CN113630590A - Image stripping method - Google Patents
Image stripping method Download PDFInfo
- Publication number
- CN113630590A CN113630590A CN202010383090.2A CN202010383090A CN113630590A CN 113630590 A CN113630590 A CN 113630590A CN 202010383090 A CN202010383090 A CN 202010383090A CN 113630590 A CN113630590 A CN 113630590A
- Authority
- CN
- China
- Prior art keywords
- pixel
- value
- compensated
- color
- calculating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000008030 elimination Effects 0.000 claims description 15
- 238000003379 elimination reaction Methods 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 description 29
- 238000004042 decolorization Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 3
- 241000023320 Luma <angiosperm> Species 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- OSWPMRLSEDHDFF-UHFFFAOYSA-N methyl salicylate Chemical compound COC(=O)C1=CC=CC=C1O OSWPMRLSEDHDFF-UHFFFAOYSA-N 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N9/00—Details of colour television systems
- H04N9/64—Circuits for processing colour signals
- H04N9/646—Circuits for processing colour signals for image enhancement, e.g. vertical detail restoration, cross-colour elimination, contour correction, chrominance trapping filters
Abstract
The invention provides an image color band removing method, which comprises the following steps: in a plurality of adjacent pixels in a default distance range in a straight line direction passing through the pixel to be compensated, a first comparison pixel and a second comparison pixel are calculated according to the color original values of the color components of the plurality of adjacent pixels and the pixel to be compensated before the debanding process, a first compensation value is calculated according to a first difference value between the color original values of the first comparison pixel and the pixel to be compensated, a first pixel distance between the first comparison pixel and the pixel to be compensated and a preset pixel distance range, and a second compensation value is calculated according to a second difference value between the color original values of the second comparison pixel and the pixel to be compensated, a second pixel distance between the second comparison pixel and the pixel to be compensated and the preset pixel distance range, so as to calculate the color component values of the pixel to be compensated after the debanding process.
Description
Technical Field
The invention relates to an image debanding method.
Background
In the known image processing technology, in order to produce a low-volume image, some color details are often omitted in a quantization procedure and an encoding procedure of the image processing, or the image is compressed to produce a low-volume image. However, since color detail is reduced, color bands are often generated on low-volume images, i.e. the images have abnormal color gamut boundaries, and therefore the color bands of the images need to be eliminated to improve the image definition. In the conventional color band elimination technique, if one pixel needs to be subjected to color band elimination, the operation is performed on the pixels of the whole image, so that the color band elimination can be performed on one pixel, which results in a large amount of operation, too long operation time, too much operation resources, and too much operation amount, which causes the hardware complexity of the color band elimination to increase, that is, the production cost cannot be reduced.
Disclosure of Invention
In some embodiments, there is provided an image debanding method comprising: in a first pixel straight line direction of a pixel to be compensated passing through an image signal, calculating a first comparison pixel located in a first lateral direction of the pixel to be compensated and a second comparison pixel located in a second lateral direction of the pixel to be compensated based on a first difference between color original values of the first comparison pixel and the pixel to be compensated, a first pixel distance between the first comparison pixel and the pixel to be compensated and the preset pixel distance range, in a first adjacent pixel within a preset distance range of the pixel to be compensated according to the first adjacent pixel and the color original value of the pixel to be compensated before a debanding process, calculating a first compensation value based on a second difference between color original values of the second comparison pixel and the pixel to be compensated, a second pixel distance between the second comparison pixel and the pixel to be compensated and the preset pixel distance range, and calculating the color component value of the pixel to be compensated in the color component after the color band elimination according to the first compensation value, the second compensation value and a first random value.
Drawings
FIG. 1 is a block diagram of an embodiment of an image processing apparatus employing an image processing circuit for color removal band according to the present invention;
FIG. 2 is a diagram of one embodiment of an image debanding method performed by the image debanding processing circuit of FIG. 1;
FIG. 3 is a flow chart of one embodiment of an image debanding method according to the present invention;
FIG. 4 is a circuit diagram of one embodiment of the image debanding band processing circuit of FIG. 1;
FIG. 5 is a diagram illustrating an embodiment of a method for calculating a compensation value by the comparison pixel calculation circuit of FIG. 4;
FIG. 6A is a flowchart of one embodiment of the image debanding method of FIG. 3;
FIG. 6B is a flowchart following one embodiment of the image debanding method of FIG. 6A;
FIG. 7 is a diagram of another embodiment of an image debanding method performed by the image debanding band processing circuit of FIG. 1.
