TW530487B - Color interpolation processor and its color interpolation calculating method - Google Patents

Abstract

The present invention relates to a color interpolation processor and its color interpolation calculating method, and particularly to a color interpolation processor and its color interpolation calculating method applied in a real time image signal processing system using charge couple device (CCD) to perform sampling. The present color interpolation calculating method performs color interpolation by the edge directions weighting concept and local gain method based on that the brightness intensity is primarily determined by the intensity of green color, thereby achieving the purpose of improving interpolation quality. Furthermore, the calculation complexity of the color interpolation calculating method of the present invention is not high. When implementing the color interpolation calculating method of the present invention in an image signal processing system, the required cost is low and thus the cost of the product is greatly reduced.

Description

530487 A7

Field of the Invention: relates to color interpolation processor and basin * (1 p〇latiGn life) of one color-based method of calculating the present invention, particularly, "U Μ ^ Wu member (Charge Couple Device; CCD) for taking '1豕 (Sampling) real-time (Real Time) image signal processing system, a color interpolation processing! And its color interpolation calculation method to improve the efficiency of color interpolation ^ π η vivid processing, j reduce the color Insertion processing time and implementation cost. BACKGROUND OF THE INVENTION: Most digital color photography systems, the old β3, Erxinle, and Yuanduyu use CCD sensors (Sensors) for image sampling. CCD # υυ uses the Photoelectric Effect to convert the incident light that is sensed into electrical signals. Then these electrical 彳 § numbers are converted by analog / digital and digitalized to facilitate image processing and recording. And to reduce The size of the sensor, the color filter array (CFA) specification is one of the most commonly used sampling formats at present. 0 ------------- install · —— (Please read the precautions on the back first (Fill in this page again) The Digital Consumer Sampling Co., Ltd. printed by CCD as the image capturing unit of the Consumer Property Cooperative of the Ministry of Intellectual Property Bureau can be divided into three parts. The first part includes the image processing involved in the CCD sampling system, such as optical Compensation for black level, defect blocking, white balance, and automatic This paper size applies Chinese National Standard (CNS) A4 (210 X 297 mm) $ 530487 A7 and adjustment of brightness contrast V. Description of the invention (white balance, for various CFA performs & ^ _ Ba shirt k number separation and interpolation. Through these processes, color image information corresponding to full color format of every ~ ^ & pixels of the screen can be obtained. Then, after correction and landscaping, ^ ^ Level processing, such as lens flash compensation, color correction, gamma correction, edge field, heart boundary enhancement, etc. 0 When using the CFA sampling format to pull-, hum, a mother sampling point has only one color, that is, the full color system (that is, red, green One-third of the two colors), in order to make up the structure of the colored shirt at each sampling point, complex and time-consuming calculations must be performed, and the two missing colors are interpolated at each sampling point. In order to achieve the purpose of improving the resolution of the sampled image. ------- II ------- (Please read the precautions on the back before filling out this page) Printing Office The explicit interpolation means determining the unknown value of the function in some known sampling points. There are many traditional interpolation methods, such as Nearest Neighbor interpolation, Bilinear interpolation, cubic Interpolation methods such as B-spline (Cubic B-spline) interpolation and Cubic Convolution interpolation. However, these traditional interpolation methods each have their own disadvantages. For example, the nearest neighbor interpolation method and bilinear interpolation method are fast, but the quality of interpolation is generally low, and the images after interpolation often appear severe. Blur, so it does not provide good image quality. At present, video images are increasingly demanding high resolution and high contrast. Nearest neighbor interpolation and bilinear interpolation are no longer applicable. The operations of the cubic B-spline interpolation method and the cubic convolution interpolation method are set by the paper. The paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 530487 A7

Multiple calculations are required for many parameters in 4 ', so the calculation process is too complicated. In the π image signal processing system,' 胄 can provide accurate and good quality interpolation ', but because the calculation is complex and time-consuming, the cost of the product will be increased during implementation. It is also not suitable for real-time digital color imaging systems. Moreover, in the digital color imaging system using the CCD and CFA sampling formats, when the traditional interpolation method is used, the image after interpolation is prone to color fringing and edge blurring, which is the main disadvantage of the traditional interpolation method. It is color noise and color quality, but many slow interpolations are not calculated. The cost of the large image is increased, and the difference is also because the punching area is increased by a small amount of two. The processing system is processed after interpolation. Improve the domain and more. These systems for recording these colors are more based on the image quality methods of the traditional time and the traditional methods. Although these interpolation methods are time-consuming operations and resources in the calculation process. When the operation of interpolation and interpolation that should be interpolated greatly decreases, if applied, the method can change the architectural parameters, which requires high time for calculation. Failure to improve edge sharpness, such as edge-sensitive interpolation, is too complicated, and When it is done dozens of times, it is tedious, and it should be a product that can be used to quickly reduce and reduce the sensitivity of the image. If a sample point is needed for the calculation of the product, the real-time calculation will make ------ ------ ^ --- I ---- ^ ------- II ^ (Please read the unintentional matters on the back before filling out this page) Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs The general description of the eyesight, the middle and the back of the hair, the format of the image at the forensic office, AFC, and, the French yuan single camera is inserted as the system D, CC, on the paper Zhang scale is applicable to China National Standard (CNS) A4 specification (210 X 297 meals) 530487 A7 _ B7 V. Description of invention () (Please read the precautions on the back before filling this page) If the nearest point is interpolated, double line Interpolation methods, such as sexual interpolation, cubic B-spline interpolation, and cubic convolution interpolation, all failed to provide effects. Image processing systems, especially real-time image processing systems, have a good interpolation quality and fast interpolation operations, making the product fail to have both good interpolation quality and performance, and low cost. The main purpose of the present invention is to provide a color interpolation processor and a color interpolation calculation method, and more particularly, it is related to the application in a real-time image signal processing system for capturing images with a CCD. Since the color interpolation calculation method of the present invention has low calculation complexity and can provide good interpolation quality, implementing the color interpolation calculation method of the present invention in an image signal processing system requires low cost and product cost. It is greatly reduced. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs According to the above-mentioned purpose, the present invention provides a color interpolation processor and a color interpolation calculation method thereof, and in particular, it relates to a real-time image signal applied to acquire images with a CCD. Processing system. Since the density of brightness (Y) is mainly determined by the green color, the color interpolation calculation method of the present invention mainly uses the concept of edge weights and local gain to perform green color interpolation operations to improve the image quality after interpolation. Clarity. Furthermore, since the color interpolation calculation method of the present invention is not complicated, the interpolation operation is fast and suitable for application to a real-time image signal processing system. In practice, only two buffers are needed to record the information in the calculation process. Effectively reduce the cost of implementation. This paper size applies to China National Standard (CNS) A4 (210 x 297 mm) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 A7 B7 V. Description of the invention () Brief description of the drawings: Preferred embodiment of the present invention The following figures will be supplemented in the following explanatory texts for more detailed description, wherein: FIG. 1 shows an image data array obtained by CFA sampling according to an embodiment of the present invention. FIG. 2 shows an image data array after CFA sampling according to another embodiment of the present invention. FIG. 3 illustrates an instant color interpolation processing system to which an embodiment of the present invention is applied. Fig. 4 is a flow chart showing the internal operation of the color interpolation processor according to Fig. 3 in an embodiment of the present invention. FIG. 5 is a flowchart of a core calculation according to an embodiment of the present invention. Fig. 6 is a calculation flow chart calculated based on the common parameters in Fig. 4. Fig. 7 is a flow chart showing the calculation based on the horizontal and vertical differences in Fig. 4. Fig. 8 is a schematic diagram showing the separation processing of the horizontal and vertical difference signal flows according to Fig. 7 in an embodiment of the present invention. FIG. 9 is a schematic diagram of the system timing and the first timing and the second timing corresponding to FIG. 8 in an embodiment of the present invention. This paper size applies to China National Standard (CNS) A4 specification (210 X 297 mm) ------ ----- installed ------- order --------- line (Please read the notes on the back before filling out this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 A7 B7 V. Description of the invention () Figure 10 shows the edge weights in one embodiment of the present invention Calculation calculation flow chart. Fig. 11 is a diagram showing the unintended relative timing of the system according to the edge weight calculation process of Fig. 10. Figure 12 is a flowchart showing the calculation of the local mean of B based on the mean of Figure 4. Fig. 13 is a flowchart showing the calculation of the mean value of R and the mean value of G calculated from the mean value of Fig. 4. Explanation of drawing numbers: 10 Real-time color interpolation processing system 12 Image signal 14 Input 16 Analog-to-digital converter 18 Color interpolation processor 20 Vertical and horizontal synchronization signals 22 Buffer 24 Buffer 26 m Outlet 50 m Inlet 52 Common parameter calculation 54 Horizontal and vertical difference calculation 56 Mean calculation 58 Edge weight calculation 60 Interpolation item selection 62 Parallel processing 64 Output 100 Output 102 m Output 104 Signal flow 106 Signal flow 108 Signal flow 1 10 Signal flow This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm) nni I tn n ϋ i an nnn 1 · n (inmnnn OJ · nn I nn -ϋ n · (Please read the notes on the back first (Fill in this page again) 530487 A7 _Έ7 V. Description of the invention () Printed by the Intellectual Property Bureau of the Ministry of Economic Affairs, printed by the Consumer Cooperative 1 1 Signal flow 1 14 Multiplexer 116 Divider 118 Register 120 Multiplier 122 Multiplier 124 Multiplexer 126 Multiplexer 128 Register 130 Sampled data value in the second column of the image data array 200 Adder 202 Adder 204 Multiplexer 206 Multiplexer 208 System timing 210 Subtractor 212 Horizontal and vertical difference signal flow 214 Multiplexer 216 Divider 218 First timing 220 Second timing 222 Register 230 Vertical difference signal flow 240 Horizontal difference signal flow 300 Adder 302 Signal flow 304 Register 306 Divider 308 Multiplier 3 10 more Tool 3 12 Accumulator 400 Adder 402 Adder 404 Adder G Green Color

