CN113129389B - Method for judging moire, method for inhibiting moire and circuit system - Google Patents

Abstract

In the method, brightness values of a plurality of pixels in an image are obtained, a plurality of key pixels for judging the moire type can be selected from a detection window, a moire response value of the plurality of key pixels and a plurality of adjacent pixels corresponding to the key pixels respectively is calculated for each pixel through the detection window, brightness values of the key pixels and the adjacent pixels are compared through the detection window, the comparison result is counted to judge brightness characteristics of each pixel, and then the moire position and the moire type in the image can be confirmed according to the moire response value and the counted result. Then, color noise reduction is performed on the plurality of pixels determined to be moire.

Description

Method for determining moiré, method for suppressing moiré and circuit system

Technical Field

The present invention relates to a technology for determining moiré patterns, and more particularly to a method for detecting moiré patterns in different directions, determining moiré patterns by comparing moiré feature statistics, and a method for suppressing the determined moiré patterns and a circuit system for implementing the method.

Background technique

Moiré Pattern is one of the common imaging defects in digital images. The general reason for the formation of moiré patterns is that when the photosensitive components in the image sensor (such as CCD or CMOS) obtain image data, such as taking photos with a digital camera, shooting videos with a digital video camera, or obtaining images with a scanner, the processed image is subject to high-frequency interference, and irregularly colored and shaped stripes appear on the image, which are called moiré patterns.

In addition, when shooting objects with dense patterns (such as textiles, highly repetitive lines, display screens, etc.), if the pixel sampling frequency of the photosensitive component is close to the spatial frequency of the patterns on the object, low-frequency texture will be generated on the image. At the same time, due to the use of Bayer filters, the sampling rates of red, green and blue visible light are different, and the moiré patterns are often accompanied by color noise, which is inconsistent with the actual results seen by the human eye.

In order to solve the moiré phenomenon, an optical low-pass filter is added to the lens when manufacturing the camera. Although it helps to reduce the moiré phenomenon, it will lose some image details. Another solution is to perform post-processing in the image signal processor (ISP). The location where the pixel brightness change is low and the color change is high can be judged as the location where the moiré occurs. Color compensation can be made based on the adjacent pixels, but this may cause the problem of reduced saturation.

Thus, the methods disclosed in the prior art may not be able to accurately detect moiré patterns. If there is a misjudgment, the color quality of the image will be reduced, and the color compensation mechanism may also be limited by hardware limitations and cannot completely eliminate the moiré patterns.

Summary of the invention

The invention discloses a method for determining moiré patterns, a method for suppressing moiré patterns and a circuit system for implementing the method. One of the purposes is to use image processing technology to process moiré patterns in digital images, including detecting the position of moiré patterns. One of the purposes is to reduce the color noise at the position where the moiré patterns occur, so that the imaging result is more in line with the visual perception of the human eye.

According to an embodiment of the method for determining moiré, brightness information of multiple pixels in an image is first obtained, such as a brightness value in a YUV (brightness-chroma-intensity) color space, or an average value of three color channel values in an RGB (red-green-blue) color space. Next, a detection window is set, and multiple key pixels for determining the type of moiré are selected in this detection window. A moiré response value of the multiple key pixels in the detection window and the corresponding multiple adjacent pixels is then calculated for each pixel one by one. This can be used to determine whether the image has moiré features.

Afterwards, for each pixel, the detection window is used to compare the brightness values of multiple key pixels and the corresponding multiple adjacent pixels, and the comparison results are statistically analyzed to determine the brightness characteristics of each pixel. Therefore, the position and type of moiré in the image can be confirmed based on the moiré response value and the statistical results.

Preferably, in the step of calculating the moiré response value, a weight mask is provided in the detection window, and this weight mask can be designed according to the type of moiré to be determined, including assigning higher weight values to multiple key pixels and assigning lower weight values to multiple neighboring pixels, and then multiplying the multiple key pixels and the corresponding multiple neighboring pixels by the weight values to calculate the moiré response value.

Furthermore, the moiré response value calculated for each pixel can be compared with the first threshold to obtain the brightness change of the pixel and its neighboring pixels, and then further determine whether the image has moiré characteristics. In addition, when comparing the brightness information of multiple key pixels and the corresponding multiple neighboring pixels, a second threshold can be introduced to confirm the brightness characteristics of each pixel.

The moiré pattern is a type of moiré pattern in horizontal and vertical directions, moiré pattern in a main diagonal direction, or moiré pattern in a sub-diagonal direction.

Furthermore, color moiré suppression can be performed on relevant pixels judged to be moiré, including mapping multiple pixels judged to be moiré to a chromaticity-density plane in the brightness-chromaticity-density color space, and then suppressing the pixel colors within a color suppression range in the chromaticity-density plane to grayscale. Furthermore, the chromaticity-density plane is further distinguished into a color suppression progressive range, and a suppression factor is set for the distance between the pixel color within the color suppression progressive range and a coordinate center of the chromaticity-density plane. Within this color suppression progressive range, color suppression can be performed according to the suppression factor.

According to an embodiment of the circuit system, it may be a digital image processor, in which the method for determining moiré and the method for suppressing moiré are executed.