Detailed Description
Fig. 1 is a circuit block diagram of an embodiment of an image processing apparatus 1 using an image color removal processing circuit 11 of the present invention. Referring to fig. 1, the image processing apparatus 1 includes an image debanding determination circuit 10 and an image debanding processing circuit 11 coupled to the image debanding determination circuit 10. The image debanding circuit 10 receives the input image signal S1, determines whether the pixels of the input image signal S1 need to be debanding to generate a determination result S2, and if the determination result S2 indicates that the debanding process needs to be performed on any pixel of the input image signal S1, the image debanding circuit 11 performs the debanding process on the pixel. In some embodiments, if the image debanding determination circuit 10 determines that some pixels of the input image signal S1 are less than the default distance from the image boundary, the determination result S2 is no; if the image debanding determination circuit 10 determines that some pixels of the input image signal S1 are more than the default distance from the image boundary, the determination result S2 is yes, but not limited thereto.
FIG. 2 is a diagram of an embodiment of an image debanding method performed by the image debanding processing circuit 11. As shown in fig. 2, an image block of the input image signal S1 includes a plurality of pixels, a pixel straight-line direction H1 (hereinafter referred to as a first pixel straight-line direction H1) which can be a horizontal direction passes through one of the pixels P1 and a plurality of adjacent pixels P11 and P12, the plurality of adjacent pixels P11 and P12 (hereinafter referred to as first adjacent pixels P11 and P12) are located within a predetermined pixel distance range centered on the pixel P1 in the first pixel straight-line direction H1, and "W" in the example of fig. 2 represents the predetermined pixel distance range. Moreover, on two sides of the first pixel straight-line direction H1 relative to the pixel P1, the first pixel straight-line direction H1 includes a first lateral direction H11 and a second lateral direction H12, the first lateral direction H11 passes through the first neighboring pixel P11 located within the preset pixel distance range, and the second lateral direction H12 passes through the first neighboring pixel P12 located within the preset pixel distance range.
Taking the first color component of the color components as an example, referring to fig. 1 to 3 in combination, when the image debanding determination circuit 10 determines that the debanding process needs to be performed on the first color component of the pixel P1 (hereinafter referred to as the pixel P1 to be compensated), in the first side direction H11, the image debanding processing circuit 11 calculates one of the plurality of first neighboring pixels P11 as a first comparison pixel according to the color original value of the pixel P1 to be compensated before the debanding process at the first color component and the color original value of the first neighboring pixel P11 before the debanding process at the first color component (step S01); also, in the second side direction H12, the image debanding processing circuit 11 calculates one of the plurality of first neighboring pixels P12 as a second comparison pixel based on the color original value of the aforementioned pixel P1 to be compensated before the debanding process in the first color component and the color original value of the first neighboring pixel P12 before the debanding process in the first color component (step S02).
In some embodiments, the first color component may be luma (Y), blue chroma (Cb), red chroma (Cr), red (R), green (G), or blue (B).
In some embodiments, in step S01, the image color removal processing circuit 11 may calculate a difference between the color original values of the pixel P1 to be compensated and the first neighboring pixels P11 on the first side H11, and determine whether each of the differences is smaller than a predetermined difference, where the predetermined difference is not zero. When the difference between the color original values of the first neighboring pixels P11 and the pixel P1 to be compensated is smaller than the default difference, the image color removal band processing circuit 11 selects the pixel P1 with the largest distance to the pixel P1 as the first comparison pixel (e.g., the first neighboring pixel P11 with a distance of three pixels) in the first neighboring pixels P11. The pixel distance between the first comparison pixel and the pixel P1 to be compensated is referred to as a first pixel distance.
In step S02, the image color removal processing circuit 11 calculates a difference between the first color component of the pixel P1 to be compensated and the color original value of each first neighboring pixel P12 on the second side H12, and determines whether each difference is smaller than the preset difference. When the difference between the color original values of the first neighboring pixels P12 and the pixel P1 to be compensated is smaller than the default difference, the image color removal band processing circuit 11 selects the pixel P1 with the largest distance to the pixel P12 as the second comparison pixel (e.g., the first neighboring pixel P12 with a distance of two pixels). The pixel distance between the second comparison pixel and the pixel to be compensated P1 is referred to as a second pixel distance.
