TW391133B - Method and apparatus for interpolation with dynamic profile scanning lines using one-dimensional pixel array moving detection - Google Patents

Abstract

The present invention relates to a kind of conversion between image data from interlaced scanning and image data from progressive scanning, particularly a method and apparatus using a dynamic profile scanning line for interpolation to improve the image quality of a display system. The invention provides an easy method and apparatus which employs simple moving detection step and dynamic profile interpolation to convert the image data of interlaced scanning format into the image data of progressive scanning format. Therefore, the invention uses lesser hardware and buffer cost to execute interlaced-progressive conversion; furthermore, the invention can proceed moving detection by inspecting one-dimensional pixel array of adjacent image fields. Thus, the invented interlaced-progressive conversion operation is simpler than the conventional method; furthermore, the invention provides clearer, less glittering images and maintains good edges.

Description

Printed by the Consumers' Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the Invention (1) Field of the Invention The present invention relates to the conversion of image data in an interlaced scanning format and image data in a progressive scanning format, and particularly relates to a dynamic contour Method and device for interpolating scanning lines to improve image quality of a display system. * Background of the Invention The conventional NTSC (National Television Standards Committee) color television system uses an interlaced scanning system to reduce the need for transmission bandwidth. The interlaced video data is displayed at half frame (or a field) every 1/60 second, and another field is displayed at 1/60 second. As shown in FIG. 1, it illustrates a video format with interlaced scanning of an upper field and a lower field in the conventional art. The upper field contains all the odd horizontal lines, and the lower field contains all the even horizontal lines. However, the conventional interlaced scanning method has many visual defects, such as flickering, blurring, and jagged contours. Deinterlacing can overcome the above problems and improve the appearance of the image by converting the interlaced scanned video format to a progressive scanning format. In addition, many applications (such as PC monitors, projection, and high-definition televisions) use progressive scanning technology. To convert the interlaced video format to progressive scan format, the missing pixels in each field (such as the odd horizontal line in the upper field or the even horizontal line in the lower field) are derived from the interlaced scanning video signal. produce. One method of interlaced-progressive conversion is a static technique. This static technology uses the same conversion method regardless of the image source or content. This method makes L: \ EXl \ 54 \ 54249.DOC-4 — This paper size applies to the Chinese National Standard (CNS) A4 specification (210X297 mm) -------: --rl ^ ----- -Order ------ Quan (please read the notes on the back before filling in this page) Printed by A7 _____B7 _ from the Central Government Bureau of the Ministry of Economic Affairs, Shellfish Consumer Cooperative Ⅴ. Description of the invention (2) Copy by line, vertical filtering And map occasions and technology. Lines duplicate each horizontal line in the repeating picture ’field to create a complete frame. Vertical filtering creates a missing line by traversing multiple adjacent lines, and in the case of the picture, the lines are taken from the previous field and inserted into the current field to construct the I * box. However, this method will cause visual loss, for example, moving objects in the video frame will have jagged edges. To overcome these shortcomings, adaptive de-interlacing technology came into being. This adaptive de-interlacing technique uses dynamic analysis to select the best method for different parts of the image. This adaptive de-interlacing technique first checks the current field and one or more previous fields to determine the moving content. If the sequence of the image moves little or does not move, the pixels of the line to be interpolated are not from the previous The field decision; if the sequence of the images is large, an edge dynamic space interpolation method is applied to construct the line to be interpolated. However, the above-mentioned method using adaptive de-interlacing technology has a complicated calculation process to determine the moving content and pixels to be interpolated. US Patent No. 5,019,903 discloses a method that only considers how to interpolate the pixels to be interpolated without consideration. Technology of overall motion detection circuit. U.S. Patent No. 5,473,383 discloses a technique that uses an additional buffer to store a signal representing a mobile signal. U.S. Patent No. 5,592,23 1 discloses a technique of using a median filter to select a moving signal and applying 5-edge interpolation. U.S. Patent No. 5,532,75 1 discloses a technique that uses two fields in the same frame to detect movement. U.S. Patent No. 5,339,109 discloses a 3-edge interpolation method that uses out of order, but it does not consider L: \ EXt \ 54V54249.DOC — 5 — This paper standard applies the Zhongguanjia standard rate ([just 8 4 specifications (2 丨 〇 < 297 mm) ~ ----.-------- ¢ ------, 耵 ------ ^ (Please read first Note on the back again ^ This page) A7 B7 printed by the Consumer Cooperatives of the Central Bureau of Standards of the Ministry of Economic Affairs 5. Description of the invention (3) Consider the sharpness of the edges. However, improper contour interpolation will make the image blur In view of the shortcomings of conventional techniques, the present invention provides a method by which

