CN100594719C - Image edge enhancement device and method - Google Patents

Abstract

An image edge enhancement device and a method thereof are used for adjusting the brightness value of a pixel. The image edge enhancement device comprises a first edge judgment circuit, a second edge judgment circuit and a processing circuit. The first edge judgment circuit extracts the pixel edge in a first direction and outputs a first edge value. The second edge judgment circuit extracts the pixel edge in a second direction and outputs a second edge value. The processing circuit determines an operation mode according to the brightness values of a plurality of surrounding pixels of the pixel, the first edge value and the second edge value, and outputs the adjusted brightness value of the pixel according to the operation mode. The adjusting method can effectively prevent the noise in the image from being amplified and prevent the same image edge from being repeatedly strengthened by different edge judging circuits.

Description

Image edge enhancement device and method

technical field

The present invention relates to an image processing device and its method, and in particular to an image edge enhancement device and its method.

Background technique

Image edge enhancement technology is to adjust the contrast or brightness of the edge of the object in the display screen to make the outline of the image clear, so that the object is more obvious, and the image is clearer and clearer under the sense of the human eye. Please refer to FIG. 1 , which is a schematic diagram of a pixel and its surrounding pixels. In image edge enhancement, it is judged whether the pixel p _{, i, j} is the edge of the object and the brightness is adjusted. If the pixel p _{i, j} is the object edge, it is compared with the surrounding pixels p _{i-1, j +1} , p _{i, j+1} , p _{i+1, j+1 ,} p _{i-1, j , p i+1, j} , p _{i-1 , j-1} , p _{i, j-1} and p The contrast of _{i+1, j-1} is higher.

Please refer to FIG. 2 , which is a structural diagram of an existing edge strengthening circuit. The edge enhancement circuit 200 includes register units 210 ( 1 )-( 6 ), a high pass filter 220 , a high pass filter 230 , a low pass filter 240 and a query unit 250 .

The register unit 210(1) is used to register the pixel value x _{i, j+1} , the register unit 210(2) is used to register the pixel value x _{i+1, j+1} , and the register unit 210(3) is used to register the pixel value x _{i, j} , the register unit 210(4) is used to register the pixel value x _{i+1, j} , the register unit 210(5) is used to register the pixel value x _{i, j-1} , and the register unit 210(6) is used to register the pixel value Value x _i+1,j-1 . Pixel values x _i,j+1 , x _i+1,j+1 , x _i,j , x _i+1,j , x _i,j-1 and _{x i+1,j-1} are each pixel p _{i , values of j+1} , p _{i+1, j+1} , p _i,j , p _i+1,j , p _i,j-1 and _{p i+1,j-1} in a specific color gamut.

The high-pass filter 220 receives pixel values x _i-1,j , x _i,j and x _i+1,j to output a horizontal direction value H1. The high-pass filter 230 receives pixel values xi _,j-1 , xi _,j and _xi,j+1 to output a vertical edge value V1. The low-pass filter 240 receives the vertical edge value H1 and the horizontal edge value V1, and outputs a processed value S1 after low-pass filtering. After receiving the processed value S1, the query unit 250 checks its look-up table to output the enhanced pixel value _xi,j ' of the pixel value xi _,j .

However, under the above structure, if the pixel p _{i, j} happens to be the vertical edge and the horizontal edge of the object, then the pixel p _{i, j} will be repeatedly regarded as a point on the edge by the high-pass filter 220 and the high-pass filter 130, Instead, the brightness contrast between the pixel p _{i, j} and the surrounding pixels is excessively increased. When the pixel p _{i, j} is noise, the judgment may be wrong in the above structure, so that the part of the picture that is not the edge of the object will increase the contrast and produce more obvious noise points, making the picture look rough.

Contents of the invention

In view of this, the purpose of the present invention is to provide an image edge enhancement device and method thereof, which can avoid misjudging the edge of an object and over-emphasizing the edge of the object.