The image color band removal processing circuit 11 calculates a compensation value (hereinafter, referred to as a first compensation value) based on a difference value (hereinafter, referred to as a first difference value) between the color original values of the first comparison pixel and the pixel to be compensated P1, the first pixel distance, and the preset pixel distance range (step S03), and calculates another compensation value (hereinafter, referred to as a second compensation value) based on a difference value (hereinafter, referred to as a second difference value) between the color original values of the second comparison pixel and the pixel to be compensated P1, the second pixel distance, and the preset pixel distance range (step S04). The image debanding circuit 11 calculates the color component value of the pixel P1 to be compensated in the first color component after the debanding process according to the first compensation value, the second compensation value and a random value (hereinafter referred to as the first random value) (step S05). Thus, after the image debanding circuit 11 performs the debanding process on the pixel P1 to be compensated, the image debanding circuit 11 may perform the debanding process on the other pixels to be compensated of the input image signal S1 and generate the output image signal S3 after performing the debanding process on all the pixels to be compensated.
Therefore, the image debanding processing circuit 11 calculates the color component value of the pixel P1 to be compensated in the first color component according to the neighboring pixels P11 and P12 in the first pixel linear direction H1 with the pixel P1 to be compensated as the center, and the image debanding processing circuit 11 does not need to perform calculation according to the color original value of each pixel of the input image signal S1, thereby greatly reducing the calculation amount of the debanding processing, greatly reducing the calculation time of the debanding processing, reducing the circuit complexity of the image debanding processing circuit 11, and reducing the production cost of the image processing apparatus 1.
In some embodiments, as shown in fig. 4, the image debanding band processing circuit 11 includes a comparison pixel first calculation circuit 111, a weighted average first calculation circuit 112, and a first random number summing circuit 113, the weighted average first calculation circuit 112 being coupled between the comparison pixel first calculation circuit 111 and the first random number summing circuit 113.
After receiving the input image signal S1, the first calculating circuit 111 of the comparison pixel performs steps S01 and S02 to calculate the first comparison pixel and the second comparison pixel according to the predetermined difference. Next, in step S03, the first calculating circuit 111 for pixel comparison can calculate the first compensation value according to the triangle theorem. In detail, referring to fig. 5, fig. 5 illustrates two triangular shapes T1 and T2, and the two triangular shapes T1 and T2 are approximately triangular. In some embodiments, the length of the base and the height of the triangle T1 are the predetermined pixel distance range and the first difference value, respectively, and the length of the base and the height of the triangle T2 are the first pixel distance and the first offset value, respectively. Therefore, the comparison pixel first calculation circuit 111 can calculate the first compensation value according to the triangle theorem using equation 1.1, and the first compensation value calculated by the comparison pixel first calculation circuit 111 can include a decimal.
Similarly, in step S04, the first calculating circuit 111 for pixel comparison can calculate the second compensation value according to the triangle theorem. As shown in fig. 5, the two triangles T3 and T4 are similar to each other in a triangle, in some embodiments, the length of the triangle T3 is the predetermined pixel distance range and the second difference, and the length of the triangle T4 is the second pixel distance and the second compensation value. Therefore, the comparison pixel first calculation circuit 111 can calculate the second compensation value according to the triangle theorem using equation 1.2, and the second compensation value calculated by the comparison pixel first calculation circuit 111 can include a decimal.
In some embodiments, referring to fig. 4 and fig. 6A and 6B, in step S05, the weighted average first calculating circuit 112 calculates a first weighted average value according to the color original value, a first weight value, a first compensation value, a second weight value and a second compensation value of the pixel P1 to be compensated in the first color component (step S051). Wherein, the first weight value corresponds to the first pixel distance, and if the first pixel distance is larger, the first weight value is smaller; the second weight value corresponds to a second pixel distance, and the second weight value is smaller if the second pixel distance is larger. The weighted average first calculation circuit 112 may calculate a first weighted average according to equation 1.3 in step S051, and the first weighted average may include decimal numbers.
(wgt 1: (X1+ ori) + wgt 2: (X2+ ori))/(wgt1+ wgt2) … formula 1.3, wherein ori represents the color original value of the pixel P1 to be compensated in the first color component, wgt1 represents the first weight value, and wgt2 represents the second weight value.
In addition, in step S05, the first random number summing circuit 113 generates a first value of the first random number value from a first random number range corresponding to the first difference value, and generates a second value of the first random number value from a second random number range corresponding to the second difference value (step S052). For example, the first range of random numbers may be
To
The first random number summing circuit 113 may randomly generate one of the first values as a first random number value according to the first random number range, the first value of the first random number value may include a decimal number; second range of random numbers(may be)
To
The first random number summing circuit 113 may randomly generate one of the second values as the first random number value based on the second random number range, and the second value of the first random number value may include the fractional number. In some embodiments, the first value and the second value of the first random number are equal, but not limited thereto.