I's motion 'detection and adaptive contour interpolation to transform interlaced images into progressive images. Therefore, an object of the present invention is to accomplish interlaced-to-progressive conversion with less hardware and buffer costs. The present invention performs motion detection by viewing a one-dimensional pixel array of adjacent fields. If the pixel to be interpolated is the stable part of the shadow-image, its 値 will be replaced by the corresponding 値 of the previous field; otherwise, the pixel to be interpolated is interpolated using adaptive contour interpolation. The adaptive contour interpolation interpolates the pixels to be interpolated based on the smallest 値 of the three sides in the edge direction. For video formats with MPEG-2 (Moving Picture Expert Groups-2), the present invention also provides a memory architecture for error-progressive conversion to reduce the need for buffers. Therefore, the video image of the present invention has a clearer image with better edges and less flicker. In addition, the computational complexity required by the present invention is simpler than the conventional technique. Since the present invention only uses a one-dimensional pixel array to detect and determine the mobile signal, the required hardware cost of the buffer is also lower than the conventional technique. Brief description of the drawings In order to illustrate other objects and advantages of the present invention, the following preferred embodiments are described in conjunction with the description of the drawings as follows, wherein: FIG. 1 shows the format of an interlaced scan video image of a conventional technique; L: \ E5CI \ 54 \ 54249.DOC — 6 — This paper size is applicable to China National Standard (CNS) A4 specification (210X297 mm) I ------.--- installation -------- ---- ^ (Please read the notes on the back before ^ writing this page) A7 B7 V. Description of the invention (4) 2 is a block diagram of the interlaced-gradual conversion disclosed by the preferred embodiment of the present invention ; Figure 3 is not a block diagram of the interlaced-progressive conversion of adaptive contour interpolation disclosed in the present invention; __% Figure 4 shows an example of a procedure for determining a moving signal disclosed in the present invention; Figure 5 shows a flowchart of adaptive contour interpolation disclosed in the present invention; Figure 6 shows a diagram of interlaced-progressive conversion of a video sequence with a moving image expert group (MpEG) disclosed in the present invention. Example; FIG. 7 is a block diagram of a memory architecture for a mpeg video format disclosed in a preferred embodiment of the present invention; Figure 8 shows an example of the memory and content update procedure for the mpeg video format disclosed in the present invention. The main component symbols in _ 处 _ ±. Please read the precautions on the back before / write this page Ordered by the Central Economic and Technical Bureau of the Ministry of Economic Affairs of the Bayan Consumer Cooperatives 2 1 Graphic input 34 Round interpolation 22 Video capture 3 5 Switch 23 MPEG decoder 3 6 Comparator 24 Interlaced-progressive conversion 3 7 Threshold 値 25 Mixer 7 1 Memory 1 26 VGA monitor 72 Memory 2 3 1 Line memory 7 3 Memory 3 3 2 First field memory 74 Memory 4 3 3 Second field memory

