Image processing circuit and image processing method

Abstract

The present disclosure relates to an image processing circuit and an image processing method. The present disclosure discloses an image processing method, comprising the following steps: receiving image data; for a plurality of consecutive pixels in the image data, calculating a plurality of consecutive three-point incremental component values ​​based on the brightness difference between any two adjacent pixels of the plurality of pixels, and then calculating a plurality of consecutive three-point incremental left component values ​​and a plurality of consecutive three-point incremental right component values ​​based on the plurality of consecutive three-point incremental component values; calculating a plurality of adjusted brightness differences based on the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of consecutive three-point incremental left component values, and the plurality of consecutive three-point incremental right component values; and generating a plurality of adjusted brightness values ​​based on the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences.

Description

Technical Field

[0001] The present invention relates to image processing, and more particularly to an image processing method for improving image clarity. Background Art

[0002] Traditional image magnification techniques use splines to calculate pixel values ​​after magnification. However, this method can cause irregular shapes on the image's oblique edges. Using different image magnification techniques to mitigate this phenomenon typically requires significant computational effort, increasing software processing time and hardware manufacturing costs. Furthermore, when the image magnification ratio is high, the magnified image will have a blurred edge transition, requiring even greater computational effort to approximate the original image's clarity.

[0003] In addition, if the image edge has a wide transition band, that is, the image content includes a plurality of pixels with continuous increasing / decreasing brightness, the quality of subsequent image magnification or other image processing will also be affected. Summary of the Invention

[0004] Therefore, one of the objectives of the present invention is to provide an image processing method that can process a plurality of pixels with continuously increasing / decreasing brightness with a relatively low computational load, so as to facilitate related image processing operations.

[0005] In one embodiment of the present invention, an image processing circuit is disclosed, comprising a receiving circuit, a component value calculation circuit, a brightness difference adjustment circuit, and an output circuit. The receiving circuit is configured to receive image data; for a plurality of consecutive pixels in the image data, the component value calculation circuit calculates a plurality of consecutive three-point incremental component values ​​based on the brightness difference between any two adjacent pixels of the plurality of pixels, and then calculates a plurality of consecutive three-point incremental left component values ​​and a plurality of consecutive three-point incremental right component values ​​based on the plurality of consecutive three-point incremental component values; the brightness difference adjustment circuit is configured to calculate a plurality of adjusted brightness differences based on the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of consecutive three-point incremental left component values, and the plurality of consecutive three-point incremental right component values; and the output circuit is configured to generate a plurality of adjusted brightness values ​​based on the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences.

[0006] In another embodiment of the present invention, an image processing method is disclosed, which includes the following steps: receiving image data; for a plurality of continuous pixels in the image data, calculating a plurality of continuous three-point incremental component values ​​according to the brightness difference between any two adjacent pixels of the plurality of pixels, and then calculating a plurality of continuous three-point incremental left component values ​​and a plurality of continuous three-point incremental right component values ​​according to the plurality of continuous three-point incremental component values; calculating a plurality of adjusted brightness differences according to the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of continuous three-point incremental left component values ​​and the plurality of continuous three-point incremental right component values; and generating a plurality of adjusted brightness values ​​according to the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 FIG. 1 is a schematic diagram of an image processing circuit according to an embodiment of the present invention.

[0008] Figure 2 A schematic diagram showing the original brightness of image data and the adjusted brightness of the output image is shown.

[0009] Figure 3 FIG. 4 is a flow chart of an image processing method according to an embodiment of the present invention.

[0010] Figure 4 FIG. 4 is a schematic diagram of an image processing circuit according to another embodiment of the present invention. DETAILED DESCRIPTION

[0011] Figure 1 FIG. 1 is a schematic diagram of an image processing circuit 100 according to an embodiment of the present invention. Figure 1 As shown, the image processing circuit 100 includes a receiving circuit 110, a component value calculation circuit 120, a brightness difference adjustment circuit 130, and an output circuit 140. In this embodiment, the image processing circuit 100 is used to receive image data (image frame) Din and perform image processing to generate an output image Dout. The output image Dout is processed by a back-end processing circuit 102 and then transmitted to a display panel 104 for display.

