JP3687458B2 - Compression decoding method and compression decoding apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
  The present inventioncompressionDecryptionMethod andIn particular, the present invention relates to a compression / decoding device, and in particular, can change the format of a component video signal to perform compression encoding with high image quality, and can also perform compression decodingcompressionDecryptionMethod andAnd a compression decoding apparatus.
[0002]
[Prior art]
A digital video signal (hereinafter referred to as 4: 2: 2 image signal) of a 4: 2: 2 component encoding method in which each sampling frequency of the two types of color difference signals is ½ that of the luminance signal. Each sampling frequency of the two types of color difference signals is ½ that of the luminance signal, but a digital moving image signal (hereinafter referred to as 4: 2: 0 component coding system) that transmits the color difference signals in a line sequential manner. (4: 2: 0 image signal), and the bus line control is complicated when transferring the 4: 2: 0 image signal and additional information necessary for compression encoding of the moving image to the data compression unit. An image compression / encoding apparatus that does not allow this has been known (Japanese Patent Laid-Open No. 8-46519).
[0003]
In this conventional image compression coding apparatus, a 4: 2: 2 image signal reproduced by a D1-VTR that performs recording and reproduction in a known D1 format is converted into a 4: 2: 0 image signal by a specification conversion circuit. While being written in the frame memory, it is supplied to the additional information detection unit to detect additional information (motion vector information and mode determination information necessary for motion compensation prediction) necessary for compression encoding of a moving image. The multiplexer outputs the 4: 2: 0 image signal from the frame memory alternately in the odd field and the even field, and in the odd field, the luminance signal pixel and the color difference signal pixel are alternately arranged for each pixel, In the even field, the luminance signal and the additional information from the additional information detection unit are taken out as a data string alternately arranged for each pixel and transferred to the data compression unit. In the data compression unit, the 4: 2: 0 image signal and the additional information are separated by the demultiplexer, and the 4: 2: 0 image signal is compression-coded using the additional information.
[0004]
In this conventional image compression coding apparatus, when the 4: 2: 2 image signal format is converted to 4: 2: 0, the portion where the color difference signal pixel data of one screen is reduced by half is effectively used. Parameter information necessary for compression encoding of 4: 2: 0 images such as MPEG is multiplexed.
[0005]
Next, the compression method MPEG used here will be described. MPEG is the name of an organization that examines video coding standards established in 1988 by ISO / IEC JTC1 / SC2 (International Organization for Standardization / International Electrotechnical Standards Meeting Technical Committee 1 / Technical Committee 2, current SC29) ( It is an abbreviation for “Moving Picture Experts Group”. MPEG1 (MPEG Phase 1) is a standard for storage media of about 1.5 Mbps, and is used for low-transfer rates for video conferencing and video telephony for JPEG and service integrated digital network (ISDN) for the purpose of still image coding. H. for video compression. It inherits the basic technology of H.261 (CCITT SGXV, standardized by the current ITU-T SG15), and introduces a new technology for storage media. These were established in August 1993 as ISO / IEC 11172. MPEG2 (MPEG Phase 2) is an ISO / IEC 13818, November 1994 H.264 standard for the purpose of general-purpose standards so as to be compatible with various applications such as communication and broadcasting. It is established as 262.
[0006]
The format of the MPEG input image is basically 4: 2: 0 in the generally used profile of MP @ ML of MPEG1 and MPEG2. When the 4: 2: 2 component format defined in the ITU-R recommendation 601 is schematically represented on the screen, the color difference indicated by a black circle with respect to the pixel of the luminance signal indicated by the white square in FIG. While the pixel of the signal is halved in the horizontal direction, the 4: 2: 0 component format of MPEG2 described above is a pixel of the luminance signal indicated by a white square as shown in FIG. On the other hand, the color difference signal pixels indicated by black circles are halved in each of the vertical and horizontal directions. In FIG. 14, the pixel position indicates the phase.
[0007]
Further, the 4: 2: 0 component format of MPEG1 is a color difference signal indicated by a black circle with respect to a pixel of a luminance signal indicated by a white square, as shown in FIG. 14C, in contrast to the 4: 2: 2 component format. Is the same as the MPEG2 4: 2: 0 component format in that the pixels are ½ times in the vertical and horizontal directions, but the MPEG2 4: 2: 0 component format is the color difference signal. Pixel position (phase) is different.
[0008]
In any case, the 4: 2: 0 component signal has the same vertical resolution as that of the 4: 2: 2 component signal, and the number of pixels of the color of the digital data is also halved. As is well known, the human eye thins out the number of pixels of the color difference signal at the time of input of MPEG coding based on the experimental result that the recognition ability of the color resolution is smaller than the luminance.
[0009]
The encoded part of MPEG is created by combining several technologies. FIG. 15 is a block diagram showing an example of an MPEG image compression encoding apparatus. In the figure, the input image is decoded by the motion compensated predictor 1, and the time redundant portion is reduced by taking the difference between the motion compensated predicted image and the input image by the subtraction circuit 2. There are three modes of prediction from the past and the future. These can be switched and used for each 16 pixels × 16 pixels MB (macroblock). The prediction direction is determined by the picture type given to the input image. There are P picture, B picture, and I picture. There are two modes in which the prediction from the past and the MB are independently encoded without prediction, and the P picture is present. In addition, there are four modes in which prediction from the future, prediction from the past, prediction from both, and four modes independently encoded exist. It is the I picture that all MBs are encoded independently.
[0010]
In motion compensation (MC), a motion vector is pattern-matched for each MB to detect a motion vector with half-pel accuracy, and is predicted after shifting by the amount of motion. The motion vector has a horizontal direction and a vertical direction, and is transmitted as additional information of the MB together with the MC mode indicating where the prediction is from. A group from an I picture to a picture before the next I picture is called a GOP (Group Of Picture), and when it is used in a storage medium or the like, generally about 15 pictures are used.
[0011]
The differential image signal extracted from the subtraction circuit 2 is subjected to orthogonal transformation in the DCT unit 3. Discrete cosine transform (DCT) is an orthogonal transform that discretely transforms an integral transform with a cosine function as an integral kernel into a finite space. In MPEG, two-dimensional DCT is performed on an 8 × 8 DCT block obtained by dividing MB into four. In general, a video signal has many low-frequency components and few high-frequency components. Therefore, when DCT is performed, coefficients are concentrated in a low frequency.
[0012]
The quantized image data is subjected to quantization by the quantizer 4. This quantization is based on a value obtained by multiplying a value obtained by weighting an 8 × 8 two-dimensional frequency called a quantization matrix with a visual characteristic and a value called a quantization scale for multiplying the whole by a scalar, and using a quantized value as a DCT coefficient. Divide by the digitized value. When inverse quantization is performed by the decoder, a value approximating the original DCT coefficient is obtained by multiplying by the quantized value.
[0013]
The quantized data is variable length encoded by the VLC unit 5. Of the quantized values, the direct current (DC) component uses DPCM (Differencial Pulse Code Modulation) which is one of predictive coding. Also, AC (AC) components are zigzag scanned from low to high, with zero run length and effective coefficient value as one event, and Huffman coding that assigns codes with a short code length from those with high appearance probability Is done. The variable-length encoded data is stored in the temporary buffer 6 and output as encoded data at a predetermined transfer rate.
[0014]
The generated code amount for each macroblock of the output data is supplied to the code amount controller 7, and the error code amount of the generated code amount with respect to the target code amount is fed back to the quantizer 4 to quantize the scale. The amount of code is controlled by adjusting. The quantized image data is inversely quantized by the inverse quantizer 8, inversely DCTed by the inverse DCT device 9, and then temporarily stored in the image memory 11 through the adder circuit 10, and then in the motion compensated predictor 1. , Used as a reference decoded image for calculating a difference image. The output signal of the motion compensation predictor 1 is input to the subtraction circuit 2 and the addition circuit 10.
[0015]
In the case of video, the encoded bit stream output from the buffer 6 has a variable amount of code for each picture. This is because MPEG uses information conversion such as DCT, quantization, and Huffman coding, and at the same time, it is necessary to adaptively change the code amount allocated to each picture in order to improve image quality. This is because the motion compensated prediction is performed, so that the entropy of the encoded image itself changes greatly, such as encoding the input image as it is in some cases and encoding a difference image that is the difference between the predicted images in some cases. In this case, in many cases, the code amount is controlled while allocating the entropy ratio of the image and keeping the limitation of the buffer. The limitation of this buffer is to encode the buffer on the decoding device side so that neither overflow nor underflow occurs, and it is specified as VBV (Video Buffering Verifier) in MPEG.
