KR100215789B1 - Apparatus and method for driving pip in digital tv - Google Patents

Abstract

An apparatus and method for implementing a PIP screen of a digital TV, the method comprising: detecting I, P, and B frames by checking a picture head of an input video signal for a sub picture, and a DC value for all frames regardless of the detected frames. The data is erased according to the DC value calculation result or the PIP screen is configured by summing the DC value of the previous reference frame which is motion compensated with respect to the DC value of the current frame and the calculated DC value of the current frame. It is very good for PIP application because it can recover the excellent picture which the motion is not damaged in spite of the original picture quality even though the video decoder of simplification is very simple, especially when trying to implement multiple PIP screens simultaneously It can be pursued.

Description

Digital TV's PIP Implementation Device and Method

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to digital TV, and more particularly, to a PIP implementation apparatus and method for digital TV capable of obtaining a high quality picture-in-picture (PIP) screen with minimal overhead in digital TV broadcasting. will be.

In general, a PIP screen of a digital TV is to form a small sub-screen at the corner of the TV main screen, that is, a part, for the purpose of channel searching in TV broadcasting.

Depending on the application, it may be necessary to display multiple PIP screens, so the simpler and economic the PIP implementation circuit, the greater the advantages.

Hereinafter, a PIP implementation apparatus and a method of a digital TV according to the prior art will be described with reference to the accompanying drawings.

1 is a block diagram showing a PIP implementation apparatus of a digital TV according to the prior art, and FIG. 2 is an operation flowchart showing a PIP implementation method of a digital TV according to the prior art.

First, referring to FIG. 1, a configuration of a first decoding unit 11 which restores a compressed mother screen video signal to an original video signal; After reconstructing the variable-length encoded video signal for a certain length, extracting data for a specific frame (I-frame), and extracting the DC value required for the parent / child picture from it. A second decoder 12 for outputting the signal; An interpolation and attenuation unit which receives the reduced image data output from the second decoder 12 and interpolates or attenuates the image data according to the window size, and restores the image data for the sub picture to the enlarged or reduced image data. (13); A storage unit 14 storing the image data for the sub picture output from the interpolation and attenuation unit 13; A switching unit 18 for synthesizing the mother screen image data output from the first decoder 11 and the child screen image data output from the memory 14; A control unit l5 for outputting a window size signal and a window pan for controlling the own screen; An address generator (16) for supplying an address signal to the memory (14) to output valid sub-screen image data according to the window size signal; And a moving generator 17 for supplying a timing signal to the switching unit 18 and the address generator 16 under the control of the window fan signal.

An operation of the configuration will be described.

It is assumed that the compressed video bitstream is simultaneously input to the A channel and the B channel.

At this time, the video bitstreams of the A and B channels refer to only the result obtained by correcting the error information by demodulating the signal transmitted on the transmission line and extracting only the information about the image from the transport bitstream.

The main channel, which is the A channel, that is, the channel for the screen that the viewer wants to view on the parent screen, and the B channel means the subchannel, that is, the channel for the screen that the viewer wants to selectively view on the own screen.

Although one B channel is illustrated in FIG. 1, two or more B channels may exist.

If there is no PIP function and only the main channel video is to be viewed, only the A channel signal is restored to the original video signal through the first decoder 11.

The first decoder 11 is a block for decoding an image compressed in a method such as MPEG, which can be replaced by any known protector.

The video signal input through the subchannel, that is, the B channel, is restored to a code of a constant length like other decoders through a variable length decoder (not shown). If the video signal is encoded to compensate for motion, In the variable length decoder, only data for a frame without motion, that is, an I-frame, is extracted.

In other words, if multiple P-frames and B-frames (motion-compensated frames) follow one I-frame, as in the case of MPEG, all data for these frames are removed and only the data for the I-frame is extracted.

Only the DC value required for the PIP is extracted from the extracted I-frame data. In the case of MPEG, this means a DC value for 8 × 8 blocks.

By doing so, the variable length decoder can be implemented with a simpler structure than a normal image data decoder, i.e., a circuit related to motion compensation becomes unnecessary, and thus a code table generated is unnecessary.

Only the table which decodes data for an I-frame in variable length code is needed.

In general, the DC value is composed of a code, not a variable length, and the DC value is restored to the reduced image data. At this time, the reduced image data is the image reduced by the horizontal and vertical size of the DCT block. Image data reduced by 1/8.

