JPH01245686A - Picture receiver - Google Patents

Abstract

PURPOSE:To raise the detecting capacity of a transmitting mode and to improve the picture quality of a reproducing picture by setting a threshold value corresponding to a transmitting mode change to the picture element data of respective picture element blocks received by a mode memory means and a threshold value setting means. CONSTITUTION:A block distortion operating means 90, which operates a block distortion to a previous picture on respective picture element blocks, a mode memory means 91, which stores the transmitting mode of the respective picture element blocks of the previous picture, a threshold value setting means 92, which refers to the block distortion, channel quality and the transmitting mode of the previous picture stored in the mode memory means 91 and sets a threshold value to respective transmitting mode changes between the previous picture and a present picture, and an evaluating means 84, which evaluates the transmitting mode of the respective picture element blocks of the present picture according to the threshold value, are provided. Thus, the transmitting mode change, of which picture quality improving effect is the highest, is selected maximumly, the change which has possibility to make the picture quality worse can be made not to be selected much, and the improvement of the picture quality and a data compressing ratio by a third mode to use the data of the previous picture can be exhibited maximumly.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an image receiving device, and more specifically, the present invention relates to an image receiving device, and more specifically, it divides one screen of image information into a plurality of blocks, and processes each block according to a predetermined rule. The present invention relates to an image receiving device in a system that assigns one of a plurality of transmission modes to transmit image information.

[Conventional technology]

When transmitting information such as image information, the problem is how to reduce the amount of transmitted information to faithfully reproduce the original image, and a wide variety of transmission methods have been proposed for this purpose. For example, there is an adaptive variable density sampling method that appropriately changes the sampling density, that is, the density of transmitted information. This method focuses on the fact that there are high and low spatial frequency regions in the boat fishing image, and increases the sampling density in the high spatial frequency regions and lower the sampling density in the low spatial frequency regions. This reduces the amount of information to be transmitted.

In this method, in order to keep the transmission amount constant across the screen, the screen is divided into areas with high and low sampling density based on spatial frequency and other criteria. Low areas may be assigned to low-density sampling areas, and while this may be bad for moving images, the deterioration in image quality becomes noticeable in still image areas due to the high resolution of the human eye.

On the other hand, a transmission system has been proposed that takes advantage of the fact that continuous images, such as regular television signals, have a correlation even in the time axis direction, and aims to improve their image quality. In other words, correlation in the time axis direction is taken into account for image information that has a two-dimensional spread, and since there is no need to update pixel data on the receiving side at least for static areas of the screen, transmission of static areas is possible. The aim is to omit this, increase the transmission density in other areas accordingly, and thereby improve image quality. This method that utilizes the correlation in the time axis direction is hereinafter referred to as three-dimensional TAT (Time Axis Tra
nsformation) method.

In this three-dimensional TAT method, one screen is divided into multiple pixel blocks, and each pixel block is Under predetermined rules, there are two modes: a mode in which all constituent pixel data is transmitted (hereinafter referred to as C mode), and a mode in which only data of a basic part of the constituent pixels (basic pixels) is transmitted (hereinafter referred to as C mode). C mode), and a mode that reuses the data of the corresponding block of the screen that has already been transmitted (hereinafter referred to as C mode).
The receiving side performs signal processing corresponding to each mode to restore the image. A signal indicating transmission mode assignment is separately transmitted as mode information. In addition, in the p mode, only basic pixel data is transmitted as in the C mode, and if the received data is the same as the corresponding pixel data on the previous screen, it can be determined that it is the p mode instead of the C mode. This conventional example will be explained below.

FIG. 4 shows a schematic block diagram of a transmitter in the three-dimensional TAT system. Note that an analog transmission line is used as the transmission line. The analog image signal to be transmitted is sent to the A/D converter 10.
is converted into a digital signal by A/D converter 10
All pixel data from is applied to a thinning circuit 12, where the data of pixels other than the basic pixels are thinned out. That is, the thinning circuit 12 outputs C-mode pixel data. The interpolation circuit 14 interpolates the pixel data thinned out by the thinning circuit 12 to reproduce data of all pixels of the block. The block distortion calculation circuit 16 includes an A/D converter 10
The pixel data outputted from the interpolation circuit 14 is compared with the interpolated data outputted from the interpolation circuit 14, and block distortion DC of C mode transmission is calculated for each pixel block.

