JP2002014696A - Device and method for control - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a device and a method for control in which signal quality is further improved by improving the quality of coded signals. SOLUTION: A bit reserver level setting means 108 is provided to set a bit reserver level 'a', which is a target value of the number of bits to be stored in a bit reserver 106 in an adaptable manner, from a transmission bit rate 'b', the number of bits stored in a bit reserver 106 and the number of request bits necessary to code a present process frame. In accordance with the level 'a' which is set in an adaptable manner, the number of bits to be stored in the reserver 106 is controlled with a bit reserver control means 107 to prevent the quality deterioration of coded signals caused by the reduction in the number of bits useable for coding over a long time duration in an especially low bit rate and to further improve the signal quality.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control apparatus and a control method for controlling the number of coded bits per frame when a video or audio signal is coded per frame and transmitted at a fixed bit rate.

[0002]

2. Description of the Related Art Numerous methods have been devised and put into practical use for video and audio signal encoding devices using digital signal processing. As a representative example, MPEG (Moving Pi) has been proposed.
culture Experts Group) standard. The MPEG standard employs an irreversible coding method for reducing and coding a part of information of original video and audio in order to realize high quality at a low bit rate. In particular, if it is necessary to transmit at a very low bit rate with respect to the original, many parts of the original information must be reduced, and it is very difficult to determine which part of the information should be reduced. It becomes difficult.

For this reason, in the MPEG standard, based on the characteristics of human perception, information of a part that is perceived sensitively is left, information of a part that is not perceived is reduced, and the bit rate is reduced as much as possible without reducing quality. Can be greatly reduced. As described above, the use of human perceptual characteristics is very effective in reducing the overall bit rate, but on the other hand, it is perceived sensitively when viewed partially, for example, in units of encoded processing frames. It can be said that the information that can be reduced is small in the frame including many parts (the number of bits required for encoding is large), and the information that can be reduced is large in the frame including the part that is not much perceived (the number of bits required for encoding is small). The required number of bits will be non-uniform in frame units.

On the other hand, in a device for transmitting coded bits, it is required that a bit signal per unit time flowing through a transmission path, that is, a bit rate, is constant because the complexity of the device can be reduced. Often required. In order to fill such a gap in characteristics between the encoding device and the transmission device, a buffer is provided between the encoding device and the transmission device, and the encoded bit signal output from the encoding device is temporarily stored in the buffer. A method of extracting an encoded bit signal from a buffer at a fixed bit rate and transmitting the encoded bit signal is used.

One example is an MPEG-2 AAC (ISO) which is one of the audio coding systems included in the MPEG standard.
/ IEC IS-13818-7 Advanced
A detailed description will be given along the method used in Audio Codec) with reference to FIG.

In MPEG-2 AAC, encoding means 1
A buffer called a bit reservoir 106 is provided between 01 and a transmission unit (not shown). The size of the buffer is determined by the number of channels of the audio signal to be encoded.
6144 × 2 = 1228 for 4-bit stereo signal
8 bits.

[0007] Of the reserved buffer area, a portion corresponding to the number of bits allocated per coding processing frame when the fixed bit rate is considered is a buffer area (frame buffer 105) used during transmission. , A bit reservoir 1 for actually storing encoded bits
The size of 06 is obtained by subtracting a portion corresponding to the number of bits per frame from the reserved buffer.

For example, a sampling frequency of 44100H
When encoding a stereo signal of z at 128 kbps, AAC processes 1024 samples as one frame, so the number of bits allocated per frame FRAME_BITS is: FRAME_BITS = 128000 × 1024/44100 = 2972 bits...・ (1)

The size of the bit reservoir BR_SI
ZE is BR_SIZE = 12288−FRAME_BITS = 9316 bits (2)

In the equation (1), fractional bits below the decimal point are generated. The fractional bits are corrected by a method such as adding the fractional bits and increasing FRAME_BITS by an integral value when the fractional value is exceeded. .

If the number of bits used in the encoding process on a frame-by-frame basis and the number of bits transmitted between frames are always equal, after the encoded bits of a certain frame are written in the area of the frame buffer 105, By the time the encoding processing of the frame is completed, transmission of all encoded bits in the frame buffer 105 is completed.

Therefore, when the encoding process for the next frame is completed, the frame buffer 105 is empty, and the encoded bits can be written into the buffer area again. However, if the number of coded bits in each frame is different, the number of coded bits to be output may be too large or small with respect to the size of the frame buffer 105.

