Image edge enhancement apparatus and method thereof

This application claims the benefit of Taiwan application Serial No. 94132726, filed Sep. 21, 2005, the subject matter of which is incorporated herein by reference.

1. Field of the Invention

The invention relates in general to an image processing apparatus and a method thereof, and more particularly to an image edge enhancement apparatus and a method thereof.

2. Description of the Related Art

By adjusting the contrast or brightness of the edge of an object in a display frame, the edge enhancement technology produces a clear outline of an image and makes the object more highlighted and the frame more clearly manifested. Referring to FIG. 1 , a pixel and its adjacent pixels are shown. During edge enhancement, whether a pixel p i,j is at an edge of an object is determined, and the brightness of the pixel is adjusted. If the pixel p i,j is at the edge of an object, then the contrast of the pixel is enabled to be higher than its adjacent 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 .

Referring to FIG. 2 , a structural diagram of a conventional edge enhancement circuit is shown. The edge enhancement circuit 200 comprises six registers 210 ( 1 )˜( 6 ), a high pass filter 220 , a high pass filter 230 , a low pass filter 240 and an inquiry unit 250 .

The register 210 ( 1 ) is used for temporarily storing the pixel value x i,j+1 . The register 210 ( 2 ) is used for temporarily storing the pixel value x i+1,j+1 . The register 210 ( 3 ) is used for temporarily storing the pixel value x i,j . The register 210 ( 4 ) is used for temporarily storing the pixel value x i+1,j . The register 210 ( 5 ) is used for temporarily storing the pixel value x i,j−1 . The register 210 ( 6 ) is used for temporarily storing the pixel value x i+1,j−1 . The 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 respectively the values of the pixels p i,j+1 , p i+1,j+1 , p i,j , p i+1,j , p i,j−1 and p i+1,j−1 at the predetermined color gamut.

The high pass filter 220 receives the pixel values x i−1,j , x i,j and x i+1,j to output a horizontal edge value H 1 . The high pass filter 230 receives the pixel values x i,j−1 , x i,j and x i,j+1 , to output a vertical edge value V 1 . The low pass filter 240 receives a vertical edge value V 1 and a horizontal edge value H 1 , and then outputs a processed value S 1 . After the inquiry unit 250 receives the processed value S 1 , an enhanced pixel value x i,j ′ of the pixel value x i,j is looked up form the look-up table to be outputted.

According to the above structure, if the pixel p i,j happens to be at a vertical or horizontal edge of an object, then the pixel p i,j will be regarded by the high pass filter 220 and the high pass filter 230 as a pixel at an edge twice. Consequently, the brightness and the contrast of the pixel p i,j opposite to the adjacent pixels are over-enhanced. If the pixel p i,j is a noise, the above structure may be misjudged, causing spots to occur at non-edge parts of the object in the frame and making the frame looked speckled.

It is therefore an object of the invention to provide an image edge enhancement apparatus and a method thereof capable of avoiding over-enhancing an object edge when the edge is determined by mistake.

The invention achieves the above-identified object by providing an image edge enhancement apparatus used for adjusting the brightness of a pixel. The image edge enhancement apparatus comprises a first edge judgment circuit, a second edge judgment circuit and a process circuit. The first edge judgment circuit retrieves the edge data of the pixel in a first direction to output a first edge value. The second edge judgment circuit retrieves the edge data of the pixel in a second direction to output a second edge value. The process circuit determines an operation mode of the pixel according to the brightness of a plurality of adjacent pixels around the pixel, the first edge value and the second edge value to output an adjusted brightness of the pixel according to the operation mode.

The invention further achieves the above-identified object by providing an image edge enhancement method. At first, the edge data of the pixel is retrieved in a first direction to output a first edge value. Next, the edge data of the pixel is retrieved in a second direction to output a second edge value. Afterwards, an operation mode is determined according to the brightness of a plurality of adjacent pixels around the pixel, the first edge value and the second edge value. At last, an adjusted brightness of the pixel is outputted according to the operation mode.

Other objects, features, and advantages of the invention will become apparent from the following detailed description of the preferred but non-limiting embodiments. The following description is made with reference to the accompanying drawings.

FIG. 1 illustrates a pixel and its adjacent pixels;

FIG. 2 illustrates a structural diagram of a conventional edge enhancement circuit;

FIG. 3 illustrates a structural diagram of an image edge enhancement apparatus according to a preferred embodiment of the invention;

FIG. 4A illustrates a possible image distribution when the pixel E is a noise type pixel;

FIG. 4B illustrates a possible image distribution when the pixel E is an angled-edge type;

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

FIG. 6 illustrates step 53 of the method flowchart; and

FIG. 7 illustrates step 54 of the method flowchart.