The operation principle of the variable length decoder is shown in FIG. 2, and this principle can be applied to any existing variable length decoder.

In other words, first check the picture header (S1O1), and then check whether it is an I-frame (S102) and if it is I-frame data, restore the variable length code through the variable length decoder (S103) or continue to the next. Check the picturehead at and skip until an I-frame is detected.

After the above-mentioned variable length decoding process, to extract only the DC value, it is checked whether it is a DC value (S104). If it is not the DC value, the data is erased (S106), and if it is a DC value, data is output (S105).

The reduced image data output through the second decoder 12 is supplied to the interpolation and attenuation unit 13 to adjust the size of the sub picture.

In other words, when a large size screen is desired in the process of reproduction, the data reduced through interpolation is restored to a larger size image data, and when a small size screen is desired, it is attenuated again.

In this case, the interpolation may be calculated by simply using a zoom or a calculation process between adjacent pixels, and the applied interpolation and attenuation circuit may be any known technique.

When the image data of the desired size is made, it is stored in the storage unit 14. The small sized sub picture data stored in the storage unit 14 is input from the first decoding unit 11 by the switching unit 18. It is synthesized with video data of the parent screen.

At this time, the switching unit 18 is controlled by the timing signal output from the timing generating unit 17. The timing generating unit 17 determines the switching point of the mother screen video signal and the sub screen video signal. At the same time, the address generator 16 determines the time at which the address is output to output the data of the storage 14.

The controller 15 outputs to the timing generator 17 a control signal for outputting a window pan, i.e., the position and window size of the own screen image, that is, the size of the own screen, and information desired by the producer and the user. The timing generator 17 then outputs a timing signal into which the sub-screen video data is inserted.

In addition, the control unit 15 outputs the window size information to the interpolation and attenuation unit 13 so as to generate a user screen image data having a size desired by the user, and the timing information unit 17 also provides the size information. Outputs

In response to this, the timing signal output from the timing generator 17 outputs valid video data for the sub-screen from the storage 14.

If a P-frame or a B-frame follows consecutively after an I-frame such as MPEG, the P-frame or B-frame is removed, so the I-frame is reduced / until the next I-frame occurs. Duplicate output

At this time, since the data for the I-frame is already stored in the storage unit 14, it is possible to output as many frames as desired.

However, when the interpolation and attenuation unit 13 does not exist in FIG. 1, as shown in FIG. 2, the output of the second decoding unit 12 is recorded in a memory composed of RAM having a predetermined capacity.

In the case of a 2x zoom image, the address generator 16 generates an address so that data of one address is repeatedly read, and when it is reduced to 1/2, the address generator 16 skips the addresses in the storage unit 14 one by one. Let's do it.

In this way, the screen can be doubled or enlarged.

The PIP implementation apparatus and method of the digital TV according to the prior art are to add a dedicated PIP circuit to the receiver and recover the PIP screen using only the DC component of I-frame data in the compressed video bit stream as the circuit. It is possible to get very simple and cheap PIP circuit by eliminating the circuit part related to DCT and motion compensation, but the frame is about 4,5 because the frame is decoded and displayed by decoding only I-frame data. There is a problem that can only be obtained.

Accordingly, an object of the present invention is to solve the above-mentioned problems according to the prior art, and an object of the present invention is to decode an I-frame in the same manner as in the prior art, but adds minimal data to the coded bit stream at the encoder side. The present invention provides a digital TV PIP implementation that decodes B frames so that a cumulative error does not occur to form a PIP screen.

Another object of the present invention is to provide a method corresponding to the apparatus.

1 is a block diagram showing a PIP implementation of a digital TV according to the prior art

2 is an operation flowchart showing a PIP implementation method of a digital TV according to the prior art;

Figure 3 is a block diagram showing a PIP implementation of a digital TV according to the present invention

FIG. 4 is a detailed block diagram of the second decoding unit shown in FIG. 3. FIG.

5 is a view for explaining a DC value interpolation method of a reference frame according to the present invention;

6 is an operation flowchart showing a PIP implementation method of a digital TV according to the present invention.

Explanation of symbols for main parts of the drawings

21,22: decoder 23,26: storage

24: control unit 25: format conversion unit

27: switching part

A feature of the present invention is to implement a PIP screen using all DC values for I, P, and B frames of an input video signal.