[0012] During the operation of the image processing circuit 100, the receiving circuit 110 receives image data Din. The component value calculation circuit 120 calculates a correlation component value for three consecutive pixels in the image data Din with increasing / decreasing brightness. These correlation component values ​​may include component values ​​for two consecutive pixels with increasing / decreasing brightness, component values ​​for three consecutive pixels with increasing / decreasing brightness, and component values ​​for four consecutive pixels with increasing / decreasing brightness. Next, the brightness difference adjustment circuit 130 adjusts the brightness differences between some pixels in the image data Din based on the correlation component values. Finally, the output circuit 140 generates an output image Dout based on the adjusted brightness differences.

[0013] For detailed explanation, an example is given below. It is assumed that the image data Din includes N consecutive pixels located in the same column or row of a frame. For ease of explanation, nine consecutive pixels are used for the following description. It is assumed that the pixel values ​​corresponding to the pixels at positions '0', '1', '2', '3', '4', '5', '6', '7', and '8' of the image data Din (hereinafter, the pixel values ​​are described as "brightness") are '0', '0', '2', '5', '8', '11', '14', '16', and '16', respectively. For details on the operations of the component value calculation circuit 120, the brightness difference adjustment circuit 130, and the output circuit 140, please refer to Table 1 below.

[0014]

[0015]

[0016] Table 1

[0017] In the operation of component value calculation circuit 120, the luminance difference between two pixels is first calculated, i.e., (0, 2, 3, 3, 3, 3, 2, 0) as shown in Table 1. Next, component value calculation circuit 120 calculates two consecutive incremental component values ​​based on the same-sign components of the two luminance differences. The same-sign components of the two luminance differences can be the median of the two luminance differences and '0'. For example, for the pixel at position '2', the two adjacent luminance differences are '2' and '3', respectively. Therefore, the two consecutive incremental component values ​​for the pixel at position '2' are '2', the median of the three numbers (0, 2, 3). For the pixel at position '3', the two adjacent luminance differences are '3' and '3', respectively. Therefore, the two consecutive incremental component values ​​for the pixel at position '2' are '3', the median of the three numbers (0, 3, 3). Next, the component value calculation circuit 120 calculates a consecutive three-point incremental component value based on the same-sign components of the two consecutive two-point incremental component values. The same-sign components of the two consecutive two-point incremental component values ​​may be the median of the two consecutive two-point incremental component values ​​and '0.' For example, for pixels at positions '2' and '3', their consecutive two-point incremental component values ​​are '2' and '3', respectively. Therefore, the consecutive three-point incremental component value between the pixels at positions '2' and '3' is '2', the median of the three numbers (0, 2, 3). For pixels at positions '3' and '4', their consecutive two-point incremental component values ​​are '3' and '3', respectively. Therefore, the consecutive three-point incremental component value between the pixels at positions '3' and '4' is '3', the median of the three numbers (0, 3, 3). Next, the component value calculation circuit 120 calculates a continuous four-point incremental component value based on the same-signed components of the two consecutive three-point incremental component values. The same-signed components of the two consecutive three-point incremental component values ​​can be the median of the two consecutive three-point incremental component values ​​and '0'. For example, for the pixel at position '2', the three-point incremental component values ​​on its left and right sides are '0' and '2', respectively. Therefore, the continuous four-point incremental component value of the pixel at position '2' is '0', the median of the three numbers (0, 0, 2). For the pixel at position '3', the three-point incremental component values ​​on its left and right sides are '2' and '3', respectively. Therefore, the continuous four-point incremental component value of the pixel at position '3' is '2', the median of the three numbers (0, 2, 3).