According to the object of the present invention, an image edge enhancement device is provided for adjusting the brightness value of a pixel. The image edge enhancement device includes a first edge judging circuit, a second edge judging circuit and a processing circuit. The first edge judging circuit outputs a first edge value after extracting the pixel edge in a first direction. The second edge judging circuit outputs a second edge value after extracting the pixel edge in a second direction. A processing circuit determines the edge type of the pixel according to the brightness values of a plurality of surrounding pixels of the pixel, and determines an operation mode based on the sign of the first edge value, the sign of the second edge value and the edge type of the pixel, and then according to Mathematical mode to output the adjusted luminance value of the pixel.

According to another object of the present invention, an image edge enhancement method is proposed. Firstly, a first edge value is output after pixel edge is extracted in a first direction. Next, output a second edge value after extracting the edge of the pixel in the second direction. Afterwards, an operation mode is determined according to the luminance values, the first edge value and the second edge value of a plurality of surrounding pixels of the pixel. Finally, output the adjusted luminance value of the pixel according to the operation mode. Wherein, the step of determining the operation mode of the pixel includes: determining the edge type of the pixel according to the brightness values of the surrounding pixels; and according to the sign of the first edge value, the sign of the second edge value, and The edge type of the pixel determines the operation mode.

In order to make the above-mentioned purpose, features, and advantages of the present invention more comprehensible, a preferred embodiment will be described in detail below together with the accompanying drawings.

Description of drawings

FIG. 1 is a schematic diagram of a pixel and its surrounding pixels.

FIG. 2 is a structural diagram of a conventional edge strengthening circuit.

FIG. 3 is a structural diagram of an image edge enhancement device according to a preferred embodiment of the present invention.

FIG. 4A is a schematic diagram of a possible image distribution when the pixel E is noisy.

FIG. 4B is a schematic diagram of a possible image distribution when the pixel E is an edge-bending type.

FIG. 5 is a flowchart of an image edge enhancement method according to a preferred embodiment of the present invention.

FIG. 6 is a flowchart of the method in step 53 .

FIG. 7 is a flowchart of the method in step 54 .

Description of reference symbols

P _i-1,j+1 , P _i,j+1 , P _{i+1,j+1 ,} P i _-1,j , P _{i,j ,} P _i+1,j , P _{i-1,j -1} , P _{i, j-1} , P _{i+1, j-1} : Pixels

200: Edge strengthening circuit

210(1)-(6), 310(1)-(6): storage unit

220, 230: high pass filter

240: Low pass filter

250: query unit

300: image edge enhancement device

321, 322: edge judgment circuit

330: Processing circuit

331: Computing unit

332: query unit

Detailed ways

Please refer to FIG. 3 , which is a structural diagram of an image edge enhancement device according to a preferred embodiment of the present invention. The image edge enhancement device 300 is used to adjust the brightness value of a pixel, which is illustrated as a pixel p _i,j in this embodiment. The image edge enhancement device 300 includes register units 310 ( 1 )-( 6 ), an edge judging circuit 321 , an edge judging circuit 322 and a processing circuit 330 .

Please refer to Figure 1, pixel p _{i, j} is the target to be processed for image edge enhancement, surrounding pixel p _{i, j} are 8 surrounding pixels p _{i-1, j+1} , p _{i, j+1} , p _{i+1, j+1} , p _{i-1, j} , p _{i+1, j , p i-1, j-1 ,} p _{i, j-1} , _{p i+1, j-1} , the surrounding pixels and The nearest pixel p _{i, j} is the boundary pixel p _{i, j+1} , p _{i-1, j} , p _{i, j-1} , p _{i+1, j} , and the rest of the surrounding pixels are critical pixels, such as : p _{i-1, j+1} , p _{i+1, j+1} , p _{i-1, j-1} , p _{i+1, j-1} .

The register units 310(1)-(6) are used to register the brightness values of the pixel p _i,j and several surrounding pixels, for example, the brightness values of the pixel p _i,j and five surrounding pixels. Please refer to Figure 1 and Figure 3 at the same time. The register unit 310(1) is used to register the pixel value y _{i, j+1} , the register unit 310(2) is used to register the pixel value y _{i+i, j+1} , and the register unit 310(3) is used to register the pixel value y _{i, j} , the register unit 310(4) is used to register the pixel value y _{i+1, j} , the register unit 310(5) is used to register the pixel value y _{i, j-1} register unit 310(6) is used to register the pixel value y _{i+1, j-1} . Pixel value y _{i-1, j+1} , y _{i, j+1} , y _{i+1, j+1} , y _{i-1, j} , y _{i, j} , y _{i+1, j} , y _{i-1 , j-1} , y _{i, j-1} and y _{i+1, j-1} are pixels p _{i-1, j+1} , p _{i, j+1} , p _{i+1, j+1} , p _{i -1, j} , p _{i, j} , p _{i+1, j} , p _{i-1, j-1} , p _{i, j-1} and _{p i+1, j-1} in a specific color gamut, this specific The color gamut in this embodiment is the Y color gamut.