Detailed description of the invention L: \ EXT \ 54V54249.DOC The standard size of the original paper is applicable to the Chinese national standard (CNS &gt; A4 specification (21 ××: 297 mm). Printed by the Consumer Cooperatives of the Central Bureau of the Ministry of Economic Affairs. Explanation (5) Please refer to FIG. 2 and FIG. 2 + Jin Shi Guo — “-Jin, Ma, block diagram of interlaced progressive conversion disclosed in the preferred embodiment of the present invention. Take the video from the video or decode it by reading The video signal decoded by the device is a video signal in the interlaced scanning format. It is a video signal reading format that simultaneously displays the non-interlaced scanning format (22 or 23) <VGj graphic signal (21) in—for example, the scanning format vga display㈣ The signal of the above 'interlaced scanning format' or 23) is converted by -interlaced and progressive conversion ', and is a signal of the progressive scanning format. The video signal of the video signal is mixed with the graphic signal (21) and the mixer (25). It is displayed again on the monitor (26). ## Refer to FIG. 3 ′ The tf shown in FIG. 3 is a block diagram of the interlaced-progressive conversion of the adaptive edge-interpolatlon disclosed in the present invention. For each pixel of the line to be interpolated in the field (where 丨 indicates (A field of a first polarity (32)) and the motion detection is first used to determine whether a pixel is in a moving area of the image. The motion detection is compared by _comparator⑽ and ^^ 丨The threshold is (37). The result of the comparison is then passed through-the switcher (35) never decides whether to make a mapping occasion or contour interpolation to generate pixels. The motion detection architecture uses two consecutive fields "丨 and i + 1, the two fields are in the field! Before and after the picture field, which Bu! And i + 1 represents a field of the second polarity (33). That is, when the field 丨 corresponds to the lower field, the fields 1-1 and 1 + 1 are the upper field; and when the field 丨Corresponding to the field above, fields i-1 and i + 1 are fields below. Please refer to FIG. 4. FIG. 4 shows an example of an unlimited graph, which is used to explain how the present invention determines how to determine the moving signal of the pixel to be interpolated x2. The moving signal is based on the inspection field 丨 the previous field 丨 _ i and r ----- ^-(Please read the precautions on the back before this page} Order

