means for detecting a total generated code amount which is a total of code amounts generated as a result of the moving image data being compression-encoded by the encoding means; recording the total generated code amount; The difference value between the total generated code amount at the target average rate, which is the bit rate to be targeted for maintaining the image quality of the recording mode determined from the available capacity and the recording mode related to the image quality set by the user, is When at least one threshold value is exceeded, the compression ratio in the encoding means is increased stepwise according to the threshold value.
And a changing means for changing with the hysteresis characteristic .

According to a seventh aspect of the present invention, there is provided the moving picture encoding apparatus according to the sixth aspect , further comprising a filter for removing a predetermined component from the moving picture data and supplying the same to the encoding means, Is characterized in that the compression ratio is changed by changing a rate of removing the predetermined component in the filter.

[0019] The invention according to claim 8 relates to the moving picture encoding apparatus according to claim 7, wherein the filter from the moving picture data, noise or, in the current video data and the past video data It is characterized by removing minute changes.

According to a ninth aspect of the present invention, there is provided the moving picture coding apparatus according to the seventh or eighth aspect , wherein the filter removes an oblique component of the moving picture data from the moving picture data. And

The invention according to claim 10 is the invention according to claims 6 to 9
In the moving picture coding apparatus according to any one of the above, the changing means changes the compression ratio in accordance with a property of the moving picture data.

[0022]

The storage medium according to the invention of claim 11 is:
A video encoding program for causing a computer to realize the function according to any one of claims 1 to 10 is stored.

[0024]

According to the structure of the present invention, moving image data can be encoded with a simple and inexpensive structure and in a short time without deteriorating the image quality even in the latter half of the scheduled recording time.

[0025]

Embodiments of the present invention will be described below with reference to the drawings. The description will be specifically made using an embodiment. A. First Embodiment First, a first embodiment will be described. FIG. 1 is a block diagram showing an electrical configuration of a moving picture coding apparatus according to a first embodiment of the present invention. The moving picture coding apparatus of this example includes a filter 11, a coding unit 12, a recording medium 13
, A generated code amount detection means 14 and a code amount control means 15. The filter 11 is composed of a noise reduction filter that removes uncorrelated noise and minute changes by utilizing the correlation between adjacent frames, and the control signal S supplied from the code amount control unit 15. Based on _C1 , noise and the like are removed from the input moving image data and output as noise removal data. The filter 11, as shown in FIG.
, A converting means 22, an adding means 23, and a frame memory 24.

The subtraction means 21 subtracts the moving image data input in units of frames from the noise removal data one frame before supplied from the frame memory 24, and outputs a subtraction result. Converting means 22 includes, for example, a ROM, a control signal based on the _{S C1} level of the noise removal is set to one of level _L 0 ~L _6, data conversion level _L 0 ~L ₆ that has been set Referring to the table, the subtraction result is converted into predetermined data and output. Addition means 2
3 adds the data supplied from the conversion means 22 and the noise removal data one frame before supplied from the frame memory 24, and outputs the addition result as noise removal data. The frame memory 24 is composed of a semiconductor memory such as a RAM, for example, and stores and outputs noise removal data for one frame.

FIG. 3 shows each level L of the conversion means 22.
₀ shows an example of a portion of the data conversion table ~L every _six. 3 (1) is an example of a data conversion table of the level L ₀ (omitted value 2 or more is the same as value 1), the input data (subtraction result) is directly output data of the conversion means 22, That is, no processing is performed. The numerical values represent the values after quantization. 3 (2) is an example of the level L ₁ data conversion table, the value of the input data is 0 and 10
In the above case, the output data is used as it is. When the value of the input data is 1, 2, 8, and 9, a value smaller by 1 than the value is set as the output data. When the value of the input data is 3 to 7, the value is set as the output data. A value smaller by 2 is used as output data. 3 (3) is an example of the level L ₂ data conversion table, it becomes output data when the value of the input data is 0 and 15 or more, when the value of the input data is 1 and 14 than that value A value smaller by 1 is used as the output data, and when the value of the input data is 2, 3, 12, and 13, it is 2
Is smaller than the output data, and the value of the input data is 4 to
In the case of 6 and 9 to 11, a value smaller by 3 than that value is set as output data, and when the value of input data is 7 and 8, a value smaller by 4 is set as output data. FIG.
(4) is an example of the level L ₃ data conversion table, it becomes output data when the value of the input data is 0 and 20 or more, when the value of the input data is 1 and 19 is only 1 than the value When the value of the input data is 2, 3, 17, and 18, a smaller value is used as the output data.
Output data, and the value of the input data is 4,
In the case of 5, 15, and 16, a value smaller by 3 than that value is set as output data, and the value of the input data is 6, 7, 13, and 1.
In the case of 4, a value smaller by 4 than that value is set as output data, and when the value of the input data is 8 to 12, a value smaller by 5 than that value is set as output data. Hereinafter, although not shown, the data conversion tables of the levels L _{4 to} L ₆ are similar to the above-described data conversion tables of the levels L _{1 to} L _3. The relationship between the input and output data is set so as to increase the difference from the output data, that is, to enhance the characteristics of the filter.