A feature of the digital TV PIP apparatus according to the present invention is characterized by comprising: decoding means for outputting decoded image data by reconstructing an input sub-screen image signal into a DC value irrespective of a frame; First storage means for temporarily storing the video data decoded by the decoding means and outputting the image data according to an output control signal; According to the scan selection control signal and the screen size selection control signal provided with the image data output from the first storage means, the DC value in the space or the DC value on the time axis are interpolated to have the same scanning method and frame rate as the main screen. Format converting means for converting the format to an appropriate size; A switching unit for synthesizing and displaying the image data of the format-converted sub picture and the image data of the decoded main picture according to a control signal; And control means for providing the one control signal, the scan selection control signal, and the screen size selection control signal.

A feature of the digital PIP implementation method according to the present invention includes a decoding step of outputting decoded image data by reconstructing an input image signal for a sub picture only into a DC value regardless of a frame; After reconstructing the decoded image data only by the DC value, the data is temporarily stored and output according to one output control signal, and the output image data is interpolated in space according to a scan selection control signal and a screen size selection control signal. A format conversion step of interpolating the DC value on the time side to have an appropriate size with the same scanning method and frame rate as the main picture; Synthesizing and displaying the image data of the format-converted sub picture and the decoded main picture according to one control signal.

Another feature of the method of the present invention includes the steps of: detecting a corresponding frame (I, P, B) by checking a picture head of an input image signal for a sub picture; calculating a DC value irrespective of the detected frame; And erasing data according to the DC value calculation result or summing the DC value of the previous reference frame whose motion compensation is performed with respect to the DC value of the current frame and the calculated DC value of the current frame to form a PIP screen. Is in.

Hereinafter, an apparatus and method for implementing a digital TV according to the present invention will be described with reference to the accompanying drawings.

3 is a block diagram showing a PIP implementation of a digital TV according to the present invention, FIG. 4 is a detailed block diagram showing a second decoding unit shown in FIG. 3, and FIG. 5 is a view of a reference frame according to the present invention. It is a figure for demonstrating the DC value interpolation method.

First of all, instead of adding minimal data to the compressed data stream compressed in digital TV broadcasting, the present invention provides a simple PIP decoder so as to obtain a small sub picture without any distortion.

The data added to the video encoding sequence in the present invention are DC values of a reference frame required for motion compensation of these data for every P-frame data, and there is no problem for a general video decoder to ignore these data.

The configuration of the present invention is a video encoder that compresses the additional data as described above and inserts it in a video encoding bit stream, a compressed video encoding sequence in which such additional data is properly inserted, and a PIP having excellent motion characteristics by using such a new encoding sequence. It is composed of a decoder dedicated to PIP that decodes the screen.

Adding new data to the compressed video bitstream or compressing the added data is not very important or new, so let's focus on how the PIP-only decoder works using the added data. would.

Here, the PIP implementation apparatus of the digital TV according to the present invention shown in FIG. 3 will be described in detail.

First, referring to FIG. 3, the configuration thereof includes a first decoder 21, a second decoder 22, a controller 24, a first storage 23, a format converter 25, and a second. It consists of a storage part 26 and a switching part 27.

The first decoder 21 restores the compressed mother screen video signal input through one channel (A channel) to the original video signal and outputs the original video signal to the switching unit 27. Information such as a method and frame rate is output to the control unit 24.

The second decoder 22 restores the input image signal for a sub picture to a predetermined length, and then reconstructs the sub picture image signal for reconstruction only with a DC value regardless of the type (I, P, or B frames) of the frame. The control unit 24 outputs the information to the first storage unit 23 to the first storage unit 23 and outputs information such as an image scanning method and a frame rate to the control unit 24 in the case of the B channel.

The control unit 24 determines the synchronization signal and information on the scanning method of the mother / sub picture video data output from the first decoding unit 21 and the second decoding unit 22 and the size of the sub picture. A control signal corresponding to the switching unit 27, the first and second storage units 23 and 26, and the format conversion unit 25 is provided.

The first storage unit 23 stores the decoded sub picture video data reconstructed only with a DC value irrespective of the type (I, P, or B frames) output from the second decoder 22. Outputs to the format conversion unit 25 according to the output control signal of the control unit 24.

