WO2016095541A1 - Image processing method, device, system and computer storage medium - Google Patents

Abstract

Disclosed in an embodiment of the present invention are an image processing method, device, system and computer storage medium, the image processing method comprising: obtaining image data, the image data being image data processed according to a preset zoom-processing; when it is determined that the image data satisfies a preset condition, obtaining pixel points of a preset size in the image data sequentially to constitute matrix image data and processing the matrix image data according to a first processing, the first processing being a detail sharpening processing, an edge sharpening processing and a smoothing processing.

Description

Image processing method, device, system and computer storage medium Technical field

The present invention relates to image processing technologies, and in particular, to an image processing method, apparatus, system, and computer storage medium.

Background technique

In the image processing process, the original image is lost due to operations such as scaling the original image, thereby causing the output image to be blurred. To this end, the output image needs to be enhanced to make the output image clearer. Among them, the enhancement processing method for the image includes sharpening processing and smoothing processing.

In the prior art, the commonly used sharpening processing methods mainly include: a differential method and a high-pass filtering method; commonly used smoothing methods mainly include: an interpolation method, a linear smoothing method, a convolution method, and the like. The above sharpening processing method and smoothing processing method generally have the following problems: 1. Only global processing is performed, and each area in the video is not effectively divided to perform enhancement processing for a specific area; 2. The enhancement strength cannot be flexibly adjusted. 3, inconvenient to achieve hardware flow.

Summary of the invention

In order to solve the existing technical problems, embodiments of the present invention provide an image processing method, apparatus, system, and computer storage medium, which can effectively divide each area in a video to perform enhancement processing for a specific area, thereby facilitating hardware implementation. .

To achieve the above objective, the technical solution of the embodiment of the present invention is implemented as follows:

An embodiment of the present invention provides an image processing method, where the method includes:

Obtaining image data; the image data is image data processed according to a preset scaling processing manner;

Determining, when the image data meets the preset condition, sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to a first processing manner; Processing mode, edge sharpening processing or smooth processing.

As an embodiment, the determining that the image data meets a preset condition comprises: determining that the image data meets a preset condition when the image data is image data processed by three differential amplification processing methods; Correspondingly, the sequentially acquiring the pixel points of the preset size in the image data to form the matrix image data and processing the matrix image data according to the first processing manner comprises: sequentially acquiring the pixels of the preset size in the image data. Forming matrix image data and processing the matrix image data in a detail sharpening process;

And determining that the image data meets the preset condition, the method includes: determining that the image data meets a preset condition when the image data is image data processed by the B-spline amplification processing manner; And sequentially acquiring the pixel points of the preset size in the image data to form matrix image data and processing the matrix image data according to the first processing manner, including: sequentially acquiring a matrix of pixel points of a preset size in the image data. Image data and processing the matrix image data in an edge sharpening process;

And determining that the image data satisfies a preset condition, including: when the image data is image data processed in a reduction processing manner, determining that the image data meets a preset condition; correspondingly, the And sequentially acquiring the pixel data of the preset size in the image data to form the matrix image data, and processing the matrix image data according to the first processing manner, comprising: sequentially acquiring pixel points of the preset size in the image data to form matrix image data; The matrix image data is processed in a smoothing process.

As an implementation manner, the sequentially acquiring pixels of a preset size in the image data to form matrix image data and processing the matrix image data according to a detail sharpening processing manner, including: starting position of the image data Starting to sequentially acquire 3×3 pixels in the image data Brightness values to form first matrix data;

Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

Presetting a first threshold, a second threshold, and a third threshold, and dividing the luminance range into four intervals according to the first threshold, the second threshold, and the third threshold; wherein the first threshold, the The second threshold and the third threshold are both greater than 0 and less than 255;

Determining, when the first gradient value of the central pixel point is in the second interval composed of the first threshold and the second threshold, performing sharpening processing on the first matrix data.

As an implementation manner, the sequentially acquiring pixel points of a preset size in the image data to form matrix image data and processing the matrix image data according to an edge sharpening processing manner, including: starting position of the image data Starting to sequentially acquire luminance values of 3×3 pixel points in the image data to form first matrix data;

Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

Pre-configuring a fourth threshold, and dividing the luminance range into two intervals according to the fourth threshold; wherein the fourth threshold is greater than 0 and less than 255;

Determining, when the first gradient value of the central pixel point is in the interval formed by the fourth threshold and 255, performing sharpening processing on the first matrix data.

As an implementation manner, the sharpening the first matrix data includes: a second gradient of a central pixel point of the first matrix data acquired according to a preset first Laplacian operator value;

And obtaining, according to the second gradient value, a sharpening value of the central pixel point according to a preset sharpening manner; wherein the preset sharpening manner comprises: obtaining an absolute value of the second gradient value, and obtaining an average value a parameter; limiting a maximum value of the first parameter to obtain a second parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point;

When the second gradient value is greater than zero, the luminance value of the central pixel point + the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point;

When the second gradient value is less than zero, the luminance value of the central pixel point - the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point.

As an implementation manner, the sequentially acquiring pixel points of a preset size in the image data to form matrix image data and processing the matrix image data in a smoothing manner includes: starting from a starting position of the image data Obtaining 3×3 pixel points in the image data to form first matrix data;

A smoothing result of the first matrix data is obtained according to a pre-configured second Laplacian and a smoothing intensity parameter.

An embodiment of the present invention further provides an image processing apparatus, where the apparatus includes: an acquiring unit and an image enhancing unit; wherein

The acquiring unit is configured to acquire image data; the image data is image data processed according to a preset scaling processing manner;

The image enhancement unit is configured to, when determining that the image data acquired by the acquiring unit meets a preset condition, sequentially acquire pixel image data of a preset size in the image data to form matrix image data, and process the processing according to the first processing manner. The matrix image data is described; the first processing mode is a detail sharpening processing mode, an edge sharpening processing mode, or a smoothing processing mode.

