Image processing method, device, system, equipment and medium
By acquiring visual characteristic information of images and using the JND model to reduce the grayscale level of images, the problem of excessive data volume in near-eye display technology is solved, and data storage requirements are reduced while ensuring visual effects.
Patent Information
- Application Number
- CN202110130720.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-01-29
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2041-01-29
Smart Images

Figure CN112767558B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image processing, and in particular to image processing methods, devices, systems, equipment and media. Background Art
[0002] In order to provide a seamless immersive or personal experience, near-eye display technologies such as Virtual Reality (VR), Augmented Reality (AR) and Mixed Reality (MR) have emerged.
[0003] In the near-eye display technology, it is necessary to generate an image to be displayed, and then transmit relevant data of the image to be displayed to a display device, so that the image to be displayed can be displayed on the display device.
[0004] However, at present, in order to make the image display effect satisfy the visual perception of the human eye, the data volume of the image to be displayed is often large. The large amount of image data to be displayed will cause the near-eye display system to face problems such as excessive data transmission bandwidth. Therefore, a technical solution is needed to reduce the size of image data while ensuring the visual perception of the human eye. Summary of the invention
[0005] The embodiments of the present application provide an image processing method, apparatus, system, device and medium, which can reduce the size of image data while ensuring the visual experience of the human eye.
[0006] In a first aspect, an embodiment of the present application provides an image processing method, the method comprising:
[0007] acquiring a first image;
[0008] Determining visual characteristic information of a target pixel in the first image;
[0009] Determine the JND value of the target pixel by using the visual characteristic information and the preset JND model;
[0010] Determine a target grayscale level corresponding to the JND value of the target pixel by using a preset correspondence relationship between the JND value and the grayscale level, wherein the target grayscale level of the target pixel is less than or equal to an original grayscale level of the target pixel in the first image;
[0011] Based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is downscaled and quantized to obtain a second image.
[0012] In a second aspect, an embodiment of the present application provides an image processing device, the device comprising:
[0013] An image acquisition module, used for acquiring a first image;
[0014] An information determination module, used to determine visual characteristic information of a target pixel in the first image;
[0015] A JND value determination module, used to determine the JND value of the target pixel by using visual characteristic information and a preset just noticeable difference (JND) model;
[0016] A grayscale determination module, used to determine a target grayscale corresponding to the JND value of a target pixel by using a preset correspondence between the JND value and the grayscale, wherein the target grayscale of the target pixel is less than or equal to an original grayscale of the target pixel in the first image;
[0017] The image determination module is used to quantify the grayscale value of each pixel channel of the target pixel based on the target grayscale level to obtain a second image.
[0018] In a third aspect, a micro display system is provided, comprising:
[0019] A microdisplay controller, configured to execute the image processing method provided in the first aspect or any optional implementation of the first aspect, generate a control signal of a target image, and output the control signal to the microdisplay;
[0020] The micro display is used for displaying a target image based on a control signal.
[0021] According to a fourth aspect, an image processing device is provided, comprising:
[0022] a processor and a memory storing computer program instructions;
[0023] The processor reads and executes computer program instructions to implement the image processing method provided by the first aspect or any optional implementation manner of the first aspect.
[0024] In a fifth aspect, a computer storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the image processing method provided by the first aspect or any optional implementation of the first aspect is implemented.
[0025] The image processing method, device, system, equipment and medium of the embodiments of the present application can determine the target grayscale level of the target pixel according to the correspondence between the preset JND value and the grayscale value after determining the JND value of the pixel. Then, based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is quantized to obtain a second image. Since the fewer the grayscale levels of the pixels in the image, the smaller the amount of data required to store the image, and the JND value can characterize the maximum tolerance of the human eye to image distortion. Determining the target grayscale level of the target pixel by the JND value can reduce the amount of image data while ensuring the subjective perception of the human eye. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0027] Figure 1 1 is a schematic diagram of a curve showing the relationship between JND and background brightness provided in an embodiment of the present application;
[0028] Figure 2 is a schematic diagram of a JND and contrast relationship curve provided in an embodiment of the present application;
[0029] Figure 3 This is a schematic diagram of the distribution of photoreceptor cells in the retina of a human eye provided in an embodiment of the present application;
[0030] Figure 4 is a schematic diagram of a curve showing the relationship between JND and eccentricity angle provided in an embodiment of the present application;
[0031] Figure 5 is a schematic diagram of a JND-depth relationship curve provided in an embodiment of the present application;
[0032] Figure 6 It is a flowchart of an image processing method provided in an embodiment of the present application;
[0033] Figure 7 is a flowchart of another image processing method provided in an embodiment of the present application;
[0034] Figure 8 It is a flowchart of another image processing method provided in an embodiment of the present application;
[0035] Fig. 9 is a flowchart of another image processing method provided in an embodiment of the present application;
[0036] Fig.10 is a logic diagram of an exemplary target error value dispersed to adjacent pixels provided in an embodiment of the present application;
[0037] Fig.11 is a structural schematic diagram of an image processing device provided in an embodiment of the present application;
[0038] Fig.12 is a system architecture diagram of a micro display system provided in an embodiment of the present application;
[0039] Fig.13 A schematic structural diagram of an embodiment of an image processing device provided in the third aspect of the present application. DETAILED DESCRIPTION
[0040] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.
