JPH10136358A - Image decoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image decoder in which field freeze is conducted for freeze processing when flurring in a moving image is a problem and frame freeze is conducted for freeze processing when the deterioration of resolution a still image is a problem. SOLUTION: A display control section 17 of the image decoder is provided with a changeover circuit 32 that selects frame freeze or field freeze in response to information denoting whether or not image data in an image input stream is a progressive image. As to the information denoting whether or not image data in an image input stream is a progressive image, a value of a progressive frame in a picture header extension in an input stream is used.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image decoding apparatus, and more particularly to an apparatus for decoding a digitally compressed image.

[0002]

2. Description of the Related Art When digitally represented image data is transmitted or stored, encoding is performed to reduce the amount of data. As an encoding method, there is a method of reducing the degree of redundancy by utilizing temporal or spatial correlation of image information (image data). As a method using the temporal correlation, there are a method of encoding a difference between two screens (frames), a method of detecting a motion of an image, and performing a motion compensation.
In addition, as a method of utilizing spatial correlation, the image is divided into blocks of a predetermined size (for example, in the vertical direction and in the horizontal direction each of 8 pixels), and the data in the blocks is orthogonally transformed.
There is a method in which transform coefficients are scan-converted (for example, rearranged from low-frequency components to high-frequency components) and variable-length coding is performed. An image coding method (hereinafter abbreviated as MPEG2), which is a standard defined by the Moving Picture Experts Group (MPEG), uses the above two methods in combination. MPEG2 recommendations are described in ISO / IEC13818-2. The image coding method by MPEG is as follows. The image is encoded in the order of a predetermined frame with reference to the preceding and following frames. First, forward prediction is for predicting the current frame from past frames, backward prediction is for predicting the current frame from future frames, and bidirectional prediction is for predicting the current frame from both past and future frames. That. Frames that are coded without using inter-frame prediction are intra-frames (I-frames), frames that are coded by forward prediction are forward-predicted frames (P frames), and frames that are coded by bi-directional prediction are bi-directional prediction. It is called a frame (B frame). The prediction method in the P frame and the B frame is determined for each macro block. That is, P
The macroblock in the frame is an intra macroblock without motion compensation or a forward prediction macroblock, and the macroblock in the B frame is any one of an intra macroblock, a forward prediction macroblock, a backward prediction macroblock, and a bidirectional prediction macroblock. become. Now, input frames in the order of I frame I0, B frame B1, B frame B2, P frame P3, B frame B4, B frame B5, P frame P6, B frame B7, B frame B8, P frame P9 are encoded. Consider the case. In order to encode a B-frame, the reference frames on both sides must first be encoded. Therefore, the order of encoding is I0, P3, B
1, B2, P6, B4, B5, P9, B7, B8. Conversely, I
In order to decode a stream in the order of 0, P3, B1, B2, P6, B4, B5, P9, B7, B8, decoding is performed in this order, and I0, B
1, B2, P3, B4, B5, P6, B7, B8, and P9 are displayed in this order.
The operation of rearranging the encoded input images in the order of display is called reordering.