[0016]
FIG. 16 is a block diagram showing an example of an apparatus for decoding encoded data compressed and encoded by MPEG. In the figure, encoded data compressed and encoded by MPEG is subjected to variable length decoding by a VLD unit 15 and then multiplied by a quantization width by an inverse quantizer 16 to obtain a value approximating the original DCT coefficient. Then, the signal is supplied to the inverse DCT unit 17 and subjected to inverse DCT to be locally decoded.
[0017]
Also, the motion vector and the prediction mode extracted from the inverse quantizer 16 are supplied to the motion compensation predictor 18 together with the decoded data from the image memory 20, and thereby output the image data subjected to motion compensation prediction. The adder 19 decodes image data equivalent to the image data input to the encoding device by adding the data from the inverse DCT device 17 and the image data subjected to motion compensation prediction from the motion compensation predictor 18. The decoded data is supplied to the image memory 20 and output to the outside.
[0018]
[Problems to be solved by the invention]
In the image compression coding apparatus compliant with the conventional MPEG system shown in FIG. 15, as described above, compression coding is performed in the 4: 2: 0 format. On the other hand, DVC (Digital Video Cassette) is a typical home digital VTR. DVC is a compression method different from MPEG, and has an SD standard for the current method and an HD standard for HDTV. In the SD method, the input image format is a 4: 1: 1 component format in NTSC. . In the 4: 1: 1 component format, the pixel of the color difference signal indicated by the black circle is horizontal with respect to the pixel of the luminance signal indicated by the white square as shown in FIG. The direction is ½ times (¼ times the input image), and the vertical direction is the same.
[0019]
Here, consider a case where a 4: 2: 2 image signal is recorded in a DVC, and further output from the DVC and copied to an application using MPEG coding. When image data encoded by DVC is decoded, the pixel interval of the color difference signal is ¼ in the horizontal direction with respect to the pixel interval of the luminance signal because the DVC format is 4: 1: 1. . When this signal is encoded again with the 4: 2: 0 format MPEG encoding, the vertical direction of the 4: 1: 1 image signal is further halved, and the number of pixels is 4: 2: 0 image signal. Despite the presence of the same number of color difference signal pixels, the same color difference signal band as that of the 4: 1: 0 image signal, that is, the number of pixels of the color difference signal is 1 / H in the horizontal direction with respect to the number of pixels of the luminance signal. 4, the vertical direction is equivalent to a signal format of ½.
[0020]
FIG. 17A shows a conceptual diagram of the number of pixels of a 4: 2: 2 image signal, and FIG. 17B shows a conceptual diagram of the number of pixels of a 4: 2: 0 image signal. When the conceptual diagram of the number of pixels is shown in FIG. 5C, the image signal obtained by re-encoding the 4: 1: 1 image signal with the MPEG encoding device is 4: 1: compared with the 4: 2: 2 image signal. The pixel of the 4: 1: 0 image signal as shown in FIG. 17D, which has both horizontal and vertical bad conditions of the color difference signals (Cb, Cr) of one image signal and 4: 2: 0 image signal. It means that it is in the same band as an image that has only a number. As described above, there is a problem in that the deterioration of the color difference signal becomes significant by encoding the format having a different color difference signal format.
[0021]
Note that the conventional image compression coding apparatus described in JP-A-8-46519 discloses a color difference of one screen when a 4: 2: 2 image signal format is converted to a 4: 2: 0 image signal format. By effectively using the portion where the signal pixel data has been reduced by half, the parameter information necessary for compression encoding of 4: 2: 0 image signals such as MPEG is multiplexed, thereby adding the necessary amount as a system. Although there is an effect of not complicating the bus line control when transferring information to the data compression unit, this conventional apparatus does not disclose any means for reducing the deterioration of the color difference signal of the image.
[0022]
The present invention has been made in view of the above points. A compression encoding method, a compression decoding method, and a compression code that can reduce deterioration of a color difference signal of image encoded data when compression encoding is performed by changing a component format. An object of the present invention is to provide an encoding device and a compression decoding device.
[0023]
[Means for Solving the Problems]
  In order to achieve the above object, the present inventionIn the compression decoding method, when an input image signal in a component format of 4: 1: 1 is displayed on one screen, the pixels of the color difference signal in the input image signal arranged in the horizontal direction in a 4-pixel cycle for each line. Among them, when four adjacent lines are taken as one set, the pixels of the color difference signal on the second or third line in each set are shifted to the right by 2 pixels and vertically to a position on 1 or 2 pixels. Then, the pixel of the color difference signal on the fourth line is shifted to the position of 2 pixels to the right and 1 or 2 pixels in the vertical direction, so that the image signal in the 4: 2: 0 component format image signal A first step of performing pixel replacement in which the color difference signal is shifted to a pixel position equivalent to a pixel position displayed on one screen, and the pixel-replaced image signal is compressed in a component format of 4: 2: 0. A second step of performing compression encoding by the encoding device to be encoded, and a third step of obtaining a decoded signal by decoding the compression-encoded image signal by a decoding device that decodes the image signal in a 4: 2: 0 component format And when the decoded signal is displayed on one screen, the pixel shift opposite to the pixel shift in the first step is applied to the pixels of the color difference signal arranged in the horizontal direction in the decoded signal in a two-pixel cycle. And a fourth step of performing pixel replacement in which the color difference signal in the input image signal of the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen.It is characterized by that.
[0024]
  In addition, the present inventionCompression decoding methodIn order to achieve the above purpose,When an input image signal of a component format of 4: 1: 1 is displayed on one screen, when a total of 32 adjacent pixels of 8 pixels in the horizontal direction and 4 pixels in the vertical direction are combined as a set, a cycle of 4 pixels for each line. Among the chrominance signal pixels in the input image signal that are arranged in the horizontal direction and there are eight for each set, four chrominance signal pixels adjacent in the vertical direction are on the second or third line. Pixels of a certain color difference signal are shifted to the right by two pixels and a position one or two pixels in the vertical direction, and pixels of the color difference signal on the fourth line are two pixels to the right and vertical. For the remaining four pixels adjacent to each other in the vertical direction, the color difference signal pixels on the first line and the color difference signal on the third or second line are shifted. Pixels are shifted 2 pixels to the right In addition, by shifting each color difference signal pixel on the second or third line and the fourth line in the vertical direction to the pixel position of the color difference signal after the shift in each vertical direction, A first step of performing pixel replacement in which a color difference signal in an image signal of a component format of 2: 0 is shifted to a pixel position equivalent to a pixel position displayed on one screen, and the image signal after pixel replacement Second step of performing compression encoding by an encoding device that performs compression encoding in a 4: 2: 0 component format, and decoding for decoding a compression-encoded image signal in a 4: 2: 0 component format A third step of obtaining a decoded signal by decoding by the apparatus, and a color difference arranged in a two-pixel cycle in the horizontal direction in the decoded signal when the decoded signal is displayed on one screen The pixel position where the color difference signal in the input image signal in the 4: 1: 1 component format is displayed on one screen by performing the pixel shift opposite to the pixel shift in the first step for the pixels of the signal No. 1 And a fourth step of performing pixel replacement to be displayed at the same pixel position.
[0031]
  Furthermore, in order to achieve the above object, the compression decoding apparatus of the present invention includes:When an input image signal of a component format of 4: 1: 1 is displayed on one screen, four adjacent lines among the pixels of the color difference signal in the input image signal arranged in a horizontal direction with a period of four pixels for each line. , The pixels of the color difference signal on the second or third line in each set are shifted to the position two pixels to the right and one or two pixels vertically, and the fourth line The pixel of the upper color difference signal is shifted to the position two pixels to the right and one or two pixels in the vertical direction, so that the color difference signal in the image signal of the 4: 2: 0 component format is on one screen. A first pixel replacer that performs pixel replacement shifted to a pixel position equivalent to the pixel position displayed in (1) and an image signal that has undergone pixel replacement by the first pixel replacer are 4: 2: 0. Component An encoding device that compresses and encodes with a mat; a decoder that receives a compressed encoded image signal that has been compressed and encoded by the encoding device, decodes the signal in a 4: 2: 0 component format, and outputs a decoded signal; When the decoded signal is displayed on one screen, by performing a pixel shift opposite to the pixel shift by the first step on the pixels of the color difference signal arranged in the horizontal direction in the decoded signal in a two-pixel cycle, The same image as the image signal input to the encoding device by performing pixel replacement in which the color difference signal in the input image signal in the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen. And a second pixel replacement unit that outputs a signal.