To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are only for reference and description and are not intended to limit the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a functional block diagram showing an embodiment of a circuit system for implementing a method for determining and suppressing moiré;

FIG2 is a flow chart showing an embodiment of a method for determining and suppressing moiré;

FIG3 is a schematic diagram showing an embodiment of determining the direction of moiré in a method for determining moiré;

4A and 4B are schematic diagrams showing horizontal and vertical moiré patterns, respectively;

5A to 5C are schematic diagrams showing a method for determining horizontal and vertical moiré patterns;

6A to 6C are schematic diagrams showing a method for determining moiré in the main diagonal direction;

7A to 7C are schematic diagrams showing a method for determining moiré in a sub-diagonal direction;

FIG8 is a schematic diagram showing an embodiment of a color space in which a method of suppressing moiré is performed; and

FIG. 9 is a flow chart illustrating an embodiment of a method of suppressing moiré.

Detailed ways

The following is an explanation of the embodiments of the present invention through specific embodiments. Those skilled in the art can understand the advantages and effects of the present invention from the contents disclosed in this specification. The present invention can be implemented or applied through other different specific embodiments, and the details in this specification can also be modified and changed in various ways based on different viewpoints and applications without departing from the concept of the present invention. In addition, the drawings of the present invention are only simple schematic illustrations and are not depicted according to actual sizes. The following embodiments will further explain the relevant technical contents of the present invention in detail, but the disclosed contents are not intended to limit the scope of protection of the present invention.

It should be understood that, although the terms "first", "second", "third", etc. may be used herein to describe various components or signals, these components or signals should not be limited by these terms. These terms are mainly used to distinguish one component from another component, or one signal from another signal. In addition, the term "or" used herein may include any one or more combinations of the associated listed items depending on the actual situation.

According to the embodiments of the method for determining moiré, the method for suppressing moiré and the circuit system disclosed in the specification, in the embodiment of the method for determining moiré, moiré in different directions is mainly detected based on the moiré response value, that is, the distribution characteristics of pixels in the image are used to obtain the location where the moiré occurs, and then the moiré feature comparison value is statistically calculated to determine the type, and finally, in the method for suppressing moiré, color noise reduction is performed on the pixels determined to be moiré.

The types of moiré can be roughly divided into horizontal and vertical moiré, main diagonal and sub-diagonal moiré, etc. For example, the circuit system can analyze each pixel (or sampled pixel) channel in the image one by one to determine whether the pixel has horizontal, vertical or positive/sub-diagonal moiré, and then further suppress it. It is mentioned here that because the characteristics of moiré are repeated and dense lines, the brightness characteristics such as horizontal, vertical, and diagonal can be used as the conditions for judging moiré.

First, refer to Figure 1, which is a functional block diagram of an embodiment of a circuit system for implementing a method for determining and suppressing moiré. The circuit system may be a digital image processor or implemented by a computer system. The main circuits of the circuit system include a processor and a memory, wherein the processor executes the method for determining and suppressing moiré. According to the function, the method may include a color space conversion unit 102, a moiré response value calculation unit 103, a moiré feature statistics unit 104, a moiré determination unit 105, and a moiré suppression unit 106 implemented by software or hardware. The following will describe the process of the circuit system executing the method for determining moiré and the method for suppressing moiré. Please refer to the method embodiment process shown in Figure 2, as well as the function of each unit.

After the circuit system shown in FIG. 1 receives the original image 101 (step S201, FIG. 2), which may be a raw image file (RAW) or a pixel value of a color space, a color space conversion unit 102 converts the image into a specific color space that can perform determination and suppression of color moiré (step S203), such as a YUV (brightness-chroma-density) color space, an RGB (red-green-blue) color space, etc., in particular, any color space that can obtain pixel brightness information. The Y value of each pixel in the YUV color space represents the brightness value, and the RGB color space can be the average value of the three color channel values of each pixel after color restoration as the basis for determining the brightness.

Then the circuit system can determine a detection window according to the hardware computing capability, and determine pixel by pixel whether each pixel is the location where moiré occurs. The method is to select multiple key pixels for determining the type of moiré in the detection window first, and calculate each pixel one by one through the detection window to obtain the distribution characteristics of the key pixels and their neighboring pixels in each area (corresponding to the detection window size) in the image to obtain the location of the moiré. In the embodiment, the moiré response value calculation unit 103 in the circuit system can apply the weight values of each pixel set in the detection window, and use the brightness value of the pixel to calculate the moiré response value in the horizontal, vertical or diagonal direction one by one (step S205). The moiré response value in each direction can determine whether each pixel belongs to an edge pixel (edge pixel) that meets the moiré feature. Afterwards, the moiré feature statistics unit 104 in the circuit system obtains the moiré feature relationship between several key pixels for determining the direction of moiré and neighboring pixels in each detection window, which can be the brightness value of the comparison pixel, and the moiré feature comparison value of the adjacent pixels is counted (step S207).

Next, the moiré judgment unit 105 of the circuit system can check the moiré response value and the moiré feature comparison value obtained in the above steps according to the threshold value set by the system to determine whether the current pixel is part of the moiré (step S209), and then the moiré suppression unit 106 performs color suppression on the pixel determined to be part of the moiré, wherein the color of the pixel in the moiré range can be suppressed to grayscale, or different suppression ratios can be set according to different degrees, and then the color of the pixel is suppressed to grayscale (step S211), and finally the result is output, that is, the output image 107 after moiré suppression is output (step S213).

Reference may be made to FIG. 3 for a schematic diagram of an embodiment of determining the direction of moiré in a method for determining moiré.