FIG. 5 shows an example of the configuration of an image decoding apparatus for decoding data encoded by such a method. FIG.
, The decoding process is executed by the buffer control unit 51, the variable length decoding unit 52, the scan converter 53, the inverse quantizer 54, the inverse DCT unit 55, and the motion compensation image reproducing unit 56. A memory 520 is used for the decoding device, and a buffer memory 5
21 and the frame memory (the three I, P, and B frame memories described above. The I frame is not always written in the I frame memory, and the P frame may be written. In some cases, an I frame may be written instead of a P frame being written.
23 is called a P frame memory) 522, 523, 524
Consists of Further, the display control unit 57 reads the display data from the memory 520 and outputs a decoded image. The decryption processing unit 50 includes a memory control unit 58, and the memory 520
Is transferred via 58. Next, the operation will be described. The input bit stream is stored in the buffer memory 521 under the control of the buffer memory control unit 51. Data is read from the buffer memory 521 for each frame, and the read data is variable-length decoded by the variable-length decoder 52. Although not all data is variable-length coded, it is assumed that fixed-length codes are also decoded by the variable-length decoder 52.
Next, after the scan converter 53 rearranges the order of the data, the data is inversely quantized by the inverse quantizer 54. next,
The inverse DCT unit 55 performs an inverse discrete cosine transform. When the inter-frame difference is received, the motion-compensated image reproducing unit 56 reads out the previously decoded reference data from the predicted frame memory 522 or 523 (in FIG. However, there is also a case where data is read out from one of them), and after adding the received data, the reproduced image is written to the predicted frame memory 522 or 523 or 524.
(FIG. 5 shows a case where data is written to the B frame memory 524. The I frame memory 522 and the P frame memory 523
In some cases). When the data encoded in the frame is received, the received data is written to the predicted frame 522 or 523 as it is. The reproduced image is reproduced as described above. Further, any one of the reproduced images of the I, P, and B frames is read from 522, 523, and 524, and the display data is output. A display control signal is sent from 57 to 58 for reading the display data.

[0004]

By the way, there is a case where it is desired to freeze the display screen (display the same screen repeatedly twice or more) during the MPEG image decoding operation. There is a case where the user wants to keep the screen being broadcast and keep it, or sometimes wants to keep the screen on a specific screen of the moving image player. The former may be called a freeze, and the latter may be called a pause. In the present invention, all of the cases where the display screen is stopped including these cases are called a freeze.

[0005] On the other hand, screen display includes a method called an interlaced display and a method called a non-interlaced (or progressive) display. In the former, one screen (frame) is divided into two fields and displayed, and the scanning lines of the first field and the second field are alternately arranged. The latter scans one screen from top to bottom. FIG. 6A shows an example of the interlaced display. In this figure, for simplification, it is assumed that one frame is composed of ten scanning lines, that is, one field is composed of five scanning lines. Actually, in NTSC, one frame is 525 scanning lines (480 lines are displayed), that is, 262.5 scanning lines are displayed in one field (240 lines are displayed). In PAL, One frame is composed of 625 scanning lines (576 lines are displayed), that is, one field is composed of 312.5 scanning lines (288 lines are displayed). In FIG. 6A, odd-numbered scanning lines are scanned first to form one field, and then even-numbered scanning lines are scanned to form the next one field. The former is called a top field because it is displayed one scanning line higher than the latter, and the latter is called a bottom field. The interval between the two fields is about 1/60 second in NTSC and 1/50 second in PAL. If the display image is stationary, it is displayed as shown in FIG. 6 (A), but if it is moving, movement appears between the fields as shown in FIG. 6 (C).

[0006] In the case of freezing, a method of repeatedly displaying the same frame and a method of repeatedly displaying the same field can be considered. In the method of displaying the same frame repeatedly, in the case of a moving image, FIG.
Are alternately displayed, and the screen appears to be blurred. Conversely, in the method of repeatedly displaying the same field which is one of the two fields, only one field is displayed as shown in FIG. 6B, so that the resolution in the vertical direction deteriorates. As described above, there is a disadvantage that the moving image is blurred in the conventional frame freeze, and the resolution of the still image is deteriorated in the field freeze.

Accordingly, it is an object of the present invention to provide a field freeze for a freeze when moving image blurring is a problem and a frame freeze for a freeze when resolution of a still image is degraded. To provide an image decoding device that can be controlled in the same manner.

[0008]

In order to achieve the above object, an image decoding apparatus according to the present invention provides a frame freeze in response to information indicating whether or not image data in an image input stream is a progressive image. And a switching circuit for switching between the field freeze and the field freeze. More specifically, the information as to whether the image data in the input stream is a progressive image includes a progressive frame in the picture header extension in the input stream.
(progressive_frame) value is used. The switching circuit of the display control unit performs frame freeze when the value of the progressive_frame is 1, and performs field freeze when the value of the progressive_frame is 0.
Controls data read from the frame memory.