[0032]
  Furthermore, in order to achieve the above object, the compression decoding apparatus of the present invention includes:When an input image signal of a component format of 4: 1: 1 is displayed on one screen, when a total of 32 adjacent pixels of 8 pixels in the horizontal direction and 4 pixels in the vertical direction are combined as a set, a cycle of 4 pixels for each line. Among the chrominance signal pixels in the input image signal that are arranged in the horizontal direction and there are eight for each set, four chrominance signal pixels adjacent in the vertical direction are on the second or third line. Pixels of a certain color difference signal are shifted to the right by two pixels and a position one or two pixels in the vertical direction, and pixels of the color difference signal on the fourth line are two pixels to the right and vertical. For the remaining four pixels adjacent to each other in the vertical direction, the color difference signal pixels on the first line and the color difference signal on the third or second line are shifted. Pixels are shifted 2 pixels to the right In addition, by shifting each color difference signal pixel on the second or third line and the fourth line in the vertical direction to the pixel position of the color difference signal after the shift in each vertical direction, A first pixel replacement unit that performs pixel replacement in which a color difference signal in an image signal of a component format of 2: 0 is shifted to a pixel position equivalent to a pixel position displayed on one screen, and a first pixel replacement An image signal that has been subjected to pixel replacement by the encoder in a 4: 2: 0 component format, and a compression-coded image signal that has been compression-encoded by the encoder. : A decoder that decodes in a component format of 0 and outputs a decoded signal, and a color difference signal arranged in a two-pixel cycle in the horizontal direction in the decoded signal when the decoded signal is displayed on one screen The pixel position at which the color difference signal in the input image signal of the 4: 1: 1 component format is displayed on one screen is obtained by performing a pixel shift opposite to the pixel shift in the first step on And a second pixel replacement unit that performs pixel replacement to be displayed at the same pixel position and outputs the same image signal as the image signal input to the encoding device.
[0035]
DETAILED DESCRIPTION OF THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings. First, the concept of the present invention will be described with reference to FIG. In the conceptual diagram of the number of pixels in FIG. 17, the area indicates the number of pixels per screen. In the 4: 2: 2 (component format) image signal, as shown in FIG. Compared with the number of pixels of the luminance signal Y, the number of pixels of Cb and Cr is ½ with respect to the horizontal direction, and the vertical direction is the same. In the 4: 2: 0 image signal, the number of pixels of the color difference signals Cb and Cr on one screen is 1 in the horizontal direction compared to the number of pixels of the luminance signal Y, as shown in FIG. / 2 and 1/2 in the vertical direction.
[0036]
In the 4: 1: 1 image signal, the number of pixels of the color difference signals Cb and Cr on one screen is compared with the number of pixels of the luminance signal Y as shown in FIG. It is 1/4 and the vertical direction is the same. Furthermore, in the 4: 1: 0 image signal, the number of pixels of the color difference signals Cb and Cr in one screen is 1 in the horizontal direction as compared with the number of pixels of the luminance signal Y, as shown in FIG. / 4 and 1/4 in the vertical direction.
[0037]
Accordingly, as can be seen from FIGS. 17B and 17C, the number of pixels (area) of the color difference signals (Cb, Cr) of the 4: 2: 0 image signal and the 4: 1: 1 image signal are equal. In this embodiment, paying attention to this, pixel replacement is performed so that the color difference signal of the 4: 1: 1 component format is converted as if it was a color difference signal of the 4: 2: 0 component format. It is a thing.
[0038]
That is, as described above, when the original compression-encoded image data is an image signal having a 4: 1: 1 component format and is converted into an image signal having a 4: 2: 0 component format, Although the number of pixels in the vertical direction of the color difference signal of the 4: 1: 1: 1 image signal is further halved and the number of pixels is the same as that of the 4: 2: 0 image signal, the number of pixels is the same. As in the case of the 4: 1: 0 image signal, the signal format is such that the number of pixels of the two color difference signals is 1/4 in the horizontal direction and 1/2 in the vertical direction with respect to the number of pixels of the luminance signal. It becomes equivalent.
[0039]
Therefore, in this embodiment, in order to avoid this, the pixel is replaced, and the color difference signal in the 4: 1: 1 image signal is as if it was the color difference signal in the 4: 2: 0 image signal. Convert. This is a method in which the phase of the actual image is ignored, but the phase returns if this inverse transformation is always performed when outputting. In addition, since compression is involved, there is a possibility that encoding degradation will spread, but it can be sufficiently avoided by increasing the encoding rate in order to minimize degradation.
[0040]
For example, since the DVC is about 25 Mbps and the 4: 1: 1 component format, when this output compression encoded image data is recorded on the DVD using the MPEG2 compression of the 4: 2: 0 component format at about 8 Mbps, etc. If the embodiment is not used, no matter how high the encoding rate is encoded, the band of the 4: 1: 0 component format is used in the format, and the vertical resolution becomes about half. In the case of a moving image, field processing is performed to halve the number of pixels in the vertical direction in the field, so that the vertical resolution characteristic in the moving image band becomes ¼. However, by using the present invention, for a general image, encoding deterioration hardly occurs at an encoding rate of 8 Mbps with MPEG2 compression. Therefore, it is possible to perform decoding while maintaining the vertical resolution of the 4: 1: 1 image signal by returning the pixel replacement of the present invention in the reverse procedure.
[0041]
Next, a method for converting a color difference signal in the 4: 1: 1 component format into a color difference signal in the 4: 2: 0 component format (hereinafter referred to as “color difference signal 411/420 pixel replacement method”) is specifically described. explain. The color difference signal 411/420 pixel replacement method is roughly divided into two types: a method of replacing within a frame picture and a method of replacing within a field picture.
[0042]
FIGS. 1A and 1B are explanatory diagrams of the first and second embodiments of the image compression coding method according to the present invention, both of which are methods for replacing the color difference signal 411/420 pixels in the field picture. Show. FIGS. 1A and 1B both show pixel positions on the screen of a 4: 1: 1 image signal, white squares indicate luminance signal pixels, black circles indicate color difference signal pixels, and gray. These circles indicate the pixels of the color difference signal after being replaced with pixel positions where only the luminance signal exists. The actual 4: 1: 1 image signal is, for example, 360 pixels (0 to 359 in the horizontal direction) × 480 lines (0 to 479 in the vertical direction) on the screen. This represents a portion of 8 lines (the same applies to FIG. 2 described later).
[0043]
In the color difference signal pixel replacement method of FIG. 1A, when the pixel number counted from 0 is divisible by 4, and the line number is divided by 4, the pixel of the color difference signal at the pixel position in the first field is 2 When the pixel number is divisible by 4 and the line number is divided by 4, the remainder of the color difference signal at the pixel position in the second field is shifted by 2 pixels to the right and 2 pixels upward. The pixels are shifted by two pixels to the right and two pixels upward (however, pixel number 0 is divisible by 4).
[0044]
Therefore, for example, the pixel number 0 in the first field and the third color difference signal in the vertical direction (line number 2) are divisible by 4 as indicated by the arrow 21 in FIG. Since there are two remaining pixel positions divided by 4, it is shifted by two pixels to the right and two pixels upward and shifted to the pixel position of pixel number 2 and line number 0. Similarly, for example, the pixel number 0 in the second field and the fourth color difference signal in the vertical direction (line number 3) are divisible by 4 as indicated by the arrow 22 in FIG. Divide by 4 and there are three remaining pixel positions, so it is shifted to the right by 2 pixels and up by 2 pixels, and shifted to the pixel position of pixel number 2 and line number 1.
[0045]
With this pixel replacement method, the pixel numbers that are divisible by 4 in the 4: 1: 1 image signal are all divisible by 4 for each pixel of the color difference signal that is present in 2 or 3 remaining line numbers when divided by 4. As a result, the pixel position is shifted to the pixel position of the remaining line number where no color difference signal exists, and as a result, it is replaced with a pixel arrangement equivalent to the pixel position of the color difference signal of the 4: 2: 0 image signal.