The edge judging circuit 321 receives pixel values y _i-1,j , y _i,j , y _i+1,j , extracts the edge of pixel p _i,j in the first direction and outputs the first edge value H2, the first direction For example, the horizontal direction. The edge judging circuit 322 receives pixel values y _{i, j-1} , y _{i, j} , y _{i, j+1} , and outputs a second edge value V2 after extracting the edge of pixel p _{i, j} in the second direction. For example, the vertical direction. The edge judging circuit 321 and the edge judging circuit 322 are, for example, finite impulse response high-pass filters (high-pass finite impulse response filter, HP FIR filter).

The processing circuit 330 according to the luminance values of the surrounding pixels of the pixel p _{i, j} , namely y _{i-1, j+1} , y _{i, j+1} , y _{i+1, j+1} , y _{i-1, j} , y _{i+1, j} , y _{i-1, j-1} , y _{i, j-1} and y _{i+1, j-1} , the first edge value H2 and the second edge value V2 determine the operation of pixel p _{i, j} mode to output the adjusted luminance value y _i,j ' of pixel p i _,j . The processing circuit 330 includes a calculation unit 331 and a query unit 332 . The calculation unit 331 determines the edge type of the pixel p _{i, j} according to the luminance values of surrounding pixels of the pixel p _{i, j} , and then according to the sign of the first edge value H2, the sign of the second edge value V2 and the pixel p _i, The edge type of _j has an operation mode. The calculation unit 331 then generates the pixel calculation value S2 according to the calculation mode of the pixel p _i,j . The query unit 332 corresponds to a look-up table according to the calculated pixel value S2 to generate the adjusted luminance value y _i,j ′ of the pixel p _i,j .

The calculation unit 331 determines the operation mode of the pixel p _{i, j} as follows: the calculation unit 331 first sets four calculation values y _{i, j+1} ', y _{i-1, j} ', y _{i, j-1} ' and y _{i+1, j} ', to determine the light and shade relationship between pixel p _{i, j} and the four surrounding boundary pixels.

Therefore, the boundary pixel calculation values y _{i ,j+1} ', y _i-1,j ', y _i,j-1 ' and y _{i+1,j' of pixel p i,j} are as follows:

y _i,j+1 '=y _i,j+1 -(y _i-1,j+1 +y _i+1,j+1 )/2;

y _{i-1, j} '= y _{i-1, j} - (y _{i-1, j+1} + y _{i-1, j-1} )/2;

y _{i, j-1} ' = y _{i, j-1} - (y _{i-1, j-1} + y _{i+1, j-1} )/2;

y _{i+1, j} '=y _{i+1, j} - (y _{i+1, j-1} + y _{i+1, j+1} )/2;

Then, depending on the relationship between the calculated values y _i,j+1 ', y _i-1,j ', y _i,j-1 ' and y _i+1,j ' and a critical value Th to determine the corresponding judgment value C1, C2, C3 and C4 are 0 or 1. Th is determined by the user of the shadow edge enhancement device, and the value of Th will affect the degree of noise suppression. If y _i,j+1 '<Th, then C1 is 1; if y _i,j+1 '>Th, then C1 is 0. The values of the judgment values C2, C3 and C4 can also be deduced by analogy.