Therefore, if it is determined from the above information that the image data in the input stream is not a progressive image (ie, that the image data in the input stream is an interlaced display), the switching circuit of the display control unit sets the Since the same field which is one of the two fields is repeatedly displayed as the freeze, the problem of the blurring of the freeze of the moving image is solved.
On the other hand, when it is determined from the above information that the image data in the input stream is a progressive image (that is, the image data in the input stream is non-interlaced display), the switching circuit of the display control unit sets the frame freeze. Since two fields of one frame are alternately displayed, the problem of the resolution degradation of the freeze of the still image is solved.

[0010] By the way, the progressive_frame information in the picture header extension of the input stream is originally introduced for use in a stream which has been subjected to hierarchical scalable sequence. This hierarchical coding provides high-definition television broadcasting compatible with current television broadcasting. Therefore, in this high-definition television broadcast, only the difference from the current television broadcast signal is added to the current television broadcast signal and transmitted. On the high-definition television receiver side, the difference signal is processed by a spatial filter. Transmission and reception of high-definition television broadcasting compatible with current television broadcasting becomes possible. Furthermore, this prog
The ressive_frame information is defined so as to be finally equal to the information of chroma_420_type indicating the color difference format. That is, this information makes it possible to switch the color difference data interpolation method.
As described above, in the related art, the progressive_frame information is for specifying the hierarchical coding or the chrominance format, and is not for specifying the freeze method. It is used for switching control. As a result, a good freeze screen can be obtained without requiring new information.

Further, in another specific embodiment of the present invention, in addition to the mode for switching between the frame freeze and the field freeze using the value of the progressive_frame information, the frame freeze is unconditionally performed regardless of the value of the progressive_frame information. And a mode for unconditionally setting the field freeze irrespective of the value of the progressive_frame information, and means for selecting these modes are provided.

[0012]

FIG. 1 shows a first embodiment of the present invention.
3 will be described with reference to FIG.

FIG. 1 shows the configuration of an image decoding apparatus according to the present embodiment. In FIG. 1, a buffer control unit 11, a variable length decoding unit 12, a scan converter 13, an inverse quantizer 14, an inverse DCT
The decoding process is executed by the unit 15 and the motion compensation image reproducing unit 16. A memory 120 is used for the decoding device, and a buffer memory 121 and a frame memory (three I, P, and B frame memories. An I frame is not always written in the I frame memory, and a P frame may be written. On the contrary, the P frame is not always written in the P frame memory, and the I frame may be written.
123 is called a P frame memory) 122, 123, 12
Consists of four. Further, the display control unit 17 reads display data from the memory 120 and outputs a decoded image. Further, the decoding processing unit 10 has a memory control unit 18, and data transfer between the decoding processing unit 10 and the memory 120 is performed by the memory control unit 1.
8 is performed. Next, the operation will be described. MP
The input bit stream that is the digital image data of the EG is stored in the buffer memory 121 under the control of the buffer memory control unit 11. The data read from the buffer memory 121 is variable-length decoded by the variable-length decoder 12. Although not all data is variable-length coded, fixed-length codes are also decoded by the variable-length decoder 12. Next, after the order of the data is rearranged by the scan converter 13, the data is inversely quantized by the inverse quantizer 14. Next, the inverse DCT unit 15 performs an inverse discrete cosine transform. When the inter-frame difference is received, the motion-compensated image reproducing unit 16 reads the previously decoded reference data from the predicted frame memory 122 or 123 (in FIG. 1, the reference data is read from the I frame memory 122 and the P frame memory 123). However, there is also a case where data is read out from one of them), and after adding the received data, the reproduced image is written into the predicted frame memory 122 or 123 or 124.
(FIG. 1 shows a case where data is written to the B frame memory 124. The I frame memory 122 and the P frame memory 123
In some cases). When the data encoded in the frame is received, the received data is written to the predicted frame 122 or 123 as it is. The reproduced image is reproduced as described above. Further, any one of the reproduced images of the I, P, and B frames is read from 522, 523, and 524, and the display data is output. A display control signal is sent from the display control unit 17 to the memory control unit 18 for reading the display data. The screen freeze is performed by repeatedly displaying data from any of the frame memories 122, 123, and 124 that is displayed last. In the present embodiment, progressive_frame information serving as a control signal is sent from the variable length decoder 12 to the display control unit 17, and the display control unit 17 is controlled based on this information.