Next, the first random number summing circuit 113 calculates the color component value of the pixel P1 to be compensated in the first color component after the debanding process according to the first weighted average, the first difference, the second difference, the first value of the first random number and the second value (step S053). For example, the first random number summing circuit 113 may generate a summation result (hereinafter, referred to as a first summation result) according to equation 1.4 to calculate a color component value of the pixel to be compensated P1 at the first color component after the debanding process, and the first random number summing circuit 113 may carry the first summation result to an integer number without including a decimal number.
wm1+ d1 RV11+ d2 RV12 … formula 1.4, where wm1 represents a first weighted average, RV11 represents a first value of the first random value, and RV12 represents a second value of the first random value. If the first and second values of the first random number are equal, equation 1.4 can be reduced to wm1+ (d1+ d2) RV 11.
In some embodiments, as shown in fig. 7, the second pixel straight-line direction H2 may also pass through the pixel P1 and pass through a plurality of adjacent pixels P21, P22 (hereinafter referred to as second adjacent pixels P21, P22) of the pixel P1, and the second pixel straight-line direction H2 may be perpendicular to the first pixel straight-line direction H1. The second neighboring pixels P21 and P22 are located within a predetermined pixel distance range centered on the pixel P1 in the second pixel straight line direction H2. Moreover, on two sides of the second pixel straight-line direction H2 relative to the pixel P1, the second pixel straight-line direction H2 includes a third lateral direction H21 and a fourth lateral direction H22, the third lateral direction H21 passes through a second neighboring pixel P21 located within the preset pixel distance range, and the fourth lateral direction H22 passes through a second neighboring pixel P22 located within the preset pixel distance range. When the image debanding determining circuit 10 determines that the debanding process is required to be performed on the first color component of the pixel P1 to be compensated, the image debanding processing circuit 11 may further calculate another two compensation values (hereinafter referred to as a third compensation value and a fourth compensation value) according to the color original values of the first color component of the second neighboring pixels P21 and P22 before the debanding process, and further determine how to calculate the color component values of the pixel P1 to be compensated after the debanding process on the first color component according to the third compensation value and the fourth compensation value, that is, the image debanding processing circuit 11 further considers the second neighboring pixels P21 and P22 in the other direction (i.e., the second pixel straight direction H2), so as to enhance the debanding effect of the input image signal S1.
In detail, referring to fig. 4 and fig. 6A and 6B, the image processing circuit 11 further includes a second calculating circuit 114 for comparing pixels, a second calculating circuit 115 for weighted average, and a second random number summing circuit 116, wherein the second calculating circuit 115 for weighted average is coupled between the second calculating circuit 114 for comparing pixels and the second random number summing circuit 116. When the image debanding determining circuit 10 determines that the debanding process needs to be performed on the first color component of the pixel P1 to be compensated, the second comparison pixel calculating circuit 114 calculates the difference between the color original values of the pixel P1 to be compensated and each of the second neighboring pixels P21, and determines whether each of the differences is smaller than the preset difference. When the difference between the color original values of one of the second neighboring pixels P21 and the pixel P1 to be compensated is smaller than the predetermined difference, the second calculating circuit 114 calculates the aforementioned one of the second neighboring pixels P21 as a third comparison pixel (step S07), for example, the second calculating circuit 114 calculates the second neighboring pixel P21 which is one pixel away from the pixel P1 to be compensated as the third comparison pixel.
Moreover, the second calculating circuit 114 of the comparison pixel calculates a difference between the first color component of the pixel P1 to be compensated and the color original value of each of the second adjacent pixels P22, and determines whether the difference is smaller than the predetermined difference, when the difference between the color original values of one of the second adjacent pixels P22 and the pixel P1 to be compensated is smaller than the predetermined difference, the second calculating circuit 114 of the comparison pixel calculates the one of the second adjacent pixels P22 as a fourth comparison pixel (step S08), for example, the second calculating circuit 114 of the comparison pixel can calculate the second adjacent pixel P22 which is four pixels away from the pixel P1 to be compensated as the fourth comparison pixel.