R color color color

B color color blue ---------------- 1 ---- order --------- * 5 ^ (Please read the precautions on the back before filling this page ) Explain in detail that the paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 530487 Printed by the Consumers ’Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 — —-—__ B7_ V. Description of the invention Specifications sampling, so each sampling point is only one-third of the full color system (ie, three colors of red, green, and blue). In order to complement the complete color structure, 'the missing two colors must be interpolated at each sampling point. Since the resolution of the image is mainly determined by the density of brightness (Luminance Density), the density of brightness (γ) can be defined by equation (1): Y = 0.59G + 0.1 lB-f 0.3 R (1) by the equation ( 1) It can be seen that the density of brightness mainly depends on the signal of green color (hereinafter referred to as G), so the g signal has a larger weight (Weight), so when calculating the color interpolation, the interpolation of G is more than the red color (Hereinafter referred to as R) interpolation and blue color (hereinafter referred to as B) interpolation are important. Please refer to FIG. 1, which shows an image data array sampled by CFA in one embodiment of the present invention, and the image data is an image data array formed by sampling data of 4 rows and 9 rows, and the first row And the third column is composed of R sampling data values and G sampling data values, and the second and fourth columns are composed of G sampling data values and B sampling data values. Since the interpolation of 0 is more important than the interpolation of R and B, it is first assumed that the sampled data value of the image is B ”(as shown in Figure 1, in the second and fourth rows of the image data array). Intersection), explain the calculation method of G's interpolation value. In order to improve the quality of interpolation, the parameters of the edge of Bn need to be calculated. First, calculate the vertical difference (ΔΠ and horizontal difference (research)) of By, such as equations ⑺ and Equation ⑺ shown II ---- · 1 III --- Order --- ------ (Please read the precautions on the back before filling out this page) (2) 530487 A / B7 V. Description of the invention ( △ F4 = | G 1 4-G3 4 丨 AHS = IG23-G25! (3) The vertical difference (△ &) is obtained by subtracting the sampled data value from the sampled data value and then taking its absolute value, and the horizontal The difference (Δ ^ 5) is obtained by subtracting the G23 sampled data value from the G23 sampled data value and then taking its absolute value. Then use the concept of Edge Directions Weighting to calculate the initial interpolation value of G in Bn. ((¾), as shown in equation (4), if Δ / ί = 0 and △ factory = 0 [[j AH + AV 2 otherwise G {^ = AH x ΔΚ + AVx-

AH (4) AH: AH + AV ΔΚ 77P- + AVx (l- AH + AV v

AH + ΔΚ AV

AH + AV Printed by Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs where G24 represents the initial value of G in b24, a / = (G25 + G23) / 2 and ZF- (Gi4 + G34) / 2 respectively represent horizontal directions The average value (that is, the horizontal average value) and the vertical direction average value (that is, the vertical average value). It can be seen from equation (4) that when the difference in a certain direction is large, the weight in that direction decreases, so that Avoid image blur. If the vertical difference (△ 〇 and the horizontal difference (Δ / 〇 are zero at the same time), the initial value of the interpolation of 0 is obtained by using the average of the horizontal average and the vertical average. ----- If-- IIII -------- Order --------- (Please read the precautions on the back before filling this page) 10 530487 A7 V. Description of the invention (

Then calculate the local mean value of B 24 from / (heart), as shown in Fang Xiwang formula (5) (5) where 4 is the enclosed value of B in ~ in the first figure. Then G's interpolated final value ((42)) is G's interpolated initial value (like LV interpolated initial value ((/ 24) multiplied by the sampled data value from, and then divided by the local mean (¾), such as the equation (6) shows LM {B2A) (6) where Bm / ZA ^ 24) is the local gain at B24. The interpolated value of R calculated in ~ 'is first calculated by the average of four points (ie, Ri5, R13, R35, and R33), as shown in equation (7)-4 (7) ---- -1 ΙΙΙΙΙΙΙΙ »— — — — — — II ^« — — 111 — — (Please read the notes on the back before filling out this page) Printed by the Intellectual Property Office of the Ministry of Economic Affairs and Consumer Cooperatives and then interpolated by R (々μ) Π can be obtained as shown in equation (g). This paper size applies the Chinese National Standard (CNS) A4 specification (210 x 297 mm) 530487 A7 B7 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of invention () R24 = LM (R24): LM (Bu) (8) In the same way, the final interpolation value of G at R 3 3 () is as shown in equation (9). G3 (32) =-^ 33