The output of A/D converter 10 is also connected to frame memory 18.
is also applied. The frame memory 18 functions as a delay means for one screen. In other words, frame memory 18
, all the data of the previous screen is stored, and the block distortion calculation circuit 20 compares the pixel data of the current screen and the pixel data of the previous screen from the frame memory 18 for each corresponding block, and calculates the block distortion. Calculate . This block distortion D2 indicates the degree of similarity in the time axis direction.

The comparison circuit 22 compares DC and D2 and selects selection mode data DC and 1. which indicates which block distortion is smaller. At the same time, the smaller block distortion is output as the block distortion D1 for determining transmission mode assignment. In other words, the comparison circuit 22 determines for each pixel block whether C mode transmission or p mode transmission has smaller block distortion, and if DC>D, it does not select C mode and Dc In the case of Dp, p mode is not selected. However, if p-mode transmission is performed on the current screen for pixel blocks that have been transmitted in C-mode on the previous screen, no improvement in image quality can be expected, so p-mode transmission is performed only on pixel blocks that are transmitted in C-mode on the previous screen. Do it like this.

For blocks for p-mode transmission, only basic pixel data is transmitted. Then, on the receiving side, it is compared with the basic pixel data of the same block in the previous screen, and if they are the same, it is determined to be P mode and the data from the previous screen is used, and if they are different, it is determined to be C mode and the image is restored by interpolation processing. do.

In addition, for pixel blocks that are being transmitted in C mode on the previous screen, if they are transmitted in p mode on the current screen, no improvement in image quality can be expected. This will be p-mode transmission.

After the block distortion of all the pixel blocks constituting one screen is calculated in this way, each pixel block is set to a transmission mode so that the transmission amount of one screen, that is, the compression ratio is constant (for example, 1/2). The pixel data of each pixel block is sequentially transmitted according to the transmission mode. That is, the mode determination circuit 24 performs assignment according to the output of the comparison circuit 22 and outputs a determination signal. The switch 26 is controlled according to the determination signal, and for each pixel block, in the case of C mode, the A/D converter 1 is connected via the buffer 28.
All pixel data starting from 0 is supplied to the D/A converter 32 through the buffer 30 in the case of c and p modes, and basic pixel data from the thinning circuit 12 is supplied. The D/A converter 32 converts the digital pixel data into an analog signal and sends it to the transmission path.

In the case of C mode transmission, there is no need to rewrite the data in the corresponding pixel block of the frame memory 18, so according to the determination signal output from the mode determination circuit 24,
The frame memory 18 is write-protected.

The determination signal output from the mode determination circuit 24 is also C
It is sent out to the transmission path via the buffer 34 as mode information indicating whether mode transmission is being performed or not.

FIG. 5 shows mode distribution for block distortions DC and Dp. The larger the movement of the block, the higher it is located on the Dp axis.
Furthermore, blocks with higher definition, that is, blocks with higher two-dimensional frequency, are located to the right on the DC axis. Since C or p mode is designated by the magnitude relationship between DC and Dp, the area above the straight line De=D in Figure 5 is basically C mode, and the area below the line is p mode. . Also, the value of the pixel block glue located at Xc is the value on the DC axis when a perpendicular line is drawn to the DC axis,
The value of the pixel pro-tsukori located at Become. If the D1 axis is provided in FIG. 5 and the threshold value T1 for selecting the C mode is considered, then on the DC1Dp axis, the threshold value T2 becomes. In other words, pixel blocks with rapid movement and high definition are C
mode.

Fix the data compression rate for the entire screen to 1/2, for example, and
, If the pixel data to be transmitted in the p mode is 174, which is the constituent pixel data of a pixel block, then the total number of pixel blocks to be transmitted in the C mode is 173.

That is, as shown in FIG. 6, a total of 173 blocks are transmitted in C mode, and the remaining blocks are transmitted in P or C mode depending on the block distortion Dc9Dp. FIG. 6 shows changes in the distribution ratio of transmission modes depending on the time correlation of images. 6th
The straight line portion on the right side of the figure corresponds to the case where there is no correlation between the front and rear screens, and corresponds to performing TAT compression in the horizontal and vertical directions of the screen. The left side of FIG. 6 corresponds to the case where a complete still image is transmitted, and the resolution is the same as when all pixel blocks are transmitted in C mode. The mode distribution ratio for the screen consisting of e and c is on the broken line indicated by A in the figure.

p is indicated by the length of the line segment divided by each region. The position of dotted line A depends on the degree of time correlation of image information.