[0046]
Further, the color difference signal pixel replacement method of FIG. 1 (B) is the case where the pixel number counted from 0 is divisible by 8, and the line number is divided by 4 to leave 2 and the color difference signal at the pixel position in the first field is left. The pixel is replaced so that it is shifted 2 pixels to the right and 2 pixels upward, and when the pixel number is divided by 8, the remainder is 4 and the line number is divisible by 4. The color difference at the pixel position in the first field The pixel of the signal is shifted to the right by two pixels, and then the two pixels below are shifted by two pixels to the current position (up). Also, when the pixel number is divisible by 8 and the line number is divided by 4 and the remainder is 3, the pixel of the color difference signal at the pixel position in the second field is shifted by 2 pixels to the right and by 2 pixels upward To do. Further, the pixel of the color difference signal at the pixel position of the second field, which is four remainders when the pixel number is divided by 8, and one remainder when the line number is divided by 4, is shifted by two pixels to the right, and then 2 This is a method of shifting the next lower pixel to the current position (upward) by 2 pixels (provided that 0 is divisible by 4).
[0047]
Accordingly, for example, the first color difference signal in the vertical direction (line number 0) at pixel number 4 is at the pixel position in the first field where the pixel number is divisible by 8 and the line number is divisible by 4. B) is shifted to the right by two pixels as indicated by an arrow 23 and is replaced with the position of line number 0 by pixel number 6 where no chrominance signal pixel originally exists, and subsequently, pixel number 4 and line number 0 The pixel of the chrominance signal in the first field of pixel number 4 and line number 2 two pixels below is shifted as indicated by the arrow 24 in FIG. 1B, and the current chrominance signal pixel no longer exists. Shift to the pixel position of pixel number 4 and line number 0 of the position. In addition, the color difference signal pixels whose pixel number is divisible by 8 and whose line number is divided by 4 or remaining 2 or 3 are shifted in the same manner as in FIG. By this pixel replacement method, as shown in FIG. 1B, the 4: 1: 1 image signal is replaced with a pixel array equivalent to the pixel position of the color difference signal of the 4: 2: 0 image signal.
[0048]
FIGS. 2A and 2B are explanatory diagrams of the third and fourth embodiments of the image compression coding method according to the present invention, both of which are chrominance signal 411/420 pixel replacement methods in a frame picture. Show. FIGS. 2A and 2B both show the pixel positions of the 4: 1: 1 image signal on the screen. The white squares indicate the luminance signal pixels, the black circles indicate the color difference signal pixels, and gray. These circles indicate the pixels of the color difference signal replaced with the pixel positions where the pixels of the color difference signal do not exist.
[0049]
In the color difference signal pixel replacement method of FIG. 2A, when the pixel number counted from 0 is divisible by 4, the line number is divided by 2, and the pixel of the color difference signal at the 1 or 3 remaining pixel position moves to the right. In this method, the pixel is shifted by 2 pixels and shifted upward by 1 pixel (however, pixel number 0 is divisible by 4). Accordingly, for example, the second color difference signal in the vertical direction (line number 1) at pixel number 0 is divided by 4 and divided by 2 as indicated by an arrow 26 in FIG. Since there is one more pixel position, it is shifted by two pixels to the right and one pixel upward and shifted to the pixel position of pixel number 2 and line number 0. Similarly, for example, the pixel number 0 and the fourth color difference signal (line number 3) in the vertical direction are divisible by 4 and the line number is divided by 2, as indicated by an arrow 27 in FIG. Therefore, the pixel position is shifted by two pixels to the right and one pixel upward, and shifted to the pixel position of pixel number 2 and line number 2.
[0050]
With this pixel replacement method, each pixel of the color difference signal existing in the line number that is 1 or 3 remaining when divided by 4 is divided by 4, and each pixel of the color difference signal that is present in the remaining line number when divided by 4 is 2 remaining pixel numbers that are 2 The line number that can be divided is shifted to a pixel position where no color difference signal exists, and as a result, a pixel arrangement equivalent to the pixel position of the color difference signal of the 4: 2: 0 image signal is obtained.
[0051]
Further, the color difference signal pixel replacement method in FIG. 2B is a pixel position where the pixel number counted from 0 is at a pixel position divisible by 8, and when the line number is divided by 4, the color difference signal at the remaining pixel position is 1 or 3. The pixel is replaced so that it is shifted by 2 pixels to the right and by 1 pixel upward, and when the pixel number is divided by 8, there are 4 more pixels of the color difference signal at the pixel position where the line number is divisible by 2. This is a method of shifting right by two pixels and then shifting the next lower pixel to the current position (up) by one pixel (provided that 0 is divisible by 8).
[0052]
Therefore, for example, the first color difference signal in the vertical direction (line number 0) at pixel number 4 is at a pixel position where the pixel number is divided by 8 and there are four remainders, and the line number is divisible by 2. Therefore, FIG. As shown by an arrow 28, the pixel is shifted by 2 pixels to the right and is replaced with the position of line number 0 by pixel number 6 where no chrominance signal pixel originally exists, and subsequently, the pixels of pixel number 4 and line number 0 are changed. The pixel of the color difference signal of the pixel number 4 and the line number 1 that is one lower is shifted as indicated by an arrow 29 in FIG. 2B, and the pixel number 4 of the current position where the color difference signal pixel no longer exists, Shift to the pixel position of line number 0. In addition, the color difference signal pixels at the pixel number divisible by 8 and the line number divided by 2 and remaining one are shifted in the same manner as in FIG. With this pixel replacement method, as shown in FIG. 2B, the color difference signal of the 4: 1: 1 image signal becomes a pixel array equivalent to the pixel position of the color difference signal of the 4: 2: 0 image signal.
[0053]
Next, a method of converting a color difference signal of 4: 2: 0 component format into a color difference signal of 4: 1: 1 component format (hereinafter referred to as “color difference signal 420/411 pixel replacement method”) is specifically described. explain. The chrominance signal 420/411 pixel replacement method performs the reverse operation of the chrominance signal 411/420 pixel replacement method. This method is also broadly divided into a method of replacing within a frame picture and a method of replacing within a field picture. There are two types.
[0054]
FIGS. 3A and 3B are explanatory diagrams of the first and second embodiments of the image compression decoding method according to the present invention, both of which are methods for replacing the color difference signal 420/411 pixel in the field picture. Show. 3 (A) and 3 (B) both show pixel positions of the 4: 2: 0 image signal on the screen, white squares indicate luminance signal pixels, black circles indicate color difference signal pixels, and gray Indicates a pixel of the color difference signal after replacement. The actual 4: 2: 0 image signal has 360 pixels (0 to 359 in the horizontal direction) × 480 lines (0 to 479 in the vertical direction), for example, on the screen. This represents an 8-line portion (the same applies to FIG. 4 described later).
[0055]
The color difference signal pixel replacement method of FIG. 3A is when the pixel number counted from 0 is divided by 4 and left by two, and the pixel of the color difference signal in the first field whose line number is divisible by 4 is 2 to the left. When the pixel number is divided by 4 and left by 2 and the line number is divided by 4 and left by 1 and left by 2 pixels, the pixel number is shifted down by 2 pixels and down. This is a method of shifting by two pixels (however, pixel number 0 is divisible by 4).
[0056]
Therefore, for example, the first color difference signal in the vertical direction (line number 0) at pixel number 2 is at a pixel position where the pixel number is divided by 4 and the remainder is 2 and the line number is divisible by 4. FIG. As indicated by an arrow 31, the pixel is shifted to the left by two pixels and downward by two pixels to be shifted to the pixel position of pixel number 0 and line number 2. Similarly, for example, the color difference signal of pixel number 2 in the second vertical direction (line number 1) has a pixel number divided by 4 to give two remainders as shown by an arrow 32 in FIG. Since it is at the remaining pixel position divided by 4, it is shifted by 2 pixels to the left and by 2 pixels downward and shifted to the pixel position of pixel number 0 and line number 3. With this pixel replacement method, the color difference signal of the 4: 2: 0 image signal replaced by the method of FIG. 1A is replaced with a pixel array equivalent to the pixel position of the color difference signal of the 4: 1: 1 image signal. .
[0057]
3B, when the pixel number counted from 0 is divided by 8 and 2 is left, and the line number is divisible by 4, 2 pixels to the left and 2 pixels to the bottom. When the pixel number is divided by 8 and left to be 4 and the line number is divisible by 4, it is shifted down by 2 pixels, and then the pixel on the 2 right is the current Shift one pixel to the position (to the left). Further, when the pixel number is divided by 8 and 2 is left, and when the line number is divided by 4 and 1 is left, the pixel is shifted to the left by 2 pixels and down by 2 pixels. Also, when the pixel number is divided by 8 and there are four remainders, and when the line number is divided by 4 and one remainder, the lower pixel is shifted by 2 pixels, and then the pixel on the right is moved to the current position ( (To the left) This is a method of shifting by two pixels (provided that 0 is divisible by 4).