As an embodiment, the image enhancement unit includes a first image enhancement unit configured to determine that the image data meets a preset condition when the image data is image data processed in a three-time difference amplification processing manner; And configured to sequentially acquire pixel points of a preset size in the image data to form matrix image data, and process the matrix image data according to a detail sharpening processing manner;

And/or, the image enhancement unit further includes a second image enhancement unit configured to determine that the image data meets a preset condition when the image data is image data processed in a B-spline enlargement processing manner; And configured to sequentially acquire pixel points of a preset size in the image data to form matrix image data, and process the matrix image data according to an edge sharpening processing manner;

And/or the image enhancement unit further includes a third image enhancement unit configured to determine that the image data satisfies a preset condition when the image data is image data processed in a reduction processing manner; Obtaining pixel points of a preset size in the image data to form matrix image data and processing the matrix image data in a smoothing manner.

As an embodiment, the first image enhancement unit is configured to sequentially acquire luminance values of 3×3 pixel points in the image data to form first matrix data from a starting position of the image data; Obtaining, by a Sobel operator, a first gradient value of a central pixel point of the first matrix data; preconfiguring a first threshold, a second threshold, and a third threshold, according to the first threshold, the second threshold, and the The third threshold divides the luminance range into four intervals; wherein the first threshold, the second threshold, and the third threshold are both greater than 0 and less than 255; determining that the first gradient value of the central pixel is in the When the first threshold and the second threshold are in the second interval, the first matrix data is sharpened.

As an embodiment, the second image enhancement unit is configured to sequentially acquire luminance values of 3×3 pixel points in the image data to form first matrix data from a starting position of the image data; a Sobel operator obtains a first gradient value of a central pixel point of the first matrix data; a fourth threshold is pre-configured, and the luminance range is divided into two intervals according to the fourth threshold; wherein the fourth threshold is greater than 0 is less than 255; determining that the first gradient value of the central pixel point is sharpened in the interval formed by the fourth threshold and 255.

In an embodiment, the first image enhancement unit and the second image enhancement unit are both configured to acquire a central pixel point of the first matrix data according to a pre-configured first Laplacian operator. a second gradient value; obtaining, according to the second gradient value, a sharpening value of the central pixel point according to a preset sharpening manner; wherein the preset sharpening manner comprises: taking an absolute value of the second gradient value And averaging to obtain a first parameter; limiting a maximum value of the first parameter to obtain a second parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point; When the second gradient value is greater than zero, the luminance value of the central pixel point + the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point; when the second gradient When the value is less than zero, the luminance value of the central pixel point - the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point.

As an embodiment, the third image enhancement unit is configured to sequentially acquire 3×3 pixel points in the image data to form first matrix data from a starting position of the image data; according to a pre-configured The second Laplacian and the smoothing intensity parameter obtain a smoothed result of the first matrix data.

The embodiment of the present invention further provides an image processing system, where the image processing system includes: an enhancement module, a register configuration module, and a configuration synchronization module; wherein the enhancement module includes the image processing device according to the embodiment of the present invention;

The register configuration module is configured to configure parameters required in the enhancement module;

The configuration synchronization module is configured to synchronize parameters of the register configuration module configuration to the enhancement module.

As an implementation manner, the enhancement module includes: an input module, a first line cache module, a second line cache module, a reg register, a read cache module, a logic enhancement module, and an output module;

The input module is configured to receive image data, and input an Nth row and an N+1th row of the image data into the first row cache module and the second row cache module, respectively; After the input of the Nth line of the image data is completed, input the N+1th line of the image data;

The read cache module is configured to be configured when the Nth row of the image data is input to the first row cache module and the N+1th row data of the image data is input into the second row cache module Reading the image data of the preset size from the first line buffer module; when the N+1th line of the image data is input to the second line buffer module and the image data is When the N+2 line data is input into the first line buffer module, the preset size image data is read from the second line buffer module; wherein, when the image data is N+2 When the row data is initially input into the first line buffer module, a preset amount of image data in the Nth row of data of the image data is input into the reg register; and is further configured to be according to the first line cache module Reading image data of a preset size read from the second line buffer module, reading the image data of the preset size, and the preset number of image data in the reg register to generate matrix data; , N is an integer;

The logic enhancement module is configured to perform enhancement processing on matrix data in the degree cache module; the enhancement processing includes a sharpening process and a smoothing process;

The output module is configured to output matrix data processed by the logic enhancement module.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the image processing method according to the embodiment of the invention.

The image processing method, device, system and computer storage medium provided by the embodiments of the present invention acquire image data; the image data is image data processed according to a preset scaling processing manner; and when the image data is determined to meet a preset condition And sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to a first processing manner; the first processing manner is a detail sharpening processing manner, an edge sharpening processing manner, or Smooth processing. In this manner, by adopting the technical solution of the embodiment of the present invention, the image data area is divided into matrix image data of a preset size, thereby effectively dividing each area in the video to perform enhancement processing for a specific area, and facilitating implementation of the present invention. The hardware flow implementation of the image processing method provided by the example.

DRAWINGS

FIG. 1 is a schematic flowchart diagram of an image processing method according to an embodiment of the present invention;

2 is a schematic diagram of first matrix data according to an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an image processing apparatus according to an embodiment of the present invention; FIG.

4 is a schematic structural diagram of an image processing system according to an embodiment of the present invention;

5(a) to 5(e) are diagrams showing the data flow of an image processing method according to an embodiment of the present invention.

detailed description

The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

Embodiments of the present invention provide an image processing method. 1 is a schematic flowchart of an image processing method according to an embodiment of the present invention; as shown in FIG. 1, the image processing method includes:

Step 101: Acquire image data; the image data is image data processed according to a preset scaling processing manner.