FIG. 7 shows a schematic block diagram of a receiving device corresponding to the transmitting device shown in FIG. 4. The analog pixel signal transmitted from the transmitter shown in FIG. 4 is converted into a digital signal by the A/D converter 70. The switch 72 is controlled by the received mode information, and when each pixel block is being transmitted in C mode, the output of the A/D converter 70 is
In other cases, the output of the interpolation circuit 74 is selected.

The interpolation circuit 74, like the interpolation circuit 14 in FIG. 4, is a circuit that interpolates thinned-out pixel data during C mode transmission from basic pixel data. Therefore, switch 72 always outputs the full pixel data for each pixel block, which is written to full pixel frame memory 73. switch 76
In the case of C mode transmission, the output of the thinning circuit 78 which extracts basic pixel data from the pixel block of C mode transmission is selected, and in other cases, the output of the A/D converter 70 is selected. select.

Therefore, the output of switch 76 is always elementary pixel data, which is written to elementary pixel frame memory 80.

The block distortion calculation circuit 82 calculates the difference between the basic pixel data from the switch 76 and the basic pixel data of the previous screen from the frame memory 80, and outputs the sum of the differences for each pixel block. Output of circuit 82 (hereinafter referred to as block distortion D
b) is compared with a threshold value TH in a comparator circuit 84. D
If b is smaller than TH, it is determined that the pixel block has been transmitted not in C mode but in P mode. Based on the reception mode information and the output of the comparison circuit 84, the arithmetic circuit 86 applies a p identification signal to the frame memory 73.80, indicating whether the mode is p mode or not. This allows the frame memory 7'3.8 to be used for pixel blocks transmitted in p-mode.
Rewriting of 0 is prohibited, and the data from one screen ago will remain as is. In this way, the frame memory 73
The data is updated and read out to the D/'A converter 88. A high resolution analog video signal is obtained from the D/A converter 88.

[Problem to be solved by the invention]

However, in such a transmission system, if the receiving device does not accurately detect the mode of pixel blocks for p-mode transmission, the image quality improvement effect cannot be obtained. That is, consider the change in transmission mode between the current screen (current screen) and the screen immediately before it (previous screen). A pixel block in C mode on the previous screen is in P mode on the current screen (e = p
), an improvement effect can be expected in some cases, but no improvement effect can be expected in C-+p, and therefore, in the conventional example described above, this format is prohibited. On the other hand, if a block that transmits in C mode on the current screen is transmitted in e or p mode on the previous screen (
That is, 6−+ (or p −+ C) may occur. In the conventional example described above, a signal indicating whether the transmission mode is C mode or non-e mode is transmitted as the transmission mode information.
mode transmission may be mistakenly determined to be p-mode transmission; in such a case, 6-+ p or p-p
However, the image quality of the video portion deteriorates.

Furthermore, in the case of a mode change of p-'p or 6-+p, which corresponds to a still image part, when the p mode of the current screen is mistakenly determined to be the C mode, p-'C or e-+1C
As a result, the image quality improvement effect as a still image cannot be obtained.

The reason for such misjudgments is that, in order to be compatible with the two-dimensional TAT method that does not utilize correlation in the time axis direction, in p mode, basic pixel data is transmitted in the same way as in C mode, and on the receiving side, Distinguish between C mode and p mode based on the magnitude of block distortion -13--1-----
-- This is because the mode information of C or p mode has low credibility. In particular, when the same pixel block is continuously determined to be in p mode on the receiving side, there is a problem that if there is even one error in pixel data on the way, the error will be propagated as is. Conversely, if the same pixel block is continuously transmitted in p-mode on the transmitting side, once it is no longer in p-mode due to impulsive noise,
C mode playback with noise must be continued after the noise is superimposed.

Therefore, it is an object of the present invention to provide an image receiving device that does not cause such problems.

[Means to solve the problem]

The image receiving device according to the present invention has a first mode in which image information of one screen is divided into a plurality of pixel blocks and all of its constituent pixel data is transmitted to each pixel block, data of basic pixels among the constituent pixels. A second mode that transmits the data of the previous screen and a third mode that reuses the data of the previous screen are assigned, and the transmission mode of the pixel signal and each pixel block in each transmission mode is the first mode or other mode. An image receiving device in a transmission system that continuously transmits a plurality of temporally consecutive image groups by transmitting transmission mode information indicating whether the transmission mode is Block distortion calculation means for calculating block distortion for the previous screen, mode memory means for storing the transmission mode of each pixel block of the previous screen, block distortion by the block distortion calculation means, transmission path quality, and the mode memory means. a threshold setting means for setting a threshold for each transmission mode change between the previous screen and the current screen by referring to the transmission mode of the previous screen stored in the current screen; and evaluation means for evaluating the transmission mode of each pixel block.