After determining the judgment values C1 , C2 , C3 and C4 , the type of the pixel p _{i, j} is determined according to the judgment values C1 , C2 , C3 and C4 . In this embodiment, the pixels p _{i, j} are classified into noise type, edge turning type and single-directional object edge. Please refer to FIG. 1 and FIG. 4A at the same time. FIG. 4A is a schematic diagram of a possible image distribution when the pixel p _i,j is noise. To avoid drawing confusion, in FIG. 4A and FIG. 4B , the pixel codes in FIG. 1 are not added, and the corresponding positions of each pixel are described. In Fig. 4A, any three of the judgment values C1, C2, C3 and C4 are 0, and the judgment formula is:

((C1&C2&C3)|(C1&C2&C4)|(C1&C3&C4)|(C2&C3&C4))=0;

If the result of this judgment formula is 0, the pixel p _{i, j} is noise. In this case, the pixel p _{i, j} is less likely to be an object edge, and is more likely to be high-frequency noise or noise. In this case, it will not be judged as an object edge, and no enhancement action will be taken.

Please refer to FIG. 1 and FIG. 4B at the same time. FIG. 4B is a schematic diagram of a possible image distribution when the pixel p _i,j is an edge turning type. Then in the situation in Fig. 4B, the judgment formulas of judgment values C1, C2, C3 and C4 are:

((C1|C2)&(C3|C4))=0;

At this time, the pixel p _{i, j} is more likely to belong to the edge of the object in the picture, and may be the turning point of the edge of the picture object. Under the processing of the existing image edge enhancement circuit, it will be detected as an edge by the horizontal and vertical high-pass filters at the same time, and two enhancement actions will be performed, resulting in an over-enhancement result.

If the pixel p _{i, j} is not noise type or edge turning type, it is determined as an object edge in a single direction. Pixels p _{i, j} are highly likely to be object edges, but they are only horizontal or vertical edges, so they can be treated as general object edges.

After knowing the edge type of pixel p _{i, j} , further determine the pixel p i, j according to the sign of the first edge value H2, the sign of the second edge value V2 and the edge type of pixel p _{i , j} operation mode.

If the pixel p _{i, j} is noise type, the brightness of the pixel p _{i, j} is not adjusted. If the pixel p _{i, j} is of edge turning type, in order to prevent the pixel p _{i, j} from being emphasized in the horizontal direction and the vertical direction at the same time. Then, according to whether the first edge value H2 and the second edge value V2 have the same sign, the operation mode when the pixel P _i,j is an edge turning type is determined. When the first edge value H2 and the second edge value V2 have the same sign, then:

S2=(V2+H2)/4;

When the first edge value H2 and the second edge value V2 are different signs, then:

S2=(V2+H2)/2;

If the pixel p _{i, j} is an edge in a single direction, only horizontal or vertical edges will be detected, so one of the first edge value H2 and the second edge value V2 will be close to 0. Still according to the polarity adjustment and correction method of the first edge value H2 and the second edge value V2, but the problem of over-enhancement is less likely, the divisors are changed to 2 and 1 respectively. When the first edge value H2 and the second edge value V2 have the same sign, then:

S2=(V2+H2)/2;

When the first edge value H2 and the second edge value V2 are different signs, then:

S2=(V2+H2).

Please refer to FIG. 5 , which shows a flowchart of an image edge enhancement method according to a preferred embodiment of the present invention. Firstly, a first edge value H2 is output after extracting the edge of pixel p _{i, j} in a first direction, the first direction is, for example, the horizontal direction, as shown in step 51 . Next, a second edge value V2 is output after extracting the edge of the pixel p _{i, j} in a second direction, such as a vertical direction, as shown in step 52 . Then, the calculation mode of the pixel p _{i, j} is determined according to the luminance values of the surrounding pixels of the pixel p _i , j, the first edge value H2 and the second edge value V2, as shown in step 53 . Finally, the adjusted luminance value of the pixel p _i,j is output according to the operation mode of the pixel p _i,j , as shown in step 54 .

Please refer to FIG. 6 , which is a flow chart of the method in step 53 . First, the edge type of pixel p _{i, j} is determined according to the brightness values of surrounding pixels, as shown in step 61 . Next, the operation mode of the pixel p i _{, j is determined according to the sign of the first edge value H2, the sign of the second edge value V2, and the edge type of the pixel p i , j} , as shown in step 62 . .

Please refer to FIG. 7 , which shows a flow chart of the method in step 54 . First, a pixel calculation value is calculated according to the operation mode of the pixel p _i,j , as shown in step 71 . Afterwards, a look-up table (look-up table, LUT) is generated according to the calculated value of the pixel to generate the adjusted luminance value of the pixel p _i,j , as shown in step 72 .