FIG3 shows a region of a 5x5 pixel array in an image. In actual implementation, an appropriate pixel array may be selected according to hardware resources. In this embodiment, the Y value (brightness value) of the pixel in the YUV color space is used as the calculation basis. The pixel currently performing the judgment in the 5x5 pixel array is Y(i,j), which is used as the origin of the region to mark the adjacent pixels. For example, the horizontal pixels of the current pixel Y(i,j) include: Y(i,j-2), Y(i,j-1), Y(i,j), Y(i,j+1) and Y(i,j+2); the vertical pixels include: Y(i-2,j), Y(i-1,j), Y(i,j), Y(i+1,j) and Y(i+2,j); the main diagonal (upper left and lower right) pixels include: Y(i-2,j-2), Y(i-1,j-1), Y(i,j), Y(i+1,j+1) and Y(i+2,j+2).

FIG4A is a schematic diagram showing a pixel array with vertical moiré. In a detection window, the central pixel is the current pixel 40. According to the characteristics of the vertical moiré, the characteristic of the vertical moiré is that the brightness values of the pixels in the vertical direction are closer than those in other directions (a threshold value can be set to determine the degree of proximity). In addition to the pixels in the vertical direction, the brightness value difference between the current pixel 40 and other adjacent pixels (left, right, upper left, lower left, upper right, lower right) should be relatively large (a threshold value can be set to determine the difference size). Therefore, this feature can be used to find vertical moiré.

FIG4B is a schematic diagram of a pixel array with horizontal moiré, in which the central pixel is the current pixel 40′. Similarly, in FIG4B , the characteristic of the horizontal moiré is that the pixel brightness values in the horizontal direction are relatively close, and except for the pixels in the horizontal direction, the brightness value difference between the current pixel 40′ and other adjacent pixels (upper, lower, upper left, lower left, upper right, lower right) is relatively large. Therefore, this feature can also be used to find horizontal moiré.

Embodiment 1: Horizontal and vertical moiré patterns.

Based on the above-mentioned characteristics of horizontal and vertical moiré, as shown in FIGS. 5A to 5C , the method for determining moiré is to determine whether there is horizontal or vertical moiré by calculating the difference in brightness (such as Y value) between the current pixel 50 and the adjacent oblique (diagonal) pixels (called key pixels). FIG. 5A uses the central pixel as the current pixel 50, and the brightness difference between the central pixel and the oblique pixels (indicated by arrows) as the basis for determining horizontal or vertical moiré. FIG. 5B is another representation method, which requires obtaining the brightness difference between several key pixels in the 5x5 pixel array and their oblique pixels (indicated by arrows) as the basis for determining horizontal or vertical moiré. FIG5C shows an example of a weight mask set for a key pixel and its neighboring pixels in a detection window. The weight mask is designed according to the type to be determined, and a higher weight value can be assigned to the key pixel when calculating the moiré response value. In this embodiment, the weight values are 4 and 1. The lower pixel 50 can be set with the highest weight value 4, and the remaining key pixels are set with a weight value of 1 as coefficients of the calculation formula. Conversely, the diagonal neighboring pixels of the key pixel are assigned a lower weight value (-2 in this embodiment), thereby emphasizing the spatial distribution characteristics of the image.

Referring to the pixels marked in Figure 3 and Figures 5A to 5C, Formula A represents the calculation of the horizontal or vertical moiré response value of the specific area corresponding to the detection window for each pixel in the image using the detection window. Formula A shows that the first half of the formula "(4*Y(i,j)+Y(i-2,j)+Y(i,j+2)+Y(i+2,j)+Y(i,j-2))" is used to calculate the sum of the brightness values of several key pixels in a certain image area, and in particular, it can be multiplied by the weight value shown in Figure 5C; the second half of Formula A "(2*Y(i-1,j-1)+2*Y(i-1,j+1)+2*Y(i+1,j+1)+2*Y(i+1,j-1))" is used to calculate the sum of the brightness values of the obliquely adjacent pixels of the key pixel (which can be multiplied by the weight value shown in Figure 5C). Subtracting the previous and next formulas of formula A and taking the absolute value will give the brightness gradient in this pixel area. When the brightness gradient of the entire image is calculated, the larger the difference (absolute value), the more prominent the horizontal or vertical moiré in the image. Otherwise, it means there is no obvious moiré. This difference is expressed as the moiré response value in the horizontal and vertical directions (Moire_HVEdge), which is to obtain the edge characteristics of the pixel to determine whether it has moiré characteristics. It also confirms that moiré will occur when the sampling frequency of the pixel is consistent with or close to the spatial frequency of the pattern.

Moire_HVEdge＝|(4*Y(i,j)+Y(i-2,j)+Y(i,j+2)+Y(i+2,j)+Y(i,j-2))-(2*Y(i-1,j-1)+2*Y(i-1,j+1)+2*Y(i+1,j+1)+2*Y(i+1,j-1))|(Formula A)

Next, using the detection window, the brightness values of multiple key pixels and the corresponding adjacent pixels in the pixel area where each pixel is located are compared one by one, and statistics are performed to detect the edge and determine it as a moire. Formula B and Formula C represent the statistical horizontal or vertical moire feature comparison value. According to the pixel brightness value indicated in Figure 1, if the current pixel brightness value Y(i,j) is a relatively high brightness value, as shown in Formula B, where KHV represents the difference in pixel values set by the user, it is determined according to the actual ambient light source as a threshold for confirming that the brightness value difference between the key pixels (Y(i,j), Y(i-2,j), Y(i,j+2), Y(i+2,j), Y(i,j-2)) in the horizontal and vertical directions and their adjacent key pixels is greater than (the brightness value is at least greater than KHV) and can be judged as a moire. Among them, the brightness value comparison and statistics can be performed only for the pixels confirmed as part of the moire through the moire value response value.