Here, the first storage unit 23 matches the speed because the speed of the video data for the sub picture output from the second decoder 22 may be different from the speed of the data output from the format converter 25. It is a storage unit to play a role of buffering.

The format converter 25 properly interpolates the image data for the sub picture output from the first storage unit 23 according to the screen size control signal and the control signal for the scanning method provided from the controller 24. It interpolates on the time axis and converts the format to the appropriate screen size with the same scanning method and frame rate as the mother screen on which the PIP screen is output.

The format-converted video data is output to the second storage unit 26 according to an output control signal of the controller 24 and stored.

In this case, the second storage unit 26 serves as a buffer to match the output time point of the video data output from the format converter 25 and the time point at which the data is displayed on one screen. It is necessary.

The video data for the sub picture stored in the second storage unit 26 is output to the switching unit 27 according to the output control signal of the control unit 24.

The switching unit 27 receives the mother screen image data output from the first decoding unit 21 and the sub-screen image data output from the second storage unit 26 to receive the mother screen image data of the control unit 24. According to the display position control signal of the child screen, the child screen (PIP screen) is placed at an appropriate position of the mother screen, and data of two channels are appropriately selected and displayed on one screen.

Here, since the PIP circuit according to the present invention decodes using only the DC value regardless of the type of frame in the video bit stream, it is not meaningful to make the decoded sub picture large.

Therefore, when the size and position combinations of the predetermined sub-screens are basically made only, the format converter 25 and the control unit 24 of FIG. 3 can be sufficiently implemented as logic circuits.

In this case, the second decoder 22 may include: a VLD 22a for variable length decoding of the video signal for the sub picture input through the channel B; An inverse quantizer 22b for inversely quantizing the variable-length decoded video signal into a DC value irrespective of a frame; A frame memory 22d for storing a frame of a variable-length decoded video signal output from the VLD 22a; The motion compensation unit 22c compares the video signal stored in the frame memory with the dequantized video signal to compensate for the motion and outputs the motion compensation unit 22c to the second storage unit 23.

Here, the size of the frame memory 22d has a size capable of storing the second frame in a frame composed only of a DC value.

FIG. 6 is a flowchart illustrating the operation of the second decoding unit 22 shown in FIG. 4. Referring to FIG. 6, an operation method of the PIP-only decoding unit, that is, the second decoding unit 22 will be described with reference to FIG. 6.

First, a picture head of an input video signal of a sub picture is checked (S201), and it is determined whether it is an I-frame (S202).

If it is an I-frame, it is determined whether the DC value is to extract only the DC value (S203).

As a result of the determination, if the DC value is outputted (S204), if not the DC value is erased (S205).

If the determined frame is not an I-frame, it is determined whether it is a P-frame (S206). If it is a P-frame, it is determined whether it is a DC value to extract only a DC value (S207).

If it is not the DC value, the data is erased (S211). If the DC value, the P-frame is a motion compensated reference frame (Reference Frame) is input (S208).

When the reference frame is input, the DC value of the reference frame and the DC value of the P-frame are summed (S209) to output data (S210).

On the other hand, if the input frame is not a P-frame, it is naturally a B-frame.

Therefore, in order to extract only the DC value for the B-frame, it is determined whether it is a DC value (S212). If it is not the DC value, data is erased (S216). If the DC value is interpolated, the reference frame is interpolated (S213).

In this case, since the B-frame is not used for motion compensation of another frame, the DC value of the reference frame can be interpolated as shown in FIG. 6 because interpolation of the reference frame does not significantly affect image quality deterioration.

When the reference frame of the B-frame is interpolated, the DC value of the interpolated reference frame and the DC value of the B-frame are added (S214) to output data (S215).

That is, in the above method, since the I-frame data has no motion compensation, its DC value can be obtained as its compressed bit string.

However, since P and B-frames are motion compensated, accurate DC values cannot be obtained without knowing the DC values of the reference frames.

In addition, the DC value of this reference frame cannot be accurately inferred from any DC frame output from the PIP decoder, that is, the second decoder 22 in the past.

Only approximations can be obtained by appropriate algorithms.

In the case of taking an error and calculating an approximation, since the error accumulates in one group of pictures (GOP) and the quality deteriorates as it goes backward, the data is accurate no matter how much P-frame is used as the reference of another frame. You must recover it.

Therefore, the DC value of the reference frame is transmitted even if some bits are transmitted for the P-frame.