The bit reservoir 106 absorbs such an excess or deficiency of coded bits in a frame unit, and always transmits coded bits having a frame buffer size per frame, that is, transmits at a fixed bit rate. In principle, when the number of coded bits in a certain frame is larger than the frame buffer size, the coded bits in a portion exceeding the frame buffer size are stored in the bit reservoir 1.
06, and conversely, when the number of coded bits is smaller than the frame buffer size, the coded bits stored in the bit reservoir 106 are taken out, and
By performing the operation of adding until the frame buffer 105 is full, it is possible to control the number of coded bits transmitted in frame units so as to be constant at the frame buffer size.

[0014]

However, in the conventional method, the upper limit value of the bit signal stored in the bit reservoir is limited only by the bit reservoir size. Cannot be allocated, which may result in degradation of the quality of the encoded signal.

Since the size of the bit reservoir is a fixed value determined in advance by a standard or the like for compatibility of the transmitting and receiving apparatus, the transmission apparatus intended for use at a high transmission bit rate has a low transmission capacity. When used at the bit rate, the size of the bit reservoir relative to the transmission bit rate may be large. As a result, the time required to store the bit signal in the bit reservoir is inevitably increased, and the time required for encoding at a bit rate lower than the original transmission bit rate is increased. May lead to deterioration of the product.

SUMMARY OF THE INVENTION The present invention has been made in view of such circumstances, and by performing optimal bit allocation for encoding at a low bit rate, the quality of an encoded signal is improved to improve the signal quality. It is an object of the present invention to provide a control device and a control method that can be performed.

[0017]

In order to solve the above-mentioned problems, a control device according to the present invention comprises: an encoding unit for encoding an input signal in a frame unit having a predetermined time length; Buffer means for storing bit signals; calculating means for calculating the number of bits required to encode the input signal in frame units; and number of bits used by the coding means based on the calculated required number of bits and the buffer level Control means for controlling the buffer level so as to fall within a predetermined value range, and adaptively controlling the predetermined value range of the buffer level based on the transmission bit rate and the required number of bits. And so on.

With this configuration, the coded bit signal is temporarily stored, the number of bits required to encode the input signal in frame units is calculated, and the calculated required number of bits and the buffer level are calculated. By adjusting the number of bits used by the encoding means as a reference, the buffer level is controlled so as to fall within a predetermined value range,
By adaptively controlling the range of the predetermined value of the buffer level based on the transmission bit rate and the required number of bits, and performing optimal bit allocation for encoding, especially at a low bit rate, the encoded signal is Quality can be improved.

According to the control device of the present invention, a buffer means stores a coded bit signal, which is an output of the coding means, in frame units, and a bit for transmitting / receiving a surplus or deficient bit signal in cooperation with the frame buffer. It is configured to have a reservoir.

With this configuration, the coded bit signal output from the coding means is stored in the buffer in frame units, and the exchange of the surplus or insufficient bit signal is performed by the bit reservoir in cooperation with the frame buffer. In particular, by allocating the optimal bits for encoding at a low bit rate,
The quality of the encoded signal can be improved.

In the control device according to the present invention, the control means is provided with bit reservoir level setting means for setting a bit reservoir level based on a transmission bit rate, a bit reservoir, and an output from the calculating means, and the bit of the set bit reservoir is provided. The number of usable bits is transmitted to the encoding means based on the level, the output of the bit reservoir, and the output from the calculation means.

With this configuration, the bit reservoir level is adaptively set, and the number of usable bits is transmitted to the encoding means based on the set bit level of the bit reservoir, the output of the bit reservoir, and the calculated output. This makes it possible to improve the quality of the coded signal by allocating bits optimally for coding, especially at low bit rates.

[0023] The control device of the present invention comprises encoding means for encoding an input signal in frame units of a time length determined on the transmission side, transmission-side buffer means for storing encoded bit signals, Calculation means for calculating the number of bits required to encode the input signal in frame units, and coding means based on the calculated required number of bits and the buffer level indicating the amount of bit signals stored in the buffer. By controlling the number of bits to be controlled, the buffer level is controlled so as to fall within a predetermined range, and the predetermined range of the buffer level is adaptively controlled based on the transmission bit rate and the required number of bits. Receiving means for storing the coded bit signal sent from the transmitting buffer means on the receiving side; and The obtained decoding means decodes and outputs the encoded bit signal, and configured to include a.