In FIG. 1, the encoding means 12 comprises:
The noise removal data supplied from the filter 11 is converted into a control signal S _C2 supplied from the code amount control means 15 by using a variable rate encoding method such as the MPEG method.
Is compressed and coded into coded data having a code amount based on, and is output. The recording medium 13 is an optical disk, a magneto-optical disk,
It is made of a magnetic drum, a magnetic tape, or the like, and stores encoded data supplied from the encoding means 12. The generated code amount detecting means 14 detects the total generated code amount of the coded data supplied from the coding means 12 and supplies it to the code amount control means 15. The code amount control unit 15 outputs a control signal _SC2 for controlling the generated code amount of the encoding unit 12 based on the target average rate supplied from the outside and the generated code amount supplied from the encoding unit 12. The generated code amount is supplied to the encoding means 12, and when the difference value between the total generated code amount supplied from the generated code amount detecting means 14 and the total generated code amount at the target average rate exceeds a predetermined threshold value, the filter is performed. 11
It supplies a control signal S _C1 for changing the setting of the level of the noise removal filter 11.

Here, the target average rate is not the bit rate obtained by dividing the recordable capacity of the recording medium as in the third conventional example described above by the scheduled recording time, but the recordable capacity of the recording medium 13. And a recording mode (high-quality mode, normal-quality mode, low-quality mode, etc.) set by the user, and is a bit rate to be targeted in order to maintain the image quality of the recording mode. FIG. 4 shows an example of the time characteristic of the bit rate of the encoding means 12 which is varied by the control signal _SC2 . In this embodiment, as in the first conventional example (see FIG. 11), the bit rate is not converged to a substantially constant value for each predetermined period regardless of the content of the moving image data. By setting an upper limit ULT and a lower limit DLT experimentally obtained for each of the recording modes with respect to the target average rate OAR, and controlling the target average rate OAR to fall within the range, the degree of freedom in controlling the bit rate is increased to some extent. I have it.

Next, the operation of the moving picture coding apparatus having the above configuration will be described with reference to the flowchart shown in FIG. First, the code amount control unit 15 proceeds to step SP1, generates a control signal _SC1 for setting an initial characteristic (for example, level L ₃ ) of the filter 11, and generates the control signal _SC1.
And calculates a target average code amount based on a target average rate supplied from the outside, generates a control signal _SC2 indicating the target average code amount, supplies the control signal _SC2 to the encoding means 12, and then proceeds to step SP2. move on. Thereby, the filter 1
1, the control signal based on the S _C1, the level L ₃ as the initial characteristic is set to remove the noise or the like from the moving image data input at that level L ₃ to the coding unit 12 as noise removal data Supply. Therefore, the encoding unit 12 converts the noise removal data supplied from the filter 11 based on the control signal _SC2 supplied from the code amount control unit 15 using a variable rate encoding method such as the MPEG method. The encoded data having the target average code amount is compression-encoded and recorded on the recording medium 13. Further, the generated code amount detecting means 14 detects the total generated code amount of the coded data supplied from the coding means 12 and supplies it to the code amount control means 15 at predetermined time intervals.