L: \ EXI \ 54 \ 54249.DOC

Printed by A7 B7 of the Consumer Cooperatives of the Central Procurement Bureau of the Ministry of Economic Affairs 5. Description of the invention (6) '~~~~ The content of the one-dimensional pixel array of the field i + 1 after the field (three pixels in this example) And get. For example, as for the pixel χ2 to be interpolated, a two-pixel (three.pixei) array corresponding to field 1-1 contains pixels M, ^^, and "and a three-pixel array corresponding to field i + 1 includes Pixels y, "2," and "3. The absolute sum of the differences between the two pixel arrays is generated according to the following equation: ^ Hl-bl | + [j2-b2 | + [j3-b3 | / Is compared with a predetermined threshold 値. If the sum signal / is less than the predetermined threshold 値, the pixel to be interpolated χ2 is set as a stable part in the image. The pixel to be interpolated * pixels値 χ2 is replaced by the corresponding pixel 値 b2 in the field i_ 1. If the sum signal is greater than the predetermined threshold 値, it means that there is substantial movement in the part of the image, and the pixel 値 χ2 is predicted by contour interpolation In order to reduce the calculation cost, the contour interpolation method disclosed in the present invention considers three edges across the pixel to be interpolated. For example, for the pixel X2 to be interpolated, the three edges are vertical contours e 2 respectively -f 2 and two diagonal contours ei-f3 and e3-fl. Please refer to FIG. 5, which shows the adapting wheel disclosed by the present invention Flow chart of contour interpolation method. First, the edges of the two diagonal edges ei_f3 and e3_n are calculated first (51), and then the edges of the vertical edges e2-f2 are calculated (52). The edge direction is then It can be selected, and the selection criterion is selected from the smallest one of the three absolute differences 値 (ie | ehf3 |, | e2-f2 丨, and | e3-fl |). If two or more edges have the same The absolute value of the smallest difference, the edge direction is selected as the edge (53) that is calculated last and has the smallest value. After the edge direction is selected, the pixels to be interpolated are compared by comparing the edges L: \ EXI \ 54 \ 54mDOC-9 _ This paper size is applicable to China National Standard (CNS) A4 specification (210X297 mm): _ 丨 .---- install -------- order ------ line (谙 read the first Note on this page) A7 B7 V. Invention description (値 and a predetermined threshold 门 are determined (54). If the edge 値 is greater than a predetermined threshold 门 (55), the pixel to be interpolated 値 x2 is at the edge One of the two pixels 方向 of the direction and a sharper image edge is generated. Therefore, one to ensure that a sharper image will not have Interpolation motions that result in blurred edges are not performed. If the edge 値 is less than a predetermined threshold (56), the pixel to be interpolated x2 is interpolated to two pixels in the edge direction 値Additive sum (average of two pixels 値) ^ For example, if the edge direction is and its edge 値 is less than a predetermined threshold 値, the pixel to be interpolated 値 is equal to (e2 + f2) / 2. If the edge is The edge in the direction is greater than the predetermined threshold. 'The pixel to be interpolated, X2 is the two pixels in the edge direction'. For example, if the edge direction is e2-f2 and its edge 値 is greater than a predetermined threshold 値, the pixel 値 x2 to be interpolated is equal to the pixel 値 62 or f2. For a video signal with MPEG format, the interlaced-progressive conversion occurs after the MPEG decoding program (as shown in Figure 2), so the interlaced-progressive conversion is required The buffer is shared with the MPEG decoder (23). Please refer to FIG. 6, which is an example of a diagram of interlaced-progressive conversion with MPEG video sequence disclosed in the present invention. Each of the 1-, B-, and P- graphics includes an upper field and a lower field. As shown in FIG. 6 ', its display order is similar to τν. When only two consecutive B · graphics exist between I-and P _ graphics (this situation occurs most often in the MPEG video data stream system), four memory blocks have been used to meet the buffer requirements . Please refer to FIG. 7, which is a block diagram of a memory architecture for an MPEG video format according to a preferred embodiment of the present invention. The memory architecture uses four memory blocks. Two memory blocks are used for interlaced-progressive conversion ’and the other two memory blocks are -10

L: \ EJn \ 54 \ 54249DOC This paper size is applicable to China National Standards (CNS) 8-4 specifications (210X297 mm) Printed by the Shellfish Consumer Cooperative of the Central Standards Bureau of the Ministry of Economic Affairs A7 --- —__ B7 V. Description of the invention ( 8) It is used to store the MpEG graphic data to be decoded. -Please refer to FIG. 8, which is a video for MpEG disclosed by the present invention