[0041] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.
[0042] In the prior art, the overall grayscale level of the image to be displayed is consistent, for example, the grayscale values of each channel of the pixel points of the entire image are all within the range of 0 to 255. However, due to the uniqueness of the physiological structure of the human eye, the human eye has different abilities to distinguish grayscale levels in different areas of the same image. For example, when the human eye looks at a picture, the larger the visual angle of the human eye, the weaker the human eye's ability to distinguish grayscale levels. Therefore, for images to be displayed with consistent overall grayscale levels, although the subjective perception of multiple images by the human eye is better, there is a visual perception redundancy phenomenon related to grayscale levels, and the amount of data for images to be displayed with consistent overall grayscale levels is large.
[0043] Therefore, a solution is needed that can reduce the number of images while ensuring the subjective perception of the human eye.
[0044] Based on this, the embodiments of the present application provide an image processing method, device, system, equipment and computer storage medium, which are suitable for near-eye display devices such as VR, AR or MR. In an embodiment of the present application, after acquiring the first image and determining the JND value of the target pixel, the target grayscale level corresponding to the JND value can be determined, and then based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is quantized to obtain a second image. Since the fewer the grayscale levels of the pixels in the image, the smaller the amount of data required to store the image, and the JND value can characterize the maximum tolerance of the human eye to image distortion. Determining the target grayscale level of the target pixel by the JND value can reduce the amount of image data while ensuring the subjective perception of the human eye.
[0045] In order to better understand the present application, the embodiments of the present application explain and illustrate the concepts of JND, visual masking, brightness masking, contrast masking, foveal masking, depth masking, grayscale level and grayscale quantization in detail.
[0046] (1) JND.
[0047] Just noticeable difference (JND), also known as the difference threshold, is used to indicate the minimum change in stimulation perceived by the human visual system. It also reflects the maximum tolerance of the human eye to image distortion.
[0048] (2) Visual masking.
[0049] Visual masking refers to the fact that the human eye has different resolutions of images due to the influence of various visual characteristics. In other words, JND is affected by various visual characteristics, such as brightness masking, contrast masking, foveal masking, and depth masking.
[0050] Next, various types of masking will be described in turn.
[0051] (3) Brightness masking.
[0052] Brightness masking means that different background brightness will affect the human eye's ability to distinguish images. Figure 1 : is a schematic diagram of a curve showing the relationship between JND and background brightness provided in an embodiment of the present application. Figure 1 The curve in is used to show the JND variation with background brightness when the depth is 1.0 and the eccentricity angle is 20 degrees (°).
[0053] pass Figure 1 It can be seen from the curve that the JND value is larger in the area with smaller background brightness value and in the area with larger background brightness value, and the JND value is smaller in the middle area between the area with smaller value and the area with larger value.
[0054] (4) Contrast masking.
[0055] Contrast masking describes a phenomenon in which when multiple different stimuli exist at the same time, the human eye can only distinguish some of them. Figure 2 : is a schematic diagram of a JND and contrast relationship curve provided in an embodiment of the present application. Figure 2 The curve in is used to show the variation curve of JND with contrast when the depth is 1.0, the eccentricity angle is 20°, and the background brightness is 140. The curve shows that contrast masking increases the JND value, and the greater the contrast, the greater the JND value.
[0056] pass Figure 2 It can be seen from the curve shown that the JND value increases with the increase of contrast.
[0057] (5) Foveal masking.