The image processing method of the embodiment is applied to the original image data after the preset scaling processing mode is processed, that is, the original image data is processed by the preset scaling processing manner to obtain the image data described in this embodiment; The zoom processing method includes an enlargement processing manner and a reduction processing manner; the amplification processing manner further includes: a three-time difference amplification processing method and a B-spline amplification processing manner.

Step 102: When it is determined that the image data meets a preset condition, sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to a first processing manner; the first processing manner Sharpen the processing, edge sharpening, or smoothing.

Here, the determining that the image data satisfies a preset condition comprises: determining that the image data meets a preset condition when the image data is image data processed in a three-time difference amplification processing manner; correspondingly, Obtaining the matrix image data of the preset size in the image data to form the matrix image data and processing the matrix image data according to the first processing manner, comprising: sequentially acquiring pixel points of the preset size in the image data to form matrix image data. And processing the matrix image data according to a detail sharpening processing manner;

And/or, determining that the image data meets a preset condition, including: when the image data When the image data processed by the B-spline processing method is used, it is determined that the image data meets a preset condition; correspondingly, the pixel points of the preset size in the image data are sequentially acquired to form matrix image data, and Processing the matrix image data in a processing manner, comprising: sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to an edge sharpening processing manner;

And determining that the image data satisfies a preset condition, including: when the image data is image data processed in a reduction processing manner, determining that the image data meets a preset condition; correspondingly, the And sequentially acquiring the pixel data of the preset size in the image data to form the matrix image data, and processing the matrix image data according to the first processing manner, comprising: sequentially acquiring pixel points of the preset size in the image data to form matrix image data; The matrix image data is processed in a smoothing process.

In this embodiment, the sharpening processing manner includes a detail sharpening processing manner and an edge sharpening processing manner; and the corresponding sharpening processing manner may be performed according to different zoom processing manners performed before the image data. Specifically, the sequentially acquiring the pixel points of the preset size in the image data to form the matrix image data and processing the matrix image data according to the detail sharpening processing manner includes: sequentially acquiring from the starting position of the image data 3×3 pixel points in the image data to form first matrix data;

Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

Presetting a first threshold, a second threshold, and a third threshold, and dividing the luminance range into four intervals according to the first threshold, the second threshold, and the third threshold; wherein the first threshold, the The second threshold and the third threshold are both greater than 0 and less than 255;

Determining, when the first gradient value of the central pixel point is in the second interval composed of the first threshold and the second threshold, performing sharpening processing on the first matrix data.

And sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to an edge sharpening processing manner, including: starting from the image data Starting from the start position, sequentially acquiring 3×3 pixel points in the image data to form first matrix data;

Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

Pre-configuring a fourth threshold, and dividing the luminance range into two intervals according to the fourth threshold; wherein the fourth threshold is greater than 0 and less than 255;

Determining, when the first gradient value of the central pixel point is in the interval formed by the fourth threshold and 255, performing sharpening processing on the first matrix data.

Specifically, FIG. 2 is a schematic diagram of first matrix data according to an embodiment of the present invention; as shown in FIG. 2, wherein Z1, Z2 to Z9 represent luminance values of respective pixel points. The pixel points corresponding to the Z1, Z2 to Z9 may be any 3×3 pixel matrix in the image data, and the first matrix data is the first row to the third row in the image data. The matrix data consisting of the third column of pixels.

Wherein, the Sobel operator weights the values corresponding to the current row or column, and then averages and differentiates. Specifically, the Sobel operator expression is as follows:

G(i,j)=|S _x |+|S _y | (1)

Among them, S _x and S _y represented by the expression (2) and the expression (3) respectively represent the horizontal gradient and the vertical gradient of the Sobel operator.

Obtaining a horizontal gradient value gx and a vertical gradient value gy of the luminance value Z5 of the central pixel point of the first matrix data according to the expression (2) and the expression (3), respectively; the horizontal gradient value gx and the vertical gradient value Gy satisfies the following expression (4) and expression (5), respectively:

Gx=(Z7+2×Z8+Z9)-(Z1+2×Z2+Z3) (4)

Gy=(Z3+2×Z6+Z9)-(Z1+2×Z4+Z7) (5)

After obtaining the horizontal gradient value gx and the vertical gradient value gy, when the absolute value of the horizontal gradient value gx is greater than the absolute value of the vertical gradient value gy, the absolute value of the horizontal gradient value gx is taken as a first gradient value g of the central pixel, ie g=abs(gx); conversely, when the absolute value of the horizontal gradient value gx is smaller than the absolute value of the vertical gradient value gy, the vertical gradient value gy The absolute value is taken as the first gradient value g of the central pixel point, i.e., g = abs (gy).

In the detail sharpening processing mode, since the range of the luminance value is (0 to 255), the first threshold T1, the second threshold T2, and the third threshold T3 are pre-configured based on the luminance value range, the first threshold T1, the second threshold T2 and the third threshold T3 are all in the brightness value range (0-255); the brightness range is divided according to the first threshold T1, the second threshold T2 and the third threshold T3 The four sections are the first section (0 to T1), the second section (T1 to T2), the third section (T2 to T3), and the fourth section (T3 to 255). In this embodiment, the first matrix data falling into the second interval (T1 to T2) is sharpened only for the first gradient value g; and when the first gradient value g falls into the The first interval (0 to T1) and the first matrix data falling in the third interval (T2 to T3) are not subjected to image processing; when the first gradient value g falls within the fourth interval (T3 to 255) The first matrix data is subjected to anti-aliasing processing; the anti-aliasing processing is the same as the prior art, and details are not described herein again. The value of the first threshold T1, the second threshold T2, and the third threshold T3 may be pre-configured according to a specific situation or an empirical value; for example, the first threshold T1 is 32, and the second threshold T2 is 96. The third threshold T3 is 128. Correspondingly, when the first gradient value g falls within the interval (32-96), the first matrix data corresponding to the first gradient value g is sharpened. .