In step SP2, the code amount control means 15
Generates a new control signal _SC2 based on the target average rate and the generated code amount newly supplied from the encoding unit 12, supplies the control signal _SC2 to the encoding unit 12, and generates a control signal _SC2 from the generated code amount detection unit 14. After inputting the total generated code amount supplied every predetermined time, the process proceeds to step SP3. Step SP
3, the code amount control unit 15 sets the generated code amount detection unit 1
After calculating the difference value between the total generated code amount supplied from 4 and the total generated code amount at the target average rate, the process proceeds to step SP4.

In step SP4, the code amount control means 15
Determines whether the difference value is greater than the upper threshold UTH (in this case, the upper threshold UTH ₄ ), which is one level higher than the current level. Such determination is made for the following reason. That is, as shown by the curve in FIG.
In the vicinity (boundary between the level L ₃ and level L ₄ are in the example of FIG. 6) adjacent the level of the boundary, when the total generated code amount is repeatedly increased or decreased, it is varied immediately react with the characteristics of the filter 11, Flicker-like noise occurs. Therefore, in this embodiment, the upper threshold UTH and the lower threshold DTH (the upper threshold UTH ₄ and the lower threshold DTH _{4 in} the example of FIG. 6) are provided at the boundary between adjacent levels, and the total generated code amount is once set to the upper threshold UTH. After becoming larger, the characteristic level of the filter 11 is not changed to a lower level even if it becomes smaller than the upper threshold UTH again.
Only when it becomes smaller, the characteristic level of the filter 11 is changed to a lower level, and the total generated code amount is temporarily reduced to the lower threshold DT.
After becoming smaller than H, the level of the characteristic of the filter 11 is not changed to the upper level even if it becomes larger than the lower threshold DTH again, and the level of the characteristic of the filter 11 is not changed to the upper level until it becomes larger than the upper threshold UTH. The level is changed, that is, the hysteresis characteristic is provided.

If the decision result in the step SP4 is "YES", that is, the difference value is the upper threshold value UTH one level higher than the current level (in this case, the upper threshold value UT
If H ₄ ), the code amount control unit 15 proceeds to step SP5. In step SP5, the code amount control unit 15 generates a control signal _SC1 for changing the setting of the noise removal level of the filter 11 to a level slightly higher (in this case, the level L ₄ ), and supplies the control signal _SC1 to the filter 11. After that, the process returns to step SP2. Thereby, the filter 11
Since There supplied to the coding means 12 as altered level L ₄ more removed to the amount of information is reduced noise removed data noise such as from the moving image data input in the encoding unit 12, the filter 11 _Is compressed and encoded into coded data of a code amount based on the control signal _SC2 newly supplied from the code amount control means 15 using a variable rate coding method, and the resultant data is recorded on the recording medium 13. Record. Further, the generated code amount detecting means 14 detects the total generated code amount of the coded data supplied from the coding means 12 and supplies it to the code amount control means 15 at predetermined time intervals.

On the other hand, if the result of the determination in step SP4 is "N
O ”, that is, the difference value is an upper threshold value UTH that is one level higher than the current level (in this case, the upper threshold value U
If TH ₄₎ or less, the code amount control means 15,
Without doing anything, the process proceeds to step SP6. In step S6, the code amount control unit 15, the difference value is (in this case, the lower threshold DTH ₃₎ below threshold DTH the current level determines whether the difference is less than. If the result of this determination is "YES", the code amount control means 15 proceeds to step SP7.
In step SP7, the code amount control unit 15 generates a control signal _SC1 for changing the setting of the noise removal level of the filter 11 to one level lower (in this case, the level L ₂ ), and supplies the control signal _SC1 to the filter 11. After that, the process returns to step SP2. As a result, the level L at which the filter 11 is changed
₂ is supplied to the encoding unit 12 as noise-removed data in which the amount of information is increased by removing less noise and the like from the moving image data input in step ₂ , the encoding unit 12 removes the noise supplied from the filter 11. The data is compressed and coded into a coded data of a code amount based on the control signal _SC2 newly supplied from the code amount control means 15 using a variable rate coding method, and is recorded on the recording medium 13. Further, the generated code amount detecting means 14 outputs the coding means 1 every predetermined time.
2 and detects the total generated code amount of the coded data supplied from 2 and supplies it to the code amount control means 15.