The image edge enhancement device and method disclosed in the above-mentioned embodiments of the present invention have an accurate judgment procedure for the object edge, so as to appropriately enhance the object edge. It can avoid misjudgment caused by the inability to accurately filter out the edge of the object in the picture, such as the situation where the edge is over-enhanced or the noise is enhanced.

In summary, although the present invention has been disclosed as above with a preferred embodiment, it is not intended to limit the present invention. Anyone skilled in this art can make various modifications without departing from the spirit and scope of the present invention. Changes and modifications, so the scope of protection of the present invention should be defined by the scope of the appended patent application.

Claims (27)

1. An image edge strengthening device, in order to adjust a pixel p _{i, the brightness value of j} , this image edge strengthening device comprises: A first edge judging circuit, which extracts the edge of the pixel in a first direction and outputs a first edge value H2; A second edge judging circuit, which extracts the edge of the pixel in a second direction and outputs a second edge value V2; and A processing circuit, which determines the edge type of the pixel according to the brightness values of a plurality of surrounding pixels of the pixel, and determines the positive or negative of the first edge value H2, the positive or negative of the second edge value V2, and the edge type of the pixel An operation mode, and output the adjusted luminance value of the pixel according to the operation mode. 2. The image edge enhancement device according to claim 1, wherein the processing circuit further comprises: a calculation unit, which calculates a pixel calculation value S2 after determining the calculation mode; and A lookup unit, corresponding to a lookup table according to the calculated value S2 of the pixel to generate the adjusted brightness value of the pixel. 3. The image edge enhancement device according to claim 1, wherein the image edge enhancement device further comprises a plurality of register circuits to register the brightness values of the pixel p _{i, j} and a plurality of surrounding pixels, wherein the plurality of surrounding pixels Pixels are contained within the surrounding pixels. 4. The image edge enhancement device according to claim 3, wherein the pixel p _i,j has 8 surrounding pixels p _i-1,j+1 , p _i,j+1 , p _{i+1,j+ 1} , p _{i-1, j} , p _{i+1, j} , p _{i-1, j-1} , p _{i, j-1} and p _{i+1, j-1} , the image edge enhancement device also includes 6 The register circuit is used to register the luminance values of the pixel and 5 surrounding pixels. 5. The image edge enhancement device according to claim 4, wherein the surrounding pixels include a plurality of boundary pixels p _i,j+1 , p _i-1,j , p _i,j-1 and p _{i+1 , j} , and multiple adjacent pixels p _{i-1, j+1} , p _{i+1, j+1} , p _{i-1, j-1} and p _{i+1, j-1} , and the boundary pixel p _i,j+1 , p _i-1,j , p _i,j-1 and p _i+1,j have the closest distance to the pixel p _i,j , and the processing circuit is based on the surrounding pixels p _{i- 1, j+1} , p _{i, j+1} , p _{i+1, j+1 ,} p _{i-1, j} , p _{i+1, j} , p _{i-1, j-1 ,} p _{i, j- 1} and the brightness value of p _{i+1, j-1} to determine the edge type of the pixel p _{i, j} . 6. The image edge enhancement device according to claim 5, wherein the processing circuit is based on the brightness values y _i,j+1 , y _i-1,j , y _i,j-1 and y of the boundary pixels _{i+1, j} and the brightness value y _{i-1, j+1 of the critical pixel, y i+1} , _{j+1 ,} y _{i-1, j-1} and y _{i+1, j-1} are determined A plurality of boundary pixel calculation values y _{i, j+1} ', y _{i-1, j} ', y _{i, j-1} ' and y _{i+1, j} ', and the calculation values of the boundary pixels are calculated according to the following equation Decide: y _i,j+1 '=y _i,j+1 -(y _i-1,j+1 +y _i+1,j+1 )/2; y _{i-1, j} '= y _{i-1, j} - (y _{i-1, j+1} + y _{i-1, j-1} )/2; y _i,j-1 '=y _i,j-1 -(y _i-1,j-1 +y _i+1,j-1 )/2; and y _{i+1, j} '=y _{i+1, j} - (y _{i+1, j-1} + y _{i+1, j+1} )/2. 