[0058] Foveal masking refers to the phenomenon that the human eye has strong resolution in the foveal area and weak resolution in the peripheral area. This is caused by the distribution of photoreceptor cells on the retina of the human eye, where the distribution of photoreceptor cells is highly uneven. Figure 3 This is a schematic diagram of the distribution of photoreceptor cells in the retina of a human eye provided in an embodiment of the present application. Figure 3 The figure shows the distribution of photoreceptor cells on the retina within the visual angle range of 5° to 50°, where each hexagonal unit represents the same number of photoreceptor cells.
[0059] like Figure 3 As shown in the figure, the closer to the fovea, that is, the smaller the viewing angle, the densest distribution density of photoreceptor cells. Since cone cells have a strong resolution for strong light, the closer to the fovea, the higher the visual acuity. The farther from the fovea, the larger the viewing angle, the sparser the distribution density of cells, and correspondingly, the lower the visual acuity.
[0060] Based on the above physiological structure of human eyes, when the human eye views an image, Figure 4 : is a schematic diagram of a curve showing the relationship between JND and eccentricity angle provided in an embodiment of the present application. Figure 4 The curve in FIG. 1 is used to show the variation of JND with eccentricity angle when the depth is 1.0 and the background angle is 140.
[0061] like Figure 4 As shown in Figure 3, JND remains basically unchanged in the range of 0° to 5°, then begins to increase, changes gently in the range of 15° to 25°, and begins to increase sharply after exceeding 40°.
[0062] (6) Depth masking.
[0063] Depth masking describes the difference in human eye perception at different depths. In stereoscopic vision, the human eye has a comfort zone and an discomfort zone. Within the comfort zone, the visual acuity of the human eye decreases as the viewing distance increases, and the eye pays more attention to nearby objects. These are all manifestations of depth masking.
[0064] Figure 5 : is a schematic diagram of a JND-depth relationship curve provided in an embodiment of the present application. Figure 5 The curve in FIG. 1 is used to show the JND variation with depth when the eccentricity angle is 20° and the background angle is 140°.
[0065] like Figure 5 As shown in the figure, JND gradually increases in the depth range of 0.5 to 1, which is consistent with the fact that visual acuity decreases with increasing viewing distance. In the depth range less than 0.5, the smaller the depth value, the larger the JND. This may be because the imaging is too close and the human eye begins to enter the discomfort zone, resulting in a decrease in visual acuity.
[0066] (7) Grayscale levels and grayscale quantification.
[0067] For an image with a consistent overall grayscale, the grayscale refers to the number of grayscale levels in an image from the darkest to the brightest. Each pixel in the image can present multiple colors after quantization. The more grayscale levels there are, the more delicate the image effect will be and the stronger the image layering will be. If the brightness range of the pixel is quantized by n bits, the grayscale of the pixel from black to white will contain 2 n Optionally, the value of n can be any positive integer less than or equal to 8.
[0068] In the embodiment of the present application, for images with inconsistent overall grayscale levels, the grayscale level refers to the number of grayscale levels divided between the darkest and brightest channels of each pixel. For example, if the first image in the embodiment of the present application is a grayscale image, the number of channels of the pixels in the image is 1, that is, if the value of n is 8, the value of the pixel can be between 0 and 255. If the first image in the embodiment of the present application is an RGB image (that is, an image with three channels representing the three colors of red, green, and blue respectively), the number of channels of each pixel in the image is 3. If the grayscale level of the pixel channel of a certain pixel is 2 n , and the value of n is 8, then the value of each pixel channel of the pixel can be between 0 and 255.
[0069] Table 1 shows the quantization of different gray levels in the embodiment of the present application.
[0070] Table 1
[0071] Grayscale Quantized value of gray level <![CDATA[2 8 Level]]> 0、1、2、……、253、254、255 <![CDATA[2 7 Level]]> 0、2、4、……、250、252、254 <![CDATA[2 6 Level]]> 0、4、8、……、244、248、252 <![CDATA[2 5 Level]]> 0、8、16、……、232、240、248 <![CDATA[2 4 Level]]> 0、16、32、……、208、224、240 <![CDATA[2 3 Level]]> 0、32、64、……、160、192、224 <![CDATA[2 2 Level]]> 0、64、128、192 <![CDATA[2 1 Level]]> 0、128
[0072] After fully understanding the above concepts, in order to better understand the present application, the image processing methods, devices, equipment and media according to the embodiments of the present application will be described in detail below in conjunction with the accompanying drawings. It should be noted that these embodiments are not intended to limit the scope of the present application.
[0073] Figure 6 is a flow chart of an image processing method provided in an embodiment of the present application. Figure 6 As shown, the image processing method includes S510 and S520. Optionally, the execution body of each step of the image processing method 600 may be an image processing device independently provided with the near-eye display device, or may be an image processing module inside the near-eye display device, which is not limited.