[0018] After calculating the two-point increment component values, the three-point increment component values, and the four-point increment component values, the component value calculation circuit 120 calculates the three-point increment left component value and the three-point increment right component value based on the three-point increment component values ​​and the four-point increment component values. Specifically, the component value calculation circuit 120 subtracts the four-point increment component value on the left from each of the three-point increment component values ​​to obtain the three-point increment left component value. For example, for the three-point increment component value '2' between pixels at positions '2' and '3', the three-point increment left component value is '2' (i.e., 2-0=0); for the three-point increment component value '3' between pixels at positions '3' and '4', the three-point increment left component value is '1' (i.e., 3-2=1). The component value calculation circuit 120 further subtracts the four consecutive incremental component values ​​on the right from each three consecutive incremental component value to obtain three consecutive incremental right component values. For example, for the three consecutive incremental component value '2' between the pixels at positions '3' and '4', its three consecutive incremental right component values ​​are '0' (i.e., 3-3=0); for the three consecutive incremental component value '3' between the pixels at positions '4' and '5', its three consecutive incremental right component values ​​are '1' (i.e., 3-2=1).

[0019] Next, the brightness difference adjustment circuit 130 calculates the adjusted brightness difference based on the brightness difference, the three-point consecutive incremental left component value, and the three-point consecutive incremental right component value. Specifically, for each brightness difference, the brightness difference adjustment circuit 130 adds the three-point consecutive incremental left component value and the three-point consecutive incremental right component value to the brightness difference, and subtracts the three-point consecutive incremental left component value from the brightness difference on the left side of the brightness difference, and subtracts the three-point consecutive incremental right component value from the brightness difference on the right side of the brightness difference. In other words, the calculation method of the brightness difference adjustment circuit 130 can be regarded as moving the brightness difference inward, using the contents of Table 1 as an example: according to the three-point consecutive incremental left component value '2' between the pixels at positions '2' and '3', the brightness difference '2' between the pixels at positions '1' and '2' is moved to between the pixels at positions '2' and '3'; according to the three-point consecutive incremental left component value '1' between the pixels at positions '3' and '4', the brightness difference '2' between the pixels at positions '2' and '3' is moved to between the pixels at positions '2' and '3'. The luminance difference '1' is moved between pixels at positions '3' and '4'; the right component value '2' is incremented based on three consecutive points between pixels at positions '5' and '6' to move the luminance difference '2' between pixels at positions '6' and '7' to move between pixels at positions '5' and '6'; and the right component value '1' is incremented based on three consecutive points between pixels at positions '4' and '5' to move the luminance difference '1' between pixels at positions '5' and '6' to move between pixels at positions '4' and '5'. Based on the above calculation results, the luminance difference adjustment circuit 130 can generate an adjusted luminance difference (0, 0, 4, 4, 4, 4, 0, 0).

[0020] Then, the output circuit 140 generates adjusted pixel values ​​'0', '0', '0', '4', '8', '12', '16', '16', and '16' at positions '0' to '8' according to the adjusted brightness difference.

[0021] Figure 2 A schematic diagram showing the original brightness of the image data Din and the adjusted brightness of the output image Dout is shown. Figure 2 As shown, the transition band of the adjusted brightness (i.e., positions '2' to '6') is smaller than the transition band of the original brightness (i.e., positions '1' to '7'), which means that the adjusted brightness has better clarity and sharper edges, so that the back-end circuit 102 can maintain clearer edges without causing edge blurring problems when performing other operations, such as image magnification.

[0022] The above embodiments are described using pixels with continuously increasing brightness. A person skilled in the art will be able to understand how to implement an example of pixels with continuously decreasing brightness after reading the above embodiments, so the relevant details are not repeated here.

[0023] The above embodiment describes the magnification operation of multiple pixels in a single direction, such as the horizontal magnification of image data Din. Subsequently, component value calculation circuit 120 and brightness difference adjustment circuit 130 can further magnify multiple pixels in another direction (e.g., the vertical direction of image data Din) to complete the magnification of the entire image data Din and generate output image Dout. Since multiple pixels located at the edge of an image typically have strictly increasing / decreasing pixel values, the operation of the above embodiment can effectively and reliably reduce edge blurring caused by the image magnification process. Furthermore, due to the relatively simple calculations, only a relatively low amount of software or hardware computing circuitry is required, thereby reducing design and manufacturing costs.