[0058]
Therefore, for example, the first color difference signal in the vertical direction (line number 0) at pixel number 4 is at a pixel position where the pixel number is divided by 8 and left by 4 and the line number is divisible by 4, so FIG. As indicated by an arrow 33, the pixel is shifted by 2 pixels to the lower position and replaced with the position of line number 2 by pixel number 4 where no color difference signal pixel originally exists, and subsequently, the pixels of pixel number 4 and line number 0 are changed. The pixel of the chrominance signal of pixel number 6 and line number 0 to the right two is shifted as shown by the arrow 34 in FIG. 3B, and the pixel number 4 at the current position where the chrominance signal pixel no longer exists, Shift to the pixel position of line number 0. Further, the chrominance signal pixel whose pixel number is divided by 8 to be 2 and the line number is divisible by 4 or is left is shifted in the same manner as in FIG. By this pixel replacement method, the color difference signal of the 4: 2: 0 image signal in which the pixel is replaced by the method of FIG. 1B is changed to the color difference of the 4: 1: 1 image signal as shown in FIG. It is replaced with a pixel array equivalent to the pixel position of the signal.
[0059]
FIGS. 4A and 4B are explanatory diagrams of the third and fourth embodiments of the image compression decoding method according to the present invention, both of which are the color difference signal 420/411 pixel replacement methods in the frame picture. Show. 4A and 4B show the pixel positions of the 4: 2: 0 image signal on the screen, and the white squares indicate the pixels of the luminance signal, the black circles indicate the pixels of the color difference signal, and gray Indicates a pixel of the color difference signal after replacement.
[0060]
The color difference signal pixel replacement method in FIG. 4A is when the pixel number counted from 0 is divided by 4 and left by 2 and when the line number is divisible by 2, 2 pixels to the left and 1 pixel to the bottom. This is a method of replacement so as to shift (however, pixel number 0 is divisible by 2).
[0061]
Accordingly, for example, the first color difference signal in the vertical direction (line number 0) at pixel number 2 is at a pixel position where the pixel number is divided by 4 and the remainder is 2 and the line number is divisible by 2, so that FIG. As indicated by an arrow 36, the pixel position is shifted to the left by two pixels and shifted downward by one pixel to be shifted to the pixel position of pixel number 0 and line number 1. Similarly, for example, the color difference signal of pixel number 2 in the third vertical direction (line number 2) is at a pixel position where the pixel number is divided by 4 and the remainder is 2 and the line number is divisible by 2. FIG. ) Is shifted to the left by two pixels and downward by one pixel, and is shifted to the pixel position of pixel number 0 and line number 3. By this pixel replacement method, the color difference signal of the 4: 2: 0 image signal in which the pixel is replaced by the method of FIG. 2A becomes a pixel array equivalent to the pixel position of the color difference signal of the 4: 1: 1 image signal. Replaced.
[0062]
Also, the color difference signal pixel replacement method of FIG. 4B is when the pixel number counted from 0 is divided by 8 and 2 is left, and when the line number is divisible by 2, 2 pixels to the left and 1 to the bottom. If the pixel number is divided by 8 and the remainder is 4 and the line number is divisible by 2, the pixel number is shifted to the bottom by one pixel, and then the pixel on the two right is This is a method of shifting two pixels to the current position (to the left) (provided that 0 is divisible by 2).
[0063]
Therefore, for example, the color difference signal of pixel number 4 in the first vertical direction (line number 0) is at a pixel position where the pixel number is divided by 8 to be 4 and the line number is divisible by 2, so that FIG. As indicated by an arrow 38, the pixel is shifted by one pixel to the bottom and replaced with the position of line number 1 by pixel number 4 where no color difference signal pixel originally exists, and subsequently, the pixels of pixel number 4 and line number 0 are replaced. The pixel of the chrominance signal with pixel number 6 and line number 0 two right is shifted as indicated by the arrow 39 in FIG. 4B, and the pixel number 4 at the current position where the chrominance signal pixel no longer exists, Shift to the pixel position of line number 0. Further, the color difference signal pixels at the positions where the pixel number is divided by 8 to be two and the line number is divisible by 2 are shifted in the same manner as in FIG. By this pixel replacement method, the color difference signal of the 4: 2: 0 image signal in which the pixel is replaced by the method of FIG. 2B is changed to the color difference of the 4: 1: 1 image signal as shown in FIG. 4B. The pixel arrangement is equivalent to the pixel position of the signal.
[0064]
In the above description, for the sake of convenience, it has been described that the color difference signal 411/420 pixel replacement is performed in the image compression encoding and the color difference signal 420/411 pixel replacement is performed in the image compression decoding. Of course, the color difference signal 411/420 pixels may be replaced by compression decoding, and the color difference signal 420/411 pixels may be replaced by image compression encoding.
[0065]
Next, embodiments of the compression encoding apparatus and compression decoding apparatus according to the present invention will be described with reference to the drawings. FIG. 5 shows a block diagram of a first embodiment of a compression coding apparatus according to the present invention. In the figure, the 4: 1: 1 image signal is input to the color difference signal 411/420 format converter 41 and the color difference signal 411/420 pixel replacement unit 42, respectively.
[0066]
The color difference signal 411/420 format converter 41 oversamples the horizontal direction of the input 4: 1: 1 image signal to double the number of horizontal pixels. In addition, in the vertical direction, filtering is performed to halve the band, and then the number of sampling is halved. The chrominance signal 411/420 pixel replacement unit 42 is set to any one of the chrominance signal 411/420 pixel replacement methods described in conjunction with FIGS. 1 (A) and 1 (B) and FIGS. 2 (A) and 2 (B). Alternatively, the color difference signal 411/420 pixels are replaced by a predetermined method.
[0067]
The color difference signal 411/420 pixel replacement unit 42 separates the color difference signal from, for example, the input 4: 1: 1 image signal, and sequentially writes it into the memory under the control of the address controller. The stored color difference signal data is read from the memory and written again in the memory in the order of addresses so as to realize the color difference signal 411/420 pixel replacement method described above, and the color difference signal data read from the memory is input 4: 1: 1. The luminance signal data separated from the image signal is combined and output.
[0068]
The output image signals of the color difference signal 411/420 format converter 41 and the color difference signal 411/420 pixel replacer 42 are supplied to a switch circuit (S / W) 44, respectively, where a switching signal from the user interface 43 is used. The image signal of the method specified by the user is selected. In this case, when decoding using the decoding apparatus of the present invention described later, the 4: 2: 0 component format equivalent from the color difference signal 411/420 pixel replacer 42 is set to 4: 2: The S / W 44 is switched so that the 0 image signal is selected. On the other hand, when a digital signal compressed and encoded by an existing decoding apparatus that does not have a pixel replacement unit according to the present invention is decoded, a 4: 4 band extracted from the chrominance signal 411/420 format converter 41 is used. The S / W 44 is switched so that an image signal in the 1: 0 component format is selected.
[0069]
The image signal selected by the S / W 44 is input to the MPEG encoder 45, and is compressed and encoded in accordance with the 4: 2: 0 component format MPEG system. The MPEG encoder 45 has a known configuration shown in FIG. The input image signal of the MPEG encoder 45 is originally in a 4: 1: 1 component format. However, when the pixel replacement is performed by the color difference signal 411/420 pixel replacer 42, it is as if the color difference signal is 4. : 2: 0 component format color difference signal is converted as if it were a 4: 2: 0 image signal by encoding as in the prior art. It is taken out. The bit stream encoded and extracted by the MPEG encoder 45 is supplied to the buffer memory and formatter 46.
[0070]
On the other hand, the synchronization control unit 47 has a timer for a predetermined period, and outputs a trigger signal to the synchronization signal generator 48 every predetermined period. This predetermined period is selected, for example, 1 / 29.97 seconds, which is one frame period of the NTSC image signal. When the synchronization signal generation unit 48 detects the input of the trigger signal from the synchronization control unit 47, the synchronization signal generation unit 48 is set as an area where data unrelated to video and audio can be embedded in the first MPEG syntax, such as user_data. A fixed pattern synchronization signal that can be uniquely identified in advance and that indicates the presence of the identification code of the present embodiment in a predetermined area is generated.
[0071]
For example, the video stream picture layer of MPEG1 is defined as shown in FIG. 12, and after transmitting a user data start code (user_data_start_code) before the slice layer, user data (user_data) is transmitted in units of 8 bits. A mechanism that can do this is defined. Also, in the transport stream system layer such as MPEG2, as shown in FIG. 13, when the transport private data flag (transport_private_data_flag) is set to 1, it can be clearly indicated that private data (private_data) exists. The private data (private_data) of the data length set in the transport private data length (transport_private_data_length) can be transmitted under the restriction that the data length also does not protrude from the transport packet. ing.