With this configuration, the bit signal encoded on the transmission side is stored in the buffer, and the number of bits required to encode the input signal in frame units is calculated. By adjusting the number of bits used by the encoding means on the basis of the buffer level and the buffer level, the buffer level is controlled so as to fall within a predetermined range, and the predetermined range of the buffer level is defined as the transmission bit rate and the transmission bit rate. Adaptive control is performed based on the required number of bits to perform optimal bit allocation for encoding, especially at low bit rates. On the receiving side, the received encoded bit signals are temporarily stored in a buffer, and then encoded. The bit signal is decoded and output. In this way, by optimizing the bit allocation for encoding at the transmitting side, especially at low bit rates, and improving the quality of the encoded signal, decoding can be performed without changing the conventional receiving apparatus. The quality of the converted signal can also be improved.

In the control device according to the present invention, the encoding means performs irreversible encoding processing on the video or audio signal based on the human perception limit, and the calculation means performs encoding based on the human perception limit. It is configured to calculate necessary bit signals.

With this configuration, an irreversible encoding process based on a human perception limit is performed on a video or audio signal, and a bit signal required for encoding is calculated based on the human perception limit. In this way, it is possible to reduce the deterioration of the quality due to the encoding process, and at the same time, to optimize the bit distribution for encoding, especially at a low bit rate, and improve the quality of the encoded signal.

The control device of the present invention is configured to control the control means so that an image or audio signal can be encoded and transmitted at an arbitrarily specified fixed bit rate.

With such a configuration, control is performed so that an image or audio signal can be encoded and transmitted at an arbitrarily specified fixed bit rate, and an optimal bit distribution for encoding is especially obtained at a low bit rate. Thus, the quality of the encoded signal can be improved.

When the bit signal stored in the buffer is equal to or smaller than a specified value, the control device according to the present invention sets the number of bits added at a fixed rate to the required number of reference bits as the number of added bits. If the number of bits is smaller than the value, the number of bits subtracted at a fixed rate is set as the number of subtraction bits, and the encoding process is performed based on the number of addition bits and the number of subtraction bits.

With this configuration, when the bit signal stored in the buffer is equal to or more than the specified value, the number of bits added at a fixed rate to the required number of reference bits is set as the number of added bits, By performing the encoding process based on the number of added bits and the number of subtracted bits with the number of bits subtracted at a fixed rate when the value is equal to or less than the specified value as the number of subtracted bits, optimal bit distribution particularly for encoding at low bit rates Thus, the quality of the encoded signal can be improved.

According to the control method of the present invention, a process of encoding an input signal in frame units of a time length determined on the transmission side, a step of storing encoded bit signals on the transmission side, Calculating the number of bits required to encode the input signal; and calculating the number of bits used by the encoding means with reference to the calculated required number of bits and a buffer level indicating the amount of bit signals stored in the buffer. Adjusting the buffer level to fall within a predetermined range, and adaptively controlling the predetermined range of the buffer level based on the transmission bit rate and the required number of bits. A receiving side for storing the coded bit signal transmitted from the transmitting side; and a decoding and outputting the coded bit signal stored on the receiving side. A step, in which so as to include a.

In this way, the transmitting side encodes the input signal in frame units of a predetermined time length, stores the encoded bit signal, and encodes the input signal in frame units. To control the buffer level to fall within a predetermined range by adjusting the number of bits used by the encoding means based on the calculated required number of bits and the buffer level. At the same time, the buffer level is adaptively controlled based on the transmission bit rate and the required number of bits based on the transmission bit rate and the required number of bits, so as to perform optimal bit allocation for encoding, particularly at a low bit rate, The coded bit signal sent from the receiving side is stored, and the coded bit signal stored on the receiving side is decoded and output. In this way, the optimal bit allocation for encoding at the transmitting side, especially at low bit rates, and by improving the quality of the encoded signal, the decoding can be performed without changing the conventional receiving side apparatus. The quality of the obtained signal can also be improved.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The main object of the present invention is to improve the signal quality over a long period of time by controlling coded bit signals in frame units when transmitting video and audio at a fixed bit rate. It is in.

(Embodiment 1) FIG. 1 is a block diagram showing a configuration of a control device according to Embodiment 1 of the present invention.

The coding bit rate control device 100 includes coding means 101 for coding an input signal in frame units of a time length determined by the transmitting side 109, and buffer means 102 for storing coded bit signals. And calculating means (necessary bit number calculating means) 103 for calculating the number of bits required to encode the input signal in frame units, and the calculated required number of bits and the amount of bit signals stored in the buffer. By adjusting the number of bits used by the encoding means on the basis of the buffer level, the buffer level is controlled so as to fall within a predetermined range, and the predetermined range of the buffer level is defined as the transmission bit rate and the required bit rate. And control means 104 for adaptively controlling based on the number of bits.

The receiving side 110 has a frame buffer 1 for receiving and buffering the output coded bit h from the transmitting side.
11, a frame buffer 111, a bit reservoir 112 for exchanging surplus bits n and an insufficient bit o, and decoding means 113 for decoding and outputting the buffer output of the frame buffer 111.