Of course, even in this case, DC values can be compressed by an appropriate method, and additional data such as intra-coded macro blocks are not necessary.

However, since the B-frame is a frame that is not used for motion compensation of another frame, interpolation of the reference frame as described above does not significantly affect image quality deterioration.

Accordingly, the DC value of the reference frame can be interpolated by the method proposed by the present invention as shown in FIG. 5.

Referring to FIG. 5, let B be the reference point of the motion vector and H be the end point of the motion vector.

In FIG. 5, B, O, P, Q, and R are all DC values of the reference frame stored in the frame memory 22d.

At this time, the DC value of H can be approximated by the following equation.

[Equation 1]

In Equation 1, DC [] represents a DC value and S represents an area.

Equation (1) holds true even when any two neighboring areas are zero or any three neighboring areas are zero.

For example, if S3 = S4 = 0

[Equation 2]

Same as

PIP implementation device and method of a digital TV according to the present invention is very excellent for PIP application because it can recover the excellent picture that the motion is not damaged in spite of the original picture quality even though the existing video decoder is greatly simplified. When implementing multiple PIP screens at the same time, there is an advantage that both economical and image quality can be pursued.

Claims (9)

In a digital TV implementation of a PIP having a decoder for restoring a compressed mother screen video signal to an original video signal, the input video signal for a child screen is reconstructed to be a DC value regardless of a frame. Decoding means for outputting decoded video data; First storage means for temporarily storing the video data decoded by the decoding means and outputting the image data according to an output control signal; According to the scan selection control signal and the screen size selection control signal provided with the image data output from the first storage means, the DC value in the space or the DC value on the time axis are interpolated to have the same scanning method and frame rate as the main screen. Format converting means for converting the format to an appropriate size; A switching unit for synthesizing and displaying the image data of the format-converted sub picture and the image data of the decoded main picture according to a control signal; And a control means for providing the one control signal, the scan selection control signal, and the screen size selection control signal. 2. The apparatus of claim 1, wherein the decoding unit comprises: a VLD for decoding a long-length variable input video signal into a DC value regardless of a frame; An inverse quantizer for inversely quantizing a variable-length decoded video signal with the DC value; A frame memory configured to store a frame configured only with a DC value of a variable length decoded video signal output from the VLD; And a motion compensator for compensating for motion by comparing the video signal stored in the frame memory with the dequantized video signal. The digital TV as claimed in claim 1, further comprising a second storage means for buffering to synchronize an output time point of the image data output from the format conversion means and a time point at which the data is displayed on one screen. PIP implementation. The apparatus of claim 1, wherein the first storage means plays a buffering role to synchronize the speed of data output from the decoding means and the output speed of data output from the format converter. . A PIP implementation method of an implementation apparatus of a digital TV, comprising: a decoding step of outputting decoded image data by reconstructing an input image signal for a sub picture only into a DC value regardless of a frame; After reconstructing the decoded image data only by the DC value, the data is temporarily stored and output according to one output control signal, and the output image data is interpolated in space according to a scan selection control signal and a screen size selection control signal. A format conversion step of interpolating the DC value on the time axis to have an appropriate size with the same scanning method and frame rate as the main picture; And synthesizing and displaying the image data of the format-converted main screen and the input decoded main screen according to a control signal. 6. The method of claim 5, wherein the decoding comprises: inversely quantizing the decoded DC value by variable length length decoding the input sub picture image signal into a DC value regardless of a frame; Storing a frame including only DC values for the variable length decoded video signal; And a motion compensation step of compensating for the motion by comparing the stored video signal with the dequantized video signal. 6. The digital TV of claim 5, further comprising a storage step of buffering the synchronization point of time of the image data output in the format conversion step and the time point at which the data is displayed on one screen. How to implement PIP. A PIP implementation method of a PIP implementation apparatus of a digital TV, comprising the steps of: detecting a corresponding frame (I, P, B) by checking a picture head of an input video signal for a sub picture; Calculating a DC value regardless of the detected frame; And erasing data according to the DC value calculation result or adding the DC value of the previous reference frame whose motion compensation is performed to the DC value of the current frame and the calculated DC value of the current frame to form a PIP screen. PIP implementation method of the digital TV, characterized in that. 10. The method of claim 8, wherein the step of calculating the DC value and interpolating the reference frame for the B frame when the detected frame is a B frame.