Formula B:

Moire_HVCMP1＝(Y(i,j)>Y(i-1,j-1)+KHV)

Moire_HVCMP2＝(Y(i,j)>Y(i-1,j+1)+KHV)

Moire_HVCMP3＝(Y(i,j)>Y(i+1,j+1)+KHV)

Moire_HVCMP4＝(Y(i,j)>Y(i+1,j-1)+KHV)

Moire_HVCMP5＝(Y(i-2,j)>Y(i-1,j-1)+KHV)

Moire_HVCMP6＝(Y(i-2,j)>Y(i-1,j+1)+KHV)

Moire_HVCMP7＝(Y(i,j+2)>Y(i-1,j+1)+KHV)

Moire_HVCMP8＝(Y(i,j+2)>Y(i+1,j+1)+KHV)

Moire_HVCMP9＝(Y(i+2,j)>Y(i+1,j+1)+KHV)

Moire_HVCMP10＝(Y(i+2,j)>Y(i+1,j-1)+KHV)

Moire_HVCMP11＝(Y(i,j-2)>Y(i-1,j-1)+KHV)

Moire_HVCMP12＝(Y(i,j-2)>Y(i+1,j-1)+KHV)

If the current pixel brightness value Y(i,j) is a relatively low brightness value, as shown in Formula C, KHV is also used as the threshold to confirm that the difference in brightness between the key pixels (Y(i,j), Y(i-2,j), Y(i,j+2), Y(i+2,j), Y(i,j-2)) in the horizontal and vertical directions and their adjacent pixels is greater than (using KHV to increase the brightness difference threshold) to be judged as moiré.

Formula C:

Moire_HV’CMP1＝(Y(i,j)<Y(i-1,j-1)-KHV)

Moire_HV’CMP2＝(Y(i,j)<Y(i-1,j+1)-KHV)

Moire_HV’CMP3＝(Y(i,j)<Y(i+1,j+1)-KHV)

Moire_HV’CMP4＝(Y(i,j)<Y(i+1,j-1)-KHV)

Moire_HV’CMP5＝(Y(i-2,j)<Y(i-1,j-1)-KHV)

Moire_HV’CMP6＝(Y(i-2,j)<Y(i-1,j+1)-KHV)

Moire_HV’CMP7＝(Y(i,j+2)<Y(i-1,j+1)-KHV)

Moire_HV’CMP8＝(Y(i,j+2)<Y(i+1,j+1)-KHV)

Moire_HV’CMP9＝(Y(i+2,j)<Y(i+1,j+1)-KHV)

Moire_HV’CMP10＝(Y(i+2,j)<Y(i+1,j-1)-KHV)

Moire_HV’CMP11＝(Y(i,j-2)<Y(i-1,j-1)-KHV)

Moire_HV’CMP12＝(Y(i,j-2)<Y(i+1,j-1)-KHV)

Afterwards, the statistical results of the horizontal and vertical moiré features obtained by formula B and formula C are compared, such as formula D, to compare the statistical results of the current pixel brightness value Y(i,j) being relatively high and low, so as to determine whether the current pixel is relatively high or low in brightness relative to the adjacent pixels.

Formula D:

Embodiment 2: Moiré in the main diagonal direction.

For the main diagonal direction type, the above-mentioned moiré features of the diagonal direction are summarized, as shown in Figures 6A to 6C. The figure also takes a 5x5 pixel array as an example. In Figure 6A, the central pixel is the current pixel 60. The method for judging the moiré is to judge whether there is moiré in the main diagonal direction by calculating the difference in brightness value (such as Y value) between the current pixel 60 and the horizontal and vertical adjacent pixels. As shown by the arrows in the figure, the brightness value difference between the current pixel 60 and the horizontal and vertical pixels is used as the basis for judging the main diagonal moiré. Figure 6B is also another representation method. The brightness value difference between several key pixels in the 5x5 pixel array and their horizontal and vertical adjacent pixels (as indicated by arrows) is used as the basis for judging the main diagonal moiré. FIG6C shows an example of a weight mask, which shows that when calculating the moiré response value, higher weight values are assigned to key pixels, which are 4 and 1 in this embodiment. The bottom pixel 60 can be set to the highest weight value 4, and the remaining key pixels are set to a weight value of 1; conversely, the pixels in the horizontal and vertical directions of the key pixels are assigned lower weight values (-2 in this embodiment), thereby emphasizing the spatial distribution characteristics of the image.

Referring to the pixels marked in Figure 3 and Figures 6A to 6C, Formula E calculates the main diagonal moiré response value of the specific area corresponding to the detection window for each pixel in the image using the detection window. Similarly, Formula E shows that the first half of the formula "(4*Y(i,j)+Y(i-1,j+1)+Y(i-2,j+2)+Y(i+1,j-1)+Y(i+2,j-2))" in the absolute value is used to calculate the sum of the brightness values of several key pixels in a certain image area, especially it can be multiplied by the weight value shown in Figure 6C; the second half of Formula E "(2*Y(i-2,j)+2*Y(i,j+2)+2*Y(i+2,j)+2*Y(i,j-2))" is used to calculate the sum of the brightness values of the key pixels (such as the current pixel Y(i,j)) in the horizontal and vertical directions (which can be multiplied by the weight value shown in Figure 6C). Subtracting the previous and next formulas of formula E and taking the absolute value will give the brightness gradient in this pixel area. Once the brightness gradient of the entire image is calculated, the larger the difference (absolute value), the more prominent the main diagonal moiré in the image; otherwise, it means there is no obvious moiré. This difference is the moiré response value in the main diagonal direction (Moire_DIAG_POSEdge), and the edge characteristics of the pixel are obtained to determine whether it has moiré characteristics.