After the third comparison pixel and the fourth comparison pixel are calculated, the weighted average second calculation circuit 115 calculates a pixel distance (hereinafter referred to as a third pixel distance) between the third comparison pixel and the pixel to be compensated P1, and the weighted average second calculation circuit 115 calculates a pixel distance (hereinafter referred to as a fourth pixel distance) between the fourth comparison pixel and the pixel to be compensated P1. The weighted average second calculation circuit 115 calculates a third compensation value according to a difference between the color original values of the third comparison pixel and the pixel to be compensated P1 (hereinafter, referred to as a third difference), the third pixel distance, and the preset pixel distance range (step S09), and the weighted average second calculation circuit 115 may calculate the third compensation value according to the similar triangle theorem (similar to the calculation of X1 and X2 shown in the embodiment of fig. 5) in formula 1.5 in step S09.
Likewise, the weighted average second calculation circuit 115 calculates a fourth compensation value from a difference between the color original values of the fourth comparison pixel and the pixel to be compensated P1 (hereinafter referred to as a fourth difference), a fourth pixel distance, and a preset pixel distance range (step S10). The weighted average second calculation circuit 115 may calculate the fourth compensation value according to the similar triangle theorem (similar to the calculation manner of X1 and X2 shown in the embodiment of fig. 5) in formula 1.6 in step S10.
Next, the weighted average second calculation circuit 115 calculates another weighted average value (hereinafter referred to as a second weighted average value) from the color original value, the third weight value, the third compensation value, the fourth weight value, and the fourth compensation value of the first color component of the pixel P1 to be compensated (step S111). Wherein the third weight value corresponds to a third pixel distance, and the larger the third pixel distance, the smaller the third weight value; the fourth weight value corresponds to a fourth pixel distance, and the larger the fourth pixel distance, the smaller the fourth weight value. The weighted average second calculation circuit 115 may calculate a second weighted average value according to equation 1.7 in step S111, and the second weighted average value may include a decimal number.
(wgt 3: (X3+ ori) + wgt 4: (X4+ ori))/(wgt3+ wgt4) … formula 1.7, wherein wgt3 represents a third weight value and wgt4 represents a fourth weight value.
Then, the second random number summing circuit 116 generates a first value of the second random number value based on a third random number range corresponding to the third difference value, and generates a second value of the second random number value based on a fourth random number range corresponding to the fourth difference value (step S112). For example, the third random number range may be
To
In between, the second random number summing circuit 116 may randomly generate one of the first values as a first value of a second random number value according to a third random number range, the first value of the second random number value may include a decimal; the fourth random number range may be
To
In between, the second random number summing circuit 116 may randomly generate one of the second random number values as a second value of the second random number value based on the fourth random number rangeThe value may comprise a decimal number. In some embodiments, the first value and the second value of the second random number are equal, but not limited thereto.
The second random number summing circuit 116 calculates a second summing result according to the first value and the second value of the summed second weighted average, the third difference, the fourth difference, and the second random number (step S113), the second random number summing circuit 116 may calculate the second summing result according to formula 1.8 in step S113, and the second random number summing circuit 116 may carry the second summing result to an integer without a decimal.
wm2+ d3 RV21+ d4 RV22 … formula 1.8, where wm2 represents a second weighted average, RV21 represents a first value of a second random value, and RV22 represents a second value of the second random value. If the first and second values of the second random number are equal, equation 1.8 can be reduced to wm2+ (d3+ d4) RV 21.
After the first and second random number summing circuits 113 and 116 calculate the first and second summing results, respectively, the image debanding band processing circuit 11 uses the one of the first and second summing results, which has a large difference between the value and the color original value of the pixel to be compensated P1, as the color component value of the pixel to be compensated P1 in the first color component after the debanding processing. As shown in fig. 4, the image processing circuit 11 further includes a comparing circuit 117 and a selecting circuit 118, the comparing circuit 117 is coupled to the first random number summing circuit 113, the second random number summing circuit 116 and the selecting circuit 118, and the selecting circuit 118 is controlled by the comparing circuit 117 and is coupled to the first random number summing circuit 113 and the second random number summing circuit 116. The comparison circuit 117 and the selection circuit 118 receive the first summation result generated by the first random number summation circuit 113 and the second summation result generated by the second random number summation circuit 116. The comparison circuit 117 compares the first and second summation results to determine whether the difference between the first summation result and the color original value of the pixel to be compensated P1 is greater than the difference between the second summation result and the color original value of the pixel to be compensated P1 (step S12). When the comparison circuit 117 determines that the difference between the first summation result and the color original value of the pixel to be compensated P1 is greater than the difference between the second summation result and the color original value of the pixel to be compensated P1 (yes, the comparison circuit 117 outputs the comparison result to control the selection circuit 118 to output the first summation result as the color component value of the pixel to be compensated P1 at the first color component after the debanding process (step S13) to complete the debanding process of the pixel to be compensated P1; when the comparison circuit 117 determines that the second summation result is larger than the first summation result (no), the comparison circuit 117 outputs the comparison result to control the selection circuit 118 to output the second summation result as the color component value of the pixel to be compensated P1 at the first color component after the debanding process (step S14) to complete the debanding process of the pixel to be compensated P1.