In the edge sharpening processing mode, the fourth threshold T is also pre-configured based on the range of the luminance value (0 to 255), and the fourth threshold T is in the luminance value range (0 to 255); The fourth threshold T divides the luminance range into two intervals, which are a fifth interval (0 to T) and a sixth interval (T to 255). In this embodiment, only the first gradient value g falls. The sixth interval The first matrix data of (T ～ 255) is subjected to sharpening processing; and when the first gradient value g falls into the first matrix data of the fifth interval (0 to T), image processing is not performed; wherein The value of the fourth threshold T may be pre-configured according to a specific situation or an empirical value; for example, the fourth threshold T is 64, and correspondingly, when the first gradient value g falls within the interval of (64-255), Sharpening processing is performed on the first matrix data corresponding to the first gradient value g.

In this embodiment, whether the first matrix value of the first gradient value g falls within the second interval (T1 to T2) is sharpened in the detail sharpening processing manner, or is the edge sharp In the processing method, the first matrix data in which the first gradient value g falls within the sixth interval (T to 255) is sharpened, and the sharpening processing performed is the same. The sharpening process of the first matrix data includes: a second gradient value of a central pixel point of the first matrix data acquired according to a preset first Laplacian operator;

And obtaining, according to the second gradient value, a sharpening value of the central pixel point according to a preset sharpening manner; wherein the preset sharpening manner comprises: obtaining an absolute value of the second gradient value, and obtaining an average value a parameter; limiting a maximum value of the first parameter to obtain a second parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point;

When the second gradient value is greater than zero, the luminance value of the central pixel point + the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point;

When the second gradient value is less than zero, the luminance value of the central pixel point - the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point.

Specifically, the Laplacian operator in the embodiment may be pre-configured in plurality; the expression of the Laplacian operator may be as follows:

Wherein, the Laplacian operator H1 represented by the expression (6) is a conventional Laplacian operator; the Laplacian represented by the expression (7) and the expression (8) The Laplacian operator is a Laplacian operator transformed according to empirical values. In this embodiment, a suitable Laplacian operator may be pre-selected as the first Laplacian according to the type of the image data (such as a face or a landscape type) or an attribute parameter of the image data. Si operator.

Taking the first Laplacian operator as an example of H2 represented by (7), the central pixel point of the first matrix data acquired according to the first Laplacian operator H2 The second gradient value f0 satisfies the following expression:

F0=-1×Z1-1×Z2-1×Z3-1×Z4+8×Z5-1×Z6-1×Z7-1×Z8-1×Z9 (9)

Obtaining a sharpening value f3 of the central pixel point according to the second gradient value f0 in the following sharpening manner:

Step 1: f1 (Z5) = abs (f0 (Z5)) / 8;

Step 2: f2 (Z5) is the limit f1 (Z5) has a maximum value of 32;

Step three: f3 (Z5) = f2 (Z5) × enhance_ratio / 8;

Wherein, in step one, the absolute value and the average are obtained for the second gradient value f0 to obtain the first parameter f1 (Z5). Step 2 is to limit the maximum value of the first parameter f1 (Z5) to obtain a second parameter f2 (Z5) to reduce the influence of noise on the video sharpening result; wherein the maximum value may be 32, that is, The value of the first parameter f1 (Z5) is limited to a maximum of 32. Step 3 is to modify the second parameter f2 (Z5) to obtain a sharpening value f3 (Z5); wherein the enhancement_ratio is an enhancement coefficient, and the enhancement_ratio may be configured in advance to achieve adaptive adjustment sharpening strength the goal of.

Further, the brightness value of the central pixel point is sharpened according to the attribute of the second gradient value f0. When the second gradient value f0 is a positive number, the sharpening result of the luminance value Z5 of the central pixel point is f4(Z5)=Z5+f3(Z5); correspondingly, when the second gradient value f0 For In the case of a non-positive number, the sharpening result of the luminance value Z5 of the central pixel point is f4 (Z5) = Z5 - f3 (Z5).

Finally, after obtaining a sharpening result of the brightness value of the central pixel point, the method further comprises: limiting an output bit width of the sharpening result of the brightness value of the central pixel point. That is, (0 to 255) is used as a threshold to control the sharpening result f4 (Z5) of the luminance value of the central pixel point, and the final output result is obtained.

In this embodiment, when the image data is image data processed in a reduction processing manner, the image data is processed in a smoothing processing manner. The sequentially acquiring pixel points of a preset size in the image data to form matrix image data and processing the matrix image data in a smoothing manner includes: sequentially acquiring the image from a starting position of the image data. 3 × 3 pixels in the data to form the first matrix data;

A smoothing result of the first matrix data is obtained according to a pre-configured second Laplacian and a smoothing intensity parameter.

Specifically, the first matrix data shown in FIG. 2 is still taken as an example, as shown in FIG. 2, wherein Z1, Z2 to Z9 represent luminance values of respective pixel points. The pixel points corresponding to the Z1, Z2 to Z9 may be any 3×3 pixel matrix in the image data, and the first matrix data is the first row to the third row in the image data. The matrix data consisting of the third column of pixels.

The smoothing processing method described in the embodiment of the present invention is to perform smoothing processing globally. The Laplacian operator employed can be pre-configured in plurality; the expression of the Laplacian operator can be as follows:

In this embodiment, a suitable Laplacian operator may be pre-selected as the second Laplacian according to the type of the image data (such as a face or a landscape type) or an attribute parameter of the image data. Si operator.