[0024] exist Figure 1 In the embodiment, the image processing circuit 100 is mainly used to process three consecutive pixels with increasing / decreasing brightness in the image data Din. If the pixel values ​​corresponding to the pixels at positions '0' to '8' of the image data Din are not in an increasing / decreasing sequence or a strictly increasing / decreasing sequence, other circuits can be used for processing, or the original brightness can be decomposed into two groups of brightness values, one of which includes a strictly increasing / decreasing brightness value, for processing by the image processing circuit 100.

[0025] It should be noted that the calculation of brightness in the above embodiment may include pixel values ​​of multiple different colors depending on the color palette used. Taking the "red, green, and blue" color palette as an example, the calculation of multiple pixels in the above embodiment includes the calculation of red pixel values, green pixel values, and blue pixel values; taking the "brightness and color difference (YUV)" color palette as an example, the calculation of multiple pixels in the above embodiment includes the calculation of brightness, blue color difference, and red color difference; taking the "cyan, magenta, yellow, black (CMYK)" color palette as an example, the calculation of multiple pixels in the above embodiment includes the calculation of cyan pixel values, magenta pixel values, yellow pixel values, and black pixel values.

[0026] Figure 3 FIG. 1 is a flow chart of an image processing method according to an embodiment of the present invention. Referring to the description of the above embodiment, the flow of the image processing method is as follows.

[0027] Step 300: The process starts.

[0028] Step 302: Receive image data.

[0029] Step 304: Calculate a plurality of consecutive two-point incremental component values ​​according to the brightness difference between any two adjacent pixels of the plurality of pixels.

[0030] Step 306: Calculate a plurality of consecutive three-point incremental component values ​​according to the plurality of consecutive two-point incremental component values.

[0031] Step 308: Calculate a plurality of consecutive four-point incremental component values ​​according to the plurality of consecutive three-point incremental component values.

[0032] Step 310: Calculate a plurality of consecutive three-point incremental left component values ​​and a plurality of consecutive three-point incremental right component values ​​according to the plurality of consecutive three-point incremental component values ​​and the plurality of consecutive four-point incremental component values.

[0033] Step 312: Calculate a plurality of adjusted brightness differences according to the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of consecutive three-point incremental left component values, and the plurality of consecutive three-point incremental right component values.

[0034] Step 314: Generate a plurality of adjusted brightness values ​​according to the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences.

[0035] Figure 4 FIG. 4 is a schematic diagram of a circuit 400 according to an embodiment of the present invention. Figure 4 As shown, the image processing circuit 400 includes a receiving circuit 410, an image decomposition circuit 420, an image processing unit 430, an image amplification circuit 440, and an output circuit 450. In this embodiment, the image processing circuit 400 is used to receive image data (image frame) Din and perform an image amplification operation to generate an output image Dout. The output image Dout is processed by a back-end processing circuit 402 and then transmitted to a display panel 404 for display.

[0036] During the operation of the image processing circuit 400, the receiving circuit 410 receives image data Din, and the image decomposition circuit 420 divides the pixel value of each pixel in the image data Din into two parts to generate first image data Din1 and second image data Din2. The first image data Din1 includes three consecutive pixels with increasing / decreasing brightness. Next, the image processing unit 430 performs image processing on the first image data Din1 to generate processed first image data Din1'. The image amplification circuit 440 adds the processed first image data Din1' to the second image data Din2 and then performs amplification to generate amplified image data Din'. Finally, the output circuit 450 generates an output image Dout based on the amplified image data Din'.