LM (R (9) 33. where (¾) is the initial value of G interpolated to R33 and R33 / ZM (i? 33) is the local gain of R33. The calculation of (¾ 丨 > will correspond to The data can be obtained by substituting into the formula shown in equation (4). In addition, the local mean value ZM (i? 33) of R33 is shown in equation (10). LM (R33) (10) where i? 32 is the first The interpolation value of R in G32 is shown in the figure. The interpolation value of B in R33 is calculated by first averaging the four points (ie B22, B24, B42, and B44), as shown in equation (1 1) Shown LM (B33) B22 ^ 24 + ^ 42 + ^ 44 4 (11 (Please read the precautions on the back before filling this page) 12 This paper size is applicable to China National Standard (CNS) A4 (210 X 297 mm) ) $ 3〇4B7

Then the interpolation value (533) of B can be obtained as shown in equation (1 2)

R 33 LM (R33) (12) where IM (43) is obtained by using equation (1 0). The interpolation value of B and the interpolation value of R to be calculated in G23 are calculated first. Equation (13) shows that IM (G23) = + 4 4 (13) The interpolation value of R in Gw (and 23) can be obtained from equation (14) ^ 13 + ^ 33 ^ ^ 23 LM (G23) (14) The interpolated value of B in Gw (name 23) can be obtained from equation (15) II 1 — — — — — — 111 «-I ί (Please read the notes on the back before filling this page) Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs 24 5 + ο 22 5 II 23 < 5

23 G 23 Samples for G. For the example of He Ruming's description of different descriptions, this paper applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 530487 A ?. V. Description of the invention ( ) The method of interpolating B and R of the data value, the method of interpolating G and R of the sampled data value of B, and the method of interpolating G and r of the sampled data value. The method of interpolation of B is gone, so it can be easily applied to the image data array in Fig. 1 to find the corresponding color interpolation of each of the m sampled data values. Please refer to FIG. 2, the green +4 is not the image data obtained by CFA sampling in another embodiment of the present invention. The image data array in Fig. 2 is composed of sample data of columns M and N, and the image data array is composed of j π α shell material, and% is any integer not less than 3 'ft Ν is any integer not less than 6. The color interpolation calculation method provided by the present invention is not limited to the image data array as shown in FIG. 丨, but can be applied to an image data array composed of a plurality of columns and a plurality of rows to obtain an interpolation value for each color. Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs Next, the color interpolation calculation system provided by the present invention will be explained. Please refer to FIG. 3, which shows the real-time color interpolation processing system applying the present embodiment. . In the real-time color interpolation 10, the image signal 12 captured by the CCD is subjected to analog / digital conversion processing from the input terminal 14 digital converter 16 to list the sampled data values to the color interpolation processor 18, and Vertical and # 2 0 are also input to the color interpolation processor 18 to control the internal operation of the colorizer 18. When the value of the sampled data in the first column is input to the processor 18, the color interpolation processor 18 did not start to perform one of the methods of the actual processing system. The input analogy is input in units of horizontal synchronous interpolation and color interpolation processing. ' And i I--111111! ·· 1 III 1 丨 —Order · --II --- I (Please read the precautions on the back before filling out this page) 14 This paper size applies to China National Standard (CNS) A4 specifications ( (210 X 297 mm) 530487 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention () The value of the sampled data in the first column is stored in the buffer 22. Then when the sampled data value of the second column is input to the color interpolation processor 18, the color interpolation processor 18 has not started processing yet, and the sampled data value of the second column is stored in the buffer 24, Until the sampling data value of the third column is input to the color interpolation processor 18, the sampling data value of the first column and the sampling data value of the second column are input to the color interpolation processor from the buffer 22 and the buffer 24 at the same time. 1 8, and start arithmetic processing. After the corresponding color interpolation is calculated, the red (R), green (G), and blue (B) signals are output from the output terminal 26, respectively. In other words, when the sample data value of the i-th column and the sample data value of the i + 1th column are input to the color interpolation processor 18, the color interpolation processor 18 will store them in the buffer 22 and the buffer, respectively. 24 does not start processing, and the color interpolation processor 18 does not start arithmetic processing until the sampled data value of the i + 2 column is input to the color interpolation processor 18. Therefore, applying the real-time color interpolation processing system of the color interpolation calculation method provided by the present invention, only two buffers can be used for real-time signal processing, and the signals of three primary colors of R, G, and B are output, thereby greatly reducing The difficulty and cost of implementing technology. Please refer to FIG. 4, which illustrates a flowchart of the internal operation of the color interpolation processor according to FIG. 3 in an embodiment of the present invention. First, the sampled data value in the first column, the sampled data value in the i + 1th column, and the sampled data value in the i + 2th column are input from the input terminal 50 of the color interpolation processor 18, and the paper size is applicable to the Chinese country Standard (CNS) A4 specification (210 X 297 mm) -------------- ^ ----------------- ^ (Please read first Note on the back, please fill out this page again) 530487 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 ______B7_____ V. Description of the invention () Computations For Common Term 52 and Horizontal and Vertical Difference Calculation (Differential Computations For