After determining the second Laplacian, obtaining a smoothing result S of the first matrix data based on the second Laplacian; when the second Laplacian is an expression When the equation (10) is shown, the smoothing result S1 is:

S1=(Z5×(256-h)×2+Z2×h+Z8×h)/512 (13)

When the second Laplacian is represented by the expression (11), the smoothing result S2 is:

S2=(Z5×(256-h)×4+Z2×h+Z4×h+Z6×h+Z8×h)/1024 (14)

When the second Laplacian is represented by the expression (11), the smoothing result S3 is:

S3=(Z5×(256-h)×8+Z1×h+Z2×h+Z3×h+Z4×h+Z6×h+Z7×h+Z8×h+Z9×h)/2048 (15)

Here, h is a smoothing intensity, and the range of the smoothing intensity h is any one of (0 to 127), and the value of the smoothing intensity h can be pre-configured.

Finally, after obtaining a smoothing result of the brightness value of the central pixel point, the method further comprises: limiting an output bit width of the smoothing result of the brightness value of the central pixel point. That is, the smoothing result S of the luminance value of the central pixel point is controlled by using (0 to 255) as a threshold to obtain a final output result.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the image processing method according to the embodiment of the invention.

An embodiment of the present invention further provides an image processing apparatus. FIG. 3 is a schematic structural diagram of an image processing apparatus according to an embodiment of the present invention; as shown in FIG. 3, the apparatus includes: an acquiring unit 31 And an image enhancement unit 32; wherein

The acquiring unit 31 is configured to acquire image data; the image data is image data processed according to a preset scaling processing manner;

The image enhancement unit 32 is configured to, when determining that the image data acquired by the acquiring unit 31 meets a preset condition, sequentially acquire pixel image data of a preset size in the image data to form matrix image data, and according to the first processing manner Processing the matrix image data; the first processing mode is a detail sharpening processing manner, an edge sharpening processing manner, or a smooth processing manner.

Here, the image enhancement unit 32 includes a first image enhancement unit 321 configured to determine that the image data satisfies a preset condition when the image data is image data processed in a three-time difference amplification processing manner; Forming matrix image data by sequentially acquiring pixels of a preset size in the image data and processing the matrix image data according to a detail sharpening processing manner;

And/or, the image enhancement unit 32 further includes a second image enhancement unit 322 configured to determine that the image data meets a preset condition when the image data is image data processed in a B-spline enlargement processing manner. And configured to sequentially acquire pixel points of a preset size in the image data to form matrix image data, and process the matrix image data according to an edge sharpening processing manner;

And/or, the image enhancement unit 32 further includes a third image enhancement unit 323 configured to determine that the image data meets a preset condition when the image data is image data processed in a reduction processing manner; The matrix image data is composed of pixel pixels of a preset size in the image data, and the matrix image data is processed in a smoothing manner.

In the first embodiment, that is, when the image data is image data processed in a three-time difference amplification processing manner, the first image enhancement unit 321 is configured to start from the start position of the image data. And sequentially acquiring the luminance values of the 3×3 pixel points in the image data to form the first matrix data; obtaining the first gradient value of the central pixel point of the first matrix data according to the Sobel operator; preconfiguring the first threshold value a second threshold and a third threshold, dividing the luminance range into four intervals according to the first threshold, the second threshold, and the third threshold; wherein, the a threshold, the second threshold, and the third threshold are both greater than 0 and less than 255; determining that the first gradient value of the central pixel is in the second interval of the first threshold and the second threshold And sharpening the first matrix data.

Specifically, the first image enhancement unit 321 is configured to generate a second gradient value of a central pixel point of the first matrix data acquired according to a preset first Laplacian operator; based on the second gradient Obtaining a sharpening value of the central pixel in a preset sharpening manner; wherein the presetting sharpening manner comprises: obtaining an absolute value of the second gradient value and averaging to obtain a first parameter; Obtaining a maximum value of the first parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point; when the second gradient value is greater than zero, a luminance value of a central pixel point + a sharpening value of the central pixel point as a sharpening result of a luminance value of the central pixel point; and when the second gradient value is less than zero, a luminance value of the central pixel point The sharpening value of the central pixel point is a sharpening result of the luminance value of the central pixel point.

In the second embodiment, that is, when the image data is image data processed by the B-spline enlargement processing, the second image enhancement unit 322 is configured to start from the start position of the image data. And sequentially acquiring luminance values of 3×3 pixels in the image data to form first matrix data; obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator; and configuring a fourth threshold in advance And dividing, according to the fourth threshold, the luminance range into two intervals; wherein the fourth threshold is greater than 0 and less than 255; determining that the first gradient value of the central pixel is in the interval between the fourth threshold and 255 In the middle, the first matrix data is sharpened.

Specifically, the second image enhancement unit 322 is configured to acquire a second gradient value of a central pixel point of the first matrix data acquired according to a preset first Laplacian operator; based on the second gradient Obtaining a sharpening value of the central pixel in a preset sharpening manner; wherein the presetting sharpening manner comprises: obtaining an absolute value of the second gradient value and averaging to obtain a first parameter; Obtaining a second parameter as a maximum value of the first parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point; when the second gradient value is greater than zero, a luminance value of the central pixel + a sharpening value of the central pixel as a sharpening result of the luminance pixel; and when the second gradient is less than zero, a luminance value of the central pixel The sharpening value of the center pixel is used as the result of the sharpening of the brightness point.

In a third embodiment, that is, when the image data is image data processed in a reduction processing manner, the third image enhancement unit 323 is configured to sequentially acquire the image data from the start position of the image data. 3×3 pixel points in the image data are formed to constitute first matrix data; and a smoothing result of the first matrix data is obtained according to a pre-configured second Laplacian and a smoothing intensity parameter.