It should be understood that the foregoing embodiments determine the first, second, third and fourth comparison pixels according to the same predetermined difference, but not limited thereto. In some embodiments, the image color removal processing circuit 11 may determine the first, second, third and fourth comparison pixels on the first side H11, the second side H12, the third side H21 and the fourth side H22 according to different preset differences.
It should be understood that the foregoing embodiments are exemplified by the "first color component", and are not limited in number. In some embodiments, the image debanding method of the present invention may be repeatedly implemented on a plurality of color components (e.g., R, G, B triplets, but not limited thereto) of the pixel P1.
It should be understood that the foregoing embodiments are exemplified by the "pixel P1", which is not limited in number. In some embodiments, the image debanding method of the present invention may be repeatedly implemented for a plurality of pixels among the or all image patches of the input image signal S1. In some embodiments, when the image debanding method of the present invention is repeatedly implemented in a plurality of pixels, the image debanding determination circuit 10 and the image debanding processing circuit 11 perform calculation according to the color original values of the plurality of pixels, rather than according to the debanding processed values.
In some embodiments, the image de-color band processing circuit 11 may be implemented as an Application Specific Integrated Circuit (ASIC).
In summary, according to the embodiment of the image color band elimination method of the present invention, the image color band elimination processing circuit 11 calculates the compensation value of the first color component of the pixel P1 to be compensated based on the pixel P1 to be compensated as the center and the neighboring pixels within the preset pixel distance range, and the image color band elimination processing circuit 11 does not need to perform calculation based on the color component value of each pixel of the input image signal, thereby greatly reducing the calculation amount of the color band elimination processing, greatly reducing the calculation time of the color band elimination processing, and accordingly reducing the circuit complexity of the image color band elimination processing circuit 11, and reducing the production cost of the image processing apparatus 1.
Although the present invention has been disclosed in the context of embodiments thereof, it should be understood that these embodiments are not intended to limit the invention to the particular forms disclosed, but on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the invention as defined by the appended claims.
[ description of reference ]
1: image processing apparatus
10 image decolour band judging circuit
11 image decolouring band processing circuit
111 first calculating circuit for comparison pixel
112 weighted average first calculation circuit
113 first random number summing circuit
114 comparing pixel second calculating circuit
115 weighted average second calculation circuit
Second random number summing circuit 116
117 comparison circuit
118 selection circuit
P1 pixel to be compensated
P11 first neighboring pixel
P12 first neighboring pixel
P21 second neighboring pixel
P22 second neighboring pixel
S1 input image signal
S2, judging result
S3 outputting image signal
H1 first pixel straight line direction
H11 first lateral
H12 second lateral
H2 second pixel Linear Direction
H21 third lateral
H22 fourth lateral
T1-T4 triangle
S01-S05
S07-S10
S12-S14
S111-S113 step
Claims (10)
1. An image debanding method comprising:
in a first pixel straight line direction of a pixel to be compensated passing through an image signal, in a plurality of first adjacent pixels within a preset distance range of the pixel to be compensated, calculating a first comparison pixel located in a first lateral direction of the pixel to be compensated and a second comparison pixel located in a second lateral direction of the pixel to be compensated according to the plurality of first adjacent pixels and color original values of the pixel to be compensated in a color component before debanding processing;
calculating a first compensation value according to a first difference value between the first comparison pixel and the color original value of the pixel to be compensated, a first pixel distance between the first comparison pixel and the pixel to be compensated, and the preset pixel distance range;
calculating a second compensation value according to a second difference value between the second comparison pixel and the color original value of the pixel to be compensated, a second pixel distance between the second comparison pixel and the pixel to be compensated, and the preset pixel distance range; and
and calculating the color component values of the pixels to be compensated in the color components after the color components are subjected to the color band elimination processing according to the first compensation value, the second compensation value and a first random value.