An example of a memory content update procedure in a format. Memory block MEM 1 (71) and memory block MEM 4 (74) are provided for __graphics and p _ graphics use and memory block MEM 2 (72) and memory block MEM3 (73) are provided for b. Graphic use. The new I-, P- &amp; B_ graphics are stored in the corresponding memory blocks in the same type of graphics and their display order. As shown in Fig. 8, when a new frame graphic is received, the old graphic in the corresponding memory block is replaced by the new graphic. For example, when receiving and receiving p6, the data 10 is replaced with. As shown in FIG. 8, the interlaced-progressive conversion procedure for the MPEG video format t two consecutive fields start with two consecutive graphics in the display order (if the number of consecutive B-pictures is not greater than 2). For example, to interpolate fields 82, B1 and B2 are used as reference patterns. If the number of consecutive B_pictures is greater than 2, one of the memory blocks used to store the B_pictures is used to store new decoded data. Therefore, if the memory block MEM 3 (73) is used to store the decoded graphics data, the memory area is used to interpolate the field below the B-picture in the memory block MEM 2 (72). Block mem 4 (74) above the field. If the memory block MEM 2 (72) is used to store the decoded graphics data, then the B-graphic field interpolation of the memory block MEM 3 (73) uses the memory block MEM 1 ( 7 1) The field below. The present invention relates to the conversion of image data in an interlaced scanning format and image data in a progressive scanning format, and particularly relates to _ a kind of dynamic contour scanning. L: \ EX1 \ 54 \ 54249-DOO „11 _ This paper is applicable to China according to scale National Standard (CNS) A4 Specification (210X297 Meal) ------- ί --- ^! (Please read the notes on the back before writing this page) Order I unn. A7 B7 V. Description of the invention ( 9 Method and device for interpolating traces to improve _ Xuanxian's unsystematic image quality. The present invention provides a simple girl and girl, tu tjinligan # Jian Yixin method and device, which uses simple motion detection Steps and dynamic contour interpolation to convert image data with interlaced scanning format to image data with progressive scanning format. Thus, the present invention can perform interlaced-progressive conversion with less hardware and buffer costs. In addition, the present invention can Motion detection is performed by viewing a one-dimensional pixel array of adjacent fields. Therefore, the calculation of the interlaced-progressive transformation of the present invention can be simpler than the conventional technique. Furthermore, the video image of the present invention can provide clearer, Good edge retention The image with less flicker. The only embodiment described above is only to explain the principle and effect of the present invention, but not to limit the present invention. Therefore, the modifications and changes made by those skilled in the art to the above embodiments are still inconsistent. The spirit of the invention. The authority of the invention should be listed in the scope of patent application described below. Please read first

Note on the back, 4 (This booklet is bound 涑 Printed by the Male Workers Consumer Cooperatives, Central Bureau of the Ministry of Economic Affairs

L: \ EXI \ 54 \ 54249.DOC This paper size is applicable to China National Standard (CNS) A4 (210X297 mm)

Claims (1)