[0037] For detailed explanation, the following example assumes that image data Din includes N consecutive pixels located in the same column or row of a frame. For ease of explanation, the following description uses six consecutive pixels. It is assumed that the pixel values ​​corresponding to the pixels at positions '0', '1', '2', '3', '4', and '5' of image data Din are '0', '0', '5', '10', '12', and '12', respectively (the pixel values ​​are described below using brightness as an example). In this embodiment, the image decomposition circuit 420 decomposes the image data Din into first image data Din1 and second image data Din2. The brightness of the first image data Din1 is (0, 0, 2, 4, 6, 6), and the brightness of the second image data Din2 is (0, 0, 3, 6, 6, 6). Next, the image processing unit 430 processes the first image data Din1 to generate processed first image data Din1′. In this embodiment, the image processing unit 430 itself is Figure 1 In the illustrated image processing circuit 100, first image data Din1 is processed through a component value calculation circuit 120, a brightness difference adjustment circuit 130, and an output circuit 140 to generate processed first image data Din1'. Since those skilled in the art will readily understand how to calculate the first image data Din1 using the calculation method described in the embodiments of Figures 1-3, the applicant will directly present the calculation result, namely, the brightness of the processed first image data Din1' is (0, 0, 0, 6, 6, 6). In the operation of the image amplification circuit 440, the processed first image data Din1' and the second image data Din2 are added together to obtain a brightness of (0, 0, 3, 12, 12, 12). The result of the addition is then subjected to any suitable amplification operation, such as the amplification operation described in patent application number 107138896 and publication number TW202019146A, to obtain amplified image data Din'. The applicant directly lists the calculation results here. Taking a four-fold amplification as an example, the amplified image data Din' is (0, 0, 0, 0, 0, 0, 0, 3, 6, 9, 9, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12). Finally, the output circuit 450 generates an output image Dout based on the amplified image data Din'.

[0038] To briefly summarize the present invention, in the image processing circuit and the related image processing method of the present invention, the component value calculation circuit calculates multiple three-point increasing / decreasing component values, multiple three-point increasing left component values, and multiple three-point increasing right component values. This can make the adjusted brightness have better clarity and sharper edges, and can also enable the back-end circuit to maintain clearer edges when performing other operations without causing edge blurring problems.

[0039] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

[0040]

Explanation of symbols

[0041] 100: Image processing circuit

[0042] 102: Back-end circuit

[0043] 104: Display panel

[0044] 110: receiving circuit

[0045] 120: component value calculation circuit

[0046] 130: Brightness difference adjustment circuit

[0047] 140: Output circuit

[0048] 300~314: Steps

[0049] 400: Image processing circuit

[0050] 402: Backend circuit

[0051] 404: Display Panel

[0052] 410: receiving circuit

[0053] 420: Image decomposition circuit

[0054] 430: Image processing unit

[0055] 440: Image amplification circuit

[0056] 450: output circuit

[0057] Din: Image data

[0058] Din1: first image data

[0059] Din2: Second image data

[0060] Din1': first image data after processing

[0061] Din': image data after magnification

[0062] Dout: output image

Claims

1. An image processing circuit comprising: A receiving circuit for receiving image data; a component value calculation circuit coupled to the receiving circuit, wherein, for a plurality of consecutive pixels in the image data, the component value calculation circuit calculates, based on a brightness difference between any two adjacent pixels of the plurality of pixels, a plurality of consecutive three-point increasing component values ​​corresponding to component values ​​of three consecutive pixels with increasing brightness in the image data, and further calculates, based on the plurality of consecutive three-point increasing component values, a plurality of consecutive three-point increasing left component values ​​representing component values ​​on the left side of the three consecutive pixels with increasing brightness in the image data and a plurality of consecutive three-point increasing right component values ​​representing component values ​​on the right side of the three consecutive pixels with increasing brightness in the image data; a brightness difference adjustment circuit coupled to the component value calculation circuit for moving the brightness difference between any two adjacent pixels of the plurality of pixels toward the center of the plurality of pixels based on the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of consecutive three-point incremental left component values, and the plurality of consecutive three-point incremental right component values, so as to calculate a plurality of adjusted brightness differences; and An output circuit is coupled to the brightness difference adjustment circuit and is used to generate a plurality of adjusted brightness values ​​according to the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences.