2. The method of claim 1, wherein the step of calculating the first comparison pixel and the second comparison pixel comprises:
judging whether the difference value between the color original values of the first adjacent pixels and the pixel to be compensated in the first lateral direction is smaller than a preset difference value, wherein the preset difference value is not zero, so as to calculate the first comparison pixel from the first adjacent pixels in the first lateral direction; and
judging whether the difference value between the color original values of the plurality of first adjacent pixels in the second lateral direction and the pixel to be compensated is smaller than the preset difference value or not, so as to calculate the second comparison pixel from the plurality of first adjacent pixels in the second lateral direction.
3. The image debanding method according to claim 1, wherein the step of calculating the color component values of the pixel to be compensated at the color components after the debanding process comprises:
calculating a first weighted average value according to the color original value of the pixel to be compensated in the color component, a first weighted value, the first compensation value, a second weighted value and the second compensation value, wherein the first weighted value corresponds to the first pixel distance, and the second weighted value corresponds to the second pixel distance;
generating a first value of the first random number value in a random number manner according to the first difference value;
generating a second value of the first random number value in a random number manner according to the second difference value; and
and calculating a first summation result according to the first weighted average value, the first difference value, the second difference value, the first value and the second value of the first random value so as to calculate the color component value of the pixel to be compensated in the color component after the color component is subjected to the color band elimination processing.
4. The method according to claim 3, wherein the step of calculating the first compensation value comprises:
wherein the step of calculating the first compensation value comprises:
5. The method of claim 4, wherein the step of calculating the first weighted average comprises:
(wgt1 [ (X1+ ori) + wgt2 [ ((X2 + ori))/((wgt 1+ wgt2) ], wherein wgt1 represents the first weight value, wgt2 represents the second weight value, and ori represents the color original value of the pixel to be compensated in the color component.
6. The method of claim 4, wherein the step of calculating the first summation result comprises:
wm1+ d1 RV11+ d2 RV12, wherein wm1 represents the first weighted average, RV11 represents a first value of the first random number, and RV12 represents a second value of the first random number.
7. The method of claim 4, wherein the step of generating the first value of the first random number value based on the first difference value by random number is performed to
To
Randomly generating one of the first values as the first random number value; wherein generating a second value of the first random number value in a random number manner based on the second difference value is based on
To
Randomly generating one of the second values as the first random value.
8. The method of image debanding according to claim 3, further comprising:
calculating a third comparison pixel located on a third side of the pixel to be compensated and a fourth comparison pixel located on a fourth side of the pixel to be compensated according to color primitive values of the color components of the second adjacent pixels and the pixel to be compensated before a debanding process in a second pixel straight line direction passing through the pixel to be compensated, wherein the first pixel straight line direction is a horizontal direction, and the second pixel straight line direction is perpendicular to the first pixel straight line direction, among a plurality of second adjacent pixels within the preset distance range of the pixel to be compensated;
calculating a third compensation value according to a third difference value between the third comparison pixel and the color original value of the pixel to be compensated, a third pixel distance between the third comparison pixel and the pixel to be compensated, and the preset pixel distance range;
calculating a fourth compensation value according to a fourth difference value between the fourth comparison pixel and the color original value of the pixel to be compensated, a fourth pixel distance between the fourth comparison pixel and the pixel to be compensated, and the preset pixel distance range; and
and calculating the color component value of the pixel to be compensated in the color component after the color component is subjected to the color band elimination processing according to the third compensation value, the fourth compensation value and a second random value.
9. The method according to claim 8, wherein the step of calculating the color component values of the pixel to be compensated after the debanding process on the color components according to the third compensation value, the fourth compensation value and the second random value comprises:
calculating a second weighted average value according to the color original value of the pixel to be compensated in the color component, a third weight value, the third compensation value, a fourth weight value and the fourth compensation value, wherein the third weight value corresponds to the third pixel distance, and the fourth weight value corresponds to the fourth pixel distance;
generating a first value of the second random number value in a random number manner according to the third difference value;
generating a second value of the second random number value in a random number manner according to the fourth difference value; and
and calculating a second summation result according to the second weighted average value, the third difference value, the fourth difference value, the first value and the second value of the second random value so as to calculate the color component value of the pixel to be compensated in the color component after the color band elimination processing.