Horizontal / Vertical Edges) 54. Then use the result of the common parameter calculation 52 to perform the mean calculation (Mean Computations) 56 to find the local mean and surrounding mean of the corresponding color, and the result of the common parameter calculation 5 2 and the horizontal and vertical difference calculation 54 to perform the edge weight The Edge Directions Weighting Computation 5 8 »To find the final interpolation value of G. After obtaining the local mean value and peripheral mean value of each color and the final interpolation value of G, the interpolation item selection (Selection for Interpolated Components) 60 is performed, and an appropriate output signal is selected, which is then combined with the result of the common parameter calculation 52. The parallel processing (Parallel) 62 outputs a corresponding interpolation value at the output terminal 64. Please refer to FIG. 5 ′, which shows a flowchart of the core calculation of an embodiment of the present invention, and this flowchart is described based on the image data array of FIG. 1 and the internal operation flow of FIG. 4. The core calculation shown in FIG. 5 is explained by calculating the interpolation values of the sampled data values in the second column of the image data array in FIG. 1 (column κ, ^ J, and i i + l). For example, in The output terminal in Fig. 5 is an output signal stream 4 1 0, and the signal stream 4 10 is formed by an interpolation value of R of each sampled data value in the second column f, that is, the first 1 + 1 column), and The output terminal 102 in the sI ^ μ v 0 figure is the output signal stream 412, and the signal stream 412 is the corresponding G of the sampled data value in the 2nd column (i + 1th column) of Tu 0, a & The interpolated final value of β and the interpolated value of β have reached 0 (please read the precautions on the back before filling in this page). Installation • Line 16 530487 Printed by A7 B7, Consumer Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs 5. Description of Invention () As shown in Figure 5, to calculate the corresponding interpolation value of each sampled data value in the second column (ie, i +} column), at least the signal flow 1 04, signal flow 1 06, and signal flow shown in Figure 5 are required. !! 08. Signal flow 丨 丨 〇 and signal flow 1 丨 2. When calculating the interpolation values of R and G in the second row of G23 of the image data array of FIG. 1, according to equations (14) and (15), from the image data array of FIG. G23 sampled data value '3' is selected from the 2 rows of sampled data values' 3 through the control multiplexer 丨 4 to output the mean value of G23 (ZM (G23)) in the signal flow n 2 and pass through the 023 sampled data value. The divider 1 1 6 divides and sends the division result to the temporary register n 8, and then outputs to the multiplier i 20 and the multiplier 1 2 2 respectively, and then controls the multiplexer 丨 2 4 and the multiplexer at the same time. 1 2 6. From the signal stream 104, it is sent to the multiplier 120 through the register 128 and multiplied by the division result. At the output terminal 100, the interpolation value of 23 in R is output at G23, and from the signal stream 106. Send AB4 to the multiplier 122 and multiply the result of the division 'at the output terminal 102 to output the interpolation value i23 of B at g23. When calculating the interpolation value of R of B24 in the second column of the image data array in Figure 1 -4 and the final interpolation value of G (42), according to equation (4), equation (5), and equation (6) Shown 'select the b24 sampling data value from the second column, and control the multiplexer 1 1 4' to output the local mean value of B24 in the signal flow 1 1 0 (ZΜ〇δ24)) and sample the data value with B〗 4 After being divided by the divider, the divider is sent to the temporary register, which is then output to the multiplier, which is 20 and the multiplier 1 22 ', and then simultaneously controlled by the multiplexer 1 24 and the multiplexer. Device 1 26, will signal the paper standard outline + @ Θ home standard (CNS) A4 specifications (210 X 297 public meals _) ------------- install ------- -Order in —9 1_ «ϋ n an I (Please read the notes on the back before filling out this page) 530487 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 ———— —_B7 __---- V. Description of the invention () The peripheral average value of R in B24 in stream 1 12 (ζμ (Λ24)) is sent to multiplier 120 through register 128 and multiplied by the division result, and is output at the output terminal 100 and interpolated by R in Bn i24, and the signal flow 08 in the G of the B24 Initial interpolation fed to the multiplier 122 is multiplied by the result of the division, to the output terminal 102 outputs the final value for the interpolation period Shu G at the Bn). The following description is about how to generate each signal flow in Figure 5 during implementation. Please refer to Fig. 6, which shows a calculation flowchart calculated based on the common parameters in Fig. 4, and the image data array in Fig. 1 is used for explanation. The sampled data value of the first column (i.e., the ith column) and the sampled data value of the third column (ie, the i + 2) of the image data array in Figure 1 are sequentially from Ri i as shown in Figure 6. And "Start to add the corresponding input adder 200, and after the addition result is shifted two bits to the right, the common parameters such as ALi, AV2, AR3, and AV4 are sequentially output as shown in Figure 6" and this adder 200 outputs The common parameters of Fig. 5 constitute the signal flow 1 0 4 in Fig. 5, and it can be deduced from this that the red common parameter ARx = (Rlx + R3x) / 2 'and the common parameter aVx = (〇1x + (} 3x) / 2 ( That is, the vertical average value), where X is any integer not less than 1. At the same time, the image data of the first image > the sample data value of the second column (that is, the first and second columns) and the delay two time series (Delay 2 Clocks) and then input the sampled data values in the second column (ie, the i + 1 column), as shown in Figure 6, respectively, starting from G21 and sequentially adding to the corresponding input adder 2 0 2 to add, and add, 纟After shifting two bits to the right, Shiguo will output 0, 0, Ah, and Ab4 in sequence as shown in Figure 6. This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297). Li) -------- 111 ---------- Order --------- (Please read the notes on the back before filling this page) 530487 Intellectual Property Bureau of the Ministry of Economic Affairs Printed by employee consumer cooperative A7 B7 V. Description of the invention () Parameter 'Where the sampling data value in column 2 (ie, 丨 +1 column) is delayed by the two-time series input relationship, so the first system time sequence ( During the addition of the First System Clock and the Second System Clock, no addition result is generated. Therefore, the result of the addition when the first system timing and the second system timing are both “〇” And the common parameters output by this adder 202 constitute the signal flow 10 in FIG. 5 and can be deduced from this, the common parameter AHx = (G2x + g2U-2) V2 (that is, the horizontal average), and The blue common parameter ABx = (B2x + B2 (xa) / 2, where X is any integer not less than 1. Please refer to Figure 7, which is shown as calculated based on the horizontal and vertical difference in Figure 4. The calculation flow chart is based on the image data array in Figure 1. The first row (i.e., the i-th column) of the image data array in Figure 1 is sampled. The data value and the sampled data value in the second column (ie, i + 1) are input to the multiplexer 204 as shown in FIG. 7. The multiplexer 204 is controlled by the system timing 208, and the sampled data value in the first column and the The sampled data value of G is selected from the sampled data value of the second column, and is sequentially output from g2 1 to the subtractor 210 0 as shown in FIG. 7 At the same time, the third column of the image data array of FIG. 1 (that is, Column i + 2) and the sample data value of column 2 (ie, 丨 +1 column), enter the multiplexer 206 as shown in Figure 7, where the sample data value of column 2 is greater than the sample data value of column 3. The sampling data values in the column are delayed by two system timing inputs to the multiplexer 206. However, the paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ------------ -Install -------- Order --------- line (please read the notes on the back before filling this page) 530487 Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Invention After the description (), the multiplexer 20 is also controlled by the system timing 208, and the sampling data value of Q is filtered out from the sampling data value of the third column and the sampling data value of the second column, and as shown in FIG. 7 Output from 0 to the subtractor 21 sequentially. When the multiplexer 204 first starts to output Gu, the sampling data value in the second column is delayed by two system timing inputs compared to the sampling data value in the third column. The filter 206 does not output the result of screening in all schematic diagrams of an embodiment of the present invention as 0, indicating that there is no signal at the time point of the corresponding signal, or the signal is not adopted by one embodiment of the present invention. After the filtering result of the multiplexer 204 and the filtering result of the multiplexer 206 are input to the subtractor 2 10 at the same time for processing, and the subtraction result is shifted to the right by two bits, sequentially output 0, ΔK2, Δ ^ 3, △ &, Δ / ί5 and other horizontal and vertical differences constitute the horizontal and vertical difference signal flow 2 1 2, where 0 is the output of the subtractor 210 at the first system timing, and the horizontal and vertical difference signal flow 212 The vertical differences such as △ 6 and ΔΚ4 are obtained according to equation (15). △ B = | G1x-G3x | (15) Horizontal and vertical difference signal flow 2 1 2 The horizontal differences such as Δ // 3 and Δ / ί5 The quantity is obtained according to equation (16) | G2x-G2 (x.2) I (16) The K degree of this paper applies the Chinese National Standard (CNS) A4 specification (210 χ 297 mm) -------- ----Install ---- I 1 II Order --- ---- (Please