Formula E:

Moire_DIAG_POSEdge＝

|(4*Y(i,j)+Y(i-1,j+1)+Y(i-2,j+2)+Y(i+1,j-1)+Y(i+2,j-2))-(2*Y(i-2,j)+2*Y(i,j+2)+2*Y(i+2,j)+2*Y(i,j-2))|

Similarly, using the detection window, the brightness values of multiple key pixels and the corresponding adjacent pixels in the pixel area where each pixel is located are compared one by one, and statistics are performed, in order to detect edge characteristics and determine whether it is moiré. Formulas F and G represent the statistical main diagonal moiré feature comparison values. According to the pixel brightness values indicated in FIG1, if the current pixel brightness value Y(i,j) is a relatively high brightness value, as shown in Formula F, where KD represents the pixel value difference set by the user, which is determined according to the actual ambient light source as a threshold for confirming that the brightness value difference between the key pixels (Y(i,j), Y(i-2,j-1), Y(i-1,j-2), Y(i+1,j+2), Y(i+2,j+1)) in the main diagonal direction and its adjacent pixels is greater than (the brightness value is at least greater than KD) and can be determined as moiré. Among them, the brightness value comparison and statistics can be performed only for pixels that are confirmed as part of the moiré through the moiré value response value.

Formula F:

Moire_DIAG_POS CMP1＝(Y(i,j)>Y(i-2,j)+KD)

Moire_DIAG_POS CMP2＝(Y(i,j)>Y(i,j+2)+KD)

Moire_DIAG_POS CMP3＝(Y(i,j)>Y(i+2,j)+KD)

Moire_DIAG_POS CMP4＝(Y(i,j)>Y(i,j-2)+KD)

Moire_DIAG_POS CMP5＝(Y(i-2,j-1)>Y(i-2,j)+KD)

Moire_DIAG_POS CMP6＝(Y(i-2,j-1)>Y(i-1,j-1)+KD)

Moire_DIAG_POS CMP7＝(Y(i-1,j-2)>Y(i-1,j-1)+KD)

Moire_DIAG_POS CMP8＝(Y(i-1,j-2)>Y(i,j-2)+KD)

Moire_DIAG_POS CMP9＝(Y(i+1,j+2)>Y(i,j+2)+KD)

Moire_DIAG_POS CMP10＝(Y(i+1,j+2)>Y(i+1,j+1)+KD)

Moire_DIAG_POS CMP11＝(Y(i+2,j+1)>Y(i+2,j)+KD)

Moire_DIAG_POS CMP12＝(Y(i+2,j+1)>Y(i+1,j+1)+KD)

If the current pixel brightness value Y(i,j) is a relatively low brightness value, as shown in Formula G, KD is also used as a threshold to confirm that the brightness difference between the key pixel and its neighboring pixels in the main diagonal direction is greater than (using KD to increase the brightness difference threshold) and can be judged as moiré.

Formula G:

Moire_DIAG_POS’CMP1＝(Y(i,j)<Y(i-2,j)-KD)

Moire_DIAG_POS’CMP2＝(Y(i,j)<Y(i,j+2)-KD)

Moire_DIAG_POS’CMP3＝(Y(i,j)<Y(i+2,j)-KD)

Moire_DIAG_POS’CMP4＝(Y(i,j)<Y(i,j-2)-KD)

Moire_DIAG_POS’CMP5＝(Y(i-2,j-1)<Y(i-2,j)-KD)

Moire_DIAG_POS’CMP6＝(Y(i-2,j-1)<Y(i-1,j-1)-KD)

Moire_DIAG_POS’CMP7＝(Y(i-1,j-2)<Y(i-1,j-1)-KD)

Moire_DIAG_POS’CMP8＝(Y(i-1,j-2)<Y(i,j-2)-KD)

Moire_DIAG_POS’CMP9＝(Y(i+1,j+2)<Y(i,j+2)-KD)

Moire_DIAG_POS’CMP10＝(Y(i+1,j+2)<Y(i+1,j+1)-KD)

Moire_DIAG_POS’CMP11＝(Y(i+2,j+1)<Y(i+2,j)-KD)

Moire_DIAG_POS’CMP12＝(Y(i+2,j+1)<Y(i+1,j+1)-KD)

Afterwards, the statistical results of the main diagonal moiré features obtained by formula F and formula G are compared, such as formula H, to compare the statistical results of the current pixel brightness value Y(i,j) being relatively high and low, in order to determine whether the current pixel is a pixel with relatively high or low brightness.

Formula H:

Embodiment 3: Moiré in the sub-diagonal direction.

For the sub-diagonal direction type, as shown in Figures 7A to 7C, taking a 5x5 pixel array as an example, as shown in Figure 7A, the central pixel is the current pixel 70, and the method for judging moiré is to judge whether there is moiré in the sub-diagonal direction by calculating the difference in brightness value (such as Y value) between the current pixel 70 and the horizontal and vertical adjacent pixels, as shown by the arrows in the figure. Figure 7B is another representation method, where the brightness value difference between several key pixels in the 5x5 pixel array and their horizontal and vertical adjacent pixels (as shown by the arrows) is used as the basis for judging sub-diagonal moiré. Figure 7C shows an example of a weight mask proposed for judging sub-diagonal moiré, which shows that when calculating the moiré response value, a higher weight value is given to the key pixel, which is 4 and 1 in this embodiment. The current pixel 70 can be set to the highest weight value 4, and the remaining key pixels are set to a weight value of 1; on the contrary, the pixels in the horizontal and vertical directions of the key pixel are given a lower weight value (-2 in this embodiment), thereby emphasizing the spatial distribution characteristics of the image.