Patent Application No. 8? 1 丨 4324 Chinese Patent Application Amendment (October 88) Repair D8 * ' Scope of patent application: Printed by the Shell Standard Consumer Cooperative of the Central Bureau of Standards, Ministry of Economic Affairs, O: \ 54 \ 54249.DOC · '~ This paper size is applicable to the Chinese Standard for Household Standards (CNS) M (2 丨 〇χ297 公 嫠 1. Generate a progressive scan video «Heart 1 Power to Heart, including the following steps: ⑻ Select one of the first one-dimensional pixel array and one of the first image field in the first image field polarity and the second field polarity -A second one-dimensional pixel array in the second image field; 比较 compare a sum of the absolute values of the differences between the first and second arrays with a predetermined η limit to calculate a —A value between the second array; and (C) determined according to the value calculated in step ——interpolating the pixel value and interpolating the interpolated pixel value to an interlaced video image. The method of generating a progressive scanning video image, wherein the content of the one-dimensional pixel array is the pixels of the corresponding image line of the interpolated pixels. 3. The method of generating a progressive scanning video image as described in item 1 of the patent application scope 'Where if the difference between the first and the first array When the sum of the absolute values is less than the predetermined threshold value, no movement occurs between the first and second arrays. 42. For example, the method of generating a progressively scanned video image in the first item of the scope of patent application 'where if the first and When the sum of the absolute values of the differences between the second arrays is greater than the predetermined threshold value, there is a movement between the first and second arrays. 5 vAs described in item 1 of the scope of patent application, a progressive scanning video image is generated. Method, wherein if there is no movement between the first and second arrays, the interpolated pixels and values are replaced by corresponding pixel values of the first image field. O: \ 54 \ 54249, DOC- ------ install ------ order ----- line (please read the precautions on the back before filling this page) Patent Application No. 8? 1 丨 4324 Chinese Patent Application Amendment (October 88) Repair D8 * ' Scope of patent application: Printed by the Shell Standard Consumer Cooperative of the Central Bureau of Standards, Ministry of Economic Affairs, O: \ 54 \ 54249.DOC · '~ This paper size is applicable to the Chinese Standard for Household Standards (CNS) M (2 丨 〇χ297 公 嫠 1. Generate a progressive scan video «Heart 1 Power to Heart, including the following steps: ⑻ Select one of the first one-dimensional pixel array and one of the first image field in the first image field polarity and the second field polarity -A second one-dimensional pixel array in the second image field; 比较 compare a sum of the absolute values of the differences between the first and second arrays with a predetermined η limit to calculate a —A value between the second array; and (C) determined according to the value calculated in step ——interpolating the pixel value and interpolating the interpolated pixel value to an interlaced video image. The method of generating a progressive scanning video image, wherein the content of the one-dimensional pixel array is the pixels of the corresponding image line of the interpolated pixels. 3. The method of generating a progressive scanning video image as described in item 1 of the patent application scope 'Where if the difference between the first and the first array When the sum of the absolute values is less than the predetermined threshold value, no movement occurs between the first and second arrays. 42. For example, the method of generating a progressively scanned video image in the first item of the scope of patent application 'where if the first and When the sum of the absolute values of the differences between the second arrays is greater than the predetermined threshold value, there is a movement between the first and second arrays. 5 vAs described in item 1 of the scope of patent application, a progressive scanning video image is generated. Method, wherein if there is no movement between the first and second arrays, the interpolated pixels and values are replaced by corresponding pixel values of the first image field. O: \ 54 \ 54249, DOC- ------ Equipment ------ Order ----- line (please read the precautions on the back before filling out this page) -1 -1 Printed by the Bayon Consumer Cooperative, Central Bureau of Standards, Ministry of Economic Affairs B8 C8 D8 6. Application for patent scope 6. The method of generating a progressively scanned video image as described in item 1 of the patent scope 'wherein if there is movement between the first and second arrays, the interpolation is performed by The contour interpolation method is used for prediction. The method for scanning video images, wherein the prediction of the interpolated pixel value further includes the following steps: ㈤ Calculate the value of | x (k) -y (-k) 丨 'where x (k) is the interpolation in the same field The image of the image line above the pixel, the value, k is -N, ... --2, -1,1,2 ... N, and y (k) is the pixel value of the image line below the interpolation pixel in the same field; ⑹ Calculate the value of | x (〇) -y (〇) |; determine the edge direction by choosing the minimum value of | x (k) -y (-k) |, where k = -N,-&quot;- l, 0, 1 ... N; and ⑻ determine the value of the interpolated pixel by comparing the values obtained in steps (d) and (e) with a predetermined threshold value. 8. As described in item 7 of the scope of the patent application, a progressive method is described—the method of “video image”. If the edge direction value is greater than the predetermined threshold value, the pixel value to be interpolated is the pixel value in the edge direction. One. 9. If a progressive sweep is generated in item 7 of the scope of the patent application. 