read the precautions on the back before filling this page) 530487 A7 B7 V. Description of the invention (. Refer to Figure 8 for the month The drawing is a schematic diagram of the separation processing of the horizontal and vertical differential signal flows according to FIG. 7 in an embodiment of the present invention. In order to horizontally and horizontally separate the horizontal and vertical differential signal flows 2 in FIG. 7 And vertical difference of 1 point Therefore, the horizontal and vertical difference signal flow 212 is output from the subtractor 210 in FIG. 7 to the multiplexer 214 in FIG. 8. Referring to FIG. 9 ', it is shown as corresponding to the first embodiment of the present invention. Schematic diagram of system timing and first and second timing diagrams in Fig. 8. The system timing 208 in Fig. 9 is input to the divider 216 in Fig. 8. After processing, the first timing 218 is output to as much as in Fig. 8. The worker 214 and the second timing 220 are output to the register 222 of FIG. 8, wherein the cycle time of the first timing 218 and the first k sequence 2 2 0 are twice the cycle time of the system timing 208. ------------- Equipment -------- Order · (Please read the notes on the back before filling out this page) Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs at the same time Referring to FIG. 8 and FIG. 9, the multiplexer 214 is controlled by the first timing 218 in FIG. 9, so that the multiplexer 214 outputs the horizontal and vertical difference when the first timing 218 is the south level. The horizontal difference in the signal flow 2 1 2 is stored in the temporary register 222. When the first timing 218 is at a low level, the vertical difference in the horizontal and vertical difference signal flow 212 is output to the temporary storage. 222, and the output of the register 222 is controlled by the second timing 220 in FIG. 9 to obtain the vertical difference signal flow 230 and the horizontal difference signal as shown in FIG. 8 in the same system timing 208 Stream 240, where Δ // 3 in the horizontal differential signal stream 240 corresponds to the vertical differential signal stream in the first system timing. 21 This paper size is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm t) Printed by line 530487 Employee Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs Α7 Β7 V. Description of the invention (ΔK2 in 2 3 0, Δ / ί5 in horizontal difference signal flow 2 4 0 is corresponding to the vertical in the second system timing ΔK4 in the differential signal flow 230, and so on. Please refer to FIG. 10, which illustrates a calculation flowchart of edge weight calculation in one embodiment of the present invention. The horizontal differential signal flow 240 and the vertical differential signal flow 2 3 0 output from the multiplexer 2 1 4 in FIG. 8 are sent to the adder 3 0 0 in FIG. 10 for addition to obtain The signal flow 302 composed of the addition result (AJ / x +), for example: (Δ / ί3 + AF2), (Δ / ί5 + ΔΚ4), etc., and then the signal flow 302 is sent to a register controlled by the second timing 220 304 is temporarily stored for "zero check". The so-called "zero check" is to check whether ΔΚ and Δί / (in the signal flow 302 are zero, and if both are zero 'according to equation (4), then the counting step is not continued, but according to the equation (4) to calculate the initial value of G's interpolation. When Δ 匕 and Δ // (χ + 1) are not zero, the signal flow 302 continues to be sent to the divider 306, according to equation (4) As shown in the figure, the signals corresponding to the horizontal and vertical difference signal flows 2 1 2 are divided and then sent to the multiplier 3 08. At the same time, the 'signal flow 1 06 and signal flow 104 are input to the multiplexer 3 1 0, Through the control of the first timing 218, the multiplier 310 can output the horizontal average term in the signal flow 106 to the multiplier 308 at the high level of the first timing 218, according to equation (4) The output items corresponding to the divider 306 are multiplied and sent to the accumulator 3 1 2. At the low level of the first timing 218, the vertical average term in the signal stream 104 is output to the multiplier 3 08. Applicable to China National Standard (CNS) A4 specification (210 X 297 mm) according to the paper size I --- I ---- I --- · II ----- ^ -------- -* 5 ^ (Please read first Note on the back, please fill in this page again) 530487 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A7 B7 V. Description of the invention () As shown in the formula (4), the output items corresponding to the divider 306 are multiplied and sent to the accumulation The device 3 1 2 is added to the previous item temporarily stored in the accumulator 3 1 2 and is controlled by the system timing 208 to output the figure 5. It is composed of the initial value of G of each sampled data value. Signal flow 1 0 8. Please refer to Figure 11 for a schematic diagram of the relative system timing based on the edge weight calculation process of Figure 10. In the first system timing, there is no signal processing. 'Represented by "0". Then at the second system timing, and ΔΚ2 are input to the adder. At the third system timing, the addition result of ^ 3 and AF2 is obtained, and controlled by the second timing 220 , Divide (Δ // 3 + ΔΚ2) to divider 3 06 and divide it by Δκ2 in the horizontal and vertical difference signal flow 2 1 2 to obtain ΔΚ2 / (Δ // 3 + ΔΚ2), and (ΔΚ2) It is sent to the divider 306 and divided by Δ / ^ in the horizontal and vertical difference signal stream 212 to obtain Δ / / 3 / (Δ / ί3 + ΔΚ2). At the fourth system timing, the multiplexer 310 is controlled via the first timing 2 1 8 so that AΗ3 in the signal flow 106 is sent to the multiplier 3 08 and △ Κ2 / (Δ ^ 3 + ΔΚ2) and multiply, and send the multiplied result to the accumulator 312 to wait, and then control the multiplexer 3 1 0 through the first sequence 2 1 8 so that AV2 in the signal flow 104 is sent to the multiplication 308 and Δ / ί3 / (Δ // 3 + ΔΚ2) and multiply, and send the multiplied result to the accumulator 3 1 2 and accumulate with the previous term. At the same time, Δ // 5 and △ G are input to the adder. . Then through the control of system timing 208, the initial value of G (Gg) is output at G22 of B22 at the 5th system timing. At this time, 23 paper sizes are applicable to China National Standard (CNS) A4 (210 X 297 mm). (%) ≪ · ϋ IM «n am— I nim · ϋ lnn I · nnnnnnn eJe m I If I in II c Please read the notes on the back before filling out this page) 530487 Printed by the Consumers’ Cooperative of Intellectual Property Bureau of the Ministry of Economic Affairs System A7 B7 V. Explanation of the invention () The processing steps of Δ // 5 and ΔΚ4 are sequentially processed as the eight-pound 3 and δκ2 of the third system sequence, and the internal value of G from each sampled data value is continuously output from the accumulator 3 丨 2. The signal flow composed by the initial value is 1.08. Please refer to Figure 12 for a flow chart of the calculation of the local average of B based on the average of Figure 4. In order to calculate the signal flow consisting of the local mean of β in Fig. 5, firstly, the sampled data value of β is filtered from the second column of the image data array in Fig. 1, and the output terminal 102 in Fig. 5 The interpolated value of B is filtered out from the output signal stream 412. As shown in FIG. 12, the interpolated value of β at G23 and the sampled data value of B26 are sent to the adder 400 for addition and shifted right by 2 bits, and then Send the addition result and ΑΒ4 in the signal stream 106 to the adder 402 and add 2 digits to the right to get the signal flow consisting of the local average of Η in Figure 5. 〇 〇 Α4 is αβχ = (β2χ + Β2 (χ 2)) / 2 Derived. Please refer to Figure 13 for a flowchart of the calculation of the mean value of R and the mean value of G calculated from the mean value of Figure 4. In order to calculate the signal flow 1 1 2 composed of the peripheral average value of R and the average value of G in FIG. 5, the signal flow outputted by the adder 200 in FIG. 6 and the signal flow 104 delayed by two timings are sent. Add to the adder 404 to obtain the signal flow composed of the peripheral average of R and the average of G in Figure 5. Signal flow 104, signal flow 106, signal flow 108, and letter paper size in Figure 5 are applicable to China National Standard (CNS) A4 specification (210 X 297 meals) ------II丨 丨 Install ------- Order --------- line (please read the precautions on the back before filling this page) 530487 A7 V. Description of the invention (Number flow 1 1 0 and signal flow 1丨 2 are obtained by # 4 free from the above operations, and the image data array 歹 in Figure 1 can be obtained in order. 丨, —, the corresponding interpolation value of each sampled data value. Then perform the calculation in Figure 4. Interpretation, Xigu (selection of 60 for the cutting project, select the appropriate output signal, and then process it in parallel with the results of the common parameter calculation 筲 9 and double T 52 to produce a corresponding interpolation value at the output 64 The advantage of the present invention is to provide a color interpolation process H and a color interpolation calculation method. Applying the color interpolation calculation method of the present invention to an image signal processing system for capturing images with a charge-coupled element can provide good Interpolation quality, and at the same time reduce the calculation time of interpolation to improve the efficiency of interpolation processing, so it is particularly suitable Applied in the real-time image signal processing system 'can effectively reduce product costs. ------------ II ^ --- (Please read the precautions on the back before filling this page) -a- If familiar Those skilled in the art understand that the above is only a preferred embodiment of the present invention, and is not intended to limit the scope of patent application of the present invention. Any other equivalent changes made without departing from the spirit disclosed by the present invention Or “modifications” should be included in the scope of patent application below. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Stupid Ministry of Economic Affairs 5 2 This paper size applies to China National Standard (CNS) A4 (210 X 297 mm)