[0074] S610, acquiring a first image.
[0075] In S610, the first image may refer to an image to be compressed. In some embodiments, the first image may refer to an image with a consistent overall grayscale. Alternatively, the first image may refer to an image with a different overall grayscale, which is not limited.
[0076] S620: Determine visual characteristic information of a target pixel in the first image.
[0077] In S620, first, for a target pixel, the target pixel may refer to a pixel to be subjected to grayscale reduction processing. The grayscale reduction processing in the embodiment of the present application may refer to a step of re-determining the grayscale level according to the JND value of the pixel, and re-quantizing the grayscale value of each pixel channel of the pixel according to the re-determined grayscale level.
[0078] In some embodiments, the target pixels may be all pixels in the first image. In other embodiments, the target pixels may be some pixels in the first image. For example, since the human eye has a higher visual sensitivity to the central area and a lower visual sensitivity to the edge area, only the pixels in the edge area of the image may be downscaled.
[0079] Secondly, the visual characteristic information may refer to parameters that may affect the visual sensitivity of the human eye. For example, the parameters that may affect the JND value. Exemplarily, the visual characteristic information may be at least one of the parameters such as background brightness L, eccentricity angle E, depth D, and contrast C.
[0080] Specifically, the above-mentioned characteristic parameter information can be obtained by calculation. Next, the following part of the embodiment of the present application will specifically explain the calculation formulas of several parameters respectively.
[0081] First, for the background brightness L, the background brightness L(i,j) of any pixel (i,j) can be determined based on the grayscale values of its adjacent pixels, for example, it can be the weighted average of the grayscale values of its adjacent pixels. The number of adjacent pixels can be determined based on the size of the weighted template. For example, if the weighted template is m×m, it has m×m-1 adjacent pixels. m can be any positive integer, such as 5.
[0082] Specifically, the background brightness L of any pixel (i, j) is bg (i, j) satisfies the calculation formula (1):
[0083]
[0084] Wherein, a is a fixed parameter, which is related to B(x, y), for example, a is the sum of all elements of B(x, y), such as 32. Wherein, I(i-3+x, j-3+y) represents the gray value of pixel (i-3+x, j-3+y), B(x, y) is a low-pass filter, specifically, B(x, y) can be implemented as an m×m matrix.
[0085] In a specific example, if m=5, B(x, y) can be specifically expressed as formula (2):
[0086]
[0087] Second, for the eccentric angle E. The eccentric angle E(i, j) of any pixel (i, j) satisfies formula (3):
[0088]
[0089] The gaze point coordinates in the image are (i0, j0), and ppd is the angular resolution of the image displayed in the near-eye display device, which may be equal to 10, for example. The gaze point may be the center point of the image, or may be a captured gaze point of a human eye.
[0090] Third, for the depth D. The depth D(i, j) of any pixel (i, j) satisfies formula (4):
[0091]
[0092] Where b is the baseline distance between two adjacent cameras, f is the focal length of the camera, ps is the pixel size, and disp(i, j) is the disparity value between the left and right images corresponding to the pixel point (i, j).
[0093] Fourth, contrast C. The depth C(i, j) of any pixel (i, j) satisfies formula (5):
[0094]
[0095] Among them, E K (i, j) satisfies formula (6), K = H, V:
[0096]
[0097] Among them, b is a fixed parameter, which is related to G K Related, for example, can be equal to 24. G H is the Sobel horizontal gradient operator, G V is the Sobel vertical gradient operator. Specifically, G H and G V It can be implemented as an m×m matrix.
[0098] In one example, G H Satisfies formula (7):
[0099]
[0100] In one example, G V Satisfies formula (8):
[0101]
[0102] In some embodiments, the grayscale reduction process can be performed on all pixels in the image by traversing the first image using a slider. At this time, the target pixel includes the pixel in the slider. Exemplarily, the size of the slider can be m×m. Figure 7 It is a flowchart of another image processing method provided in an embodiment of the present application. Figure 7 and Figure 6 The difference is that S620 can be specifically implemented as S621 and S622.
[0103] S621 : in the process of traversing the first image by using the slider, determining visual characteristic information of reference pixels in the slider.
[0104] The reference pixel in the slider may be a pixel at the center of the slider.
[0105] Exemplarily, the slider may be moved on the image in a sequence from left to right and from top to bottom.
[0106] S622: Determine the visual characteristic information of the reference pixel as the visual characteristic information of the target pixel.