1. The method of tracing video images, if the edge direction value is less than the predetermined threshold value, the pixel values to be interpolated are the pixel values in the edge direction The average. 1〇_The method of generating a progressive scanning video image as described in item 7 of the scope of patent application, wherein if two or more values of x (k) _y (_k) | are the same, the final pair of pixels calculated is Select to predict the pixel value to be interpolated. 11. · A device for generating a progressively scanned video image, comprising: — one of the first image fields of the first image field polarity — dimension -1, (1¾ ------ 1T --- --Qi (please read the precautions on the back before filling this page) O: \ 54 \ 54249.DOC Please patent Fangu Α8 Β8 C8 D8 Printed by the Central Government Bureau of Beibei Department, Beige Consumer Cooperative Co., Ltd. Store the pixel to be interpolated. The pixel corresponding to the image line. One of the second image field polarities is one of the second image field. The second — element array is used to store the pixel to be interpolated. A comparator connected to the -dimensional pixel array to compare the sum of the absolute values of the differences between the first and second arrays with a-predetermined threshold = and produce a value; and-an interpolation device for Interpolate according to the generated value, interpolate the pixel value 0 * • as the generation of the patent application scope item u-device for progressively scanning the video image 'where if the sum of the absolute values of the difference between the first and second arrays When it is less than the predetermined threshold value, the parallel helmet movement between the first and second arrays occurs. A device for generating a progressive 'scanning video image of item U, wherein if the sum of the absolute values of the differences between the first and second arrays is greater than the predetermined threshold value, the There is a movement. 14. If a device for generating a progressive scanning video image is provided in item U of the scope of the patent application, if the movement between the first and second arrays does not occur, the inward pixel value is determined by the first The corresponding pixel value of the image field is replaced. 15 · The device for generating a progressively scanned video image in accordance with item 丨 丨 of the patent application range, wherein if there is movement between the first and the first array, the interpolated pixel value is determined by The contour interpolation method is used to predict. 'For example, the device for generating a progressively scanned video image in the scope of patent application No. 5' also includes 3- O: \ S4 \ 54249.DOC This paper size uses the bidding standard of 0 × 297. Lin-J / \ I Pack-(Please read the precautions on the back before filling this page) Thread 1 The scope of the patent application is used to calculate the value of 丨 x (k) -y (_k) 丨, f ± ψ βη 4 ^. Where x (k) is the pixel value of tens of thousands of image lines in the same field, Λτ κ is -N, ··· -2, -1, 1 2. ·· N, and y (k) are the values of interpolation in the same field; the pixels below the 10,000 image line are used to calculate | x (0 ) -y (0) | value device; The Central Consumers ’Bureau of Industry and Consumer Cooperatives of the Ministry of Economic Affairs printed a device to determine the direction of the edge, and choose this in the system. X (k) _y (-k ) | i minimum value, where k = _N, ... 10, Bu · ν; and-a comparison device for determining a pixel to be interpolated by comparing the calculated value with a predetermined threshold value value. 17. According to the patent application No.16, Yawang progressively describes the video image device 'if the edge direction value is greater than the predetermined threshold value, the prime value to be interpolated is the pixels in the edge direction One of the values. 18. For example, the device for generating a progressive scanning video image in item 16 of the claim range. If the edge direction value is less than the predetermined threshold value, the pixel to be interpolated is the average of the pixel values in the edge direction. value. 1.9. The device for generating a progressive scanning video image as described in item 16 of the scope of the patent application, wherein if two or more values of X (k) _y (_K) 丨 are the same, the last calculated pixel pair is selected to Predict the pixel value to be interpolated. 29 · —A device for storing complete image graphics for generating progressively scanned video images from interlaced scanning video images, including: a first buffer, a second buffer, a third buffer, and a fourth Buffer O: \ 54 \ 54249DOC is used to store a first decoded 1-picture; used to store a first decoded B-picture; used to store a second decoded B-picture; used to store a first Decoded P-graphics; -4-This paper size is applicable to China National Standard (CNS) A4 specification (210X297mm) .- ίι ^ ------ IT ----- ^ (Please read the Please note this page and fill in this page again) A8 B8 C8 D8 The scope of patent application among them: The previously stored p-graphics or I _ graphics are sequentially the next decoded p_ in the first buffer and the fourth buffer, respectively. The pattern or pattern is replaced, and the previously stored B- pattern is sequentially replaced by the second and third buffers. The next decoded pattern B_ is replaced by the pattern; and when the number of consecutive B- patterns is not greater than At 2 o'clock, the interlaced-progressive conversion process of ΜρΕ () video format, and three consecutive fields are displayed by two The sequence begins with the sequential pattern; when the number of consecutive B_ patterns is greater than 2, the memory blocks used to store B- patterns are used to store new decoded data. (Please read the note on the back first Please fill in this page for further information.)-Binding and printing. Printed by the Central Labor Bureau of the Ministry of Economic Affairs, Shellfish Consumer Cooperative. -5- O: \ 54 \ 54249.DOC This paper size is applicable to China National Standard (CNS) A4 (2 丨0X297 mm)