7. The image edge enhancement device according to claim 6, wherein the processing circuit calculates values y _{i, j+1} ', y _{i-1, j} ', y _{i, j-1} ' based on the boundary pixels And y _{i+1, j} ' and a critical value Th determine a plurality of corresponding judgment values C1, C2, C3 and C4, wherein the judgment values are determined according to the following judgment formula: When y _i,j+1 '<Th, C1=1; When y _i,j+1 '>Th, C1=0; When y _{i-1, j} '<Th, C2=1; When y _{i-1, j} '>Th, C2=0; When y _i,j-1 '<Th, C3=1; When y _i,j-1 '>Th, C3=0; When y _{i+1, j} '<Th, C4=1; and When y _{i+1, j} '>Th, C4=0. 8. The image edge enhancement device according to claim 7, wherein the processing circuit determines the edge type of the pixel p _{i, j} according to the judgment values C1, C2, C3 and C4, wherein the pixel p _{i, j} The edge type of is determined according to the following judgment formula: When ((C1&C2&C3)|(C1&C2&C4)|(C1&C3&C4)|(C2&C3&C4))=0, the edge type of the pixel p _i,j is noise type; When ((C1|C2)&(C3|C4))=0, the edge type of the pixel p _i,j is an edge turning type; and When ((C1&C2&C3)|(C1&C2&C4)|(C1&C3&C4)|(C2&C3&C4))=0 and ((C1|C2)&(C3|C4))=0 are not established, the edge type of the pixel p _{i, j} is single Orientation object edges. 9. The image edge enhancement device according to claim 8, wherein the edge type of the pixel p _i,j is an edge turning type, and the first edge value H2 and the second edge value V2 have the same sign, then the Pixel calculation value S2=(V2+H2)/4. 10. The image edge enhancement device according to claim 8, wherein, the edge type of the pixel p _i,j is an edge turning type, and the first edge value H2 and the second edge value V2 are different signs, then the Pixel calculation value S2=(V2+H2)/2. 11. The image edge enhancement device according to claim 8, wherein, the edge type of the pixel p _i,j is a single direction object edge, and the first edge value H2 and the second edge value V2 have the same sign, then The pixel calculation value S2=(V2+H2)/2. 12. The image edge enhancement device according to claim 8, wherein, the edge type of the pixel p _i,j is a single direction object edge, and the first edge value H2 and the second edge value V2 are different signs, then The pixel calculation value S2=(V2+H2). 13. The image edge enhancement device according to claim 1, which performs edge enhancement in the Y color gamut. 14. The image edge enhancement device as claimed in claim 1, wherein the first edge judgment circuit and the second edge judgment circuit are finite impulse response high-pass filters. 15. The image edge enhancement device as claimed in claim 1, wherein the first direction is a horizontal direction, and the second direction is a vertical direction. 16. A method for image edge enhancement, comprising: Outputting a first edge value H2 after extracting the edge of a pixel p _{i, j} in the first direction; Outputting a second edge value V2 after extracting the edge of the pixel p _{i, j} in the second direction; determining an operation mode according to the brightness values of a plurality of surrounding pixels of the pixel p _i,j , the first edge value H2 and the second edge value V2; and Outputting the adjusted luminance value of the pixel according to the operation mode, Wherein, the step of determining the operation mode of the pixel p _{i, j} includes: determining the edge type of the pixel p _i,j according to the brightness values of the surrounding pixels; and The operation mode is determined according to the sign of the first edge value H2, the sign of the second edge value V2, and the edge type of the pixel p _i,j . 17. The image edge enhancement method according to claim 16, wherein the step of outputting the adjusted brightness value of the pixel p _{i, j} according to the operation mode comprises: Calculate a pixel calculation value S2 according to the operation mode; and The calculated value S2 of the pixel corresponds to a look-up table to generate the adjusted luminance value of the pixel p _i,j . 