[Effect]

The mode memory means and the threshold value setting means can set a threshold value according to a change in transmission mode for the received pixel data of each pixel block.

In other words, it is possible to maximize the selection of transmission mode changes that have the highest image quality improvement effect, and to select less changes that may deteriorate image quality.

Therefore, the improvements in image quality and data compression rate achieved by the third mode can be maximized.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the drawings.

The transmission line for the video tape of a VTR and the electromagnetic conversion system for recording and reproducing it can be regarded as a transmission line for data transmission.
The case of TR will be explained as an example. FIG. 8 shows a block diagram of the video tape and electromagnetic conversion system. 40
is the input terminal for the video signal to be recorded on the video tape, and 41 is the PCM to be recorded on the video tape.
Signal input terminals 42 and 43 are recording modulation circuits. Standard recording mode (SP) and long-term recording mode (LP) are specified for 8mm VTR, and modulation D14.2
．． The output of 43 is recorded on video tape in SP mode or LP mode under the control of system controller 44. FIG. 2 shows the recording track pattern.

As is well known, with an 8mm VTR, part of the track is
The M-word recording area is used, and the remaining area is used as a video signal recording area, and the video signal and PCM signal are time-divisionally recorded.

Each recording signal of the PCM area and the video area is supplied to a corresponding demodulation circuit 45, 46 under the control of the system controller 44. The outputs of the demodulation circuits 45 and 46 are respectively output from known video processing circuits and PCM processing circuits (
(not shown).

Switching between SP mode and LP mode may be performed using a control signal from another device.

In this embodiment, image information, ie, pixel signals, is recorded in the video area, and transmission mode information (e or non-e) is recorded in the PCM area. The SP mode of an 8mm VTR has better recording and playback characteristics, that is, transmission characteristics, than the LP mode.
The sp mode can be considered a high quality transmission path, and the LP mode can be considered a low quality transmission path.

Therefore, the quality of the transmission path can be differentiated depending on the recording mode, and specifically, it can be obtained from the system control circuit or servo control circuit of the VTR.

Alternatively, transmission mode information for information compression by TIIT (
e or non-e) is recorded as "0thers" in the PCh area in the case of an 8mm VTR, and error detection and correction processing is performed as a digital signal during playback, so the error detection used for internal processing of the CRC at that time. By flag,
Transmission path errors may be determined and the quality of the transmission path may be determined based on the magnitude of this error rate.

FIG. 1 shows a configuration block diagram of an embodiment of the present invention.

However, in this embodiment, for simplicity, a 1-bit LP/SP discrimination signal is obtained from the system controller of the VTR to evaluate the quality of the transmission path. Further, the same components as in FIG. 7 are given the same reference numerals. Below, the differences from FIG. 7 will be mainly explained. Needless to say, the input pixel signal is played back from the video area of the video tape,
The input mode information signal is reproduced from the PCM area. 90 is a block distortion calculation circuit that calculates and outputs =18-block distortion similarly to the block distortion calculation circuit 82; 91 is a block distortion calculation circuit that calculates and outputs the input transmission mode information (e or non-e) and the output of the calculation circuit 86; The p-mode or non-p-mode identification signal determines which transmission mode (e, c, or p) for each block (note that this is a mode distinction for playback processing and is not necessarily the transmission mode at the time of transmission). ); 92 is a comparison circuit 84;
This is a threshold value setting circuit that generates a threshold value TH for.

The threshold value setting circuit 92 sets a threshold value (0 to 255 in the case of 8 bits) when the pixel block of interest shifts from each transmission mode of e, c, and p to p mode in the current screen from the previous screen.
Set. In the relationship between the previous screen and the current screen, the threshold value when the transmission mode changes to C-1p is T 肚, the threshold value when it changes to e-hp is THG, and the threshold value when it changes to p-=p is THp. shall be.

FIG. 3 shows an operation flowchart of the threshold value setting circuit 92. The operation of the threshold value setting circuit 92 will be explained with reference to FIG. In Figure 3, i (an integer from O to N) refers to the (i-1)th pixel among the (N+1) pixels that make up one screen, and mode (i) indicates the transmission mode of pixel i. e, c,
Indicates which p. Increment i sequentially from 0 to N, while m.