The buffer means 102 comprises the encoding means 101
And a bit reservoir 106 for transmitting and receiving surplus or deficient bit signals f and g in cooperation with the frame buffer 105. is there.

The control means 104 comprises: a bit reservoir level setting means 108 for setting the bit level of the bit reservoir 106 based on the transmission bit rate b, the output of the bit reservoir 106 and the output of the calculating means 103; A bit reservoir for transmitting a signal corresponding to the available bit number e to the encoding means 101 based on the reservoir level a, the output (reserved bit number) c of the bit reservoir 106 and the output (required bit number) d from the calculation means 103 And control means 107.

First, the operation of the bit reservoir control means 107 will be described in detail. Bit reservoir control means 1
Operation 07 must be controlled in consideration of the state of the receiving side of the transmission path. First, in order to perform transmission at a fixed bit rate and enable the receiving side 110 to perform decoding processing in real time, the receiving side 110 must have a frame buffer 111 of the same size as the transmitting side 109 and a transmitting side 109.
It is necessary that a bit reservoir 112 having the same size or more is provided. On the receiving side 110, coded bits are taken out from the frame buffer 111 and the bit reservoir 112 in coding processing frame units, and the decoding means 113 performs decoding processing.

On the receiving side 110, the relationship between the number of coded bits per frame and the frame buffer and bit reservoir is exactly opposite to that on the transmitting side 109,
If the number of bits used per processing frame is smaller than the frame buffer 111, the frame buffer 11
The unused encoded bits among the encoded bits received at 1 are stored in the bit reservoir 112.

If the number of bits used for each processing frame is larger than that of the frame buffer 111, the coded bits that are insufficient are taken out from the bit reservoir 112 and used. Here, if the number of encoded bits required for the bit reservoir 112 is larger than the number of encoded bits stored in the bit reservoir 112, the bit reservoir 112 cannot compensate for the insufficient encoded bits. As a result, the processing frame is decoded
Cannot be decrypted.

In order to prevent such a phenomenon from occurring, the transmitting side 109 monitors the state of the bit reservoir 112 of the receiving side 110, and controls the number of bits used in encoding based on the state. There must be. Actually, by monitoring the bit reservoir 106 of the transmitting side 109, the state of the bit reservoir 112 of the receiving side 110 can be known, so that the control of the number of bits used in encoding is realized by the processing of only the transmitting side 109. can do.

In a certain processing frame on the transmitting side 109, the number of coded bits used is stored in the frame buffer 1
05, the output coded bit h is replenished from the bit reservoir 106 of the transmitting side 109,
On the receiving side 110, the bit reservoir 10 of the transmitting side 109
The output encoded bits h that have been supplemented from 6 are stored in the bit reservoir 112 of the receiving side 110 as they are.

Conversely, on the transmitting side 109, if the number of used bits is larger than the size of the frame buffer 105 and the surplus bits f are stored in the bit reservoir 106 of the transmitting side 109, the receiving side 110 Encoded bits that are insufficient for the size of the buffer 111 are extracted from the bit reservoir 112 on the receiving side 110. The number of encoded bits is determined by the number of surplus bits f stored in the bit reservoir 106 on the transmitting side 109. be equivalent to.

From the above operating conditions, in the transmission using such a bit reservoir 106, when the number of coded bits exceeding the size of the frame buffer 105 is used in a certain processing frame, the previous processing is performed. It can be seen that encoding must be performed using a number of bits smaller than the frame buffer size in the processing frame, and the encoded bits must be stored in the bit reservoir 112 of the receiving side 110.

Further, in this case, the transmitting side 109 takes out the coded bits stored in the bit reservoir 106, so that the bit reservoir 1 of the transmitting side 109
06 must store the coded bits of the previous processing frame. It can also be seen that in order to satisfy this condition, the encoding process on the transmitting side 109 needs to be performed before the decoding process on the receiving side 110, and the transmission of encoded bits needs to be delayed.

Therefore, encoding processing for several frames is performed in advance within a range where the number of encoded bits does not exceed the size of the bit reservoir 106, and the encoded bits are stored in the bit reservoir 106 of the transmitting side 109. If the transmission of coded bits is to be started, the number of coded bits stored in the bit reservoir 112 of the receiving side 110 is determined by the code extracted from the bit reservoir 106 of the transmitting side 109 from the start of transmission of the coded bits. Equal to the number of coded bits (reserved bits).