Referring to the pixels marked in Figure 3 and Figures 7A to 7C, Formula I calculates the sub-diagonal moiré response value of the specific area corresponding to the detection window for each pixel in the image using the detection window. Similarly, Formula I shows that the first half of the absolute value formula "(4*Y(i,j)+Y(i-1,j-1)+Y(i-2,j-2)+Y(i+1,j+1)+Y(i+2,j+2))" is used to calculate the sum of the brightness values of several key pixels in a certain image area, especially it can be multiplied by the weight value shown in Figure 7C; the second half of Formula I "(2*Y(i-2,j)+2*Y(i,j+2)+2*Y(i+2,j)+2*Y(i,j-2))" is used to calculate the sum of the brightness values of the key pixels (such as the current pixel Y(i,j)) in the horizontal and vertical directions (which can be multiplied by the weight value shown in Figure 7C). Subtracting the previous and next formulas of formula I and taking the absolute value will give the brightness gradient in this pixel area. Once the brightness gradient of the entire image is calculated, the larger the difference (absolute value), the more prominent the sub-diagonal moiré in the image; otherwise, it means there is no obvious moiré. This difference is the moiré response value in the sub-diagonal direction (Moire_DIAG_POSEdge), and the edge characteristics of the pixel are obtained to determine whether it has moiré characteristics.

Formula I:

Moire_DIAG_NEGEdge＝

|(4*Y(i,j)+Y(i-1,j-1)+Y(i-2,j-2)+Y(i+1,j+1)+Y(i+2,j+2))-(2*Y(i-2,j)+2*Y(i,j+2)+2*Y(i+2,j)+2*Y(i,j-2))|

Similarly, using the detection window, the brightness values of multiple key pixels and the corresponding adjacent pixels in the pixel area where each pixel is located are compared one by one, and statistics are performed, in order to detect edge characteristics and determine them as moiré. Formula J and Formula K represent statistical sub-diagonal moiré feature comparison values. According to the pixel brightness values indicated in Figure 1, if the current pixel brightness value Y(i,j) is a relatively high brightness value, as shown in Formula J, KD represents the pixel value difference set by the user, which is determined according to the actual ambient light source, as a threshold for confirming that the brightness value difference between the key pixel and its adjacent pixels in the sub-diagonal direction is greater than the threshold value that can be judged as moiré. Among them, brightness value comparison and statistics can be performed only for pixels that are confirmed as part of the moiré through the moiré value response value.

Formula J:

Moire_DIAG_NEGCMP1＝(Y(i,j)>Y(i-2,j)+KD)

Moire_DIAG_NEGCMP2＝(Y(i,j)>Y(i,j+2)+KD)

Moire_DIAG_NEGCMP3＝(Y(i,j)>Y(i+2,j)+KD)

Moire_DIAG_NEGCMP4＝(Y(i,j)>Y(i,j-2)+KD)

Moire_DIAG_NEGCMP5＝(Y(i-2,j+1)>Y(i-2,j)+KD)

Moire_DIAG_NEGCMP6＝(Y(i-2,j+1)>Y(i-1,j+1)+KD)

Moire_DIAG_NEGCMP7＝(Y(i-1,j+2)>Y(i-1,j+1)+KD)

Moire_DIAG_NEGCMP8＝(Y(i-1,j+2)>Y(i,j+2)+KD)

Moire_DIAG_NEGCMP9＝(Y(i+1,j-2)>Y(i,j-2)+KD)

Moire_DIAG_NEGCMP10＝(Y(i+1,j-2)>Y(i+1,j-1)+KD)

Moire_DIAG_NEGCMP11＝(Y(i+2,j-1)>Y(i+1,j-1)+KD)

Moire_DIAG_NEGCMP12＝(Y(i+2,j-1)>Y(i+2,j)+KD)

If the current pixel brightness value Y(i,j) is a relatively low brightness value, as shown in formula K, KD is also used as a threshold to confirm that the brightness difference between the key pixel and its neighboring pixels in the sub-diagonal direction is greater than (using KD to increase the brightness difference threshold) and can be judged as moiré.

Formula K:

Moire_DIAG_NEG’CMP1＝(Y(i,j)<Y(i-2,j)-KD)

Moire_DIAG_NEG’CMP2＝(Y(i,j)<Y(i,j+2)-KD)

Moire_DIAG_NEG’CMP3＝(Y(i,j)<Y(i+2,j)-KD)

Moire_DIAG_NEG’CMP4＝(Y(i,j)<Y(i,j-2)-KD)

Moire_DIAG_NEG’CMP5＝(Y(i-2,j+1)<Y(i-2,j)-KD)

Moire_DIAG_NEG’CMP6＝(Y(i-2,j+1)<Y(i-1,j+1)-KD)

Moire_DIAG_NEG’CMP7＝(Y(i-1,j+2)<Y(i-1,j+1)-KD)

Moire_DIAG_NEG’CMP8＝(Y(i-1,j+2)<Y(i,j+2)-KD)

Moire_DIAG_NEG’CMP9＝(Y(i+1,j-2)<Y(i,j-2)-KD)

Moire_DIAG_NEG’CMP10＝(Y(i+1,j-2)<Y(i+1,j-1)-KD)

Moire_DIAG_NEG’CMP11＝(Y(i+2,j-1)<Y(i+1,j-1)-KD)

Moire_DIAG_NEG’CMP12＝(Y(i+2,j-1)<Y(i+2,j)-KD)

Afterwards, the statistical results of the sub-diagonal moiré features obtained by formula J and formula K are compared, such as formula L, to compare the statistical results of the current pixel brightness value Y(i,j) being relatively high and low, in order to determine whether the current pixel is a pixel with relatively high or low brightness.