18. The image edge enhancement method according to claim 17, wherein the pixel p _i,j has a plurality of surrounding pixels p _i-1,j+1 , p _i,j+1 , p _{i+1,j+ 1} , p _i-1,j , p _i+1,j , p _i-1,j-1 , p _i,j-1 and p _i+1,j-1 , and the surrounding pixels include a plurality of boundaries Pixel p _i,j+1 , p _i-1,j , p _i,j-1 and p _i+1,j , and multiple critical pixels p _i-1,j+1 , p _{i+1,j+ 1} , p _{i-1, j-1} and p _{i+1, j-1} , and the boundary pixels p _{i, j+1} , p _{i-1, j} , p _{i, j-1} and p _{i+1 , j} is the closest to the pixel p _{i, j} . 19. The image edge enhancement method as claimed in claim 18, wherein, in the step of determining the operation mode of the pixel p _i,j, it also includes: According to the luminance values y _i,j+1 , y _i-1,j , y _i,j-1 and y _i+1,j of the boundary pixels and the luminance values y _i-1,j+ of the critical pixels ₁ , y _{i+1, j+1} , y _{i-1, j-1} and y _{i+1, j-1} to determine a plurality of boundary pixel calculation values y _{i, j+1} ', y _{i-1, j} ', y _{i, j-1} ' and y _{i+1, j} ', and the calculated value of the boundary pixel is determined according to the following equation: y _i,j+1 '=y _i,j+1 -(y _i-1,j+1 +y _i+1,j+1 )/2; y _{i-1, j} '= y _{i-1, j} - (y _{i-1, j+1} + y _{i-1, j-1} )/2; y _i,j-1 '=y _i,j-1 -(y _i-1,j-1 +y _i+1,j-1 )/2; and y _{i+1, j} '=y _{i+1, j} - (y _{i+1, j-1} + y _{i+1, j+1} )/2. 20. The image edge enhancement method as claimed in claim 19, wherein, in the step of determining the operation mode of the pixel p _i,j, further comprising: Determining a plurality of corresponding judgment values according to the boundary pixel calculation values y _i,j+1 ', y _i-1,j ', y _i,j-1 ' and y _i+1,j ' and a critical value Th C1, C2, C3 and C4, wherein the judgment value is determined according to the following judgment formula: When y _i,j+1 '<Th, C1=1; When y _i,j+1 '>Th, C1=0; When y _{i-1, j} '<Th, C2=1; When y _{i-1, j} '>Th, C2=0; When y _i,j-1 '<Th, C3=1; When y _i,j-1 '>Th, C3=0; When y _{i+1, j} '<Th, C4=1; and When y _{i+1, j} '>Th, C4=0. 21. The image edge enhancement method according to claim 20, wherein, in the step of determining the operation mode of the pixel p _{i, j , the edge type of the pixel p i, j} is determined according to the following judgment formula: When ((C1&C2&C3)|(C1&C2&C4)|(C1&C3&C4)|(C2&C3&C4))=0, the edge type of the pixel p _i,j is noise type; When ((C1|C2)&(C3|C4))=0, the edge type of the pixel p _i,j is an edge turning type; and When ((C1&C2&C3)|(C1&C2&C4)|(C1&C3&C4)|(C2&C3&C4))=0 and ((C1|C2)&(C3|C4))=0 are not established, the edge type of the pixel p _{i, j} is single Orientation object edges. 22. The image edge enhancement method according to claim 21, wherein, when the edge type of the pixel p _i,j is an edge turning type, and the first edge value H2 and the second edge value V2 have the same sign, then The pixel calculation value S2=(V2+H2)/4. 23. The image edge enhancement method according to claim 21, wherein, when the edge type of the pixel p _{i, j} is an edge turning type, and the first edge value H2 and the second edge value V2 are different signs, then The pixel calculation value S2=(V2+H2)/2. 24. The image edge enhancement method according to claim 21, wherein when the edge type of the pixel p _{i, j} is a single-directional object edge, and the first edge value H2 and the second edge value V2 have the same sign, Then the pixel calculation value S2=(V2+H2)/2. 25. The image edge enhancement method according to claim 21, wherein when the edge type of the pixel p _{i, j} is a single-directional object edge, and the first edge value H2 and the second edge value V2 are different signs, Then the pixel calculation value S2=(V2+H2). 26. The image edge enhancement method according to claim 16, which is to perform edge enhancement in Y color gamut. 27. The image edge enhancement method as claimed in claim 16, wherein the first direction is a horizontal direction, and the second direction is a vertical direction.

Images