2. The image processing circuit according to claim 1 , wherein the component value calculation circuit calculates a plurality of consecutive two-point incremental component values ​​according to the brightness difference between any two adjacent pixels of the plurality of pixels, then calculates a plurality of consecutive three-point incremental component values ​​according to the plurality of consecutive two-point incremental component values, and then calculates a plurality of consecutive four-point incremental component values ​​according to the plurality of consecutive three-point incremental component values; and the component value calculation circuit calculates the plurality of consecutive three-point incremental left component values ​​and the plurality of consecutive three-point incremental right component values ​​according to the plurality of consecutive three-point incremental component values ​​and the plurality of consecutive four-point incremental component values.

3. The image processing circuit according to claim 2 , wherein the component value calculation circuit calculates the plurality of consecutive two-point incremental component values ​​based on a component of the same sign of two brightness differences between any pixel of the plurality of pixels and two adjacent pixels, wherein the component of the same sign is a median between the two brightness differences and zero.

4. The image processing circuit according to claim 2 , wherein the component value calculation circuit calculates the multiple consecutive three-point incremental component values ​​based on the same-sign component of any two adjacent consecutive two-point incremental component values, wherein the same-sign component is the median between the two adjacent consecutive two-point incremental component values ​​and zero.

5. The image processing circuit according to claim 2 , wherein for each consecutive three-point incremental component value, the component value calculation circuit subtracts the consecutive four-point incremental component value located on the left from the consecutive three-point incremental component value to obtain a corresponding consecutive three-point incremental left component value; and for each consecutive three-point incremental component value, the component value calculation circuit subtracts the consecutive four-point incremental component value located on the right from the consecutive three-point incremental component value to obtain a corresponding consecutive three-point incremental right component value.

6. An image processing method comprising: receiving image data; For a plurality of consecutive pixels in the image data, a plurality of consecutive three-point increasing component values ​​are calculated based on the brightness difference between any two adjacent pixels of the plurality of pixels to obtain correlation component values ​​of three consecutive pixels with increasing brightness in the image data, and a plurality of consecutive three-point increasing left component values ​​representing component values ​​on the left side of the three consecutive pixels with increasing brightness in the image data and a plurality of consecutive three-point increasing right component values ​​representing component values ​​on the right side of the three consecutive pixels with increasing brightness in the image data are calculated based on the plurality of consecutive three-point increasing component values; Calculating a plurality of adjusted brightness differences by moving the brightness difference between any two adjacent pixels of the plurality of pixels toward a center of the plurality of pixels based on the brightness difference between any two adjacent pixels of the plurality of pixels, the plurality of consecutive three-point incremental left component values, and the plurality of consecutive three-point incremental right component values; and A plurality of adjusted brightness values ​​are generated according to the pixel values ​​of the plurality of pixels and the plurality of adjusted brightness differences.

7. The image processing method according to claim 6 , wherein the steps of calculating the plurality of three-point incremental component values ​​based on the brightness difference between any two adjacent pixels of the plurality of pixels, and then calculating the plurality of three-point incremental left component values ​​and the plurality of three-point incremental right component values ​​based on the plurality of three-point incremental component values ​​comprise: Calculating a plurality of consecutive two-point incremental component values ​​according to the brightness difference between any two adjacent pixels of the plurality of pixels; Calculating the plurality of consecutive three-point incremental component values ​​according to the plurality of consecutive two-point incremental component values; Calculating a plurality of consecutive four-point increasing component values ​​according to the plurality of consecutive three-point increasing component values; as well as The plurality of consecutive three-point incremental component values ​​and the plurality of consecutive four-point incremental component values ​​are used to calculate the plurality of consecutive three-point incremental left component values ​​and the plurality of consecutive three-point incremental right component values.

8. The image processing method according to claim 7, wherein the step of calculating the plurality of consecutive two-point incremental component values ​​according to the brightness difference between any two adjacent pixels of the plurality of pixels comprises: The plurality of consecutive two-point incremental component values ​​are calculated according to a component of the same sign of two brightness differences between any pixel of the plurality of pixels and two adjacent pixels, wherein the component of the same sign is a median between the two brightness differences and zero.

Images