[0072]
In addition to this, there are several mechanisms for transmitting user-specific data in the MPEG system, such as transmitting by declaring a dedicated packet by setting a private stream (private_stream) in the stream id (stream_id). In the present embodiment, any mechanism may be used for the purpose of multiplexing and transmitting the synchronization signal and the identification code to the encoded data.
[0073]
Here, assuming that user_data of MPEG1 video is used, user_data_start_code is defined as 0x000001B2 before the slice layer. After the code is sent, a synchronization signal uniquely identifiable in advance indicating the presence of an identification code to be described later is transmitted in the user data area. This synchronization signal is set to a value (for example, 0x0f0f0f0f2428fdaa) determined separately from the MPEG standard, and has the purpose of allowing the decoding device side to specify the position where an identification code described later is multiplexed after this synchronization signal. . Specifically, the generation timing of the synchronization signal is determined by observing the MPEG stream input from the MPEG encoder 45 to the buffer memory and the formatter 46 while specifying a predetermined user_data area, and then setting user_data_start_code. It occurs after declaring that it is a user_data area.
[0074]
The synchronization signal generator 48 shown in FIG. 5 also outputs a signal indicating that a synchronization signal has been generated to the identification code generator 49. The identification code generator 49 determines whether the input image signal of the buffer memory and the formatter 46 is processed by the color difference signal 411/420 format converter 41 at the timing after the synchronization signal, or the color difference signal 411/420 pixel replacement unit. An identification code having a value indicating whether it has been processed in 42 is generated and supplied to the buffer memory and formatter 46.
[0075]
The value of the identification code is determined based on a signal input from the user interface 43 via the synchronization signal generator 48. For example, 1 byte is prepared as an identification code, and if 0x01, the color difference signal 411/420 pixel replacement is performed. 0x00 indicates that the color difference signal 411/420 format converter 41 has processed. The buffer memory and formatter 46 outputs to the data recording unit 50 the compressed encoded image data in which data is arranged in accordance with the format of the recording medium 51, and the digital signal composed of the synchronization signal and the identification code. Recording is performed on an optical disk or other recording medium 51.
[0076]
The color difference signal 411/420 pixel replacement unit 42 can select two or more types of color difference signal 411/420 pixel replacement methods from among the four types of color difference signal 411/420 pixel replacement methods shown in FIG. 1 and FIG. In the case of the configuration, information indicating which type of color difference signal 411/420 pixel replacement method is used is also included in the identification code.
[0077]
Next, a preferred decoding apparatus example for realizing the present invention will be described. FIG. 6 shows a block diagram of the first embodiment of the compression decoding apparatus according to the present invention. In the figure, a digital signal read from the recording medium 51 recorded by the image compression encoding apparatus of FIG. 5 is temporarily stored in the buffer memory 52, and then a synchronization signal detection unit 53, an identification code detection unit 54, and Each is supplied to the MPEG decoder 55. The synchronization signal detection unit 53 identifies a predetermined user_data area while observing the MPEG stream, detects the user_data_start_code after that, and confirms the user_data area, and then detects a fixed pattern synchronization signal. .
[0078]
The identification code detection unit 54 receives the trigger signal output from the synchronization signal detection unit 53 when the synchronization signal is detected, and detects the identification code multiplexed in a predetermined area in synchronization with the synchronization signal from the reproduced digital signal. And transmitted to the color difference signal generation method determination unit 56. The color difference signal generation method determination unit 56 determines whether the reproduced digital signal is processed by the color difference signal 411/420 format converter 41 or the color difference signal 411/420 pixel replacer 42 in FIG. Judgment is made from the input identification code, and switching control of the switch circuit (S / W) 57 is performed according to the judgment result, and the judgment result is supplied to the color difference signal output format converter 59.
[0079]
That is, if the color difference signal generation method determination unit 56 determines that the color difference signal 411/420 pixel replacement unit 42 has processed the image, the image decoded by the MPEG decoder 55 having a known configuration shown in FIG. When the data is supplied to the color difference signal 420/411 pixel replacement unit 58 and is determined to have been processed by the color difference signal 411/420 format converter 41, the image data decoded by the MPEG decoder 55 is converted into the image data. The switching circuit (S / W) 57 is subjected to switching control so as to be directly supplied to the color difference signal output format converter 59.
[0080]
The color difference signal 420/411 pixel replacement unit 58 records on the recording medium 51 in the color difference signal 420/411 pixel replacement method described in conjunction with FIGS. 3 (A) and 3 (B) and FIGS. 4 (A) and 4 (B). The color difference signal 420/411 pixel replacement is performed by a predetermined method corresponding to the color difference signal 411/420 pixel replacement method employed in the received image signal, and the original image signal input to the compression encoding device Returned to the same 4: 1: 1 image signal and output.
[0081]
That is, the color difference signal 411/420 pixel replacer 42 of the compression encoding device of FIG. 5 makes the color difference signal of the 4: 1: 1 image signal equivalent to the color difference signal array of the 4: 2: 0 image signal. The decoded signal of the signal encoded by the MPEG encoder 45 is converted into an MPEG decoder as if the color difference signal in the 4: 1: 1 image signal was a color difference signal in the 4: 2: 0 image signal. In the case of the output from 55, the color difference signal of the decoded signal is different from the phase of the original image, so that it is opposite to the color difference signal 411/420 pixel replacement method performed by the color difference signal 411/420 pixel replacer 42. It is necessary to replace the color difference signal 420/411 pixels and restore the phase of the color difference signal.
[0082]
Therefore, when the color difference signal 411/420 pixel replacement unit 42 performs the color difference signal 411/420 pixel replacement method shown in FIG. 1A, the color difference signal 420/411 pixel replacement unit 58 uses the color difference signal shown in FIG. It is necessary to perform the signal 420/411 pixel replacement method. Similarly, when the color difference signal 411/420 pixel replacement method is shown in FIG. 1B, FIG. 2A, and FIG. 3 (B), FIG. 4 (A), and FIG. 4 (B) need to perform the color difference signal 411/420 pixel replacement method.
[0083]
The color difference signal output format converter 59 is based on the determination result from the color difference signal generation method determination unit 56, which is the same as the image signal input to the compression encoding device by the color difference signal 420/411 pixel replacement unit 4: 1: When it is detected that the image signal returned to one image signal has been input, the 4: 1: 1 image signal is oversampled twice in the horizontal direction and converted into a 4: 2: 2 image signal and output. To do. On the other hand, when the color difference signal output format converter 59 detects that the image signal output from the MPEG decoder 55 is input as it is based on the determination result from the color difference signal generation method determination unit 56, the image signal Is oversampled twice in the vertical direction and output.
[0084]
Therefore, if the image signal directly input from the MPEG decoder 55 through the switch circuit 57 to the color difference signal output format converter 59 is a 4: 2: 0 image signal, the color difference signal is output as a 4: 2: 2 image signal. If the 4: 1: 0 image signal is output from the format converter 59 and is a 4: 1: 0 image signal obtained by decoding an MPEG encoded signal of the 4: 1: 1 image signal by the MPEG decoder 55, the 4: 1: 0 image signal. Will be output as is.
[0085]
Next, a second embodiment of the device of the present invention will be described. FIG. 7 shows a block diagram of a second embodiment of the compression coding apparatus according to the present invention. In the figure, the same components as those in FIG. In the first embodiment described above, it has been described that the replacement of the color difference signal pixels in the frame picture or the replacement of the color difference signal pixels in the field picture is fixedly performed in the entire screen. Then, the frame / field correlation determination unit 61 is used to switch automatically.
[0086]
That is, the 4: 1: 1 image signal is supplied to the frame / field correlation determination unit 61 in FIG. 7, where the frame / field correlation is determined for each predetermined area, and the determination result that the field correlation is large 1 is obtained, the pixels of the color difference signal are replaced in the field picture as shown in FIGS. 1A and 1B, and when the determination result that the frame correlation is large is obtained, FIG. As shown in (B), the color difference signal 411/420 pixel rewriter 42 is controlled in accordance with the determination result so as to replace the pixel of the color difference signal in the frame picture, and is converted into a 4: 2: 0 image signal.
[0087]
Also, the determination result from the frame / field correlation determining unit 61 is supplied to the identification code generating unit 62, and whether the pixel of the chrominance signal is replaced in the field picture or whether the pixel of the chrominance signal is replaced in the frame picture. It is specified using the other bit inside.