After the start of transmission, if the reserved bit signal c of the transmitting side 109 is set to be within the range of 0 to the size of the bit reservoir 106, the number of encoded bits stored in the bit reservoir 112 of the receiving side 110 Also falls within the size of the bit reservoir 112 from 0, so that real-time transmission can be performed without causing a problem that the decoding means 113 of the receiving side 110 runs short of coded bits required for decoding processing.

From the above, the operation of the bit reservoir control means 107 is based on the required bit number d
Exceeds the transmission bit rate b of the transmission path,
The number of reserved bits c is reduced in a range where the reserved bits 106 take a positive value, and the number of coded bits used in the frame encoding process is increased. If the rate is smaller than the rate, the number of coded bits used in the frame coding process is reduced, and the number c of reserved bits is increased.

Since the number c of reserved bits cannot exceed the size of the bit reservoir 106, reserved bits exceeding the bit reservoir size are transmitted to the receiving side 110 assuming that they are used. This is called a fill bit. Since the fill bit does not include coding information, it is discarded by the receiving side 110.

In general, when the transmission bit rate b of the transmission path is low, the request for the number of bits required for encoding is often higher than the bit rate of the transmission path, and the state where the reserved bit number c is 0 continues. , The bit reservoir 106
Does not work effectively. Also, the quality of a coded audio signal deteriorates as the number of coded bits that can be actually allocated is smaller than the required number of bits.

Therefore, when the degree of the required number of bits exceeding the transmission bit rate b is small, the number of bits used in encoding is reduced with respect to the transmission bit rate b, and the number c of reserved bits is increased. Keep it. Further, when the degree of transmission bit rate b exceeds, the control is performed such that the number of reserved bits c is reduced to increase the number of bits used in encoding, and the quality when the required number of bits is increased is improved. Deterioration can also be reduced.

When the number c of reserved bits is increased and becomes equal to the bit reservoir size, no more bits can be stored, and all bits corresponding to the transmission bit rate b are allocated to encoding. By performing such control, the difference between the number of bits required for encoding and the number of bits actually used for encoding is reduced,
It is possible to prevent the quality from deteriorating due to the shortage of the number of encoded bits relative to the required number of bits.

Next, the operation of the coding bit rate control device according to the first embodiment of the present invention will be described with reference to FIG. The encoding means 101 analyzes the input speech signal and calculates an evaluation scale i which is a basis for calculating the number of bits required for encoding.

The required bit number calculating means 103 calculates the number of bits required for encoding based on the evaluation scale i, and notifies the bit reservoir control means 107 of the required bit number d. At the same time, the bit reservoir control means 107 sets the reserved bit number c indicating the state of the coded bits stored in the bit reservoir 106 and the reserved bit number c, the transmission bit rate b and the required bit number d in the bit reservoir level setting means 108. Receive the bit reservoir level a obtained based on.

The bit reservoir control means 107 calculates the number of bits available in the current frame from the received necessary number of bits and the number of reserved bits, and notifies the encoding means 101 of the number. The encoding means 101 performs an encoding process using the given available bit number e, and outputs an output encoded bit h from the frame buffer 105.

If the number of output coded bits h is larger than the size of the frame buffer 105, the coded bits (surplus bits) f of the portion exceeding the size of the frame buffer 105 are stored in the bit reservoir 106, and conversely, the number is small. In this case, the bit buffer 106
Bits (insufficient bits) g satisfying 5 are taken out and the frame buffer 105 is filled. Finally, the output coded bits h of the frame buffer size stored in the frame buffer 105 are output.

Subsequently, bit reservoir level setting means 1
08 and the role of the bit reservoir level a will be described. The bit reservoir level setting means 108 calculates a target value of the number c of reserved bits considered appropriate at the set transmission bit rate b. As a calculation method, for example, at a set transmission bit rate b,
When all bits are stored in the bit reservoir 106,
It is possible to use a method in which the number of bits stored within a predetermined time is used as a target value of the reserved bit number c.

Subsequently, the calculated target value of the reserved number of bits c is corrected by the required number of required bits d and the number of bits actually stored in the bit reservoir 106, and the corrected value is calculated as a bit value. It is assumed that the reservoir level is a. However, the maximum value of the bit reservoir level a cannot be larger than the original bit reservoir size.

The reason why the bit reservoir level a is corrected is that, for example, when the required number of bits is small, the encoding means 1
01, the number of bits used also decreases, and a missing bit g occurs. At this time, if the reserved bit number c is close to the bit reservoir level a, the reserved bit number obtained by adding the generated missing bit g here It is more likely that c will exceed bit reservoir level a.