In the image processing apparatus according to the embodiment, at least one of the first image enhancement unit 321, the second image enhancement unit 322, and the third image enhancement unit 323, that is, the image processing device The first image enhancement unit 321 , the second image enhancement unit 322 or the third image enhancement unit 323 may be included, and may further include the first image enhancement unit 321 and the second image enhancement unit 322 . And the third image enhancement unit 323 may further include any two combinations of the first image enhancement unit 321, the second image enhancement unit 322, and the third image enhancement unit 323.

It should be understood by those skilled in the art that the functions of the processing units in the image processing apparatus of the embodiments of the present invention can be understood by referring to the related description of the image processing method, and the processing units in the image processing apparatus according to the embodiments of the present invention can be implemented. The function of the analog circuit of the embodiment of the present invention can be implemented by using the software of the function described in the embodiment of the present invention on the smart terminal.

In this embodiment, the first image enhancement unit 321 , the second image enhancement unit 322 , and the third image enhancement unit 323 in the acquisition unit 31 and the image enhancement unit may be used by the image processing apparatus in an actual application. Realized by a central processing unit (CPU), a digital signal processor (DSP), or a Field-Programmable Gate Array (FPGA).

4 is a schematic structural diagram of an image processing system according to an embodiment of the present invention; as shown in FIG. 4, the image processing system includes: an enhancement module 41, a register configuration module 42 and a configuration synchronization module 43; wherein the enhancement module 41 An image processing apparatus according to an embodiment of the present invention shown in FIG. 3 is included;

The register configuration module 42 is configured to configure parameters required in the enhancement module 41;

The configuration synchronization module 43 is configured to synchronize parameters configured by the register configuration module 42 to the enhancement module 41.

Specifically, the enhancement module 41 includes: an input module 411, a first row cache module 412, a second row cache module 413, a reg register, a read cache module 414, a logic enhancement module 415, and an output module 416;

The input module 411 is configured to receive image data, and input the Nth row and the N+1th row of the image data into the first row buffer module 412 and the second row cache module 413, respectively; After the input of the Nth row of the image data is completed, input the N+1th row of the image data;

The read cache module 414 is configured to complete when the Nth row of the image data is input to the first row buffer module 412 and the N+1th row data of the image data is input to the second row cache In the module 413, the image data of the preset size is read from the first line buffer module 412; when the N+1th line of the image data is input to the second line buffer module 413 and the image is completed When the N+2th row data of the data is input into the first row buffer module 412, the preset size image data is read from the second row buffer module 413; wherein, when the image data When the N+2th row data is input into the first row buffer module 412, a preset number of image data in the Nth row data of the image data is input into the reg register; Decoding the preset size image data read in the first line buffer module 412, reading the preset size image data from the second line buffer module 413, and the preset number in the reg register Image data generates matrix data; wherein N is an integer;

The logic enhancement module 415 is configured to perform enhancement processing on matrix data in the read cache module 414; the enhancement processing includes a sharpening process and a smoothing process;

The output module 416 is configured to output matrix data processed by the logic enhancement module.

The image data for the present embodiment is YUV data, YUV and RGB are different color spaces, and are configured to represent colors; the YUV data can optimize the transmission of image signals, requiring only a small amount of bandwidth (RGB requirement three) Independent video signals are transmitted simultaneously; where Y is the brightness (Luminance or Luma), which is the luminance value; and U and V are the chrominance (Chrominance or Chroma), describing the image color and saturation, configured as Specifies the color of the pixel. In this embodiment, the first line buffer module 412 and the second line buffer module 413 are configured to buffer the Y component in the YUV data, and the third line cache module and the fourth line shown in FIG. The cache module is configured to buffer the U component and the V component in the YUV data. The U component and the V component in the YUV data are not protected by the embodiment of the present invention, and are not specifically described herein.

The reg register described in the embodiment of the present invention is not shown in FIG. 4, and the reg register is only configured to buffer the Y component of two pixels, and the capacity is small.

Specifically, the parameters configured by the register configuration module 42 include: a video resolution register (resolution_reg), a video format register (format_reg) (the video format register includes YUV444, YUV422, and YUV420, etc.), and a sharpening matrix selection register. (sharp_matrix), smoothing matrix selection register (smooth_matrix), enhanced intensity coefficient register (enhance_ratio) (the highest bit of the enumeration_ratio register is the sign bit, when the sign bit is configured to 0 for the sharpening operation, and the configuration is 1 for the smoothing operation) The edge detection threshold register (threshold_reg), the interrupt register (int_reg), the clear interrupt register (clear_int_reg), and the module enable register (enable); the configuration order of the above registers is: the last configuration enable register (enable), the configuration of other registers There are no special requirements for the order.

Data processing is performed according to the selected mode. After the module is turned on (ie, the enable register) Set to high level), Y component input channel input Y component of YUV data, U component input channel input U component of YUV data, V component input channel input V component of YUV data; Y component output channel output processed YUV data The Y component, the U component output channel outputs the U component of the processed YUV data, and the V component output channel outputs the V component of the processed YUV data.

When a frame of data processing is completed, the system will issue an int interrupt signal. If the system still needs to process data, you need to clear the interrupt register to high level, clear the int interrupt; reconfigure the corresponding register to process the next frame of data. If there is no data processing, turn off the system after clearing the int interrupt (configuring the enable register low).

The data flow of the image processing method of the embodiment of the present invention will be described in detail in conjunction with the image processing system shown in FIG. 5(a) to 5(e) are diagrams showing the data flow of the image processing method according to the embodiment of the present invention. In the embodiment, an input video source having image data of 4×4 is taken as an example for describing the input. The video source is the Y component in the YUV data.