Claims (1)

530487 A8 B8 C8 D8 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 6. Application for patent scope 1. A color interpolation calculation method is used in an image processing system. The color interpolation calculation method at least includes: providing an image A data array, wherein the image data array is composed of a plurality of shirt-like sampling data sets of the image data array of M rows and N rows, wherein the image sampling data sets are composed of a first data row and a second data row The first data row is composed of a plurality of first sampling data values and a plurality of first sampling data values, and when the first sampling data values are of a third color type, the second sampling data The value is a first color type, and the first duty line is composed of a plurality of third sampling data values and a plurality of fourth sampling data values, and when the third sampling data values are the first color type The fourth sampling data value is a second color type; the one in the second image row of the first image sampling data set of the first image sampling data set is one of the image sampling data sets One of the third sampling data values is subjected to a first interpolation process, where the first interpolation processing includes at least: performing a first first color type average calculation to obtain the third sampling data value A first first color type mean value; performing a first second color type interpolation calculation to obtain a first second color type interpolation value for the third sampled data value; and performing a first third color Type interpolation calculation to obtain the first and third color type interpolation values of the third sampled data value; This paper size applies the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ^ ----- --- ^ --------- (Please read the notes on the back before filling out this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 —_J VI. Application scope of patents for these images One of the second image sampling data sets in the first data row of the second sampling data set, one of the second sampling data values is subjected to a second interpolation process, wherein the second interpolation The interpolation process includes at least: performing a second A color type mean value calculation to obtain a second first color type mean of the first sample > material value; a first second color type interpolation calculation to perform a second sample data value Interpolation of a second and second color type; and performing a second and third color type interpolation calculation to obtain a second and third color type interpolation value of the second sample lean value; sampling data sets of the images One of the fourth sampling data values in the N-1 row of the second data column of the first image sampling data set is subjected to a third interpolation processing, wherein the third interpolation The processing at least includes: performing a second color type average calculation to obtain a second color type average of the fourth sampled data value; performing a first first color type interpolation calculation to obtain the fourth sampled data A first first color type interpolation final value of the value; and a third third color type interpolation calculation to perform a third third color type interpolation value of the fourth sampled data value; Image sampling data One of the first sampled data values in the first data column of the second image sampling data set of the first data row is subjected to a fourth interpolation process, where the fourth internal Interpolate to 27 paper sizes Applicable to China National Standard (CNS) A4 specification (210 X 297mm t). I --- I-- · I ------ ^ --------- ( Please read the notes on the back before filling this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 A8 B8 C8 D8 t. The scope of patent application is small and includes: Perform a third color type mean calculation to find the first A third color type average of the sampled data value; performing a second first color type interpolation calculation to obtain a second first color type interpolation end value of the first sampled data value; and performing a third The second color type is interpolated to obtain a third second color type interpolation value of the first sampled data value. 2. The color interpolation calculation method described in item 1 of the scope of patent application, wherein the first color type is green. 3. The color interpolation calculation method described in item 1 of the scope of patent application, wherein the second color type is blue. 4. The color interpolation calculation method described in item 1 of the scope of patent application, wherein the third color type is red. 5. The color interpolation calculation method described in item 1 of the scope of patent application, wherein the value of M above is any one of the first integers not less than 3. 6 · The color interpolation calculation method described in item 1 of the scope of patent application, wherein the value of N above is any second integer not less than 6. 28 This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) ft an sn nin —ϋ n an I · m —al mmwmmw i ft fi m 1 · · mw ιμιρ Μ · * »w · * I iw (Please read the notes on the back before filling this page) 530487 Printed by A8 B8 C8 D8, member of the Intellectual Property Bureau of the Ministry of Economic Affairs and Consumer Cooperatives 6. Scope of patent application 7. As stated in the color of the scope of patent application item 1 Interpolation calculation method, wherein the above-mentioned first first color type average calculation is the second sampling data of the second sampling data values in rows N-3 of the first data column of the first image sampling data set Value, the second sampling data value of the second sampling data values in the first data row of the first image sampling data set at line N-1, the first data row of the second image sampling data set Of the second sampling data value of the second sampling data value in the N-3 line and the second sampling data value of the N-1 line in the first data column of the second image sampling data set After adding the second sampling data value 'and performing an average calculation The first color to obtain a first type of average. 8. The color interpolation calculation method as described in item 1 of the scope of the patent application, wherein the above-mentioned second first color type average calculation is in line N-2 of the second data column of the first image sampling data set The third sampling data value of the third sampling data values, the third sampling data value of the third sampling data values in the N row of the second data column of the first image sampling data set, the The third sampling data value of the third sampling data values in the second data row of the second image sampling data set and the third sampling data value of the second sampling line and the Nth row in the second data sampling data set. After the third sampling data values of the third sampling data values are added, an average calculation is performed to obtain the average value of the second first color type. 9. As for the color interpolation calculation method described in item 1 of the scope of patent application, this paper size applies to China National Standard (CNS) A4 specification (210 X 297 meals) ---- 111 -------- ------ Order · -------- (Please read the notes on the back before filling out this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 A8 B8 C8 _ D8 _______ The scope of the patent, wherein the above-mentioned second color type average calculation is the fourth sampling data value, the fourth sampling data value, the fourth sampling data value, the fourth sampling data value, the fourth sampling data value in the second data column of the first image sampling negative material group, the The fourth sampling data value of the fourth sampling data values in the N-1 row of the second data row of the first image sampling data set, and the N + th of the second data row in the first image sampling data set. The fourth sampling data value of the fourth sampling data values in line 1 and the third sampling of the third sampling data values in line N-2 of the second data column of the first image sampling data set After the first and second color type interpolation values of the data values are added, an average calculation is performed to obtain the second Mean color type. 1 〇 The color interpolation calculation method described in item 1 of the scope of the patent application, wherein the above-mentioned third color type average calculation is performed on line N-4 in the first data column of the second image sampling data set. The first sampled data value of the first sampled data values, the first sampled data value of the first sampled data values in line N-2 in the first data column of the second image sampling data set, The first sampling data values of the first sampling data values in the N row of the first data row of the second image sampling data set and the N-th of the first data rows in the second image sampling data set. After adding the second and third color type interpolation values of the second and third sample data values of the second sampled data values in 3 rows, an average calculation is performed to obtain the average value of the third color type. 1 1. The color interpolation calculation method described in item 1 of the scope of the patent application, wherein the first and second color type interpolation calculations described above take the first image to the paper size and apply the Chinese National Standard (CNS) A4 specification (210 X 297mm t) H * m— m ϋ— ϋ— 1 un 1 ^ 1 mn I HI in It If n »In HI 0% 0 m HI __1« 11 inn I (Please read the precautions on the back before (Fill in this page) 530487 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A8 B8 C8 D8 VI. The fourth sample data value of the fourth sample data value in the second data column of the patent application sample data group The four sampling data values are added to the fourth sampling data values of the fourth sampling data values in the N-1 row of the second data column of the first image sampling data group, and then divided by 2 to obtain a first average. Value, and then multiply the first average value by the third sampling data value of the third sampling data values in the N-2 rows of the second data column of the first image sampling data set, and divide by the third sampling data value. A first color type average value to obtain the first and second color type interpolation values. 