Formula L:

According to the above embodiment, in the method for determining moiré, the horizontal and vertical, positive/sub-diagonal moiré response values are first calculated, and then the brightness information of the current pixel is compared with the adjacent pixels in the horizontal and vertical, positive/sub-diagonal directions, and the comparison results are statistically analyzed to determine the brightness characteristics of the current pixel compared with the adjacent pixels. Among them, the moiré response value and the statistical results can be used to determine whether the current pixel is a moiré according to the threshold set by the system, including determining the position and type of the moiré in the image.

According to one embodiment, when the moiré response value is greater than the first threshold, it can be determined that the current pixel and its neighboring pixels have a clear brightness change, and thus can be determined to be an edge pixel of the image, that is, there is a moiré-like light and dark interval feature near this pixel; conversely, if the moiré response value is not greater than the first threshold, it cannot be determined that the current pixel has the moiré feature. When the moiré feature comparison value is greater than the second threshold, it is determined that the current pixel is relatively bright or relatively dark, and combined with the determination that the moiré response value is greater than the first threshold, it can be determined that the current pixel is part of the moiré.

After the moiré pattern in the image is determined, the circuit system can perform color noise reduction on the pixel determined to be the moiré pattern. For example, the pixel determined to be the moiré pattern is subjected to color suppression in the YUV color space. FIG8 shows the UV (chroma-density) plane in the YUV color space when Y=128. The pixel determined to be the moiré pattern is mapped in the chroma-density plane in the YUV color space. The closer to the coordinate center 80, the lower the color saturation, the closer to the grayscale, and the farther away from the coordinate center 80, the higher the color saturation, and different quadrants represent different colors.

According to the coordinate position of the color U/V value of the moiré pixel in the color space, the chromaticity-concentration plane can be divided into two or three areas, each of which undergoes different color processing. As shown in FIG8 , the central area represents a color suppression range 801, and the pixel colors falling within this color suppression range 801 are suppressed to grayscale, and then an image with suppressed color moiré is output.

According to one embodiment, the color-concentration plane may have three regions. The region between the dotted line frame and the color suppression range 801 in the figure represents a color suppression progressive range 803, so that the pixel colors within the color suppression progressive range 803 between the color suppression range 801 and the dotted line frame are adjusted to set a suppression magnification according to the distance from the coordinate center 80, so as to perform color suppression according to the suppression magnification, such as applying different grayscale levels. The suppression magnification may change linearly or nonlinearly, and the colors outside the dotted line of the color suppression progressive range 803 are not affected and maintain the original U/V value.

According to one embodiment, the color suppression range 801 is a movable rectangular area, but it must still ensure that the coordinate center 80 is covered, and in different application scenarios or according to user preferences, the rectangular area can be flexibly moved to cover the color to be suppressed. The color suppression formula is as shown in Formula M:

Cout = (Cin - 128) * (1 - supp_rate) + 128 (Formula M)

Wherein, Cin is the input U or V value, Cout is the corresponding output U or V value, and supp_rate is the suppression rate between 0 and 1. In the color suppression progressive range 803, the suppression rate is calculated based on the distance from the coordinate center 80. The farther from the center, the gradually decreasing suppression rate. The calculation method of the suppression rate from 0 to 1 can be obtained by interpolation, filtering, etc. The value 128 in formula M is based on the range of U and V values in the YUV color space in this example between 0 and 255. The conversion to the coordinate plane requires a displacement of 128. The actual implementation can be modified according to actual needs, and the calculation of the suppression rate is not limited to the listed methods.

According to the above embodiments, the process of the method for determining moiré and the method for suppressing moiré running in the circuit system first obtains the brightness information of each pixel in the received image in a color space, and then selects a detection range. The spatial distribution characteristics are obtained according to the brightness relationship between the pixels, and the edge characteristics of the pixels are obtained to determine whether there is moiré and whether the moiré is horizontal, vertical, or positive/sub-diagonal. Then, the feature comparison values of the specific type of moiré are statistically analyzed to determine whether the pixel is part of the moiré, and then color moiré suppression is performed on the pixels determined to be moiré.

In the circuit system, the process of judging and suppressing moiré can refer to FIG9. In step S901, after receiving the image, the circuit system judges the moiré of pixel processing one by one, and obtains the relationship between the current pixel and its neighboring pixel values (especially the brightness value) according to a detection window to obtain the characteristics of the image space distribution. According to the above embodiment, when the horizontal or vertical moiré is set, a key pixel in a detection window is selected, a weight value is assigned, and the horizontal and vertical moiré response values are calculated, that is, the brightness difference between the key pixel and its neighboring pixels is used as the basis for judging the direction of the moiré after comparing the first threshold.