On the other hand, if the result of the determination in step SP6 is "N
For O ", i.e., (in this case, the lower threshold DTH ₃₎ the difference value is below the threshold DTH current level is greater than or equal, the code amount control unit 15, without doing anything, the step SP2
Return to Accordingly, since the filter 11 is supplied to remove noise and the like from the moving image data input remains at level L ₃ which is set as the initial characteristics to the coding means 12 as noise removal data, the encoding means 12, Filter 1
1 is compressed and coded into coded data of a code amount based on the control signal _SC2 newly supplied from the code amount control means 15 using a variable rate coding method, and the recording medium 13 To record. Further, the generated code amount detecting means 14 detects the total generated code amount of the coded data supplied from the coding means 12 and supplies it to the code amount control means 15 at predetermined time intervals. Thereafter, the above-described processing is repeated until the remaining capacity of the recording medium 13 is exhausted. In this case, the control signal S _C1 in the filter 11
Of level switching noise removal based on, for when the difference value is positive, as shown in FIG. 7, the threshold value TH ₀ to TH ₇ and the threshold value TH ₀ of the difference value for switching to the adjacent level- stepwise reduced in accordance with elapsed time threshold UTH and lower threshold DTH on corresponding to TH _7. With this control, as shown by the curve a in FIG.
As compared with the case of the first conventional example shown by (1), it is possible to suppress going away from the target average rate OAR.

As described above, according to the configuration of this example, the first
Instead of converging the generated code amount for each predetermined cycle to a substantially constant value irrespective of the content of the moving image data as in the related art, the bit rate is set to the target average rate OAR for each recording mode. Upper limit ULT obtained experimentally
And the lower limit DLT are set and controlled so as to be within the range, so that there is a certain degree of freedom in the control of the bit rate. Even when moving image data that changes rapidly is supplied, it is possible to prevent the bit rate from deviating from the target average rate. Therefore,
It is possible to prevent mosquito noise and block noise from occurring, and to prevent a decrease in the image quality of the entire moving image. Further, according to the configuration of this example, the moving image data is encoded by one encoding process, and the code amount control unit 15 uses the control signals S _C1 and S _C2 to control the level of the filter 11 and the encoding unit 12. Since only the code amount is controlled, the second and fifth encodings performed a plurality of times are performed.
In comparison with the conventional example of Example 4 and the fourth conventional example in which the quantization step size of the quantization means constituting the encoding means is also changed, the moving picture coding apparatus can be configured simply and inexpensively, and the moving picture data can be recorded in a short time. Can be encoded. Furthermore, according to the configuration of this example, the filter 11 configured by the noise reduction filter removes components that have little influence on the reproduction of the moving image data, such as noise that does not originally need to be coded and minute changes that are difficult to perceive. By removing
Since the amount of information to be encoded is reduced, the image quality of the reproduced image does not deteriorate. Also, according to the configuration of this example,
Since the filter 11 has a hysteresis characteristic, it is possible to suppress the occurrence of flicker-like noise caused by frequent changes in the characteristic of the filter 11.

B. Second Embodiment Next, a second embodiment will be described. FIG. 9 is a block diagram illustrating an example of an electrical configuration of a filter 31 included in a moving picture encoding device according to a second embodiment of the present invention. Since the configuration of the moving picture coding apparatus other than the filter 31 is substantially the same as that of the first embodiment,
The description is omitted. The filter 31 of this example is constituted by an oblique direction component removing filter for removing oblique direction components of moving image data which is hardly perceived by human eyes, and the control signal S _C1 supplied from the code amount control unit 15.
, The oblique direction component of the moving image data is removed from the input moving image data and output as oblique direction component removal data. The filter 31 includes a horizontal low-pass filter (hereinafter, referred to as LPF) 32 and a vertical LPF 33.
It is schematically composed of

The horizontal LPF32, the delay means ₄₁ 1-4 for delaying one pixel each pixel in the horizontal direction of the video data
1 ₄ and a multiplication means ₄₂ 1 to 42 ₅ for multiplying each coefficient a1~a5 the output data of the moving image data and the delay means ₄₁ 1 to 41 _4, multiplying means 42 ₁ of the output data and the multiplier 42 ₂ outputs Addition means 43 for adding data
_1, the addition means 43 ₂ for adding the output data of the multiplying means 42 ₃ and output data of the adding means 43 _1, the adding means 4
3 and ₂ of the output data and the multiplier 42 adder means 43 ₃ for adding the output data of _4, adding means for adding the output data of the multiplying means 42 ₅ and the output data of the adding means 43 ₃ 4
It is schematically composed of 3 _4.