[0088]
The correlation determination by the frame / field correlation determination unit 61 is performed by, for example, a first difference sum (1) between the image data of one line and the image data of one line below, as indicated by 82 in the image data of the specific area 81 in FIG. (The value obtained by adding the difference values between the upper and lower pixels for one line of pixels), and similarly calculating the second difference sum with the pixels below the two lines as indicated by 83. A method of determining that the correlation is larger when the difference sum is smaller is conceivable. In FIG. 18, the first difference sum indicates the magnitude of the frame correlation, and the second difference sum indicates the magnitude of the field correlation.
[0089]
FIG. 8 shows a block diagram of a third embodiment of a compression coding apparatus according to the present invention. In the figure, the same components as those in FIGS. 5 and 7 are denoted by the same reference numerals, and the description thereof is omitted. In this embodiment, the determination result of the frame / field correlation determining unit 61 in the second embodiment of FIG. 7 is supplied to the MPEG encoder 64 instead of the identification code generating unit 62 (49) and used in MPEG. A frame / field correlation determination result is used for a macroblock (16 pixels × 16 lines). The MPEG encoder 64 receives this frame / field determination result, and the DCT unit 3 shown in FIG. 15 performs either of the frame / field based on the input result, and the motion compensation predictor 1 performs the frame / field determination. Either motion compensation prediction is performed.
[0090]
Thus, in this embodiment, after the replacement, the replacement method of the color difference signal pixels is performed on the picture in the frame or field in accordance with the determination result of determining the magnitude of the frame or field correlation in units of macroblocks. This increases the vertical correlation of the pictures and increases the encoding efficiency of MPEG compression. Accordingly, when encoding is performed at the same encoding rate, it is possible to encode with relatively high image quality.
[0091]
The determination result is a mode indicating whether the DCT which is the macroblock layer of the MPEG2 video syntax shown in FIG. 19 is performed in a frame / field (dct_type1 bit 0 in the MPEG2 video coding syntax is a frame, 1 ) Is a field), or frame / field motion compensation mode (the frame_motion_type2 bit in the MPEG2 video coding syntax is a field when the bit is 01, and a frame is when the frame is 10). You can do it.
[0092]
FIG. 9 shows a block diagram of a second embodiment of the compression decoding apparatus according to the present invention. In the figure, the same components as those in FIG. This second embodiment is an embodiment corresponding to the compression encoding apparatus shown in FIG. 8, and the reproduced digital signal from the buffer memory 52 is supplied to the MPEG decoder 68 for decoding. The DCT mode signal at that time is supplied to the color difference signal 420/411 pixel replacement unit 69.
[0093]
When the DCT mode is the field mode, the chrominance signal 420/411 pixel replacer 69 has a large field correlation, and the input image signal is considered to have undergone chrominance signal 411/420 pixel rewriting within the field picture. As shown in FIG. 3 (A) or FIG. 3 (B), when the color difference signal 420/411 pixel replacement is performed in the field picture and the DCT mode is the frame mode, the frame correlation is large, and the color difference in the frame picture Since it is considered that the signal 411/420 pixel rewriting is performed, the color difference signal 420/411 pixel replacement is performed in the frame picture as shown in FIG. 4 (A) or FIG. 4 (B).
[0094]
FIG. 10 shows a block diagram of a fourth embodiment of a compression coding apparatus according to the present invention. In the figure, the same components as those in FIG. In this embodiment, the color difference signal in the 4: 2: 0 image signal is replaced with the 4: 1: 1 image signal and encoded. That is, in FIG. 10, the 4: 2: 0 image signal is supplied to the color difference signal 420/411 format converter 71 and the color difference signal 420/411 pixel replacement unit 72, respectively.
[0095]
The chrominance signal 420/411 format converter 71 halves the number of horizontal pixels of the input 4: 2: 0 image signal and oversamples it twice in the vertical direction. The chrominance signal 420/411 pixel replacement unit 72 is arbitrarily set from among the chrominance signal 420/411 pixel replacement methods described in conjunction with FIGS. 3 (A) and 3 (B) and FIGS. 4 (A) and 4 (B). The color difference signal 420/411 pixel replacement is performed by a predetermined method.
[0096]
The output image signals of the color difference signal 420/411 format converter 71 and the color difference signal 420/411 pixel replacer 72 are supplied to a switch circuit (S / W) 44, respectively, where the output image signals are supplied by switching signals from the user interface 43. The image signal of the method specified by the user is selected. The 4: 1: 1 image signal selected by the switch circuit 44 is supplied to the DVC encoder 73, where it is encoded in a 4: 1: 1 component format compliant with the DVC standard, and then buffered. Supplied to memory and formatter 74. In this embodiment, when MPEG data is encoded as a DVC format, it can be reproduced while retaining the same horizontal band characteristics as the 4: 2: 0 component format.
[0097]
FIG. 11 shows a block diagram of a third embodiment of the compression decoding apparatus according to the present invention. In the figure, the same components as those in FIG. This third embodiment is an embodiment corresponding to the compression encoding apparatus shown in FIG. 10, and a 4: 1: 1 image signal taken out from the buffer memory 52 is decoded by the DVC decoder 76. It is supplied to the rear switch circuit 57.
[0098]
The color difference signal 411/420 pixel replacement unit 77 explained the 4: 1: 1 image signal input from the switch circuit 57 together with FIGS. 1A, 1B, 2A, and 2B. Of the color difference signal 411/420 pixel replacement methods, the color difference signal 411/420 pixel replacement is performed by one method corresponding to the encoding apparatus side.
[0099]
【The invention's effect】
As described above, according to the present invention, the input image signal in the second component format is a 4: 1: 1 component format such as DVC, and this is re-converted into a 4: 2: 0 component format such as MPEG. When re-encoding, the bandwidth of the MPEG output image signal is the same as that of the 4: 1: 1 component format without equivalently having the same bandwidth as the image having only the 4: 1: 0 signal pixel count. It is possible to reproduce while maintaining almost the band characteristics of the direction.
[0100]
Further, according to the present invention, when the 4: 2: 0 component format such as MPEG is changed and re-encoded to an application using the DVC encoding of the 4: 1: 1 component format, it is equivalently 4: 1: The DVC output image signal can be reproduced while maintaining the same horizontal band characteristics as the 4: 2: 0 component format without becoming the same band as the image having only the number of pixels of the 0 image signal. it can.
[0101]
Furthermore, according to the present invention, since pixel replacement of the color difference signal is performed in a picture having a larger correlation between frame correlation and field correlation, compression coding efficiency can be improved.
[Brief description of the drawings]
FIG. 1 is an explanatory diagram of a chrominance signal pixel replacement method in first and second embodiments of an image compression encoding method according to the present invention.
FIG. 2 is an explanatory diagram of a chrominance signal pixel replacement method in the third and fourth embodiments of the image compression encoding method of the present invention.
FIG. 3 is an explanatory diagram of a chrominance signal pixel replacement method in the first and second embodiments of the image compression decoding method of the present invention;
FIG. 4 is an explanatory diagram of color difference signal pixel replacement methods in the third and fourth embodiments of the image compression decoding method of the present invention.
FIG. 5 is a block diagram of a first embodiment of a compression encoding apparatus of the present invention.
FIG. 6 is a block diagram of a first embodiment of a compression decoding apparatus of the present invention.
FIG. 7 is a block diagram of a second embodiment of the compression encoding apparatus of the present invention.
FIG. 8 is a block diagram of a third embodiment of the compression encoding apparatus of the present invention.
FIG. 9 is a block diagram of a second embodiment of the compression decoding apparatus of the present invention.
FIG. 10 is a block diagram of a fourth embodiment of the compression encoding apparatus of the present invention.
FIG. 11 is a block diagram of a third embodiment of the compression decoding apparatus of the present invention.
FIG. 12 is a table showing a part of an MPEG1 video stream picture layer.
FIG. 13 is a table showing a part of the system layer of an MPEG2 transport stream.
FIG. 14 is a diagram illustrating pixel positions of a luminance signal and a color difference signal of a 4: 2: 2 image signal, an MPEG2 image signal, an MPEG1 image signal, and a 4: 1: 1 image signal.
FIG. 15 is a block diagram of an example of an MPEG encoder.
FIG. 16 is a block diagram of an example of an MPEG decoder.
FIG. 17 is a conceptual diagram of the number of pixels of a 4: 2: 2 image signal, a 4: 2: 0 image signal, a 4: 1: 1 image signal, and a 4: 1: 0 image signal.