Since the bit reservoir level a functions as a virtual bit reservoir size, bits in a portion exceeding the bit reservoir level a are regarded as bits having no coding information and are transmitted. ,
This will reduce the transmission efficiency. To prevent this,
When the reserved bit signal c is close to the bit reservoir level a and the required number of required bits is small, the bit reservoir level a is raised and adjusted so as to further increase the reserved bit number. When the required number of required bits increases again and the number of reserved bits decreases after the raising, the bit reservoir level a may be adjusted to be reduced.

As described above, the bit reservoir level a which is adaptively set is virtually regarded as the bit reservoir size, and the bit reservoir 106 is controlled, so that regardless of the transmission bit rate b used, By optimizing the number of bits to be allocated to the encoding process in frame units, it is possible to improve the quality of encoded and transmitted information.

(Embodiment 2) FIG. 2 is a block diagram showing a configuration of an encoding bit rate control device according to Embodiment 2 of the present invention. The difference from the bit rate device according to the first embodiment of the present invention is that a perceptual error evaluation model 201 for converting an evaluation scale i output from the encoding unit 101 into an evaluation scale m based on human perception is provided. That is the point. Other configurations are the same as those of the first embodiment, and therefore, the same components are denoted by the same reference characters, and detailed description thereof will not be repeated.

In general, in lossy compression coding of a signal, the quality of the coded signal differs depending on what kind of bit distribution is performed on the signal to be coded. As a representative index representing signal quality, an SNR (Sign) indicating a ratio of an error in encoding with respect to an original signal is used.
al-to-Noise Ratio), and in the perceptual error evaluation model 201, this SN is used as an evaluation scale.
Using R, it is possible to calculate the number of bits required to encode the input signal so as to maintain a predetermined SNR value.

Conversely, when the number of usable bits is determined, it is also possible to perform an encoding process that maximizes the SNR of the encoded signal. However, since the measure of error based on physical distance represented by SNR is different from the measure of error perceived by a real human, the quality of a signal encoded based on SNR at a certain bit rate is reduced. However, high quality is not guaranteed even as a subjective evaluation of humans.

In particular, in coding with a very high compression ratio, the difference between the evaluation scale of the error based on the physical distance and the scale of the error perceived by a real person becomes large, so that the error perceived by the human is more significant. It is necessary to use an evaluation scale close to the scale. The perceptual error evaluation model 201 includes, for example, audio encoding with a high compression rate, for example, MPEG-2.
There is a psychoacoustic model used in AAC. The psychological auditory model considers the difference in auditory characteristics depending on the frequency band, and the effect called masking, in which a small sound close in time and frequency cannot be heard for a large sound.
The evaluation scale is calculated after reducing information of a part that cannot be perceived by a human.

The perceptual error evaluation model 201 can improve the quality as a subjective evaluation by performing the bit allocation and the encoding process on the basis of the evaluation scale based on the human perceptual error. . However, when the encoding process is performed based on the evaluation scale m based on such a human perceptual error, the amount of information that can be reduced is small in a frame including many sensitively perceived parts when viewed in units of the encoding process frame. In a frame including many parts that are not perceived much (the number of bits required for encoding is large), there is much information that can be reduced (the number of bits required for encoding is small), and an error evaluation based on a physical distance is performed. The variation in the number of bits required for encoding is greater than in the case where a measure is used.

In the coding means 101 in which the required number of bits fluctuates greatly, the required number of bits becomes smaller than the transmission bit rate b when used in combination with a conventional coding bit rate control device, especially at a low bit rate. In order to fill the bit reservoir 106 by compensating for the bits used in the large frame, the ratio of frames to which only the number of bits equal to or less than the transmission bit rate b can be allocated increases. However, according to the combination with the encoding bit rate control device of the second embodiment, the bit reservoir level a
Is adaptively set, and the bit reservoir size is virtually controlled to control the bit reservoir 106, thereby optimizing the number of bits to be allocated to the encoding process in frame units regardless of the transmission bit rate used. As a result, the quality of the encoded and transmitted information can be further improved.

(Embodiment 3) FIG.3 is a block diagram showing a configuration of an encoding bit rate control device according to Embodiment 3 of the present invention. The difference from the bit rate device according to the second embodiment of the present invention is that the bit reservoir control means 10
7 in that a second bit reservoir control means 301 having a different operation is provided. The other configuration is not different from the configuration of the second embodiment. Therefore, the same configuration is denoted by the same reference numeral, and detailed description is omitted.

