JPS62272790A - Coding control circuit - Google Patents

Abstract

PURPOSE:To improve a follow-up of the action of a picture by controlling a threshold with the application of the action amount of the picture and executing a precise information generation amount control. CONSTITUTION:A generation information amount N<-1>15 which is the data amount of a preceding encoding frame is inputted in an encoding control circuit 16. A difference DELTAN26 between the preceding generation information amount N<-1>15 and a generation information amount target value N*24 is taken out by a differencial instrument 25 and increasing and decreasing amount of the threshold DELTATh28 is calculated from the difference DELTAN26 by a difference converter 27. When a preceding generation information amount N<1> is larger than a generation information amount target value N*, the threshold is increased and when said amount N<-1> is smaller than said command N* conversely, the threshold is controlled to be decreased. The calculated increasing and decreasing amount of the threshold DELTATh28 added by an adder 29 to a preceding threshold Th<-1>31 makes a new threshold Th deg.. Meanwhile in a quantizer 32, the difference DELTAN26 is quantized and a next generation information amount target value is increased or decreased in accordance with the quantization value and the threshold so as to control a next threshold properly.

Description

[Detailed description of the invention] 3. Detailed description of the invention [Industrial application field] The present invention relates to a single image encoding device.

[Conventional technology]

FIG. 7 is a block diagram showing the configuration of an encoding section of a conventional interframe block encoding device. (2) is an A/D conversion block converter that A/D converts the input analog signal and blocks the y/c separated signal into encoded blocks;
(5) is a subtractor between the blocked input signal and the previously encoded reproduced image signal, (8) is a block encoder, (IQ
7. is a transmission buffer memory that temporarily stores block encoded data and variable length encoded data of encoded control information and sends it to the transmission path aυ at a constant speed. i+3 is a frame drop controller that performs frame drop control based on the amount of information stored in the transmission buffer; @ is an encoding control circuit that controls a threshold value used as a condition for determining conditional pixel replenishment based on the amount of information stored in the transmission buffer; Nishiki is a comparator that uses the threshold value to determine whether or not to perform encoding on a block-by-block basis.

a! I is a block decoder, L2υ is an adder for the previously encoded reproduction signal and the decoded difference signal, and trowel is a frame memory.

Next, the operation will be explained. Input image analog signal (1
) in the A/D conversion block divider (2).

The input block signal (3
). The difference between this input block signal (31) and the previously encoded reproduced signal (4) which has been divided into blocks is taken by a subtractor (5) to obtain a differential block signal (6).Next, in a comparator a & , determines whether or not to perform encoding on the differential block signal peak using a threshold value Th071, and outputs a determination signal ν(7) representing the determination result.An example of the determination is shown below.

however ! = (ε1.ε2.・飄) When d>Th, C1 (encoding is performed) cl (In the case of Th, C0 (no encoding)) In block encoding (8), the above-mentioned determination signal According to ν(7), in the case of sea 1, block encoding is performed, and in the case of sea 0 and frame drop control, no encoding is performed and the differential block signal is set to 0. Next, the transmission buffer memo 1 On month 11, the block encoded signal (9) which is the output of the block encoder (8)
) and the judgment signal ν encoded with variable length are stored.

The data is sent to the transmission path 011 at a constant speed, and the amount of information o2 stored in the buffer memory is output. If the stored information amount exceeds a certain value, the frame drop controller (13) outputs a frame drop signal 041. Otherwise, the encoding control circuit (to) performs threshold control. block decoder 0
9 decodes the block encoded signal 19);
Decoded difference signal? ■Output. However, the judgment signal ν
What if is 0? -0. The previously encoded reproduced signal (4
) and the decoded difference signal (2) is stored in the frame memory (c).

Next, referring to FIG. 8, the control of the threshold value +171 and the frame drop control in the encoding control circuit will be explained. When the amount of accumulated information Q2 in the transmission buffer memo IJQl) is greater than a certain value TB, frame drop control is performed and the input image frame at that time is not encoded. The amount of accumulated information is the threshold TB
If the amount of accumulated information is large, set the threshold value +
By setting 141 to a high value, the generated information amount of the next encoded frame is reduced, and if it is small, the generated information amount is increased by setting the threshold value to a low value.
Aim to smooth out the amount of accumulated information.

[Problem that the invention seeks to solve]

Since the encoding control circuit of the conventional video encoding device is configured as described above, the amount of image movement cannot be detected because the past encoding conditions are unknown, and therefore the amount of image movement cannot be detected depending on the amount of image movement and the threshold value. There are problems in that it is difficult to control the fluctuating amount of generated information with high precision, and the ability to follow the movement of multiple images is poor.