The vertical LPF 32 delays each of the lines of the moving image data by one line by delay means 51 _{1 to} 51 _1.
4, multiplying means 5 for multiplying each coefficient b1~b5 the output data of the moving image data and the delay means ₅₁ 1 to 51 ₄
2 ₁ to 52 _5, and adding means 53 ₁ for adding the output data of the multiplying means 52 ₁ and output data of the multiplying means 52 _2,
And adding means 53 ₁ of the output data and adding means 53 ₂ for adding the output data of the multiplying means 52 _3, the adding means 53 ₃ for adding the output data of the multiplying means 52 ₄ and the output data of the adding means 53 _2, It is schematically configured from the adding means 53 ₄ for adding the output data of the adding means 53 ₃ and output data of the multiplying means 52 _5. The coefficient a1~a5 and coefficient b1~b5 is, for example, a real number from 0 to 1, similar to the filter 11 of the first embodiment, the filter characteristic is divided into six levels _L 0 ~L ₆ And each level L
₀ combination of values ~L coefficient for each ₆ a1-a5 and coefficient b1~b5 is set in advance, for example, although not shown, it assumed to be previously stored in a ROM provided inside of the filter 31. When a control signal S _C1 for selecting a certain level is supplied from the code amount control unit 15, the values of the coefficients a1 to a5 and the coefficients b1 to b5 corresponding to the level selected by the control signal S _C1 are obtained. the combination is read from the ROM, the multiplying means ₄₂ 1 to 42 ₅
And it is set to ₅₂ 1 to 52 _5.

Next, the operation of the moving picture coding apparatus having the above configuration will be described. First, operations other than the control of the filter 31 (steps SP5 and SP7 in FIG. 5) are substantially the same as those in the above-described first embodiment, and a description thereof will be omitted. Next, the control of the filter 31, for example, when strengthening the filter characteristic from the level L ₃ to level L ₄ are, the control signal S for selecting the level L ₄ from code amount control unit 15 to the filter 31 Supply _C1 . As a result, in the filter 31, the control signal S
Coefficients corresponding to the level _{L 4} selected by _C1 a1 to
The combination of the values of a5 and the coefficients b1 to b5 is the filter 3
1 is read from the internal ROM, and each of the multiplication means 42 ₁ to 4 ₁
Because it is set to 2 _5, and ₅₂ 1 to 52 _5, the horizontal direction L
In PF32 and vertical LPF 33, the horizontal and vertical components of the moving image data is deleted more than in the case of level L _3, resulting in comparison with the case oblique direction component of the moving image data of level L ₃ many are removed from the hand and is supplied to the coding unit 12 as the information amount is less oblique direction component removal data as compared with the level L _3.

On the other hand, when weakening the filter characteristic from the level _{L 3} of the level _{L 2} supplies a control signal _{S C 1} to select the level _{L 2} from the code amount control unit 15 for the filter 31. Thus, in the filter 31, the control signal combination of the value of the coefficient a1~a5 and coefficient b1~b5 corresponding to the level _{L 2} selected by _{S C1} is read out from the internal filter 31 ROM, the multiplying means 42 ₁ to 42 because it is set to ₅ and ₅₂ 1 to 52 _5, in the horizontal direction LPF32 and vertical LPF 33, the horizontal and vertical components of the moving image data is deleted less than that of level _{L 3,} the results to an oblique direction component of the moving image data is deleted less than that of level L _3, the coding unit 12 as the information amount is large oblique direction component removal data as compared with the level L ₃
Supplied to

As described above, according to the configuration of this example, the filter 31 constituted by the diagonal direction component removing filter removes the diagonal direction component of the moving image data which is hardly perceived by human eyes, thereby achieving the encoding. Since the amount of information to be reproduced is reduced, the image quality of the reproduced image does not deteriorate as in the first embodiment.

The embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to this embodiment, and changes in design and the like can be made without departing from the gist of the present invention. Even if there is, it is included in the present invention. For example, in each of the above-described embodiments, a case has been described in which the MPEG system is adopted as the encoding system of the moving image data in the encoding unit 12, but the present invention is not limited to this, and any encoding system of a variable rate may be used. Such an encoding method can be adopted. Further, in each of the above-described embodiments, the example in which the filter 11 and the filter 31 are separately provided has been described.
The configuration is not limited to this, and the filter 11 and the filter 31 may be configured to be provided in cascade. According to such a configuration, it is possible to efficiently compress and encode moving image data without deteriorating image quality. Further, in each of the above-described embodiments, no consideration is given to the properties of the input moving image data, but the present invention is not limited to this. The nature of the moving image data includes a sports program in which the appearing person or object moves violently, and a conversely, a talk program or an art introduction program in which the appearing person or object hardly moves. Therefore, the user is caused to input information relating to these by a switch operation or the like, and in the case of the filter 11, each level L _{0 to} L is set according to the property of each moving image data.
Provided ₆ data conversion table, when the filter 31 are prepared with combinations of the values of the coefficients a1~a5 and coefficient b1-b5, each level L _0, depending on the nature of the input moving image data ~L ₆ data conversion table and coefficient a1
It may be configured to select a combination of values of .about.a5 and coefficients b1 to b5. According to such a configuration, it is possible to perform compression encoding that matches the properties of moving image data.

In each of the above embodiments, each means
In the example shown above, hardware was used, but this is not a limitation.
Not. That is, the moving picture encoding device is connected to a CPU (central
Processing device), an internal storage device such as a ROM or a RAM, and F
DD (floppy disk driver), HDD (c
Disk driver), CD-ROM driver, etc.
Including an external storage device, an output unit, and an input unit.
Computer 11 and the filter 11 described above.
The subtraction means 21, the addition means 23, and the filter 31
Multiplying means 42₁~ 42₅, Multiplication means 52₁~ 5
2₅, Adding means 43 ₁~ 43₄And addition means 53₁~ 5
3₄, Encoding means 12, generated code amount detecting means 13, and
The code amount control means 15 is constituted by a CPU.
Functions as a moving picture coding program
Conductive memory and storage media such as FD, HD and CD-ROM
May be configured to be stored. In this case,
An external storage device or an external storage device constitutes the filter 11.
Conversion means 22, frame memory 24, filter
Delay means 41 constituting 31₁~ 42₄And delay means 5
1₁~ 51₄As well as a recording medium 13 for moving picture coding.
The program is read from the storage medium by the CPU,
Control the operation of U. The CPU executes the moving picture coding program.
When the system is activated, the subtraction operation constituting the filter 11 is performed.
Stage 21 and addition means 23, multiplication forming filter 31
Means 42₁~ 42₅, Multiplication means 52₁~ 52₅, Adder
Step 43₁~ 43₄And addition means 53₁~ 53₄,Coding
Means 12, generated code amount detecting means 13, and code amount control means
Functions as stage 15 for controlling moving picture coding programs
Thus, the above-described processing is executed.

[0045]

As described above, according to the configuration of the present invention, when the difference value between the total generated code amount and the total generated code amount at the target average rate exceeds at least one threshold value, Since the compression ratio in the compression encoding is changed stepwise according to the threshold value, a simple and inexpensive configuration is used.
In addition, the change in the second half of the scheduled recording time is short.
Intense moving image data can be encoded without deteriorating the image quality and further, changes the compression ratio is at least
This is done with a hysteresis characteristic for one threshold.
Therefore, the generation of flicker noise can be suppressed.
Wear. Further, according to another configuration of the present invention, it is possible to remove noise that does not originally need to be encoded, small changes that are difficult to perceive, and components that have little effect on the reproduction of moving image data, such as oblique components of moving image data. As a result, since the amount of information to be encoded is reduced, the image quality of the reproduced image does not deteriorate.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an electrical configuration of a moving picture encoding device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating an example of an electrical configuration of a filter included in the device.