FIG. 2 shows a freeze control circuit which is a part of the display control unit 17. As described in the description of the related art, since the order of frames in the input stream and the order of display are different, the control signal progressive_frame must also be reordered. This is performed by the progressive_frame reorder register 21. The progressive_frame control signal output from the reorder register 21 is output from the AND gate 2
After the polarity is inverted at 2 (a circle on the input side of the AND gate 22 indicates the polarity inversion), it is ANDed with the freeze / normal signal, and further ORed with the vertical field timing by the OR gate 23, and the top / bottom It becomes a selection signal.

FIG. 3A shows a signal waveform when the progressive_frame control signal is "L", that is, fixed at 0. The freeze / normal signal changes from "L" to "H", that is, switches from the normal mode to the freeze mode. During the normal mode, the output of the AND gate is "L", the vertical field timing becomes the top / bottom selection signal as it is, and the top field and the bottom field are displayed alternately for each field. That is, normal interlace display is performed. In the freeze mode, the inverted signal of the progressive_frame control signal becomes the output of the AND gate 22, and in this example, t
The op / bottom selection signal is fixed at "H". That is, the same field is repeatedly displayed. progressive_frame
Since the control signal "L" indicates encoding of an interlaced image, field freeze is performed, and blurring of a moving object can be stopped. In this example, top / bo
Although the ttom selection signal is fixed at "H", it goes without saying that a field freeze occurs even when the signal is fixed at "L".
As described above, when it is determined by the progressive_frame control information that the image data in the input stream is not a progressive image (that is, the image data in the input stream is interlaced display), the switching circuit of the display control unit sets Since the same field which is one of the two fields is repeatedly displayed as the freeze, the problem of the blurring of the freeze of the moving image is solved.

FIG. 3B shows a signal waveform when the progressive_frame control signal is "H", that is, when it is fixed at "1". The freeze / normal signal changes from "L" to "H", that is, switches from the normal mode to the freeze mode. During the normal mode, the output of the AND gate is "L", the vertical field timing becomes the top / bottom selection signal as it is, and the top field and the bottom field are displayed alternately for each field. That is, normal interlace display is performed. In the freeze mode, the inverted signal of the progressive_frame control signal becomes the output of the AND gate 22, and in this example, t
The op / bottom selection signal is fixed at "L". That is, even in the freeze mode, the top field and the bottom field are displayed alternately. Since “H” in the progressive_frame control signal means encoding of a progressive image,
A frame freeze is obtained, and a freeze screen with no deterioration in resolution is obtained. As described above, when it is determined by the progressive_frame control information that the image data in the input stream is a progressive image (that is, the image data in the input stream is non-interlaced display), the switching circuit of the display control unit Is to alternately display two fields of one frame as a frame freeze, so that the problem of the deterioration of the resolution of the freeze of the still image is solved. Thus, the value of the progressive_frame control information in the picture header extension is 1
Indicates that the data following this header is encoded in frame units. That is, the original image does not consist of two fields having a temporal difference but is one progressive frame of non-interlaced display. Therefore, the switching circuit of the display control unit selects the frame freeze.

Conversely, a value of 0 for the progressive_frame control information in the picture header extension indicates that the data following this header is not necessarily encoded in frame units. That is, the original image may be composed of two fields that are temporally different, and a field freeze is selected to prevent blurring.