In step S903, after processing each pixel one by one, it is determined whether the image has horizontal or vertical moiré. If it is determined that the image has horizontal or vertical moiré, color suppression is performed on the pixels at the positions where the moiré is determined to be present (step S907), and the image is output after completion (step S909); otherwise, the process continues to step S905 to determine whether the image has positive or sub-diagonal moiré. Similarly, if it is determined that the image has diagonal moiré, color suppression is performed on the moiré positions as in step S907, and the image is output after completion (step S909); otherwise, if no diagonal moiré is obtained in the image, it means that no clear moiré is found in the image, and the image is directly output (step S909). In this way, the method and circuit system for determining and suppressing moiré disclosed in the specification are implemented.

The contents disclosed above are only preferred feasible embodiments of the present invention, and are not intended to limit the scope of the present invention. Therefore, all equivalent technical changes made using the contents of the present invention's specification and drawings are included in the scope of the present invention.

【Symbol Description】

Raw Images 101

Color space conversion unit 102

Moire response value calculation unit 103

Moiré feature statistics unit 104

Moire pattern determination unit 105

Moire suppression unit 106

Output image 107

Current pixels 40,40’,50,60,70

Coordinate center 80

Color suppression range 801

Color suppression progressive range 803

Steps S201 to S213: Flow of the method for judging and suppressing moiré

Steps S901 to S909: Method flow for judging and suppressing moiré

Claims (9)

1. A method of determining moire comprising:

Obtaining brightness information of a plurality of pixels in an image;

a detection window is arranged, and a plurality of key pixels for judging the mole pattern type are selected from the detection window;

for each pixel, calculating a moire response value of the plurality of key pixels and a plurality of adjacent pixels corresponding to the key pixels respectively through the detection window, wherein the moire response value is used for judging whether the image has moire characteristics or not;

Comparing brightness information of the plurality of key pixels and a plurality of corresponding adjacent pixels by using the detection window for each pixel, and counting comparison results to judge brightness characteristics of each pixel; and

Confirming the position and type of the moire in the image according to the moire response value and the statistical result,

In the step of calculating the moire response value, a weight mask is set in the detection window, and the plurality of key pixels and the corresponding plurality of adjacent pixels are multiplied by weight values respectively to calculate the moire response value.

2. The method according to claim 1, wherein the luminance information of the plurality of pixels is a luminance value in a luminance-chrominance-density color space or an average value of three color channel values in a red-green-blue color space.

3. The method of claim 2, wherein a color space conversion is performed on the image to convert to the luminance-chrominance-density color space or the red-green-blue color space when the image is acquired.

4. The method according to claim 1, wherein in the step of calculating the moire response value, the weight mask is designed according to the type of moire to be judged, higher weight values are given to the plurality of key pixels, and lower weight values are given to the plurality of adjacent pixels.

5. The method according to claim 4, wherein the moire response value calculated for each pixel is compared with a first threshold value to obtain a brightness change between the pixel and its neighboring pixels, and further determining whether the image has moire features.

6. The method of claim 1, wherein a second threshold is introduced to verify the brightness characteristics of each pixel when comparing the brightness information of the plurality of key pixels with the brightness information of the plurality of neighboring pixels respectively corresponding to the plurality of key pixels.

7. The method according to any one of claims 1 to 6, wherein the type of moire is a moire in a horizontal and vertical direction, a moire in a main diagonal direction, or a moire in a sub diagonal direction.

8. A method of inhibiting moire comprising:

obtaining an image, converting the image into a brightness-chromaticity-concentration color space, and obtaining brightness values of a plurality of pixels in the image;

A detection window is arranged, and a plurality of key pixels used for judging the mole pattern type are selected from the detection window;

calculating a moire response value of the plurality of key pixels and a plurality of adjacent pixels corresponding to each pixel through the detection window, wherein the moire response value is used for judging whether the image has moire characteristics or not;

In the step of calculating the moire response value, a weight mask is set in the detection window, and the plurality of key pixels and the corresponding plurality of adjacent pixels are multiplied by weight values respectively to calculate the moire response value;

comparing brightness values of the plurality of key pixels and a plurality of corresponding adjacent pixels by using the detection window for each pixel, and counting comparison results to judge brightness characteristics of each pixel;

confirming the positions and types of the moire patterns in the image according to the moire response values and the statistical results;

Color noise reduction is performed on a plurality of pixels judged to be moire, wherein:

mapping a plurality of pixels determined as moire on a chrominance-density plane of the luminance-chrominance-density color space; and

And suppressing the pixel color in a color suppression range in the chromaticity-concentration plane to gray level.

9. A circuit system, comprising:

A processor and a memory, wherein the processor performs a moire determining method comprising:

obtaining an image, converting the image into a brightness-chromaticity-concentration color space, and obtaining brightness values of a plurality of pixels in the image;

A detection window is arranged, and a plurality of key pixels used for judging the mole pattern type are selected from the detection window;

calculating a moire response value of the plurality of key pixels and a plurality of adjacent pixels corresponding to each pixel through the detection window, wherein the moire response value is used for judging whether the image has moire characteristics or not;

In the step of calculating the moire response value, a weight mask is set in the detection window, and the plurality of key pixels and the corresponding plurality of adjacent pixels are multiplied by weight values respectively to calculate the moire response value;

Comparing brightness values of the plurality of key pixels and a plurality of corresponding adjacent pixels by using the detection window for each pixel, and counting comparison results to judge brightness characteristics of each pixel; and

And confirming the positions and types of the moire in the image according to the moire response value and the statistical result so as to perform color noise reduction on the pixels judged to be the moire.