As shown in FIG. 5(a), the enhancement module starts to input the first row data of the 4×4 original video source into the first row cache module;

As shown in FIG. 5(b), the first line of data input in the video source is completed, and after the first line of the second line is input to the second line buffer module, the first line is accompanied by the input of the subsequent video data. The first row of data stored in the cache module is output one by one (the output data of the video is longer than the input delay line resolution hor plus 1 pixel period, ie hor+1 cycles); wherein the first row cache module stores The data of the first two pixels in the first row of data is input to the reg memory;

As shown in FIG. 5(c), after the second line of data input is completed, the first data of the first line and the second data are stored in the reg register from the first line buffer module, and the third line is A data and a second data are respectively input to the idle position in the first line cache module; at this time, based on the currently input of the first row and the second row of the four data and the third row of the two data, can be composed A 3×3 matrix block (one pixel of the missing edge may be empty), since the image processing method in the embodiment of the present invention performs data processing on the pixel located in the middle of the matrix block, At the same time, in the absence of the edge of the pixel, image enhancement can still be performed for the pixel located in the middle; finally, the 3×3 matrix block is output from the line buffer module to the read buffer module, and then the logic enhancement module is used to 3×3 matrix blocks perform enhanced logic operations and output calculation results;

As shown in FIG. 5(d), after the data input of the third row is completed, the data of the fourth row is input to the second row cache module, and the first two pixels of the second row of data stored by the second row cache module are simultaneously The data is input to the reg memory; an operation process similar to that shown in FIG. 5(c) is performed to calculate the result of the enhancement processing;

As shown in FIG. 5(e), at this time, the enhancement module detects that the input of the last row of data has been completed. At this time, the data of the last row is read from the row buffer module, and since it cannot form a matrix of 3×3, it does not do Any processing.

Finally, after the Y component calculation in the YUV data is completed, the corresponding UV components are output together in synchronization.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention can take the form of a hardware embodiment, a software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Industrial applicability

In the embodiment of the present invention, the image data area is divided into the matrix image data of the preset size, thereby realizing the effective division of each area in the video to perform enhancement processing for the specific area, and facilitating the image processing method provided by the embodiment of the present invention. Hardware pipeline implementation.

Claims (15)

1. An image processing method, the method comprising:

   Obtaining image data; the image data is image data processed according to a preset scaling processing manner;

   Determining, when the image data meets the preset condition, sequentially acquiring pixel points of a preset size in the image data to form matrix image data, and processing the matrix image data according to a first processing manner; Processing mode, edge sharpening processing or smooth processing.
2. The method according to claim 1, wherein said determining said image data satisfies a preset condition comprises: determining said image data when said image data is image data processed in a three-time difference amplification processing manner The preset condition is met; correspondingly, the sequentially acquiring the pixel points of the preset size in the image data to form the matrix image data and processing the matrix image data according to the first processing manner includes: sequentially acquiring the image data in advance Setting pixel points of a size to form matrix image data and processing the matrix image data according to a detail sharpening processing manner;

   And determining that the image data meets the preset condition, the method includes: determining that the image data meets a preset condition when the image data is image data processed by the B-spline amplification processing manner; And sequentially acquiring the pixel points of the preset size in the image data to form matrix image data and processing the matrix image data according to the first processing manner, including: sequentially acquiring a matrix of pixel points of a preset size in the image data. Image data and processing the matrix image data in an edge sharpening process;

   And determining that the image data satisfies a preset condition, including: when the image data is image data processed in a reduction processing manner, determining that the image data meets a preset condition; correspondingly, the And sequentially acquiring the pixel data of the preset size in the image data to form the matrix image data, and processing the matrix image data according to the first processing manner, comprising: sequentially acquiring pixel points of the preset size in the image data to form matrix image data; Processing the matrix in a smoothing manner Image data.
3. The method according to claim 2, wherein the sequentially acquiring pixel points of a preset size in the image data constitutes matrix image data and processing the matrix image data in a detail sharpening processing manner, including: from the image The starting position of the data starts to sequentially acquire the luminance values of the 3×3 pixel points in the image data to form the first matrix data;

   Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

   Presetting a first threshold, a second threshold, and a third threshold, and dividing the luminance range into four intervals according to the first threshold, the second threshold, and the third threshold; wherein the first threshold, the The second threshold and the third threshold are both greater than 0 and less than 255;

   Determining, when the first gradient value of the central pixel point is in the second interval composed of the first threshold and the second threshold, performing sharpening processing on the first matrix data.
4. The method according to claim 2, wherein the sequentially acquiring pixels of a preset size in the image data to form matrix image data and processing the matrix image data in an edge sharpening processing manner comprises: The starting position of the data starts to sequentially acquire the luminance values of the 3×3 pixel points in the image data to form the first matrix data;

   Obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator;

   Pre-configuring a fourth threshold, and dividing the luminance range into two intervals according to the fourth threshold; wherein the fourth threshold is greater than 0 and less than 255;

   Determining, when the first gradient value of the central pixel point is in the interval formed by the fourth threshold and 255, performing sharpening processing on the first matrix data.
5. The method according to claim 3 or 4, wherein the sharpening the first matrix data comprises: acquiring the first matrix data according to a pre-configured first Laplacian a second gradient value of the center pixel;

   Obtaining a sharpening value of the central pixel point according to the second gradient value in a preset sharpening manner; wherein the preset sharpening manner comprises: obtaining an absolute value of the second gradient value and obtaining an average value a first parameter; limiting a maximum value of the first parameter to obtain a second parameter; multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point;

   When the second gradient value is greater than zero, the luminance value of the central pixel point + the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point;