FIG. 18 is an explanatory diagram of determination of frame correlation and feed correlation.
FIG. 19 is a diagram illustrating a macroblock layer of MPEG2 video syntax.
[Explanation of symbols]
41 Color difference signal 411/420 format converter
42 color difference signal 411/420 pixel replacer
43 User interface
44, 57 Switch circuit (S / W)
45, 64 MPEG encoder
46, 74 Buffer memory and formatter
47 Synchronization control unit
48 Sync signal generator
49, 62 Identification code generator
50 Data recording part
51 recording media
52 Buffer memory
53 Sync signal detector
54 Identification code detector
55, 68 MPEG decoder
56 Color Difference Signal Generation Method Discrimination Unit
58 Color Difference Signal 420/411 Pixel Replacer
59, 78 Color difference signal output format converter
61 Frame / field correlation determination unit
71 Color difference signal 420/411 format converter
72 Color Difference Signal 420/411 Pixel Replacer
73 DVC encoder
77 Color Difference Signal 411/420 Pixel Replacer

Claims (5)

When an input image signal of a component format of 4: 1: 1 is displayed on one screen, 4 adjacent pixels among the pixels of the color difference signal in the input image signal arranged in the horizontal direction in a 4-pixel cycle for each line. When the line is a set, the pixels of the color difference signal on the second or third line in each set are shifted to the right by 2 pixels and vertically to a position on 1 or 2 pixels. By shifting the pixels of the color difference signal on the line of 2 to the right and the position of 1 or 2 pixels in the vertical direction, the color difference signal in the image signal of the 4: 2: 0 component format is obtained. A first step of performing pixel replacement shifted to a pixel position equivalent to a pixel position displayed on one screen;
  A second step of performing compression encoding by an encoding device that compresses and encodes the pixel-replaced image signal in a 4: 2: 0 component format;
  A third step of decoding the compression-coded image signal by a decoding device that decodes the image signal in the 4: 2: 0 component format to obtain a decoded signal;
  When the decoded signal is displayed on one screen, a pixel shift opposite to the pixel shift in the first step is performed on the pixels of the color difference signal arranged in the horizontal direction in the decoded signal at a cycle of two pixels. A fourth step of performing pixel replacement in which the color difference signal in the input image signal of the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen;
  The compression decoding method characterized by including this. When an input image signal of a component format of 4: 1: 1 is displayed on one screen, when a total of 32 adjacent pixels of 8 pixels in the horizontal direction and 4 pixels in the vertical direction are combined as a set, a cycle of 4 pixels for each line. Among the color difference signal pixels in the input image signal that are arranged in the horizontal direction and there are eight for each set, the four color difference signal pixels adjacent in the vertical direction are on the second or third line. The pixel of the chrominance signal at 2 is shifted to the right by 2 pixels and the position of 1 or 2 pixels in the vertical direction, the pixel of the chrominance signal on the fourth line is 2 pixels to the right, and For the remaining four pixels that are vertically shifted to a position on one or two pixels and that are adjacent in the vertical direction, the color difference signal pixel on the first line and the color difference on the third or second line Signal pixels are 2 pixels to the right By shifting each pixel of the color difference signal on the second or third line and the fourth line in the vertical direction to the pixel position of the color difference signal after the shift and for each set. A first step of performing pixel replacement in which a color difference signal in an image signal of a 4: 2: 0 component format is shifted to a pixel position equivalent to a pixel position displayed on one screen;
  A second step of performing compression encoding by an encoding device that compresses and encodes the pixel-replaced image signal in a 4: 2: 0 component format;
  A third step of decoding the compression-coded image signal by a decoding device that decodes the image signal in the 4: 2: 0 component format to obtain a decoded signal;
  When the decoded signal is displayed on one screen, a pixel shift opposite to the pixel shift in the first step is performed on the pixels of the color difference signal arranged in the horizontal direction in the decoded signal at a cycle of two pixels. A fourth step of performing pixel replacement in which the color difference signal in the input image signal of the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen;
  The compression decoding method characterized by including this. When an input image signal of a component format of 4: 1: 1 is displayed on one screen, 4 adjacent pixels among the pixels of the color difference signal in the input image signal arranged in the horizontal direction in a 4-pixel cycle for each line. When the line is a set, the pixels of the color difference signal on the second or third line in each set are shifted to the right by 2 pixels and vertically to a position on 1 or 2 pixels. The pixel of the color difference signal on the line is 2 pixels to the right and 1 or 2 pixels in the vertical direction By shifting to the upper position, the first pixel replacement is performed in which the color difference signal in the 4: 2: 0 component format image signal is shifted to a pixel position equivalent to the pixel position displayed on one screen. A pixel replacer;
  An encoding device that compresses and encodes an image signal that has undergone pixel replacement by the first pixel replacer in the 4: 2: 0 component format;
  A decoder that receives a compression-encoded image signal that has been compression-encoded by the encoding device, decodes it in the 4: 2: 0 component format, and outputs a decoded signal;
  When the decoded signal is displayed on one screen, a pixel shift opposite to the pixel shift in the first step is performed on the pixels of the color difference signal arranged in the horizontal direction in the decoded signal at a cycle of two pixels. Thus, the color difference signal in the input image signal of the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen, and is input to the encoding device. A second pixel replacer for outputting the same image signal as the image signal;
  The compression decoding apparatus characterized by having.   When an input image signal of a component format of 4: 1: 1 is displayed on one screen, when a total of 32 adjacent pixels of 8 pixels in the horizontal direction and 4 pixels in the vertical direction are combined as a set, a cycle of 4 pixels for each line. Among the color difference signal pixels in the input image signal that are arranged in the horizontal direction and there are eight for each set, the four color difference signal pixels adjacent in the vertical direction are on the second or third line. The pixel of the chrominance signal at 2 is shifted to the right by 2 pixels and the position of 1 or 2 pixels in the vertical direction, the pixel of the chrominance signal on the fourth line is 2 pixels to the right, and For the remaining four pixels that are vertically shifted to a position on one or two pixels and that are adjacent in the vertical direction, the color difference signal pixel on the first line and the color difference on the third or second line Signal pixels are 2 pixels to the right By shifting each pixel of the color difference signal on the second or third line and the fourth line in the vertical direction to the pixel position of the color difference signal after the shift and for each set. A first pixel replacer that performs pixel replacement in which a color difference signal in an image signal in a 4: 2: 0 component format is shifted to a pixel position equivalent to a pixel position displayed on one screen;
  An encoding device that compresses and encodes an image signal that has undergone pixel replacement by the first pixel replacer in the 4: 2: 0 component format;
  A decoder that receives a compression-encoded image signal that has been compression-encoded by the encoding device, decodes it in the 4: 2: 0 component format, and outputs a decoded signal;
  When the decoded signal is displayed on one screen, a pixel shift opposite to the pixel shift in the first step is performed on the pixels of the color difference signal arranged in the horizontal direction in the decoded signal at a cycle of two pixels. Thus, the color difference signal in the input image signal of the 4: 1: 1 component format is displayed at the same pixel position as the pixel position displayed on one screen, and is input to the encoding device. A second pixel replacer for outputting the same image signal as the image signal;
  The compression decoding apparatus characterized by having. The first pixel replacement unit includes a first color difference signal storage unit that separates a color difference signal from the input image signal of the 4: 1: 1 component format and sequentially writes the difference signal into a first memory; and the input image signal Among the pixels when the image is displayed on the screen, the position of the color difference signal pixel in the input image signal is the pixel position where the color difference signal in the 4: 2: 0 component format image signal is displayed on the screen A luminance signal from the input image signal in the 4: 1: 1 component format, and a first readout means for reading out the color difference signal from the first color difference signal storage means in the order of read addresses shifted to the pixel position equivalent to A first combining unit that separates and combines the luminance signal and the color difference signal read by the first reading unit;
  The second pixel replacer includes: a second color difference signal storage unit that separates a color difference signal from the decoded signal and sequentially writes it in a second memory; and each pixel when the decoded signal is displayed on a screen. Among them, the position of the pixel of the color difference signal in the decoded signal is the 4: 1: 1 component format. Second reading means for reading out the color difference signals from the second color difference signal storage means in the order of read addresses in which the color difference signals in the input image signal of the mat are shifted to the same pixel positions as the pixel positions displayed on the screen; 5. A second synthesizing unit for separating a luminance signal from the decoded signal and synthesizing the luminance signal and the color difference signal read by the second reading unit. The compression decoding apparatus as described.

Images