Graphic rendering method based on game rendering platform

Abstract

The invention discloses a graph rendering method based on a game rendering platform, relates to the technical field of graph rendering, and solves the problem that rendering resources cannot be concentrated in key parts due to the fact that abrupt areas needing to be rendered in a graph are difficult to effectively distinguish during original rendering. Adopting mean value processing and preset fluctuation parameters to determine a feature interval, and quantizing the pixel value of the inherent interval to the feature interval; according to the quantization mode, the pixel values in the partitions can be more balanced, excessive fluctuation of the pixel values is reduced, and the visual effect of the graph is optimized. And meanwhile, the pixel points at the edge of the partition are subjected to average adjustment, and the to-be-adjusted pixel value is determined by comparing the process characteristics of different moving processes, so that the transition between adjacent partitions is more natural and smoother, and obvious boundary difference is avoided.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of graphics rendering, in particular to a graphics rendering method based on a game rendering platform. BACKGROUND

[0002] In the current booming game industry, the quality and rendering efficiency of game pictures have become key factors affecting player experience and game competitiveness. Traditional game graphics rendering technology usually adopts a unified rendering strategy to process the entire graphics, lacking precise identification and targeted optimization of different area characteristics in the graphics. This extensive rendering method not only leads to resource waste, but also fails to meet players' demand for high-quality and high-realism pictures.

[0003] In the prior art, game graphics rendering often does not fully consider the differences between pixel points in the graphics. For example, in the determination of the rendering area, it is difficult to effectively distinguish the prominent areas that need to be rendered in the graphics, so that the rendering resources cannot be concentrated in the key parts. When processing adjacent pixel points, the correlation differences between pixel points in different areas are ignored, which easily leads to the situation that the rendered picture is not coordinated or lacks details. At the same time, there is a lack of scientific and systematic method for adjusting pixel values, which cannot realize the fine optimization of picture color and details.

[0004] In addition, with the continuous improvement of game scene complexity, graphics containing a large number of fine textures and complex lighting effects put higher requirements on rendering technology. The traditional rendering method is difficult to achieve high-quality graphics rendering effect while ensuring efficiency. Therefore, there is an urgent need for a graphics rendering method that can accurately locate the rendering area, scientifically divide the pixel feature area, and optimize and adjust the pixel value, in order to improve the quality and efficiency of game graphics rendering and meet the growing visual experience needs of players. SUMMARY

[0005] In view of the shortcomings of the prior art, the present application provides a graphics rendering method based on a game rendering platform, which solves the problem that it is difficult to effectively distinguish the prominent areas that need to be rendered in the graphics during original rendering, so that the rendering resources cannot be concentrated in the key parts.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: a graphics rendering method based on a game rendering platform, comprising the following steps:

[0007] Based on a preset to-be-determined template, the to-be-rendered areas existing in a plurality of groups of graphics pre-stored in the game rendering platform are calibrated, and the specific method is as follows:

[0008] The pre-stored single group of graphics is subjected to gray scale processing to confirm the associated gray scale graphics;

[0009] The grayscale image is compared and verified with a pre-stored template to confirm the numerical differences between different pixels in the grayscale image and the white template. The pixel values ​​associated with different pixels are labeled as X. i Where i represents different pixels, using: C i =|X i -255| Confirm the difference value C associated with the corresponding pixel. i If C i If Y1 > Y1, then this pixel is marked as the point to be rendered, and Y1 is a preset value; otherwise, no marking is performed, and the area covered by several points to be rendered is marked as the area to be rendered.

[0010] Further confirm the numerical differences between different pixels within the grayscale image and the black template, using: C i =|X i -1| Confirm the difference value C associated with the corresponding pixel. i If C i If the value is greater than Y1, then this pixel is marked as the point to be rendered; otherwise, no marking is performed, and the area covered by several points to be rendered is marked as the area to be rendered.

[0011] By confirming the numerical differences between adjacent pixels within the area to be rendered, a single group of areas to be rendered is divided into multiple similar partitions. Specifically:

[0012] The grayscale value associated with each different pixel in the rendering area is confirmed and labeled as H. i A set of pixels is randomly selected from the area to be rendered as undetermined points. Starting from the undetermined points, the numerical differences between the surrounding adjacent pixels are confirmed in turn. The numerical difference is the absolute value of the difference between the pixel values ​​associated with the two adjacent points. If the numerical difference is ≤ Y2, where Y2 is a preset value, the adjacent pixels are marked as pixels of the same type. Otherwise, no marking is performed.

[0013] Based on the numerical difference, other pixels that belong to the same type of pixel as the undetermined point are confirmed, and then the confirmation is carried out outward to identify whether the adjacent points of other pixels belong to the same type of pixel. In this way, the existing pixels of the same type are marked in turn, and the specific areas of the marked pixels of the same type are marked as the same type partition.

[0014] Based on preset fluctuation parameters, the pixel values ​​of different points in each similar partition are quantized to lock the pixel values ​​to be adjusted associated with the corresponding points; then the pixel points at the outer edge of each similar partition are determined, and several adjacent pixel points associated with adjacent similar partitions are determined. The pixel value features between adjacent pixel points are adjusted by average amount to determine the pixel values ​​to be adjusted associated with the corresponding pixel points.

[0015] According to the determined pixel value to be adjusted, the RGB value associated with the designated pixel point is adjusted.

[0016] Preferably, the manner of locking the pixel value to be adjusted of each point within the same type of partition is:

[0017] The pixel values associated with a plurality of pixel points within the same type of partition are processed by averaging, the plurality of pixel points do not include the pixel points at the external edges of the same type of partition, and the average value J belonging to the same type of partition is confirmed. k wherein k represents different same type of partitions, and the characteristic interval [J k -X1, J k +X1] belonging to the same type of partition is confirmed according to a preset fluctuation parameter X1, and X1 is a preset value.

[0018] Then, the minimum value and the maximum value are selected from the plurality of pixel values associated with the plurality of pixel points within the same type of partition, and the inherent interval [Gmin, Gmax] belonging to the same type of partition is confirmed.

[0019] Gmin is quantized to (J k -X1), and Gmax is quantized to (J k +X1), based on the specific quantization process, the quantized values associated with the pixel values of the plurality of pixel points are determined in turn, and the determined quantized values are taken as the pixel values to be adjusted of the corresponding pixel points.

[0020] Preferably, the specific manner of determining the plurality of adjacent pixel points associated with the adjacent same type of partitions is:

[0021] From the plurality of same type of partitions marked, adjacent same type of partitions are determined, the pixel points at the external edges of the same type of partitions are marked as edge pixel points, and adjacent edge pixel points between adjacent same type of partitions are confirmed, and the two groups of adjacent edge pixel points belong to two adjacent pixel points in the horizontal direction.

[0022] According to the confirmed adjacent edge pixel points, the gray value columns associated with the two groups of adjacent edge pixel points are recorded as (HD o , QD o ), wherein o represents different gray value columns, different adjacent edge pixel points are associated with different gray value columns, HD o and QD o represent the pixel values of the two pixel points, respectively, and the two pixel points are located in different partitions of adjacent same type of partitions.

[0023] Each gray value column is quantized to a group of line segments, the total length of the line segment is the total value of the two groups of gray values of the gray value column, and there is a group of separation points in the line segment, which divides the line segment into a front line segment and a rear line segment, and the unit value associated with the front line segment is HDo , the unit value associated with the back line segment is QD o , adopt: CQ o = (QQ o -HQ o ) confirm the variation difference value CQ o associated with the mobile feature point;

[0024] A set of mobile feature points are calibrated in each line segment, which can move within the corresponding line segment. The part of the line segment in front of the mobile feature point is recorded as the front segment, and the part of the line segment behind the mobile feature point is recorded as the back segment. The unit value associated with the front segment is QQ o , and the unit value associated with the back segment is HQ o , adopt: CQ o = (QQ o -HQ o ) confirm the variation difference value CQ o associated with the mobile feature point;

[0025] Based on different mobile processes, the process characteristics associated with each mobile process are confirmed. The variance of a plurality of sets of variation difference values CQ o associated with the corresponding mobile process is processed to confirm the first characteristic TZ1. The variation difference value CQ o and the standard deviation value CZ o associated with the corresponding mobile process belonging to the same set of gray value columns are processed to lock the characteristic difference value associated with the corresponding gray value column, and the characteristic difference value is greater than or equal to 0. A plurality of sets of characteristic difference values generated by the corresponding mobile process are summed to confirm the second characteristic TZ2.

[0026] Adopt: Process characteristic = TZ1xC1 + TZ2xC2 to confirm the process characteristic associated with the corresponding mobile process, wherein C1 and C2 are both preset fixed coefficient factors.

[0027] From a plurality of sets of mobile processes, the specific mobile process with the minimum process characteristic is selected. QQ o associated with the specific mobile process is used as the to-be-adjusted pixel value of the pixel point associated with HD o , and HQ o is used as the to-be-adjusted pixel value of the pixel point associated with QD o .

[0028] Preferably, the specific way of associatively adjusting the RGB value associated with the specified pixel point is:

[0029] Confirm the original pixel value X i associated with the corresponding pixel point, and confirm the to-be-adjusted pixel value of the corresponding pixel point.

[0030] If the pixel value to be adjusted is greater than X i , then the RGB value is adjusted synchronously to X i , and the process stops when the difference between the pixel value associated with the adjusted RGB value and the pixel value to be adjusted is less than 3.

[0031] If the pixel value to be adjusted is less than X i , then the RGB value is adjusted synchronously to X i , and the process stops when the difference between the pixel value associated with the adjusted RGB value and the pixel value to be adjusted is less than 3.

[0032] If the pixel value to be adjusted is less than X i , then the RGB value is adjusted synchronously to X i , and the process stops when the difference between the pixel value associated with the adjusted RGB value and the pixel value to be adjusted is less than 3.

[0033] The present application provides a graphics rendering method based on a game rendering platform. Compared with the prior art, the present application has the following beneficial effects:

[0034] The present application divides the adjacent pixels into multiple similar sub-regions by analyzing the numerical differences between the adjacent pixels after determining the region to be rendered. This division method fully considers the characteristic differences of different regions within the graphics, so that the pixels within each sub-region have similar characteristics, facilitating subsequent more detailed and accurate rendering processing. In this way, the poor rendering effect caused by large pixel differences within the region can be avoided, and the quality and realism of the rendering can be improved.

[0035] For the pixels within the similar sub-region, the mean value processing and the preset fluctuation parameter are used to determine the characteristic interval, and the pixel values of the inherent interval are quantized to the characteristic interval. This quantization method can make the pixel values within the sub-region more balanced, reduce the excessive fluctuation of the pixel values, and optimize the visual effect of the graphics. At the same time, the sub-region edge pixels are adjusted by the mean value, and the pixel value to be adjusted is determined by comparing the process characteristics of different moving processes, so as to ensure that the transition between adjacent sub-regions is more natural and smooth, and obvious boundary differences are avoided. BRIEF DESCRIPTION OF DRAWINGS

[0036] Figure 1 The figure is a schematic diagram of the method of the present application;

[0037] Figure 2 The figure is a schematic diagram of the line segment quantized by the gray value column of the present application. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0039] First embodiment

[0040] Referring to Figure 1 The application provides a graphics rendering method based on a game rendering platform, comprising the following steps:

[0041] Step one, preliminary identification of a plurality of groups of graphics pre-stored in the game rendering platform, based on a preset pending template, the pending rendering area in the graphics is calibrated, specifically, the so-called preset pending template is two templates preset in advance in the original operation process, one is a pure white template, and the other is a pure black template, when the two different templates are combined with the corresponding graphics, the abrupt area associated with the corresponding graphics can be effectively displayed, so as to quickly determine the corresponding pending rendering area;

[0042] Among them, the specific way of calibrating the pending rendering area is:

[0043] Gray processing is performed on the pre-stored single group of graphics, and the gray graphics associated with the corresponding single group of graphics are confirmed (the gray value is specifically confirmed based on the inherent RGB value in the corresponding graphics, and by assigning different weights, the gray value associated with the corresponding pixel point can be effectively determined) ;

[0044] The gray graphics are compared and verified with the pre-stored pending template, which includes a black template and a white template, the pixel value associated with the white template is 255, and the pixel value associated with the black template is 1, the value difference (that is, the dark area is effectively reflected) between different pixel points in the gray graphics and the white template is confirmed: the pixel value associated with different pixel points is calibrated as X i , wherein i represents different pixel points, and C i = |X i -255| is used to confirm the difference value C i associated with the corresponding pixel point, if C i >Y1, then this pixel point is calibrated as a pending rendering point, Y1 is a preset value, and its specific value is determined by the operator according to experience, otherwise, no calibration is performed, and the area covered by a plurality of pending rendering points is calibrated as a pending rendering area;

[0045] The value difference (that is, the bright area is effectively reflected) between different pixel points in the gray graphics and the black template is further confirmed, that is, C i = |X i -1| is used to confirm the difference value C i associated with the corresponding pixel point, if C i >Y1, then this pixel point is calibrated as a pending rendering point, Y1 is a preset value, and its specific value is determined by the operator according to experience, otherwise, no calibration is performed, and the area covered by a plurality of pending rendering points is calibrated as a pending rendering area, and Y1 generally takes the value 100;

[0046] After the to-be-rendered area is specifically determined, the to-be-rendered area can be divided into a plurality of specific sub-zones with different characteristics according to the numerical value representation state between different pixel points in the to-be-rendered area, so as to facilitate the subsequent rendering process.

[0047] In step two, the numerical value difference between adjacent pixel points in the to-be-rendered area is confirmed, and based on the confirmation result, a single to-be-rendered area is divided into a plurality of similar sub-zones. Specifically, there are different individuals in a single to-be-rendered area, and there are also correlation differences between different individuals. Therefore, there are also large differences between adjacent pixel points. Therefore, in order to avoid such differences, the corresponding to-be-rendered area needs to be divided into a plurality of different sub-zones with the same characteristics.

[0048] The specific way of dividing the plurality of similar sub-zones is as follows:

[0049] The gray value associated with each different pixel point in the to-be-rendered area is confirmed and labeled as H i A group of pixel points are randomly selected from the to-be-rendered area as to-be-determined points. The numerical value difference between the adjacent pixel points around the to-be-determined points is confirmed in sequence. The numerical value difference is the absolute value of the pixel value difference between two adjacent points. If the numerical value difference is less than or equal to Y2, Y2 is a preset value, and the specific value is determined by the operator according to experience, then the adjacent pixel points are labeled as similar pixel points. Otherwise, no labeling is performed.

[0050] The other pixel points belonging to the same type of pixel points as the to-be-determined points are preferentially confirmed based on the numerical value difference, and then the confirmation is diffused to the outside. It is identified whether the adjacent points of other pixel points also belong to the same type of pixel points. In this way, the existing similar pixel points are labeled in sequence, and the specific area of the labeled similar pixel points is labeled as a similar sub-zone.

[0051] Specifically, in a to-be-rendered area, there are a plurality of different pixel points inside. There is a numerical value difference between each pixel point. When the numerical value difference is close, the corresponding ordered pixel points belong to the same type of pixel points. Otherwise, when the numerical value difference is large, the corresponding ordered pixel points do not belong to the corresponding same type of pixel points, and then belong to the second similar sub-zone. In this way, a plurality of different similar sub-zones associated with the to-be-rendered area are labeled in sequence, so that more accurate rendering processing effect can be achieved in the later stage.

[0052] Step three, determine the pixel features associated with each homogeneous partition, and then quantize the pixel values of different points according to the preset fluctuation parameter, lock the pixel values associated with the corresponding point, and the pixel points in each homogeneous partition here do not include the edge pixel points, that is, the pixel points in the outer ring are not considered, and only the pixel points in the region are rendered and optimized in this stage, and the specific way of locking the pixel values of the corresponding point is:

[0053] The pixel values associated with the pixel points in the homogeneous partition are processed by mean value, and the pixel points do not include the pixel points at the edge of the homogeneous partition, and the mean value feature J of the homogeneous partition is confirmed k , wherein k represents different homogeneous partitions, and the feature interval [J k -X1, J k +X1] of the homogeneous partition is confirmed according to the preset fluctuation parameter X1, and X1 is a preset value;

[0054] Then select the minimum value and the maximum value from the pixel values associated with the pixel points in the homogeneous partition, and confirm the inherent interval [Gmin, Gmax] of the homogeneous partition;

[0055] Quantize Gmin to (J k -X1), and quantize Gmax to (J k +X1), based on the specific quantization process, the quantized values associated with the pixel values of the pixel points are determined in turn, and the determined quantized values are used as the pixel values associated with the pixel points;

[0056] Specifically, the pixel values associated with the pixel points in the corresponding homogeneous partition are 20, 30, 40, 50 and 60, the confirmed inherent interval is [20, 60], the confirmed mean value is 40, and the preset fluctuation parameter X1 is 15, then the confirmed feature interval is [25, 55], wherein the associated feature interval is executed according to the corresponding quantization process, 20 is quantized to 25, 60 is quantized to 55, then the internal values 30, 40 and 50 are quantized synchronously, 30 is quantized to 32.5, 40 is quantized to 40 and remains unchanged, and 50 is quantized to 47.5;

[0057] In the specific quantization process, the specific way of determining the corresponding quantization value is to determine the interval range value of the inherent interval first: Z=(Gmax-Gmin), and then determine the range value of the feature interval: F=2X1, use Z÷F=LH to confirm the unit quantization value, and assume that the pixel value associated with a pixel point is XF1, the quantization value LF1=(XF1-Gmin)×LH+(J k -X1);

[0058] The pixels at the outer edge of each partition of the same type are determined, and several adjacent pixels associated with adjacent partitions of the same type are determined. The pixel value features between adjacent pixels are adjusted by a uniform amount, and the pixel value to be adjusted associated with the corresponding pixel is determined. The first sub-step of the above step three is to process the pixels in the corresponding partition so that the corresponding numerical features are fully optimized and adjusted. The second sub-step is to process the pixels at the edge of the corresponding partition so that the corresponding numerical features are optimized and adjusted synchronously.

[0059] The specific method for adjusting the average is as follows:

[0060] From the identified partitions of the same type, determine the adjacent partitions of the same type, record the pixels at the outer edge of the partitions of the same type as edge pixels, and confirm the adjacent edge pixels between adjacent partitions of the same type. Two adjacent sets of edge pixels belong to two adjacent pixels in the horizontal direction (the pixels are arranged in a nine-square grid, so two sets of adjacent pixels in the horizontal direction can be confirmed).

[0061] Based on the confirmed adjacent edge pixels, the grayscale value series associated with two sets of adjacent edge pixels is denoted as (HD). o QD o ), where 'o' represents different grayscale value columns, and different adjacent edge pixels are associated with different grayscale value columns, where HD o and QD o These represent the pixel values ​​of two pixels, which are located in different partitions of the same type of adjacent partitions.

[0062] Combination Figure 2 Each grayscale value column is quantized into a line segment. The total length of this line segment is the sum of the two grayscale values ​​in the grayscale value column. Each line segment contains a dividing point, which divides the line segment into a front segment and a back segment. The unit value associated with the front segment is HD. o The unit value associated with the latter line segment is QD. o CZ o =(HD) o -QD o Confirm the standard deviation CZ associated with the corresponding grayscale value column. o ;

[0063] Within each line segment, a set of moving feature points is marked. These moving feature points can move within the corresponding line segment. The portion of the line segment in front of the moving feature point is called the preceding segment, and the portion of the line segment behind the moving feature point is called the following segment. The unit value associated with the preceding segment is denoted as QQ. o The unit value associated with the following segment is denoted as HQ. o Using: CQ o =(QQ)o -HQ o ) confirming the variation difference CQ associated with the moving feature point o ;

[0064] Confirming the process feature associated with each moving process based on different moving processes: performing variance processing on several groups of variation differences CQ associated with the corresponding moving process o to confirm the first feature TZ1, and then performing difference processing (the difference = |CQ o -CZ o |, that is, the distance range value of the two values) on the variation differences CQ o and the standard deviation CZ o associated with the corresponding gray value column to lock the feature difference value associated with the corresponding gray value column, the feature difference value ≥ 0, and then summing up several groups of feature difference values generated by the corresponding moving process to confirm the second feature TZ2;

[0065] Adopt: Process feature = TZ1 × C1 + TZ2 × C2 to confirm the process feature associated with the corresponding moving process, wherein C1 and C2 are both preset fixed coefficient factors, the specific values of which are determined by the operator according to experience, and C1 generally takes the value 0.357, and C2 generally takes the value 0.643;

[0066] From several groups of moving processes, select the specific moving process whose process feature is at the minimum value, and take the QQ o associated with the specific moving process as the to-be-adjusted pixel value of the pixel point associated with the HD o , and take the HQ o as the to-be-adjusted pixel value of the pixel point associated with the QD o .

[0067] Step four, based on different to-be-adjusted pixel values confirmed for different pixel points in the to-be-rendered area, confirming the location of the corresponding pixel point in the original image, and performing associated adjustment on the RGB value of the original image to complete the rendering optimization process of the original image, wherein the specific way of performing associated adjustment is:

[0068] Confirming the original pixel value X i associated with the corresponding pixel point, and then confirming the to-be-adjusted pixel value confirmed for the corresponding pixel point;

[0069] If the to-be-adjusted pixel value is consistent with X i , no adjustment is needed;

[0070] If the to-be-adjusted pixel value > X i , the RGB value is synchronously increased (increased by 1 unit value each time), to the pixel value X istop when the pixel value X i the difference between the pixel value X

[0071] If the pixel value X i is less than X i stop when the pixel value X

[0072] The data in the above formula are all dimensionless, and the contents not described in detail in the specification are all the prior art known to those skilled in the art.

[0073] The above examples are only used to illustrate the technical method of the present application and not to limit it. Although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A graphics rendering method based on a game rendering platform, characterized in that, The method comprises the following steps: Based on the preset pending template, the existing rendering area of the pre-stored multiple groups of graphics in the game rendering platform is calibrated; By confirming the numerical difference between adjacent pixel points in the rendering area, a single group of rendering areas is divided into multiple similar partitions; According to the preset fluctuation parameter, the pixel values of different points in each similar partition are quantified, and the associated pixel values of the corresponding points are locked; Then, the pixel points at the external edges of each similar partition are determined, and a plurality of adjacent pixel points associated with adjacent similar partitions are determined, and the pixel value characteristics between adjacent pixel points are uniformly adjusted to determine the associated pixel values of the corresponding pixel points; According to the determined pixel values, the RGB values associated with the specified pixel points are adjusted.

2. The method of claim 1, wherein, The pending template includes a black template and a white template, and the pixel value associated with the white template is 255, and the pixel value associated with the black template is 1.

3. The method of claim 1, wherein, The specific way of calibrating the rendering area based on the preset pending template is: The pre-stored single group of graphics is subjected to gray scale processing to confirm the associated gray scale graphics; The grayscale image is compared and verified with a pre-stored template to confirm the numerical differences between different pixels in the grayscale image and the white template. The pixel values ​​associated with different pixels are labeled as X. i Where i represents different pixels, using: C i =|X i -255| Confirm the difference value C associated with the corresponding pixel. i If C i If Y1 > Y1, then this pixel is marked as the point to be rendered, and Y1 is a preset value; otherwise, no marking is performed, and the area covered by several points to be rendered is marked as the area to be rendered. Reconfirm the value difference between different pixel points in the gray-scale pattern and the black template, adopt: C i = |X i -1| Confirm the difference value C associated with the corresponding pixel point i , if C i > Y1, mark this pixel point as a to-be-rendered point, otherwise, do not perform any marking, and mark the area covered by a plurality of to-be-rendered points as a to-be-rendered area.

4. The method of claim 1, wherein, The specific way of confirming the numerical difference between adjacent pixel points in the rendering area and dividing the similar partitions is: Confirming the gray scale value associated with each different pixel point in the rendering area, and marking as H i Randomly selecting a group of pixel points from the rendering area as undetermined points, and confirming the value difference between the adjacent pixel points around the undetermined points in sequence, the value difference being the absolute value of the pixel value difference associated with two adjacent points. If the value difference is less than or equal to Y2, Y2 being a preset value, the adjacent pixel points are marked as the same type of pixel points, otherwise, no marking is performed. Based on the numerical difference confirmation and the other pixel points belonging to the same type of pixel points, the other pixel points are further confirmed, and it is identified whether the adjacent points of the other pixel points belong to the same type of pixel points, and the same is repeated for the existing similar pixel points, and the specific area of the calibrated similar pixel points is calibrated as a similar partition.

5. The method of claim 1, wherein, The way of locking the pixel values of the points in each similar partition is: The mean value of several pixels associated with a certain number of pixels within the same partition is calculated, excluding pixels at the outer edge of the partition. This determines the mean feature J belonging to this partition. k Where k represents different partitions of the same type, and the characteristic intervals belonging to this partition of the same type are determined according to the preset fluctuation parameter X1 [J] k -X1, J k +X1], where X1 is a preset value; Then, the minimum value and the maximum value are selected from the pixel values associated with the pixel points in the similar partition, and the inherent interval [Gmin, Gmax] belonging to the similar partition is confirmed; quantize Gmin as (J k quantize Gmax as (J k quantization process, the quantization values associated with the pixel values of the pixel points are determined in sequence, and the determined quantization values are taken as the to-be-adjusted pixel values associated with the pixel points.

6. The method of claim 1, wherein, The specific way of determining the adjacent pixel points associated with the adjacent similar partitions is: From the calibrated similar partitions, adjacent similar partitions are determined, the pixel points at the external edges of the similar partitions are recorded as edge pixel points, and the adjacent edge pixel points between the adjacent similar partitions are confirmed, and the two adjacent edge pixel points belong to two adjacent pixel points in the horizontal direction; According to the confirmed adjacent edge pixel points, the gray value columns associated with the two groups of adjacent edge pixel points are recorded as (HD o , QD o ), wherein o represents different gray value columns, different adjacent edge pixel points are associated with different gray value columns, HD o and QD o represent the pixel values of two pixel points, respectively, and the two pixel points are located in different partitions of adjacent same type partitions.

7. The graphics rendering method based on a game rendering platform according to claim 6, wherein, The specific way of determining the pixel values of the pixel points at the external edges of the similar partitions is: Quantize each gray value column into a group of line segments, the total length of the line segment is the total value of the two groups of gray values of the gray value column, and there is a group of separation points in the line segment, which divides the line segment into a front line segment and a rear line segment, the unit value associated with the front line segment is HD o , and the unit value associated with the rear line segment is QD o , adopt: CZ o =(HD o -QD o ) Confirm the standard deviation value CZ o associated with the corresponding gray value column. A set of mobile feature points is calibrated in each line segment, which can move within the corresponding line segment. A part of the line segment in front of the mobile feature point is recorded as a front segment, and a part of the line segment behind the mobile feature point is recorded as a rear segment. The unit value associated with the front segment is recorded as QQ o The unit value associated with the rear segment is recorded as HQ o The following is used: CQ o =(QQ o -HQ o ) to confirm the change value CQ o associated with the mobile feature point. Based on different moving processes, confirm the process characteristics associated with each moving process: several groups of variation difference CQ associated with the corresponding moving process o Perform variance processing to confirm the first feature TZ1, and then sum the variation difference CQ of the same group of gray value columns generated by the corresponding moving process to confirm the second feature TZ2 o And the standard deviation CZ o Perform difference processing to lock the feature difference associated with the corresponding gray value column, and the feature difference is greater than or equal to 0. Then sum several groups of feature differences generated by the corresponding moving process to confirm the second feature TZ2; The process characteristics associated with the corresponding mobile process are confirmed by using process characteristics = TZ1xC1+TZ2xC2, wherein C1 and C2 are both preset fixed coefficient factors; From several groups of mobile processes, select a specific mobile process whose process feature is at a minimum value, and associate QQ o As HD o The pixel value to be adjusted of the associated pixel point, and HQ o As QD o The pixel value to be adjusted of the associated pixel point.

8. The method of claim 1, wherein, The specific way of adjusting the RGB values associated with the specified pixel points is: confirming the original pixel value X associated with the corresponding pixel point i confirming the adjusted pixel value of the corresponding pixel point If the pixel value to be adjusted is identical to X i No adjustment is needed. If the pixel value to be adjusted > X i then the RGB value is simultaneously raised to the associated pixel value X after the raising i stop when the difference between the pixel value to be adjusted and the pixel value after the raising is less than 3 If the pixel value to be adjusted < X i then the RGB value is simultaneously raised to the associated pixel value X after the raising i stop when the difference between the pixel value to be adjusted and the pixel value after the raising is less than 3.

Images