This invention was made to solve the above-mentioned problems, and it performs threshold control that incorporates the amount of movement of two images, and controls the amount of information generated with high precision, thereby improving the ability to follow the movement of images. The purpose of this study is to obtain an encoding control circuit that improves the performance.

[Means for solving problems]

The encoding control circuit according to the present invention calculates the difference between the generated information amount of the previous encoded frame and the generated information amount target value using a subtractor, and sets a plurality of threshold values according to the amount of motion linearly approximated from the kernel difference value. The threshold increase/decrease value is determined using a difference converter with a characteristic of the amount of generated information.

The value obtained by adding the threshold increase/decrease value and the previous threshold value using an adder is set as a new threshold value, and the new threshold value and the quantized value of the difference value are fed back to determine the next generated information amount target value. The target value is adaptively determined by a target value calculator.

[Effect]

The encoding control circuit according to the present invention realizes the target value of the amount of generated information using a difference converter having a linear approximation threshold value versus the amount of generated information characteristic from the difference value between the amount of generated information and the target value of the amount of generated information. A threshold increase/decrease value is calculated, and an adder adds the threshold increase/decrease value and the previous threshold value to a new threshold value. -! The next threshold value is adaptively controlled by increasing or decreasing the next generation information amount target value according to the quantized value of the difference value fed back and the new threshold value.

[Embodiments of the invention]

An embodiment of the present invention will be described below.

In Fig. 2, f149 is the encoded data amount (generated information amount) of the previous encoded frame, and OG is the encoding that performs threshold control based on the characteristics of multiple thresholds and generated information amount depending on the amount of motion. It is a control circuit, (1] ~ f141° (
1η~(c) is the same as the conventional one. FIG. 1 shows the internal configuration of the encoding control circuit. (c) is a difference device that takes the difference (a) between the generated information amount and the generated information amount target value, @ is a difference converter that calculates the threshold increase/decrease value 翰 according to the difference value (a) 02 is an adder that adds the previous threshold value c111 and the threshold increase/decrease value (c) and outputs a new threshold value, 02 is a delay device, and 02 is a quantizer 2 that quantizes the difference value.
(to) is a target value calculator that adaptively determines the generated information amount target value.

The operation will be explained below. In FIG. 2, the generated information amount N'+19, which is the data amount of the previous encoded frame, is input to the encoding control circuit Oe. Other than that, it is the same as the conventional one. In FIG.

The encoding controller will be explained. The amount of generated information N-1
(Difference between IS and generated information amount target value N + c241 △
N(a) is taken by a subtractor (c), and Th(2) is calculated from the nuclear difference value ΔN(a) to a threshold increase/decrease value by a difference converter. An example of determining the threshold increase/decrease value ΔTh from the difference value ΔN will be described with reference to FIGS. 3 to 5. FIG. 3 shows the threshold value versus generated information amount characteristics according to a plurality of motion amounts. FIG. 4 shows the linear approximation threshold versus generated information amount characteristic. The linear approximation threshold versus generated information amount characteristic is expressed by the following linear equation.

N = -a -Th+ b(n) However, a>0, and although it varies depending on the value of the previous threshold Th', it is assumed to be constant for each certain range. Also b (n)
is a function of the amount of motion n of the image, and has the property that as the amount of motion n increases, b (n) also increases.

In FIG. 5, the previous threshold value Th' (31) and the previous generated information amount N'Q! 9, the threshold value vs. generated information amount characteristic for the previous motion amount n-1 can be expressed by the following equation.

b(n''-1)=N''+a-Th',゛,N=
−a・Th+(N′+a−Th′) The second motion amount n does not change during the next encoding, that is, b(n
)-b(rl') 1 generation information amount target value N
If the new threshold value for realizing ζa is Th', the threshold increase/decrease value ΔTh@ can be determined by the following equation than the above equation.

△N △Th=- However, ΔN=N''-Nl △Th = Th''-T117' In other words, if the previous generated information amount N-1 is larger than the generated information amount target value N'', the threshold value is increased, On the other hand, if the threshold value is small, control is performed to reduce the threshold value.

In FIG. 1, the threshold increase/decrease value △Th (to) determined as described above is added to the previous threshold value Th 'all by an adder (
(to) the new threshold value Th°.

On the other hand, the quantizer 03 quantizes the difference value ΔN@ and outputs it at the time of secondary encoding. Next, it will be explained with reference to FIG. In the target value calculator (to), if the quantization difference value (to) is greater than or equal to a certain value, it is assumed that one image is moving significantly.If the amount of motion increases further at the next encoding, b(n )) In preparation for b(n'''), the difference value ΔN12[9 is increased by setting the generated information amount target value N''L2J lower than the normal generated information amount target value (b), and as a result, the difference value of the difference converter - Threshold increase/decrease value △Th(2) which is the output
By increasing , a threshold value Th' that is higher than the new threshold value 5 when N"=i is output. Here, 2 Normally generated information amount target value 1 (b) is a value determined by the rate of the transmission path. Incidentally, the increase/decrease value of the first generation information amount target value with respect to the normal generation information amount target value N (b) is determined by the fed-back new threshold Th' and the fed-back quantization difference value (to).