The bit reservoir control means 107 controls the number of reserved bits from 0 to the bit reservoir level a, considering the bit reservoir level set adaptively as a virtual bit reservoir size. On the contrary,
The second bit reservoir control means 301 optimizes the bit reservoir size when the required number of bits is lower than the transmission bit rate b, and optimizes the bit reservoir level a when the required number of bits is higher than the transmission bit rate b. The number of reserved bits is controlled between 0 and the bit reservoir size, assuming that the number of reserved bits is large.

Hereinafter, the bit reservoir control means 107 and the second bit reservoir control means 30 will be described with reference to FIGS. 3 and 4.
The difference of the operation 1 will be described in detail. FIG. 4 shows how the reserved bits stored in the bit reservoir change in the encoding bit rate control device according to the third embodiment. The vertical axis is the number of reserved bits, the horizontal axis is the elapsed time, the dotted line 401 on the vertical axis is the bit reservoir size that is the absolute maximum value of the reserved bits, the bit set by the bit reservoir level setting means 108 of the third embodiment. Represents the reservoir level 402.

The state of the bit reservoir 106 by the bit reservoir control means 107 is indicated by a broken line 403 and the bit reservoir 1 by the second bit reservoir control means 301.
The state of 06 is indicated by a solid broken line 404. In this figure, for simplification of description, it is assumed that the bit reservoir level is constant in the section. The section Ta is a section in which the required number of bits is lower than the transmission bit rate b.

In the control by the bit reservoir control means 107, the number of reserved bits is the bit reservoir level 402
, The reserved bit number saturates at the bit reservoir level 402. On the other hand, in the control by the second bit reservoir control means 301, the target value of the number of reserved bits is the bit reservoir size 401
So the reserve bit is bit reservoir size 401
It is stored until. The section Tb is a section where the degree of the required number of bits exceeding the transmission bit rate is relatively large. In this section, it is necessary to reduce the reserved bits to supplement the bits used for encoding.

At this time, the second bit reservoir control means 3
In the control by 01, since more reserved bits are stored than in the control by the bit reservoir control means 107, more bits can be supplemented. Alternatively, if the same number of bits as the control by the bit reservoir control means 107 is supplemented, the bits can be supplemented over a longer section. The section Tc is a section in which the required number of bits is very large with respect to the transmission bit rate. In this section, the reserved bits are greatly reduced when either control unit is used, but the control by the second bit reservoir control unit 301 which can store more reserved bits is more advantageous.

The section Td is a section where the degree of the required bit number exceeding the transmission bit rate b is relatively small. In this section, the number of allocated bits needs to be smaller than the transmission bit rate b in order to supplement the reserved bit signal c used. In this case, since the required number of bits exceeds the transmission bit rate b, the target appropriate number of reserved bits is the bit reservoir level 402 in either of the bit reservoir control means. Therefore, since the control by the bit reservoir control means 301 approaches the bit reservoir level 402 first, the decrease in the number of bits to be allocated can be reduced earlier.

It should be noted that the required number of bits is the transmission bit rate b
If the input signal is not lower than the absolute value, the operation of the bit reservoir control means 107 and the operation of the second bit reservoir control means 301 are the same. There is almost always a section with a very small amount of information. In the control by the second bit reservoir control means 301, the bits stored in this section can be more efficiently distributed, so that the quality of the coded and transmitted information can be further improved.

(Embodiment 4) FIG. 5 is a block diagram showing a configuration of an encoding and transmitting apparatus according to Embodiment 4 of the present invention. Can be used in combination with any of the coding bit rate control devices, and can be configured to transmit coded bits at a predetermined fixed transmission bit rate. I have.

In FIG. 5, bit reservoir control means 5
Numeral 03 corresponds to a part for calculating usable bits using the evaluation scale i, the number of reserved bits, and the transmission bit rate 102 in the first to third embodiments of the present invention.

For example, taking the third embodiment as an example, the perceptual error evaluation model 201, the required bit number calculating means 102,
A bit reservoir level setting means 108 and a second bit reservoir control means 301 are included.

The encoded signal transmitted by the transmission encoded bit 504 output from the transmitting device 501 is
This is a signal encoded by the optimal bit allocation according to the encoding bit rate control device in the fourth embodiment, and has higher quality than a signal encoded and transmitted by a conventional transmission device. Further, from the viewpoint of the receiving apparatus 502, the conditions for enabling the receiving apparatus 502 to execute the decoding process in real time using the bit reservoir 507 are as described above.
Does not cause overflow or underflow. Note that reference numeral 505 in the figure denotes a frame buffer, and reference numeral 506 denotes a decoding unit.