[0107] S630, determining the JND value of the target pixel by using the visual characteristic information and a preset JND model.
[0108] In S630 , if the JND model is implemented as a JND function, the visual characteristic information may be substituted into a preset JND function to calculate a JND value of the target pixel.
[0109] In some embodiments, under the premise of ensuring image quality, in order to speed up the calculation speed, the JND model may be a JND model that takes into account both background brightness and eccentricity angle. In this case, the visual characteristic information may include background brightness and eccentricity angle.
[0110] Specifically, the JND model can be related to background brightness and eccentricity angle.
[0111] In one example, the JND model is expressed as formula (9):
[0112]
[0113] Next, the following parts of the embodiments of the present application will provide specific descriptions of several parameters respectively.
[0114] First, the first correlation parameter L1, the first correlation parameter L1 and the background brightness L of the target pixel bg Specifically, the first correlation parameter L1 of any pixel (i, j) can satisfy formula (10):
[0115]
[0116] In formula (10), the first preset parameter a1=2.046e-08, the second preset parameter a2=-9.44e-06, the third preset parameter a3=0.001437, the fourth preset parameter a4=-0.07248, and the fifth preset parameter a5=5.273.
[0117] It should be noted that the background brightness L bg The specific calculation formula of (i, j) can be found in the above formula (1), which will not be repeated here.
[0118] In addition, the first to fifth preset parameters may also be set according to specific scenarios and actual needs, and the specific values of the above preset parameters are not limited.
[0119] Second, the second related parameter E1 is related to the eccentricity angle E(i, j) of the target pixel. Specifically, the second related parameter E1 of any pixel (i, j) can satisfy formula (11):
[0120] E1=b1*E 4 (i,j)+b2*E 3 (i,j)+b3*E 2 (i,j)+b4*E(i,j)+b5 (11)
[0121] In formula (11), the sixth preset parameter b1 = 3.099e-06, the seventh preset parameter b2 = -0.0001298, the eighth preset parameter b3 = 8.477e-06, the ninth preset parameter b4 = 0.1435, and the tenth preset parameter b5 = 4.663. It should be noted that the specific calculation formula of the eccentric angle E (i, j) can be found in the above formula (3), which will not be repeated here. In addition, the sixth to tenth preset parameters can also be set according to specific scenarios and actual needs, and the specific values of the above preset parameters are not limited.
[0122] Third, the reference background brightness value E1(0). Specifically, E1(0) may be the value obtained by substituting E(i, j)=0 into formula (11).
[0123] Fourth, the third related parameter D1 and the second related parameter D1 are related to the depth value D(i, j) of the target pixel. Specifically, the third related parameter D1 of any pixel (i, j) can satisfy formula (12):
[0124] D1=c1*D 4 (i,j)+c2*D 3 (i,j)+c3*D 2 (i,j)+c4*D(i,j)+c5 (12)
[0125] In formula (12), the eleventh preset parameter c1=26.04, the twelfth preset parameter c2=-93.75, the thirteenth preset parameter c3=124, the fourteenth preset parameter c4=-66.25, and the fifteenth preset parameter c5=15.
[0126] It should be noted that the specific calculation formula of the depth value D(i, j) can be found in the above formula (5), which will not be repeated here.
[0127] In addition, the eleventh to fifteenth preset parameters may also be set according to specific scenarios and actual needs, and there is no limitation on the specific values of the above preset parameters.
[0128] Fifth, reference depth value D1(1.0). Specifically, D1(1.0) may be a value obtained by substituting D(i, j)=1.0 into formula (12).
[0129] Sixth, the first correction parameter α1 can be set according to the actual scenario and specific requirements, and will not be described in detail here. Exemplarily, the first correction parameter α1 can be equal to -1.637.
[0130] Seventh, the second correction parameter α2 can be set according to the actual scenario and specific requirements, and will not be described in detail here. Exemplarily, the second correction parameter α2 can be equal to 0.224.
[0131] In a specific example, under the premise of ensuring image quality, in order to further speed up the calculation speed, the depth value can be set to a fixed value. Specifically, the depth value in formula (9) can be set to is equal to 1.
[0132] At this time, the JND model can be expressed as formula (13):
[0133]
[0134] It should be noted that the JND model provided in the embodiment of the present application can take into account background brightness masking and foveal masking, thereby improving the matching degree between the JND model and the characteristics of the human eye. Compared with other existing JND models, under the same objective quality evaluation parameters, the subjective quality evaluation of this model is higher than that of other models, and is more in line with the visual characteristics of the human eye.