1 2. The color interpolation calculation method described in item 1 of the scope of the patent application, wherein the first and third color type interpolation calculations above are N-th in the first data column of the first image sampling data set. The first sampling data values of the first sampling data values in 2 rows and the first sampling data values of the first sampling data values in row N-2 of the first data column of the second image sampling data set The data values are added and divided by 2 to obtain a second average value, and then the second average value is multiplied by the third sampling data in row N-2 of the second data column of the first image sampling data set. The third sampled data value is divided by the first first color type average value to obtain the first third color type interpolation value. 1 3 · The color interpolation calculation method described in item 1 of the scope of the patent application, wherein the second and third color type interpolation calculations above are N-th in the first data column of the second image sampling data set. The first sampling data value of the first sampling data values in 2 rows and the first sampling data value of the first sampling data values in row N of the first data column of the second image sampling data set 31 This paper size is in accordance with China National Standard (CNS) A4 (210 X 297 mm) nnmnlnu If iunm I. Unnni I n 1 · 0, I dn · ϋ nnnn I (Please read the precautions on the back before filling this page) 530487 Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs A8 B8 C8 D8 6. Add the patent application range values and divide by 2 to get a third average value, then multiply the second average value by the second image sampling data group Divide the second sampled data values of the second sampled data values in the N-1 row of the first data column by the second first color type average to obtain the second and third color type interpolation values . 1 4. The color interpolation calculation method described in item 1 of the scope of the patent application, wherein the second and second color type interpolation calculations are N-th in the second data column of the first image sampling data set. The fourth sampling data value of the fourth sampling data values in row 1 and the fourth sampling data value of the fourth sampling data values in row N-1 of the second data column of the second image sampling data set The data values are added and divided by 2 to obtain a fourth average value, and then the fourth average value is multiplied by the second sampling data in line N-1 of the first data column of the second image sampling data set. The second sampled data value is divided by the second first color type average value to obtain the second second color type interpolation value. 15. The color interpolation calculation method according to item 1 of the scope of patent application, wherein the first first color type interpolation calculation at least includes: performing a first difference calculation to obtain a first level difference And a first vertical difference; perform a first average calculation to obtain a first horizontal average and a first vertical average; perform a first first color type interpolation initial value calculation to obtain Obtain an initial value for the first and first color type interpolation; and 32 paper sizes applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) ϋ n In mmnm--= I il I to In 11 *-- n In It i -1— 一 οI, I n fn nnn ϋ i I (Please read the precautions on the back before filling out this page) Printed by the Intellectual Property Bureau Employee Consumer Cooperative of the Ministry of Economic Affairs 530487 A8 B8 C8 D8 6. Scope of patent application Multiplying the initial value of the first first color type interpolation by the fourth sampling data value of the fourth sampling data values in the N-1 row of the second data column of the first image sampling data set, Divide by the second color type mean 'to get the first Interpolating a color type of the final value. 16. The color interpolation calculation method as described in item 15 of the scope of the patent application, wherein the above-mentioned first difference calculation includes at least: line N-2 in the second data column of the first image sampling data set The third sampling data value of the third sampling data values minus the third sampling data value of the third sampling data values in the second row of the second data column of the first image sampling data set The result is taken as an absolute value to obtain the first level difference; and the second sampled data values of the second sampled data values of the i-th row in the first data column of the first image sampling data set, The result of subtracting the second sampling data values of the second sampling data values in the first data column of the second image sampling data set from the N-1th row is taken as its absolute value to obtain the first vertical Difference. 1 7. The color interpolation calculation method described in item 15 of the scope of the patent application, wherein the first average calculation above includes at least: line N-2 in the second data column of the first image sampling data set The third sampling data value of the third sampling data value and the third sampling data of the N row in the second data column of the first image sampling data set 33 1 This paper size applies Chinese national standards ( CNS) A4 specification (210 X 297 mm) — III ------------- ^ --------- (Please read the notes on the back before filling this page) 530487 A8 B8 C8 D8 t. The result of adding the third sampling data value of the material value in the patent application range is divided by 2 to obtain the first level average value; and the first column and the material value data are the first. The two sets of samples should be sampled. The sample data should be taken as the sample. The second sample should be taken. The second sample is taken to obtain two or more, and the two should be divided. Then the result is N-end. CJN rnj, the project value data of the project, the first sample of the first group of materials, the second sample of the first group of materials, and the vertical sample of the direct company, The value of the first image is 5 T ·:-* · The method includes the calculation from the calculation to the calculation of the color value of the interpolation. Sgt. Sgt. 18 should be equal and multiplied by the multiplication value -------------- install --- (Please read the precautions on the back before filling this page) The sum of the difference is straight and the sum of the difference is divided by one and then the number is equal to the difference,. 'Pingshui1 The first-and the difference is directly at 1; the first value is two Divide the first quantity to the difference & Yu Ping S water, the first quantity of σ is straight by the L difference multiplied by the first. The first quantity can be added to the value of the second quantity and the sum value. Inner general color. Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs. Fan Li specially requested to apply for the color difference 1 2nd 2nd 1 on the description of the second line. The calculation method of the second color to the first color of the water plug package 2 is inserted into the place. Type 1 calculation 34 This paper size is applicable to China National Standard (CNS) A4 specifications (210 X 297 mm) Printed by the Intellectual Property Bureau Employee Consumer Cooperative of the Ministry of Economic Affairs 530487 A8 B8 C8 D8 Perform a second average calculation to obtain a second horizontal average and a second vertical average; perform a second first color type interpolation initial value calculation to obtain a second first color Initial value of type interpolation; and multiplying the initial value of interpolation of the second first color type by the values of the first sampled data values in the first data column of the first data column of the second image sampling data set. After the first sampled data value is divided by the third color type average value to obtain the second first color type interpolation final value. 