With reference to de(i), the threshold Tl for each transmission mode is
(Determine , TH, , THp. j is i from 0 to N
This is a variable to indicate how many times the p mode has continued until then. Dmax is the block distortion, and the maximum value of (
255) for 8 bits, no is 2Dffl
About IIX/3 (approximately 170 for 8 bits), nl is Dl,
l1IX/10 (approximately 25 for 8 bits), n2 beam,
,,, about /100 (2 to 3 for 8 bits), ns is about DffiIIIX/200. n3 indicates the step of adjusting TH.

□ Also, m is a constant with a value approximately equal to the frame frequency.

First, each variable is initialized (S1) and variable i is reset (S2). Refer to the mode memory 91 (S3), if the transmission mode of pixel i is C or e mode, j is set to O.
NishikuS4,5), assign 0 to THc, n to TH,
(S7, 8). In addition, if the transmission mode of pixel i is p mode, j is incremented (S6),
Block distortion D-19= from the block distortion calculation circuit 90, and n. Compared with S9), if Db>no, it can be determined that the movement is large, so in p mode
, thus resetting j (SIO). Next, check the quality of the transmission path (in the illustrated example, whether it is SP recording or LP recording) (S
ll). Change the variable THp according to the value of j (31
2-17). Specifically, in the case of LP, when j-〇, substitute n for TH, 513), m > j >
When it is 0, n3 is subtracted from THp (S14). Also,
In the case of SP, when j-0, n2 is assigned to THp (516), and when m>j>Q, n3 is added to TH (S17).

After steps S7, 8, 13, 14, 16.17, i
is incremented (318), and as long as i is less than or equal to N, steps S3 and subsequent steps are repeated (519). When i reaches N, processing for one screen should be completed and processing for the next screen should be started, so the system is asked whether to continue playing or stop (32, 0), and if it is to continue playing, step S2 Go to, and if it is a stop, exit (21)
.

In the case of a change in transmission mode such as C-+p, there is usually no effect of improving image quality, so T=0 is set and the change is essentially prohibited. On the other hand, in the case of a transmission mode change of 6-+p, the image quality improvement effect is high and even a bad transmission path is unlikely to lead to error propagation, so the threshold value T is set so that it is selected more often.
H, is set to be the maximum among the three types of threshold values.

In the above explanation, the difference in VTR recording mode was used as an example of comparing the quality of the transmission path, but in the case of radio wave transmission, the quality is high, whereas in the case of VTR, the quality is evaluated as low quality. Good too.

〔Effect of the invention〕

As can be easily understood from the above description, according to the present invention, even in a transmission path with large time axis fluctuations such as a VTR, the detection ability of the transmission mode for still image portions can be improved. The quality of reproduced images can be greatly improved.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of the configuration of an embodiment of the present invention, and Fig. 2 shows the recording trunk of a video tape in an 8 mm VTR.
FIG. 3 is a flowchart showing the operation of the threshold value setting circuit 92 in FIG. An explanatory diagram of mode assignment in an image information transmission system using a three-dimensional TAT method, FIG. 7 is a block diagram of a configuration of a receiving device corresponding to FIG. 4, and FIG. 8 is a schematic configuration diagram of a portion serving as a transmission path of a VTR. . 74-Interpolation circuit 78-Thinning circuit 73.80-Frame memory 84-Comparison circuit 86-・- Arithmetic circuit
90-Block distortion calculation circuit 91-Mode memory 9
2-Threshold setting circuit

Claims (1)

[Claims] The first mode divides the image information of one screen into multiple pixel blocks and transmits all of its constituent pixel data to each pixel block. The second mode transmits the data of the basic pixels among the constituent pixels. The previous screen. Transmission mode information indicating whether the pixel signal in each transmission mode and the transmission mode of each pixel block is the first mode or another transmission mode by assigning any one transmission mode of the third mode that diverts the data of An image receiving device in a transmission system that continuously transmits a plurality of temporally continuous image groups by transmitting a block that calculates block distortion for each pixel block with respect to the previous screen from received pixel signals. distortion calculation means, mode memory means for storing the transmission mode of each pixel block of the previous screen, block distortion by the block distortion calculation means, transmission path quality, and transmission mode of the previous screen stored in the mode memory means; a threshold setting means for setting a threshold for each transmission mode change between the previous screen and the current screen, and evaluating the transmission mode of each pixel block of the current screen according to the threshold set by the threshold setting means. An image receiving device comprising: evaluation means for evaluating images.