10. The method for debanding an image according to claim 9, wherein the step of calculating the color component values of the pixel to be compensated at the color components after the debanding process further comprises:
comparing the first summation result and the second summation result;
when the difference value of the first summation result from the color original value of the color component is greater than the difference value of the second summation result from the color original value of the color component, the pixel to be compensated is the first summation result at the color component value of the color component after the debanding process; and
when the difference value of the second summation result from the color original values of the color components is greater than the difference value of the first summation result from the color original values of the color components, the pixel to be compensated is the second summation result at the color component values of the color components after the debanding process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010383090.2A CN113630590A (en) | 2020-05-08 | 2020-05-08 | Image stripping method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010383090.2A CN113630590A (en) | 2020-05-08 | 2020-05-08 | Image stripping method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113630590A true CN113630590A (en) | 2021-11-09 |
Family
ID=78377266
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010383090.2A Pending CN113630590A (en) | 2020-05-08 | 2020-05-08 | Image stripping method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113630590A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6025922A (en) * | 1998-12-18 | 2000-02-15 | Electronics For Imaging | Reduction of banding in printed images |
| US20010019632A1 (en) * | 2000-02-04 | 2001-09-06 | Ricoh Company, Ltd. | Apparatus and method for forming an image by processing input image data while suppressing banding and dropout |
| CN104717474A (en) * | 2013-12-16 | 2015-06-17 | 瑞昱半导体股份有限公司 | Image processing method and module |
| US20160307298A1 (en) * | 2015-04-15 | 2016-10-20 | Apple Inc. | Debanding image data using bit depth expansion |
-
2020
- 2020-05-08 CN CN202010383090.2A patent/CN113630590A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6025922A (en) * | 1998-12-18 | 2000-02-15 | Electronics For Imaging | Reduction of banding in printed images |
| US20010019632A1 (en) * | 2000-02-04 | 2001-09-06 | Ricoh Company, Ltd. | Apparatus and method for forming an image by processing input image data while suppressing banding and dropout |
| CN104717474A (en) * | 2013-12-16 | 2015-06-17 | 瑞昱半导体股份有限公司 | Image processing method and module |
| US20160307298A1 (en) * | 2015-04-15 | 2016-10-20 | Apple Inc. | Debanding image data using bit depth expansion |
Non-Patent Citations (3)
| Title |
|---|
| 冷璐;黎明;杨小芹;: "复杂背景下疲劳条带图像预处理方法研究", 计算机工程与设计, no. 03, pages 25 - 28 * |
| 江月松;王龙奇;: "基于积分图像的红外图像降噪去条带方法", 红外, no. 07, pages 56 - 60 * |
| 石光明;王晓甜;张犁;刘哲;: "基于方向滤波器消除遥感图像孤立条带噪声的方法", 红外与毫米波学报, no. 03, pages 126 - 128 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7215366B2 (en) | Tone correcting circuit and hue correcting circuit | |
| US20070280539A1 (en) | Image Processing Method and Image Processing Device | |
| US5999230A (en) | Tone correction apparatus for digital color image signal | |
| EP1612729A1 (en) | Image processing device, method, and program | |
| JP4317619B2 (en) | Image processing device | |
| US10708562B2 (en) | Image signal processing apparatus, image processing circuit, and image processing method | |
| TWI727784B (en) | Debanding method for image | |
| US20040008381A1 (en) | System and method for converting color data to gray data | |
| CN113630590A (en) | Image stripping method | |
| EP0744864A2 (en) | Colour correction device | |
| US20070058225A1 (en) | Image processing apparatus and method | |
| US6574004B1 (en) | System and method for converting color data to gray data | |
| US9530185B2 (en) | Image processing apparatus, imaging apparatus, image processing method and storage medium | |
| JPH0722343B2 (en) | Image data compression device | |
| US6826302B2 (en) | Luminance correction of colorless low saturation regions using correction constants calculated from color saturation values | |
| CN109788261A (en) | Color displacement bearing calibration and device | |
| US20030063100A1 (en) | Apparatus and method for dithering in image processing and computer graphics systems | |
| US11379956B2 (en) | Image processing circuit and associated image processing method | |
| US6847732B1 (en) | Image error diffusion device with noise superposition | |
| JP2002344743A (en) | Image processing unit and image processing method | |
| CN112351152B (en) | Image processing circuit and related image processing method | |
| CN113965735B (en) | Image color-removing band judging method and image color-removing band judging circuit | |
| CN109819228B (en) | Color-shifting method | |
| US8885104B2 (en) | Apparatus and method for generating video signal reducing crosstalk between lightness signal and chrominance signal | |
| CN116452465B (en) | Method for eliminating JPEG image block artifact |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |