JP2006115340A - Image filter circuit and filtering processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform the conversion processing of a high efficiency encoding method and image scaling processing by one and the same circuit. <P>SOLUTION: An image filter circuit is provided with an interpolation operation processing part 13 for generating interpolation image data between image data by performing interpolation operation processing based on a phase determined by the delta of image data from image data of a 4: 2: 2 format, an adding means 14 for adding a prescribed phase shift (0. 75) between color-difference data of first image data and color-difference of second image data to a phase, and a selecting means 15 for selecting an output by the adding means 14 and supplies to the interpolation operation processing part 13 the phase with the phase shift (0. 75) added thereto in accordance with an instruction to convert the first image data into the second image data or convert the second image data into the first image data. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image filter that generates interpolated image data from image data, and more specifically, an image that realizes image scaling processing of 4: 2: 2 format image data and conversion processing of a high-efficiency encoding method with a single circuit. The present invention relates to a filter circuit and a filtering processing method.

  With the spread of terrestrial digital television broadcasting, the technology related to digital video signals has attracted attention. The video signal is composed of a luminance signal Y called a component signal and two color difference signals Cb and Cr.

  In general, human visual characteristics are less sensitive to the color difference signals Cb and Cr than the luminance signal Y. Therefore, the amount of encoded information can be reduced by using a signal obtained by thinning out the color difference signals. Therefore, as the component signal, the color difference signals Cb and Cr in the horizontal direction are ½ of the luminance signal Y with respect to the 4: 4: 4 format prepared without thinning out the color difference signals Cb and Cr with respect to the luminance signal Y. The 4: 2: 2 format is used, and the 4: 2: 0 format is used in which the color difference signals Cb and Cr in the vertical direction are half the size of the luminance signal Y.

  Specifically, in the 4: 2: 2 format, one sample of the color difference signal is arranged for two samples of the luminance signal of each scanning line, and in the 4: 2: 0 format, the luminance signal in the vertical direction. One sample of the color difference signal is arranged for two samples. A digital video signal is generated by digitally encoding such a component signal.

  Digital video signals used in broadcasting, communication, storage media, etc. are generally highly efficient encoding, so-called compression encoding, in order to enable transmission over a limited frequency bandwidth using MPEG (Moving Picture Expert Group) technology. Has been. For example, MPEG is an image-acoustic encoding method for digital storage media having a transmission rate of about 1.5 Mbit / s such as CD-ROM (Compact Disk-Read Only Memory) and HD (Hard Disk). MPEG-2 system for high-quality image data that can be used for general purposes such as 1 system, digital broadcasting, storage media, communication, and transmission at a high bit rate has been standardized and widely spread is doing.

  By the way, in MPEG-2, three formats of 4: 2: 0, 4: 2: 2, and 4: 4: 4 can be used as the luminance signal (Y) / chrominance signal (Cr, Cb) format. In particular, the 4: 2: 2 format is defined as ITU-R Recommendation 601 in the International Telecommunications Union (ITU) radio communication sector (ITU-R) and is often used for business purposes in studios and the like. The format is higher in quality than the 4: 2: 0 format. The MPEG-2 system is mainly used for high-efficiency encoding of television signals. In MPEG-1, only 4: 2: 0 format images can be handled.

  By the way, high-efficiency encoding of digital video signals by MPEG-2 method is mainly used to output high-quality video. As a video signal processing apparatus, MPEG-2 in 4: 2: 2 format is used. The circuit configuration corresponds to a digital video signal that is highly efficient encoded by two methods.

  However, since video content recorded on a CD-ROM or HD is highly efficient encoded by the MPEG-1 system, a digital video signal that has been encoded by the MPEG-1 system is processed separately. Therefore, the circuit for this will be needed and the circuit scale will increase. Therefore, the conventional video signal processing apparatus converts the MPEG-2 system digital video signal in the 4: 2: 0 format into the 4: 2: 2 format, and further converts the MPEG1 system into the MPEG-2 system. The conversion processing circuit is provided.

  Specifically, when a 4: 2: 0 format MPEG-1 digital video signal is decoded, luminance data Y and color difference data Cb and Cr arranged on the xy plane are as shown in FIG. This is first converted into 4: 2: 2 format as shown in FIG. When a 4: 2: 2 format MPEG-2 digital video signal is decoded, as shown in FIG. 14, the phase relationship is different from the phase relationship between the luminance data Y and the color difference data Cr and Cb shown in FIG. End up.

  Therefore, the phase relationship between the luminance data Y as shown in FIG. 13 and the color difference data Cb, Cr is represented by the luminance data Y obtained by decoding the MPEG-2 digital video signal as shown in FIG. It is necessary to align with the phase relationship with Cr. Therefore, in the MPEG conversion processing circuit provided in the conventional video signal processing apparatus, as shown in FIG. 15, the color difference data D2 is obtained by executing linear interpolation from two adjacent color difference data, for example, the color difference data D1 and the color difference data D2. Generate '. When the interval between adjacent data of the luminance data before interpolation (Y Source data) and the color difference data before interpolation (C Source data) is normalized to 1, the color difference data after interpolation (C destination data) becomes Dn ′ = Dn × 0.25 + D (n + 1) × 0.75 (n = 1, 2, 3,...)

  As described above, the MPEG conversion processing circuit converts the MPEG-1 digital video signal into the MPEG-2 digital video signal by aligning the phases of the luminance data Y and the color difference data Cb and Cr. .

  On the other hand, in addition to the above-described MPEG conversion processing circuit, the video signal processing apparatus includes an image scaling circuit that executes a reduction / enlargement process for reducing or enlarging an image as a function of image processing to change the size of the image. The image scaling circuit includes an interpolation filter. When the image is reduced, the number of pixels of the original image data is decreased through the interpolation filter. Conversely, when the image is enlarged, the pixels of the original image data are provided. The number is increased through this interpolation filter. In such an image scaling circuit, for example, a polyphase filter using an approximate curve algorithm such as a Cubic function is used as an interpolation filter so as to perform reduction and enlargement processing without degrading the image quality of the original image as much as possible. It is known to use (refer patent document 1).

  FIG. 16 shows a schematic configuration of a conventional video signal processing apparatus 100 including the above-described MPEG conversion processing circuit and an image scaling circuit which is a polyphase filter. As shown in FIG. 16, the video signal processing apparatus 100 includes a luminance data processing system that executes processing on luminance data and a color difference data processing system that executes processing on color difference data. ) And image scaling circuits 102 and 104, which are polyphase filters that perform image scaling processing on color difference data (C Source data) before interpolation, respectively.

  Further, as described above, the MPEG conversion processing need only be executed for the color difference data before interpolation, and therefore, the MPEG conversion processing circuit 103 is provided in the preceding stage of the image scaling circuit 104. Further, the luminance data processing system is provided with a shift register 101 for adjusting the latency (delay) due to the provision of the MPEG conversion processing circuit 103 in the color difference data processing system.

JP 2003-122338 A

  As described above, the conventional video signal processing apparatus 100 converts the digital video signal encoded with the MPEG-1 system into the MPEG-2 system when the digital video signal is reproduced. However, since the MPEG conversion processing circuit 103 and the image scaling circuit 104 for changing the image scaling are individually provided, there is a problem that the circuit scale increases.

  Therefore, the present invention has been devised to solve the above-described problems, and includes a conversion processing function of a high-efficiency encoding method provided in the above-described MPEG conversion processing circuit and an image scaling circuit. An object of the present invention is to provide an image filter circuit and a filtering processing method in which the circuit scale is greatly reduced by combining the image scaling processing function.

  In order to achieve the above object, an image filter circuit according to the present invention converts from 4: 2: 2 format image data including luminance data and color difference data to phase data determined by the enlargement ratio or reduction ratio of the image data. Interpolating arithmetic processing means for generating interpolated image data between the image data by performing interpolating arithmetic processing based on the first image data in the 4: 2: 2 format corresponding to the first high-efficiency encoding method Adding means for adding a predetermined phase shift between the color difference data and the color difference data of the second image data in the 4: 2: 2 format corresponding to the second high-efficiency encoding method to the phase data; When the conversion from the first image data to the second image data or the conversion from the second image data to the first image data is instructed, the adding means Selection means for selecting the output and supplying the phase data to which the phase shift is added to the interpolation calculation processing means, and the interpolation calculation processing means adds the phase data to the phase data to which the phase shift has been added. Based on this, interpolated color difference data between the color difference data is generated.

  In order to achieve the above-described object, the filtering processing method according to the present invention includes phase data determined by a magnification ratio or a reduction ratio of image data from 4: 2: 2 format image data including luminance data and color difference data. An interpolation calculation process for generating interpolated image data between the image data by performing an interpolation calculation process based on the first image data in the 4: 2: 2 format corresponding to the first high-efficiency encoding method An addition step of adding a predetermined amount of phase shift between the color difference data of the image data and the color difference data of the second image data in the 4: 2: 2 format corresponding to the second high-efficiency encoding method to the phase data; In response to an instruction for conversion from the first image data to the second image data, or from the second image data to the first image data, the addition is performed. And selecting the output by the step, and supplying the phase data to which the phase shift is added to the interpolation calculation processing step, and the interpolation calculation processing step includes the phase to which the phase shift is added. Interpolated color difference data between the color difference data is generated based on the data.

  In the present invention, since image scaling processing and conversion processing of a high-efficiency encoding method can be realized simultaneously with an image filter circuit that is a conventional image scaling circuit, a significant circuit reduction can be realized, This makes it possible to reduce the system latency for circuits that are no longer needed.

  Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited to the following examples, It cannot be overemphasized that it can change arbitrarily in the range which does not deviate from the summary of this invention.

{Device configuration of video signal processing device}
As shown in FIG. 16, the conventional video signal processing apparatus 100 is dedicated to each of the color difference data processing systems that execute the processes for the color difference data Cb and Cr in order to execute the MPEG conversion process and the image scaling process. An MPEG conversion processing circuit 103 and an image scaling circuit 104 were provided. In addition, since the video signal processing apparatus 100 is provided with the MPEG conversion processing circuit 103, it is essential to provide the shift register 101 in the luminance data processing system that executes processing on the luminance data Y. As a result, the circuit scale has been greatly increased.

  Here, the luminance data processing image scaling circuit 102 and the color difference data processing image scaling circuit 104 included in the video signal processing apparatus 100 will be described. The image scaling circuits 102 and 104 are, for example, polyphase filters using a Cubic function. This polyphase filter is assumed to be composed of a 4-tap FIR (Finite Impulse Response) filter.

  The polyphase filter is an interpolation filter that generates post-interpolation data (Destination data) at an arbitrary position between pre-interpolation data (Source data). For example, when a polyphase filter is used, as shown in FIG. 1, between the data src (1) and the data src (2) which are pre-interpolation data, the phase component indicated by phase from the data src (1) is obtained. Data dst (1), which is post-interpolation data, can be generated at a position shifted by a distance.

  Since the polyphase filter used here is a 4-tap filter, the number of taps, that is, four interpolation filter coefficients (Cubic coefficients) are required. As shown in FIG. 1, the interpolation filter coefficient corresponds to the impulse response at the position of the pre-interpolation data when the reference point (center) of the Cubic function, which is an approximate interpolation function, is aligned with the position where the interpolation data is generated. .

  One post-interpolation data can be generated by weighting the four pre-interpolation data using these four interpolation filter coefficients and performing an operation.

For example, as shown in FIG. 2, the interpolation filter coefficient k1 is (| x | −1) (| x | ² − | x | −) where x is a phase that is a deviation from the reference point of the Cubic function. It can be obtained by substituting it into the Cubic function 1 |). Similarly, the remaining interpolation filter coefficients k0, k2, and k3 can be calculated from the Cubic function. At this time, by matching the zero-crossing points of the Cubic function to the pitch of the data before interpolation, the data before interpolation can be normalized by 1 and can be equivalent to the phase of the data after interpolation for obtaining the phase.

  Thus, if phase is obtained, interpolation filter coefficients are calculated using the Cubic function, and post-interpolation data can be automatically generated. As shown in FIG. 1, this phase is an enlargement rate when the pre-interpolation data is enlarged or a reciprocal of the reduction rate when the reduction is performed, and is a cumulative delta that determines the pixel position of the post-interpolation data. It can be calculated from the relationship between the end of the pre-interpolation data and the offset indicating the phase shift at the end of the post-interpolation data.

  Note that the pixel position of the post-interpolation data is calculated by accumulatively adding delta, so if delta is longer than the pitch between the pre-interpolation data, the number of post-interpolation data generated is greater than the number of pre-interpolation data. Therefore, reduction processing is performed. On the other hand, when delta is shorter than the pitch between pre-interpolation data, the number of post-interpolation data to be generated is larger than the number of pre-interpolation data.

  As described above, the image scaling circuits 102 and 104, which are polyphase filters, generate four pre-interpolation data and post-interpolation data based on phase, and realize image scaling such as image enlargement or reduction. .

  In order to realize such processing, the image scaling circuits 102 and 104 are configured as shown in FIGS. 3 (a) and 3 (b), respectively.

  As shown in FIG. 3A, the image scaling circuit 102, which is a polyphase filter, is supplied from the shift register 111 that shifts the pre-interpolation luminance data, the delta accumulator 112 that accumulates the delta and obtains the phase, and the shift register 111. The cubic function block 113 for generating post-interpolation luminance data from the cubic function using the four pre-interpolation luminance data and the phase supplied from the accumulator 112.

  The image scaling circuit 104 also includes a shift register 121 that shifts the color difference data before interpolation, a delta accumulator 122 that accumulates and adds delta, obtains the phase, four color difference data before interpolation supplied from the shift register 121, and the accumulator 122. And a cubic function block 123 that generates post-interpolation color difference data from the cubic function using the phase supplied from. Since the shift register 121 alternately inputs the color difference data Cb and Cr, the number of flip-flops is larger than that of the shift register 111.

  By the way, as shown in FIG. 16, the MPEG conversion processing circuit 103 provided in the video signal processing apparatus 100 according to the prior art decodes the MPEG-1 system digital video signal to convert the 4: 2: 0 format to 4 :. The color difference data Cb and Cr of the image data shown in FIG. 13 converted into the 2: 2 format are linearly interpolated to align the phase with the luminance data Y.

  In the MPEG-1 system, only the 4: 2: 0 format is defined, but for the sake of explanation, the 4: 2: 0 format image data shown in FIG. 12 is converted, and the 4: 2: format shown in FIG. 13 is converted. The 2-format image data is referred to as the MPEG-1 4: 2: 2 format.

  As described above, the image scaling circuit 104 is a polyphase filter that can generate post-interpolation color difference data at an arbitrary position of the pre-interpolation color difference data, that is, at an arbitrary phase. Therefore, it is considered that this polyphase filter can also filter the phase of the color difference data Cb and Cr of the MPEG-2 4: 2: 2 format image data so that it matches the phase of the luminance data Y. In other words, the image can be scaled by the polyphase filter, and can be converted from the MPEG-1 system, which has been used in the MPEG conversion processing circuit 103, to the MPEG-2 system. It is done.

  Therefore, in the video signal processing apparatus shown as the best mode for carrying out the present invention, a polyphase having an MPEG conversion function as shown in FIG. 4 is used for the color difference data processing system for processing the color difference data Cb and Cr. Assume that a filter 10 is provided.

  As shown in FIG. 4, the polyphase filter 10 includes a shift register 11 that shifts color difference data before interpolation, a delta accumulator 12 that accumulates and adds delta to obtain phase, and four color differences before interpolation supplied from the shift register 11. Using the data and the phase supplied from the delta accumulator 12, a cubic function block 13 that generates post-interpolation color difference data by a Cubic function, and an adder 14 that adds a fixed value (for example, 0.75) to offset And a selector 15 that selects either offset or offset added with a fixed value and supplies the selected value to the delta accumulator 12. As described above, the polyphase filter 10 has a configuration in which the adder 14 and the selector 15 are added to the conventional image scaling circuit 104 shown in FIG.

  In the following, the adder 14 and the selector 15 are provided in the image scaling circuit 104 shown in FIG. 3B, so that the polyphase filter 10 can execute MPEG conversion in addition to the image scaling function. Will be explained.

  FIG. 5 shows a state where the horizontal luminance data Y and the color difference data Cr and Cb of the MPEG-1 4: 2: 2 format image data shown in FIG. 13 are extracted line by line. The luminance data Y is pre-interpolation luminance data, and the color difference data Cb and Cr are pre-interpolation color difference data. When such MPEG-1 format image data is converted into the MPEG-2 format by the polyphase filter 10, the interpolated color difference data generated through the polyphase filter 10 is shown in FIG. It suffices to arrange every other phase in the same phase as the pre-interpolation luminance data.

  As described with reference to FIGS. 1 and 2, when the Cubic function is used, the data before interpolation is normalized by 1 by matching the zero crossing point of the Cubic function with the pitch of the data before interpolation, and the phase is obtained. It can be equivalent to the phase of the data after interpolation. Therefore, focusing on the color difference data C3 before interpolation, it can be seen that the phase difference in the left horizontal direction should be shifted by 0.25 to obtain the color difference data C3 'after interpolation having the same phase as the luminance data Y.

  That is, if the phase shown in FIG. 2 is 0.75, if this phase and the four pre-interpolation color difference data C1, C2, C3, and C4 are used, the polyphase filter 10 uses the post-interpolation color difference data C3 ′. Can be generated.

  In the polyphase filter 10 shown in FIG. 4, 0.75 is added to offset by the adder 14, and 0.75 is added by the selector 15 when converting from the MPEG-1 system to the MPEG-2 system. When the offset is selected and MPEG conversion is not performed, the offset is selected. The same applies to a case where 0.75 is added to the phase that is the output from the delta accumulator 12, as in the polyphase filter 10 'shown in FIG.

  FIG. 5 shows a case where conversion from the MPEG-1 system to the MPEG-2 system is executed without executing the enlargement process and the reduction process with delta being 1, and the poly-types shown in FIGS. In the phase filter 10 and the polyphase filter 10 ′, if delta is set to 1, offset is set to zero, and a signal added with 0.75 is selected by the selector 15, 0.75 is supplied to the cubic function block 13 as phase. Will be. As a result, the post-interpolation color difference data filtered through the polyphase filter 10 is converted into the phase of MPEG-2 color difference data.

  As described above, the polyphase filter 10 can convert MPEG-2 4: 2: 2 format image data into MPEG-2 4: 2: 2 format image data.

  In other words, if the polyphase filter 10 gives 1 with delta as the same magnification, it simply converts the MPEG-2 4: 2: 2 format image data into the MPEG-2 4: 2: 2 format image. Convert to data.

  Further, if a value for reducing or enlarging is given to delta, normal image scaling can be executed simultaneously in addition to the conversion from the MPEG-1 system to the MPEG-2 system described above. For example, in FIG. 7, since delta = 2 is given, conversion from the MPEG-1 system to the MPEG-2 system can be executed simultaneously with the reduction process with a reduction ratio of 1/2.

{End processing}
Next, edge processing when executing image scaling will be described. When one post-interpolation data is generated from four interpolation data and image scaling is performed, that is, when interpolation calculation processing is performed, the left end and right end of the image, that is, the start point and end point of the effective image data are determined before any interpolation. It is important to generate using data. Various methods have been devised for the edge processing depending on how the image is cut out. In general, as shown in FIG. 8, the first effective image data is used in duplicate and the luminance data Y1, Y1, Interpolated luminance data Y1 ′ is generated from Y2 and Y3, and post-interpolated color difference data C1 ′ is generated from the color difference data C1, C1, C2 and C3. In the image scaling circuits 102 and 104, which are the conventional polyphase filters shown in FIGS. 3A and 3B, such edge processing is executed when image scaling is performed.

  By the way, as described above, the polyphase filter 10 of the present invention converts MPEG-1 4: 2: 2 format image data into MPEG-2 4: 2: 2 format image data. However, the edge processing of the color difference data at this time is executed as shown in FIG. That is, when the post-interpolation color difference data C1 ′ corresponding to the color difference data C1 that is the effective image data at the left end is generated, the color difference using the three color difference data C1 among the four pre-interpolation color difference data. Data C1, C1, C1, and C2 are used.

{MPEG-2 system → MPEG-1 system}
In practice, this polyphase filter 10 uses MPEG-2 system 4: 2: 2: format image data and MPEG-1 system 4: 2: 2 format image, although it is not used very often. It can also be converted to data.

  In order to convert from the MPEG-2 system to the MPEG-1 system, it is only necessary to shift the luminance data to the right by 0.25. Therefore, as shown in FIG. 10, the polyphase filter 10 has 0.25 as the initial phase. Give it. At the end, when the post-interpolation color difference data C1 ′ corresponding to the luminance data C1 that is the effective image data at the left end is generated, two of the four pre-interpolation color difference data are used. The color difference data C1, C1, C2, and C3 are used.

  That is, similarly to the conversion from the MPEG-1 system to the MPEG-2 system, if 1 is given to the polyphase filter 10 with the same delta, the 4: 2: 2 format of the MPEG-2 system is simply used. Are converted into MPEG-1 format 4: 2: 2 format image data. If a value for reducing or enlarging the delta is given, normal image scaling can be executed simultaneously in addition to the conversion from the MPEG-2 system to the MPEG-1 system described above.

{Implementation example of polyphase filter 10}
Subsequently, an implementation example of the polyphase filter 10 illustrated in FIG. 4 will be described. For example, the polyphase filter 10 is mounted on the integrated circuit 50 shown in FIG. The integrated circuit 50 shown in FIG. 11 includes an accumulate delta 51, a data shifter 52, a cubic function 53, a delay (2ck Delay) 54, a selector 55, and a delay (1ck Delay). 56.

  The accumulating delta 51 corresponds to the delta accumulator 12, the selector 15, and the adder 14 shown in FIG. 4, and generates phase_hc which is a phase by inputting hppf_delta as delta and hppf_offset as offset. The accumulating delta 51 switches the selector 15 shown in FIG. 4 to be an output from the adder 14 in accordance with the register mpeg1_mode = 1 set by a CPU (not shown). As a result, MPEG-2 4: 2: 2 format image data can be converted into MPEG-2 4: 2: 2 format image data.

  The data shifter 52 corresponds to the shift register 11 shown in FIG. 4, and the cubic function 53 corresponds to the cubic function block 13 shown in FIG. The delay 54 adjusts the latency due to the color difference data Cb and Cr, and the delay 56 adjusts the latency between the luminance data Y and the color difference data Cb and Cr.

  Note that the integrated circuit 50 shown as an implementation example of the polyphase filter 10 in FIG. 11 is an example and does not limit the present invention.

  In this way, the polyphase filter 10 is a conventional video image as shown in FIG. 16 in a color difference data processing system that executes processing for the color difference data Cb and Cr in order to execute MPEG conversion processing and image scaling processing. The functions executed by the MPEG conversion processing circuit 103 and the image scaling circuit 104 included in the signal processing apparatus 100 can be executed only by the polyphase filter 10.

  As a result, when such a polyphase filter 10 is provided, the video signal processing apparatus can perform MPEG conversion processing and image scaling processing as it is, while greatly reducing the circuit and system latency of the video signal processing apparatus. Can be reduced.

  The present invention is not limited to the type of high-efficiency encoding method such as MPEG-1 method or MPEG-2 method, but between all image data in which color difference data in 4: 2: 2 format has a phase shift. It becomes effective in the conversion of.

It is a figure for demonstrating the filtering process using a Cubic function. It is a figure for demonstrating an interpolation filter coefficient. (A) is a figure for demonstrating the conventional image scaling circuit in a luminance data processing system, (b) is a figure for demonstrating the conventional image scaling circuit in a color difference data processing system. It is a figure for demonstrating the structure of the polyphase filter shown as the best form for implementing this invention. It is a figure for demonstrating the conversion process from the MPEG-1 system to the MPEG-2 system in the polyphase filter. It is the figure shown about another structure of the same polyphase filter. It is the figure which showed a mode that the image scaling process was performed simultaneously with performing the conversion process from an MPEG-1 system to an MPEG-2 system in the polyphase filter. It is the figure shown about the edge part process accompanying the conventional interpolation calculation process. It is the figure shown about the edge part process accompanying the interpolation calculation process by the said polyphase filter. It is a figure for demonstrating the conversion process from the MPEG-2 system to the MPEG-1 system in the polyphase filter. It is the figure which showed the example of mounting at the time of mounting the polyphase filter. It is the figure which showed 4: 2: 0 format of MPEG-1 system and image data. It is the figure which showed the 4: 2: 2 format of MPEG-1 system, and image data. It is the figure which showed 4: 2: 2 format and image data of MPEG-2 system. It is a figure for demonstrating the conversion process from the MPEG-2 system to the MPEG-1 system in the conventional MPEG conversion processing circuit. It is a figure for demonstrating the conventional video signal processing apparatus.

Explanation of symbols

  10 polyphase filter, 11 shift register, 12 delta accumulator, 13 cubic function block, 14 adder, 15 selector

Claims (3)

Interpolated image data between the image data is obtained by performing interpolation calculation processing based on phase data determined by the enlargement ratio or reduction ratio of the image data from 4: 2: 2 format image data including luminance data and color difference data. Interpolating arithmetic processing means to generate,
Color difference data of the first image data in 4: 2: 2 format corresponding to the first high-efficiency encoding method and a second image in 4: 2: 2 format corresponding to the second high-efficiency encoding method Adding means for adding a predetermined amount of phase shift with the color difference data of the data to the phase data;
The output from the adding means is selected in response to an instruction to convert the first image data to the second image data, or from the second image data to the first image data. And selection means for supplying the phase data added with the amount of phase shift to the interpolation calculation processing means,
The image filter circuit, wherein the interpolation calculation processing means generates interpolated color difference data between the color difference data based on the phase data to which the phase shift is added. The first high-efficiency encoding method is MPEG-1 (Moving Picture Expert Group-1) method,
The image filter circuit according to claim 1, wherein the second high-efficiency encoding method is an MPEG-2 (Moving Picture Expert Group-2) method. Interpolated image data between the image data is obtained by performing interpolation calculation processing based on phase data determined by the enlargement ratio or reduction ratio of the image data from 4: 2: 2 format image data including luminance data and color difference data. Interpolating operation process to generate;
Color difference data of the first image data in 4: 2: 2 format corresponding to the first high-efficiency encoding method and a second image in 4: 2: 2 format corresponding to the second high-efficiency encoding method An addition step of adding a predetermined phase shift amount with the color difference data of the data to the phase data;
In response to an instruction to convert the first image data to the second image data, or from the second image data to the first image data, the output in the addition step is selected. And a selection step for supplying the phase data to which the phase shift is added to the interpolation calculation processing step,
The filtering processing method, wherein the interpolation calculation processing step generates interpolated color difference data between the color difference data based on the phase data to which the amount of phase shift is added.

Images