   When the second gradient value is less than zero, the luminance value of the central pixel point - the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point.
6. The method according to claim 2, wherein the sequentially acquiring pixel points of a preset size in the image data constitutes matrix image data and processing the matrix image data in a smoothing manner, including: from the image data Starting from the starting position, sequentially acquiring 3×3 pixel points in the image data to form first matrix data;

   A smoothing result of the first matrix data is obtained according to a pre-configured second Laplacian and a smoothing intensity parameter.
7. An image processing apparatus, the apparatus comprising: an acquisition unit and an image enhancement unit; wherein

   The acquiring unit is configured to acquire image data; the image data is image data processed according to a preset scaling processing manner;

   The image enhancement unit is configured to, when determining that the image data acquired by the acquiring unit meets a preset condition, sequentially acquire pixel image data of a preset size in the image data to form matrix image data, and process the processing according to the first processing manner. The matrix image data is described; the first processing mode is a detail sharpening processing mode, an edge sharpening processing mode, or a smoothing processing mode.
8. The apparatus according to claim 7, wherein said image enhancement unit comprises a first image enhancement unit configured to determine said image data when said image data is image data processed in a three-time difference amplification processing manner And satisfying a preset condition; further configured to sequentially acquire pixel points of a preset size in the image data to form matrix image data, and process the matrix image data according to a detail sharpening processing manner;

   And/or, the image enhancement unit further includes a second image enhancement unit configured to determine that the image data meets a preset condition when the image data is image data processed in a B-spline enlargement processing manner; And configured to sequentially acquire pixel points of a preset size in the image data to form matrix image data, and process the matrix image data according to an edge sharpening processing manner;

   And/or the image enhancement unit further includes a third image enhancement unit configured to determine that the image data satisfies a preset condition when the image data is image data processed in a reduction processing manner; Obtaining pixel points of a preset size in the image data to form matrix image data and processing the matrix image data in a smoothing manner.
9. The apparatus according to claim 8, wherein the first image enhancement unit is configured to sequentially acquire luminance values of 3 × 3 pixel points in the image data from a starting position of the image data to constitute a first a matrix data; obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator; preconfiguring a first threshold, a second threshold, and a third threshold, according to the first threshold, the first The second threshold and the third threshold divide the luminance range into four intervals; wherein the first threshold, the second threshold, and the third threshold are both greater than 0 and less than 255; determining the number of the central pixel A gradient value is sharpened by the first matrix data in a second interval composed of the first threshold and the second threshold.
10. The apparatus according to claim 9, wherein the second image enhancement unit is configured to sequentially acquire luminance values of 3 × 3 pixel points in the image data from a starting position of the image data to constitute a first a matrix data; obtaining a first gradient value of a central pixel point of the first matrix data according to a Sobel operator; preconfiging a fourth threshold, and dividing the luminance range into two intervals according to the fourth threshold; The fourth threshold is greater than 0 and less than 255; and the first gradient value of the central pixel is determined to be sharpened in the interval formed by the fourth threshold and 255.
11. The apparatus according to claim 9 or 10, wherein the first image enhancement unit and the second image enhancement unit are each configured to be obtained according to a pre-configured first Laplacian And taking a second gradient value of the central pixel point of the first matrix data; obtaining a sharpening value of the central pixel point according to the second gradient value according to a preset sharpening manner; wherein the preset sharpening The method includes: averaging the second gradient value to obtain a first parameter; limiting a maximum value of the first parameter to obtain a second parameter; and multiplying the second parameter by a preset enhancement coefficient to obtain a sharpening value of the central pixel point; when the second gradient value is greater than zero, a luminance value of the central pixel point + a sharpening value of the central pixel point is used as a luminance value of the central pixel point Sharpening the result; when the second gradient value is less than zero, the luminance value of the central pixel point - the sharpening value of the central pixel point is used as a sharpening result of the luminance value of the central pixel point.
12. The apparatus according to claim 8, wherein said third image enhancement unit is configured to sequentially acquire 3 × 3 pixel points in said image data from said start position of said image data to constitute first matrix data And obtaining a smoothed result of the first matrix data according to a pre-configured second Laplacian and a smoothing intensity parameter.
13. An image processing system, comprising: an enhancement module, a register configuration module, and a configuration synchronization module; wherein the enhancement module comprises the image processing device according to any one of claims 7 to 12;

    The register configuration module is configured to configure parameters required in the enhancement module;

    The configuration synchronization module is configured to synchronize parameters of the register configuration module configuration to the enhancement module.
14. The image processing system according to claim 13, wherein the enhancement module comprises: an input module, a first line buffer module, a second line buffer module, a reg register, a read cache module, a logic enhancement module, and an output module;

    The input module is configured to receive image data, and input an Nth row and an N+1th row of the image data into the first row cache module and the second row cache module, respectively; After the input of the Nth line of the image data is completed, input the N+1th line of the image data;

    The read cache module is configured to be configured when the Nth row of the image data is input to the first row cache module and the N+1th row data of the image data is input into the second row cache module Reading the image data of the preset size from the first line buffer module; when the N+1th line of the image data is input to the second line buffer module and the N+ of the image data When the two lines of data are input into the first line buffer module, the preset size image data is read from the second line buffer module; wherein, when the N+2 line data of the image data is When inputting into the first line buffer module, a preset amount of image data in the Nth row of data of the image data is input into the reg register; and is further configured to read according to the first line cache module And taking the preset size image data, reading the preset size image data from the second line buffer module, and the preset number of image data in the reg register to generate matrix data; wherein, N Is an integer;

    The logic enhancement module is configured to perform enhancement processing on matrix data in the degree cache module; the enhancement processing includes a sharpening process and a smoothing process;

    The output module is configured to output matrix data processed by the logic enhancement module.
15. A computer storage medium having stored therein computer executable instructions for performing the image processing method of any one of claims 1 to 6.

Images