20. The color interpolation calculation method as described in item 19 of the scope of the patent application, wherein the above-mentioned second difference calculation includes at least: line N-3 in the first data column of the second image sampling data set. The second sampling data value of the second sampling data values is subtracted from the second sampling data value of the second sampling data values in line N-1 of the first data column of the second image sampling data set. The result of the value is taken as its absolute value to obtain the second level difference; and the third of the third sampling data values of the third sampling data value in the second data column of the first image sampling data set. The sampled data value is obtained by subtracting the third sampled data value from the third sampled data value of the third sampled data value in the second data column of the second image sampling data set from the third sampled data value to obtain the absolute value. The second vertical difference. This paper size applies to China National Standard (CNS) A4 (210 X 297 mm) ------------- ^ -------- ^ -------- -(Please read the precautions on the back before filling out this page) Printed by the Consumer Cooperatives of the Intellectual Property Bureau of the Ministry of Economic Affairs 530487 AS B8 C8 D8 f. Application for patent scope 2 1. In the colors described in item 丨 9 of the scope of patent application The interpolation calculation method, wherein the above-mentioned second average calculation includes at least: the second sampling data values of the second sampling data values in the N-3 rows of the first data column of the second image sampling data set, And the result of adding the second sampling data values of the second sampling data values in the first data row of the second image sampling data set to the N-1th row and dividing by 2 to obtain the second level The average value; and the third sampling data value of the third sampling data values in the second data row of the first image sampling data set, and the third sampling data value, and the first sampling data set of the second image The result of adding the third sampling data values of the third sampling data values in line N-2 of the two data columns is divided by 2 to obtain The second vertical average. 22. The color interpolation calculation method according to item 19 of the scope of the patent application, wherein the initial calculation of the second first color type interpolation includes at least: the second horizontal average multiplied by the second vertical difference And divide by the sum of the second vertical difference and the second horizontal difference to obtain a third value; the second vertical average value is multiplied by the second horizontal difference and then divided by the second vertical difference and The sum of the second level difference to obtain a fourth value; and adding the third value and the fourth value to obtain the second 36th paper size applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 mm) nn In umnninnm ϋ I · in —I t— nn I— n sip-, · nn an nnmnt (Please read the precautions on the back before filling out this page) 530487 A8 B8 C8 D8 Color of patent application scope Type interpolation Initial value Printed by the Consumer Cooperative of the Intellectual Property Bureau of the Ministry of Economic Affairs Sample 3 Second Party Shadows This Sample-2 Second Samples The mean square image should be taken N-first calculated one taken N-first selected color one selected type calculated one-fourth The first and the third color are inserted in the first and the second columns, and some of the materials are included in the column. The color of the materials in the first and second colors is the same material value. Sample value 1; material color calculation of the value of the data obtained by the first color calculation of the second material of the calculation of the line as the first line of the data of the four rows. The plan is like inserting the first sample and inserting -2 shadow samples and inserting the value of N-1 and inserting the value of -3 into the sample. The same should be taken in the place N. The division week N-except the insert N-one of the four items. The first one of the type is the middle and the second type of the inner item. The fourth group of the first category is in the first column of the group plus the first category. In the first category, the first material is the first color column and the first material is the same color column. The color column of the latter category, the color of some materials, the value of some materials, the value of materials, the color of materials, the sample value of the third sample, the sample value of the first material, the third sample of the color, and the second material. The second fund, the third fund, the second bank, the second bank, the third bank, the first bank, the first bank, the third bank, the first bank, the first bank, the first bank, the first bank, the first bank, the first bank, the first bank, the first bank, the first bank Please take the first N, get one of the second, get the second one repeatedly, the fourth of the fourth, the fourth one of the fourth group, the first group of the first group, the first group of the second group The value of the material should be from 24 to 24, the material of the Chinese material, the value of the material of the Chinese material, the value of the material, the value of the sample, the value of the edge, and the value of the sample. (Please read the precautions on the back before filling this page) This paper size applies National Standard (CNS) A4 Specification (210 X 297 mm) Printed by the Intellectual Property Bureau of the Ministry of Economy Employees' Cooperatives 530487 A8 B8 C8 D8 々 The scope of patent application is like the N-1 in the second column of the sampling data group The fourth sampling data values of the rows are added up and divided by 4 to obtain a peripheral average value of the second color type, and then the peripheral average value of the second color type is multiplied by the second image sampling data group. After the first sampled data values of the first sampled data values in the N-2 row of the first data column are divided by the third color type average value to obtain the third second color type interpolation value. 25. A color interpolation processor, which is applied to an image processing system, wherein the image processing system includes a first buffer and a second buffer for receiving a plurality of image data from an external image data array. Row of data, the color interpolation processor includes at least: a common parameter calculation module for receiving a first row, a second row, and a third row from among the rows in the image data array Data column, and perform a common parameter calculation to calculate a plurality of common parameters; a horizontal and vertical difference calculation module for receiving the first data column, the first data column, the data column from the data columns in the image data array; The second data column and the third data column, and perform a horizontal and vertical difference calculation to calculate a plurality of horizontal differences and a plurality of vertical differences; an average calculation module for receiving calculations from the common parameter These common parameters of the module and perform an average calculation to calculate a plurality of peripheral averages and a plurality of local averages; an edge weight calculation module for receiving Calculate 38 paper sizes from this common parameter. Applicable to China National Standard (CNS) A4 (210 X 297 mm) ------------- Installation ------- Order-- ------- line (please read the precautions on the back before filling this page) 530487 A8 B8 C8 D8 t, these common parameters of the patent application module, and from the horizontal and vertical difference calculation module The horizontal difference and the vertical difference to calculate a plurality of interpolation final values; and an interpolation item selection module for receiving the interpolation final values from the edge weight calculation module , And the peripheral mean and the local mean from the mean calculation module, and perform a calculation to output a plurality of interpolation values corresponding to the first data row, the second data row, and the third data row and the Some interpolated final values. 26. The color interpolation processor described in item 25 of the scope of patent application, which further includes a parallel processing module to use the interpolation values and the interpolation final values from the interpolation item selection module as A parallel output. 27. The color interpolation processor described in item 25 of the scope of patent application, wherein the common parameters include at least a plurality of horizontal average values, a plurality of vertical average values, a plurality of red common parameters, and a plurality of blue common parameters parameter. m In mm mMW— nm ί I In m Hi I · immm 1 fn vn Order --------- line (please read the precautions on the back before filling this page) The paper size of the paper is applicable to China National Standard (CNS) A4 (210 X 297 mm)