Referring to FIG. 3 , a structural diagram of an image edge enhancement apparatus according to a preferred embodiment of the invention is shown. The image edge enhancement apparatus 300 is used for adjusting the brightness of a pixel, and is exemplified by a pixel p i,j in the present embodiment of the invention. The image edge enhancement apparatus 300 comprises six registers 310 ( 1 )˜( 6 ), an edge judgment circuit 321 , an edge judgment circuit 322 and a process circuit 330 .

Referring to FIG. 1 . The treatment of image edge enhancement is applied to the pixel p i,j . The pixel p i,j is surrounded by eight adjacent 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 . Of the adjacent pixels, the boundary pixels p i,j+1 , p i−1,j , p i,j−1 , p i+1,j are closest to the pixel p i,j , while the other adjacent pixels are threshold pixels such as p i−1,j+1 , p i+1,j+1 , p i−1,j−1 , p i+1,j−1 .

The registers 310 ( 1 )˜( 6 ) are used for temporarily storing the brightness of the pixel p i,j and a plurality of adjacent pixels such as the brightness of the pixel p i,j and five adjacent pixels. Referring to both FIG. 1 and FIG. 3 . The register 310 ( 1 ) is used for temporarily storing the pixel value y i,j+1 . The register 310 ( 2 ) is used for temporarily storing the pixel value y i+1,j+1 . The register 310 ( 3 ) is used for temporarily storing the pixel value y i,j . The register 310 ( 4 ) is used for temporarily storing the pixel value y i+1,j . The register 310 ( 5 ) is used for temporarily storing the pixel value y i,j−1 . The register 310 ( 6 ) is used for temporarily storing the pixel value y i+1,j−1 . The pixel values 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 respectively the values of the 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 predetermined color gamut. The predetermined color gamut is Y color gamut in the present embodiment of the invention.

The edge judgment circuit 321 receives the pixel values y i−1,j , y i,j , y i+1,j , and retrieves the edge data of the pixel p i,j in a first direction to output a first edge value H 2 . The first direction can be a horizontal direction for instance. The edge judgment circuit 322 receives the pixel value y i,j−1 , y i,j , y i,j+1 , and retrieves the edge data of the pixel p i,j in a second direction to output a second edge value V 2 . The second direction can be a vertical direction for instance. Examples of the edge judgment circuit 321 and the edge judgment circuit 322 include a high-pass finite impulse response filter (HP FIR filter).

The process circuit 330 determines the operation mode of the pixel p i,j to output an adjusted brightness y i,j ′ of the pixel p i,j according to the brightness of the adjacent pixels around the pixel p i,j , that is, 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 , and a first edge value H 2 and a second edge value V 2 . The process circuit 330 comprises a calculation unit 331 and an inquiry unit 332 . The calculation unit 331 determines the edge type of the pixel p i,j according to the brightness of the adjacent pixels around the pixel p i,j , and then obtains an operation mode according to whether the first edge value H 2 and the second edge value V 2 are positive and the edge type of the pixel p i,j . The calculation unit 331 generates a pixel calculation value S 2 according to the operation mode of the pixel p i,j . The inquiry unit 332 generates an adjusted brightness y i,j ′ of the pixel p i,j from a predetermined look-up table according to the pixel calculation value S 2 .

As for how the calculation unit 331 determines the operation mode of the pixel p i,j is disclosed below. The calculation unit 331 , first of all, sets four calculation values y i,j+1 ′, y i−1,j ′, y i,j−1 ′ and y i+1,j ′ to determine the brightness relationship between the pixel p i,j and the four adjacent boundary pixels.

The boundary pixel calculation values y i,j+1 ′, y i−1,j ′, y i,j−1 ′ and y i+1,j ′ of the pixel p i,j are disclosed below:

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;

And then, whether the corresponding criteria C 1 , C 2 , C 3 and C 4 are 0 or 1 is determined according to the relationship between the calculation values y i,j+1 ′, y i−1,j ′, y i,j−1 ′ and y i+1,j ′ and a predetermined threshold value Th. The threshold value Th is determined by the user of the image edge enhancement apparatus. The Th value affects how largely the noise is restricted. If y i,j+1 ′<Th, then C 1 =1. If y i,j+1 ′>Th, then C 1 =0. The values of the criteria C 2 , C 3 and C 4 are obtained likewise.

After the values of the criteria C 1 , C 2 , C 3 and C 4 are determined, the edge type of the pixel p i,j is determined according to the criteria C 1 , C 2 , C 3 and C 4 . In the present embodiment of the invention, the pixel p i,j is categorized into noise type, angled-edge type and straight-edge type. Referring to both FIG. 1 and FIG. 4A . FIG. 4A illustrates a possible image distribution when the pixel p i,j is a noise type pixel. To avoid confusion, the reference numbers of the pixels in FIG. 1 are not introduced to FIG. 4A and FIG. 4B , and the pixels are referred according to their corresponding positions. In FIG. 4A , any three of the criteria C 1 , C 2 , C 3 and C 4 are 0, and the judgment formula is:

(( C 1& C 2& C 3)|( C 1& C 2& C 4)|( C 1& C 3& C 4)|( C 2& C 3& C 4))=0.