FIG. 3 shows each level L _{0 of} the conversion means constituting the filter.
Is a diagram showing an example of a portion of a data conversion table ~L every _six.

FIG. 4 is a diagram illustrating an example of a time characteristic of a bit rate of an encoding unit that is changed by a control signal _SC2 .

FIG. 5 is a flowchart illustrating an example of an operation of controlling a filter of a code amount control unit included in the apparatus.

FIG. 6 is a diagram for explaining an example of an operation of controlling a filter of a code amount control unit included in the apparatus.

FIG. 7 is a diagram for explaining an example of an operation of controlling a filter of a code amount control unit included in the apparatus.

FIG. 8 is a diagram for explaining a difference between a relationship between an elapsed time and a total generated code amount in the present invention and the first conventional example.

FIG. 9 is a block diagram illustrating an example of an electrical configuration of a filter included in a video encoding device according to a second embodiment of the present invention.

FIG. 10 is a block diagram illustrating an electrical configuration of a moving image encoding device that is a first conventional example.

FIG. 11 is a diagram showing a relationship between elapsed time and a total generated code amount in the conventional example.

[Explanation of symbols]

 11, 31 Filter 12 Encoding means 13 Recording medium 14 Generated code amount detecting means 15 Code amount controlling means (changing means)

Claims (11)

(57) [Claims] 1. A moving image encoding method for compressing and encoding moving image data at a variable rate and recording the moving image data on a recording medium, the result of compression encoding the moving image data from the start of recording on the recording medium to the present time A target is set to maintain the image quality of the recording mode determined from the total generated code amount that is the sum of the generated code amounts and the recording mode related to the image quality set by the user and the recordable capacity of the recording medium. is a bit rate to when the difference value between the total generated code amount at the target average rate exceeds at least one threshold, and change the compression ratio in stepwise compression encoding in accordance with the threshold value, the Change of compression ratio is at least one
A moving image encoding method characterized in that a hysteresis characteristic is given to the threshold value of (i) . 2. The moving picture coding method according to claim 1, wherein the change of the compression ratio is performed by changing a rate of removing a predetermined component from the moving picture data before compression coding. . 3. The moving picture coding method according to claim 2, wherein the predetermined component is noise or a slight change between current moving picture data and past moving picture data. 4. The moving picture coding method according to claim 2, wherein the predetermined component is an oblique direction component of the moving picture data. 5. The moving picture coding method according to claim 1, wherein the change of the compression ratio is performed in accordance with a property of the moving picture data. 6. An encoding means for compressing and encoding moving image data at a variable rate and recording it on a recording medium, wherein the encoding means compresses and encodes the moving image data from the start of recording on the recording medium to the present. Generated code amount detecting means for detecting a total generated code amount which is a cumulative total of code amounts generated as a result of the conversion, a recording mode relating to the total generated code amount, a recordable capacity of the recording medium, and an image quality set by a user. If the difference value from the total generated code amount at the target average rate which is the bit rate to be targeted for maintaining the image quality of the recording mode determined from the above exceeds at least one threshold value, A moving picture coding apparatus comprising: changing means for changing a compression ratio of the coding means in a stepwise manner with a hysteresis characteristic according to a threshold value. 7. A filter for removing a predetermined component from the moving image data and supplying the removed component to the encoding unit, wherein the changing unit changes the ratio of the filter for removing the predetermined component to change the ratio. 7. The moving picture coding apparatus according to claim 6, wherein the compression ratio is changed. Wherein said filter is from the moving image data, noise or claim 7, characterized in that the removal of minute changes in the current video data and the past video data
The moving picture encoding device according to the above. Wherein said filter is from the moving picture data, moving picture encoding apparatus according to claim 7 or 8, wherein removing the oblique direction component of the moving image data. Wherein said change means moving image coding apparatus according to any one of claims 6 to 9, characterized in that to change the compression ratio according to the nature of the moving image data. 11. A storage medium storing a moving picture encoding program for realizing the functions according to any one of claims 1 to 10 in a computer.