Further, even if the value of the progressive_frame control information is 0, the image is not always a moving image, so that the frame is unconditionally frozen as in the second embodiment of the present invention described below. It is also possible to select a mode that does not degrade the resolution. In addition, 2
A progressive image is generated based on one or three fields, and is set to a value 1 of the progressive_frame control information.
In some cases, the mode can be selected unconditionally as a field freeze to prevent blurring.

Hereinafter, a second embodiment of the present invention will be described with reference to FIG. FIG. 4 is a diagram showing a configuration of a portion corresponding to FIG. 2 of the first embodiment. 4, blocks having the same reference numerals as those in FIG. 2 have the same functions as those in FIG. In this embodiment, a selector 24 is provided after the reorder register 21. The input control signals of the selector 24 are an automatic / forced mode selection signal and a field / frame selection signal. When the automatic / forced mode selection signal is "L", that is, 0, the forced mode is set, and the field / frame selection signal becomes valid. That is, when the field / frame selection signal is "L", that is, 0, the output of the selector 24 becomes "H", and the frame is frozen as in FIG. When the field / frame selection signal is "H", that is, 1, the output of 24 becomes "L", and the field is frozen as in FIG. Thus p
Force field or frame freeze, ignoring rogressive_frame information. When the automatic / forced mode selection signal is "H", that is, 1, the output of the selector 24 is the same as the output of the reorder register 21 regardless of the field / frame selection signal. That is, a field or a frame freeze is automatically selected based on the progressive_frame information as in the first embodiment. By forcibly selecting the freeze mode as in this embodiment, even if the value of the progressive_frame is 0, the frame is unconditionally frozen, and the image with little motion is frozen without lowering the resolution. Becomes possible. Conversely, a progressive image is generated based on two or three fields, and the progressive image is generated.
In some cases, it is possible to select a mode for preventing blurring as a field freeze, as it is encoded in some cases.

[0020]

According to the present invention, control is performed so that a freeze when moving image blurring is a problem is a field freeze, and a freeze when moving image resolution degradation is a problem is a frame freeze. It is possible to provide an image decoding device capable of performing the operation.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a first embodiment of the present invention.

FIG. 2 is a freeze control circuit diagram included in the display control unit according to the first embodiment of the present invention.

FIG. 3 is a diagram for explaining the operation timing of the first embodiment of the present invention.

FIG. 4 is a freeze control circuit diagram included in a display control unit according to a second embodiment of the present invention.

FIG. 5 is a conventional configuration diagram of an image decoding device.

FIG. 6 is a view for explaining freeze of an image.

[Explanation of symbols]

 10, 50: image (video) decoder, 11, 51: buffer controller, 12, 52: variable length decoder, 13, 53: scan converter, 14, 54: inverse quantizer, 15, 55: inverse DCT unit, 16, 56: motion compensation, image reproduction unit, 17, 57: display control unit, 18, 58: memory control unit, 21: progressive_frame reorder register, 22: AND gate for inverting one input, 23: OR gate, 24: selector, 120, 520: memory, 121, 521: buffer memory, 122, 522: I frame memory, 123, 523: P frame memory, 124, 524: B memory.

Claims (3)

[Claims] A storage device for storing image data of an input stream; a decoding processing unit for writing frame data formed by decoding image data read from the storage device into the storage device; A display control unit for reading the frame data to form display data, wherein the display control unit responds to information as to whether the image data of the input stream is a progressive image,
(Freeze) When generating an image, a frame freeze display operation for alternately displaying two fields of one frame in the case of a progressive image, and one of two fields of one frame in the case of a non-progressive image. An image decoding device comprising a switching circuit for switching between a field freeze display operation of repeatedly displaying the same field. 2. The image according to claim 1, wherein the switching circuit of the display control section has a function of forcibly setting a frame freeze display operation or a field freeze display operation in the image decoding device. Decoding device. 3. The decoding processing unit according to claim 1, wherein the decoding processing unit includes a variable length decoder, an inverse quantizer, an inverse discrete cosine transform unit, and a motion compensation image reproducing unit. An image decoding apparatus according to claim 1.