In addition, if a low threshold value has continued for a certain number of times in the past (n is a positive integer) or more, it is assumed that the movement of the two images is almost stationary, and the target value of the amount of generated information N'' is set to the normal target value N.
By maintaining a relatively high threshold state by continuing to lower the threshold value until the difference value ΔN exceeds a certain value, we can prepare for the start of image movement and reduce the amount of information generated when two images start moving. This suppresses the increase and prevents a drop in followability when the movement starts.

In the above embodiment, the threshold value control is performed according to the amount of information generated, but the threshold value control may be performed according to the number of encoded pixels or the encoded pixel rate. Furthermore, A/D
Of course, it is possible to provide a frame memory after the converter and perform encoding and transmission in units of image frames with frame drop control.

〔Effect of the invention〕

As described above, according to the present invention, the threshold increase/decrease value is calculated based on the linear approximation threshold value versus the generated information amount characteristic from the difference value between the generated information amount and the generated information amount target value. The addition of the previous threshold value and the previous threshold value is output as a new threshold value, and the difference value and the new threshold value are fed back to determine the target value for the amount of information to be generated next time, so the tracking performance for the movement of one image is improved. This also has the effect of simplifying the device configuration.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an encoding control circuit according to an embodiment of the invention, FIG. 2 is a block diagram showing the configuration of an image encoding device according to an embodiment of the invention, and FIG. An explanatory diagram showing the threshold value vs. generated information amount characteristics by
FIG. 4 is an explanatory diagram showing the linear approximation threshold value versus generated information amount characteristics according to the present invention, FIG. 5 is an explanatory diagram of threshold increase/decrease value calculation according to the present invention, and FIG. 6 is an explanatory diagram showing the generated information amount target according to the present invention. FIG. 7 is a block diagram showing the configuration of a conventional moving image encoding device, and FIG. 8 is an explanatory diagram of conventional threshold control. Symbol ae is an encoding control circuit, (to) is a difference device, @ is a difference converter, (c) is an adder, (to) is a delay device, and ea is a quantizer. (G) is a target value calculator. Note that in the two figures, the same reference numerals indicate the same or equivalent parts.

Claims (6)

[Claims] (1) The encoded data of the video signal, which increases or decreases depending on the movement of the subject, is stored in the transmission buffer memory for a predetermined period of time, and the encoding conditions for the next period are determined based on the amount of encoded data (amount of generated information) for the predetermined period. In a video encoding circuit that smoothes the amount of generated information through feedback control of the threshold value and transmits encoded data at a constant speed, the amount of encoded data increases as the conditional pixel replenishment determination threshold increases. For the video encoding unit that encodes to decrease the amount of information generated, a subtractor that calculates the difference between the amount of information generated previously and a target value of the amount of information generated previously, and a plurality of calculations based on the difference value according to the amount of motion of the image. a difference converter that changes to a threshold increase/decrease value based on the threshold value used in the previous encoding using a threshold value versus generated information amount characteristic, and adds the threshold increase/decrease value and the previous threshold value. and a target value calculator that calculates the next generation information amount target value based on the quantized value of the difference value and the new threshold value. Encoding control circuit. (2) The difference converter has a characteristic that linearly approximates a plurality of threshold values vs. generated information amount characteristics according to the amount of movement of the image using a linear equation, and applies a coefficient to the difference value by applying the increase/decrease value of the threshold value to the difference value. Claim 1 characterized in that the calculation is performed by multiplying
Encoding control circuit as described in section. (3) The difference converter is characterized in that a coefficient of a linear equation representing a linear approximation threshold value versus generated information amount characteristic is made variable in accordance with the previous threshold value. encoding control circuit. (4) The target value calculator performs higher threshold control by setting the generated information amount target value lower than the normal generated information amount target value determined by the transmission rate when the difference value exceeds a certain value. The encoding control circuit according to claim 1, characterized in that the encoding control circuit performs the following: (5) The target value calculator determines an increase/decrease value of the next generated information amount target value with respect to the normally generated information amount target value based on the quantized difference value and the new threshold value. An encoding control circuit according to claims 1 and 4. (6) The target value calculator lowers the generated information amount target value until the difference value exceeds a certain value when a low threshold value has been set a certain number of times in a row in the past, thereby increasing the threshold value. 2. The encoding control circuit according to claim 1, wherein the encoding control circuit maintains a control state.