If the result of the judgment formula equals 0, then the pixel p i,j is a noise. Under such circumstance, the pixel p i,j is more likely to be a high-frequency noise or spot than at an object edge. Consequently, the pixel p i,j will not be determined as an object edge, and not enhancement is applied.

Referring both to FIG. 1 and FIG. 4B . FIG. 4B illustrates a possible image distribution when the pixel p i,j is angled-edge type. In the case of FIG. 4B , the judgment formula of the criteria C 1 , C 2 , C 3 and C 4 is:

(( C 1 |C 2)&( C 3 |C 4))=0.

Meanwhile, the pixel p i,j is more likely to be at an object edge of a frame, for example, at an angled-edge of the object in the frame. If the pixel p i,j is processed by a conventional image edge enhancement circuit, the pixel p i,j will be detected as an edge by both the horizontal high pass filter and the vertical high pass filter. The pixel will be enhanced twice and become over-enhanced.

If the pixel p i,j is neither noise type nor angled-edge type, then the pixel p i,j is determined as a straight-edge of an object. The pixel p i,j which is more likely to be at an edge of an object is only at a horizontal direction or a vertical direction. And an ordinary object edge processing is applied to the pixel p i,j .

After the edge type of the pixel p i,j is obtained, the operation mode of the pixel p i,j is determined according to whether the first edge value H 2 and the second edge value V 2 are positive and the edge type of the pixel p i,j .

If the pixel p i,j is noise type, then the brightness of the pixel p i,j is not adjusted. If the pixel p i,j is angled-edge type, to avoid the pixel p i,j being enhanced in both the horizontal direction and the vertical direction, then the operation mode of the angled-edge type pixel P i,j is determined according to whether the first edge value H 2 and the second edge value V 2 are both positive or negative. If the first edge value H 2 and the second edge value V 2 are both positive or both negative, then the pixel calculation value S 2 is calculated as below:

S 2=( V 2 +H 2)/4

If the first edge value H 2 and the second edge value V 2 are not both positive or both negative, then the pixel calculation value is calculated as below:

S 2=( V 2 +H 2)/2

If the pixel p i,j is a straight-edge, then only the horizontal edge or the vertical edge will be detected. Therefore, one of the first edge value H 2 and the second edge value V 2 approximates 0. The pixel is still adjusted according to whether the first edge value H 2 and the second edge value V 2 are positive. However, the problem of over-enhancement will not occur, and the dividers are changed to 2 and 1, respectively. If the first edge value H 2 and the second edge value V 2 are both positive or both negative, then the pixel calculation value S 2 is calculated as below:

S 2=( V 2 +H 2)/2

If the first edge value H 2 and the second edge value V 2 are not both positive or both negative, then the pixel calculation value S 2 is calculated as below:

S 2=( V 2 +H 2)

Referring to FIG. 5 , a flowchart of an image edge enhancement method according to the preferred embodiment of the invention is shown. At first, the edge data of the pixel p i,j is retrieved in a first direction to output a first edge value H 2 . The first direction can be a horizontal direction as shown in step 51 . Next, the edge data of the pixel p i,j is retrieved in a second direction to output a second edge value V 2 . The second direction can be a vertical direction as shown in step 52 . Afterwards, the operation mode of the pixel p i,j is determined according to the brightness of the adjacent pixels around the pixel p i,j , the first edge value H 2 and the second edge value V 2 as shown in step 53 . At last, an adjusted brightness of the pixel p i,j is outputted according to the operation mode of the pixel p i,j as shown in step 54 .

Referring to FIG. 6 , a flowchart of step 53 of the method is shown. At first, the edge type of the pixel p i,j is determined according to the brightness of the adjacent pixels as shown in step 61 . Next, the operation mode of the pixel p i,j is determined according to whether the first edge value H 2 and the second edge value V 2 are positive and the edge type of the pixel p i,j as shown in step 62 .

Referring to FIG. 7 , a flowchart of step 54 of the method is shown. At first, a pixel calculation value is calculated according to the operation mode of the pixel p i,j as shown in step 71 . Afterwards, an adjusted brightness of the pixel p i,j is generated from a predetermined look-up table (LUT) according to the pixel calculation value as shown in step 72 .

According to the image edge enhancement apparatus and the method thereof disclosed in the above embodiments of the invention, the edge of an object is accurately determined, so that the edge of an object is appropriately enhanced accordingly. The invention avoids the edge of an object being over-enhanced and the spot being enhanced which occurs when the edge of an object in a frame is inaccurately filtered and misjudged.

While the invention has been described by way of example and in terms of a preferred embodiment, it is to be understood that the invention is not limited thereto. On the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.