Since the encoding bit rate control apparatuses of the first to third embodiments all satisfy this condition, the receiving apparatus 502 includes the transmission encoded bits of the conventional transmitting apparatus and the receiving apparatus of the present embodiment. The handling of the transmission coded bits 504 by 501 may be the same. That is, by using the transmission device incorporating the coding bit rate control means of the fourth embodiment, it is possible to transmit a higher quality signal at the same bit rate without making any changes to the conventional reception device. Becomes

[0082]

As described above, according to the present invention, it is possible to optimize the bit allocation for the encoding process, and thus to improve the quality of the encoded and transmitted signal. A control device and a control method can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of a control device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a control device according to a second embodiment of the present invention.

FIG. 3 is a block diagram illustrating a configuration of a control device according to a third embodiment of the present invention.

FIG. 4 is a diagram showing a state of a bit reservoir in the control device in FIG. 3;

FIG. 5 is a block diagram showing a configuration of an encoded transmission apparatus according to Embodiment 4 of the present invention.

FIG. 6 is a block diagram showing a configuration of a conventional coded transmission device.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 encoding bit rate control device 101 encoding means 102 buffer means 103 calculating means (necessary bit number calculating means) 104 control means 105 frame buffer 106 bit reservoir 107 bit reservoir control means 108 bit reservoir level setting means 109 transmitting side 110 receiving side Reference Signs List 111 frame buffer 112 bit reservoir 113 decoding means 201 perceptual error evaluation model 301 second bit reservoir control means 501 transmitting device 502 receiving device 505 frame buffer 506 decoding means 507 bit reservoir

Claims (8)

[Claims] An encoding means for encoding an input signal in frame units having a predetermined time length, a buffer means for storing encoded bit signals, and an encoding means for encoding the input signal in frame units. Calculating means for calculating the required number of bits, and adjusting the number of bits used by the encoding means on the basis of the calculated required number of bits and a buffer level indicating the amount of bit signals stored in the buffer. Control means for controlling the level to fall within a predetermined range and adaptively controlling the predetermined range of the buffer level based on a transmission bit rate and the required number of bits. Characteristic control device. 2. The image processing apparatus according to claim 1, wherein the buffer unit includes a frame buffer that stores an encoded bit signal output from the encoding unit in frame units, and a bit reservoir that exchanges a surplus or shortage bit signal in cooperation with the frame buffer. The control device according to claim 1, comprising: 3. A bit reservoir level setting for setting a bit reservoir level indicating an amount of bit signals stored in the bit reservoir based on a transmission bit rate and outputs from the bit reservoir and the calculating unit. 2. The apparatus according to claim 1, further comprising: means for transmitting the number of usable bits to the encoding means based on the set bit reservoir level, the output of the bit reservoir, and the output from the calculation means. Control device. 4. An encoding means for encoding an input signal in a frame unit having a time length determined on a transmission side, a transmission buffer means for storing an encoded bit signal, and an input signal in a frame unit. Calculating means for calculating the number of bits required to encode the data, and adjusting the number of bits used by the coding means based on the calculated required number of bits and a buffer level indicating the amount of bit signals stored in the buffer. Control means for controlling the buffer level to fall within a predetermined range and adaptively controlling the predetermined range of the buffer level based on the transmission bit rate and the required number of bits. A receiving buffer means for storing a coded bit signal sent from the transmitting buffer means on a receiving side, and a code stored in the receiving buffer means. And decoding means for decoding the coded bit signal and outputting the decoded bit signal. 5. The encoding means performs irreversible encoding processing on a video or audio signal based on a human perceptual limit, and the calculating means performs encoding based on a human perceptual limit. The control device according to claim 4, wherein the control unit calculates an appropriate number of bits. 6. The control device according to claim 4, wherein said control means controls so that an image or an audio signal can be encoded and transmitted at a fixed bit rate arbitrarily designated. 7. When the bit signal stored in the buffer is equal to or less than a specified value, the number of bits added at a fixed rate to the required number of reference bits is defined as the number of added bits. The encoding process is performed based on the number of added bits and the number of subtracted bits, with the number of bits subtracted at a constant rate as the number of subtracted bits.
The control device according to claim 1. 8. A step of encoding an input signal in a frame unit having a predetermined time length on a transmission side, a step of storing an encoded bit signal in a transmission side, and a step of encoding the input signal in a frame unit. Calculating the number of bits required for encoding, and adjusting the number of bits used by the encoding means based on the calculated required number of bits and a buffer level indicating the amount of bit signals stored in the buffer. Controlling the buffer level to fall within a predetermined range, and adaptively controlling the predetermined range of the buffer level based on the transmission bit rate and the required number of bits. A receiving side for storing the coded bit signal transmitted from the transmitting side, and decoding and outputting the coded bit signal stored on the receiving side; And a control method.