[0135] In other embodiments, in order to further compress the image, the JND model may be a JND model that takes into account background brightness masking, foveal masking, depth masking, and contrast masking. In this case, the visual characteristic information may include background brightness, eccentricity angle, depth, and contrast value.
[0136] At this time, the JND model is related to background brightness, eccentricity angle, depth and contrast value.
[0137] In one example, the JND model is expressed as formula (14):
[0138]
[0139] Among them, FD-JND L Please refer to formula (9), which will not be repeated here.
[0140] Among them, FD-JND C The following formula (15) can be satisfied:
[0141] FD-JND C =FD-JND L +CM (15)
[0142] Where CM is contrast masking. CM can satisfy formula (16):
[0143]
[0144] Among them, μ1 is the first preset parameter, μ2 is the second preset parameter, μ3 is the third preset parameter, μ4 is the fourth preset parameter, ν1 is the fifth preset parameter, ν2 is the sixth preset parameter, ν3 is the seventh preset parameter, ν4 is the eighth preset parameter, ω1 is the ninth preset parameter, ω2 is the tenth preset parameter, ω3 is the eleventh preset parameter, and ω4 is the twelfth preset parameter. Among them, the first preset parameter to the twelfth preset parameter can be set according to actual scenarios and specific needs, and no further details are given for this.
[0145] In a specific example, μ1 = -2.206e-05, μ2 = 0.006143, μ3 = -0.07232, μ4 = -1.06, ν1 = -1.731e-05, ν2 = 0.004102, ν3 = 0.3572, v4 = -0.65, ω1 = 3.983e-06, ω2 = -0.001496, ω3 = 0.2457, ω4 = -0.46.
[0146] It should be noted that the JND model provided in the embodiment of the present application can take into account background brightness masking, foveal masking, depth masking and contrast masking, thereby improving the matching degree between the JND model and the characteristics of the human eye. Compared with other existing JND models, under the same objective quality evaluation parameters, the subjective quality evaluation of this model is higher than that of other models, and is more in line with the visual characteristics of the human eye.
[0147] In one example, if the number of pixel channels of the target pixel is 1, the JND value can be calculated using the above formula (13) or formula (14).
[0148] In another example, if the number of pixel channels of the target pixel is multiple, Figure 8 It is a flowchart of another image processing method provided in an embodiment of the present application. Figure 8 and Figure 6 The difference is that S630 can be specifically implemented as S631 to S634.
[0149] S631, using the visual characteristic information and a preset JND model, determining a first JND value of each of the plurality of pixel channels of the target pixel.
[0150] In one embodiment, the first JND values of the plurality of pixel channels can be calculated in sequence using formula (9) or formula (13). For example, for an RGB graphic, the first JND value of the R channel calculated using formula (9) or formula (13) can be expressed as JND R , the first JND value of the G channel can be expressed as JND G , the first JND value of channel B can be expressed as JND B .
[0151] For example, the background brightness L corresponding to the three channels of RGB can be calculated using the above formula: R , L G , L B and eccentricity angle E, and then according to formula (9) or formula (13) the first JND value JND under the three channels is obtained R , JND G and JND B .
[0152] S632, determine the weight values of each of the multiple pixel channels.
[0153] In one embodiment, the sum of the weight values of the multiple channels may be equal to 1.
[0154] For example, the weighting weights can be determined according to the sensitivity of the human eye to the color of each channel. Since when R:G:B=2:1:4, the image quality can be better guaranteed, therefore, in order to improve the image quality, the weight value of the R channel can be equal to 2 / 7, the weight value of the G channel can be equal to 1 / 7, and the weight value of the B channel can be equal to 4 / 7.
[0155] S633: Calculate a weighted JND value by using the first JND values of each of the multiple pixel channels and the first JND values of each of the multiple pixel channels of the target pixel.
[0156] Exemplarily, the weighted JND value may be expressed as (just noticeable color differences, JNCD), and JNDC may represent color difference.
[0157] Specifically, JNDC can satisfy formula (17):
[0158]
[0159] S634: Determine the weighted JND value as the JND value of the target pixel.
[0160] S640, using the preset correspondence relationship between the JND value and the grayscale level, determine the target grayscale level corresponding to the JND value of the target pixel.
[0161] The target grayscale level of the target pixel is less than or equal to the original grayscale level of the target pixel in the first image.
[0162] In some embodiments, in the corresponding relationship, the JND value is negatively correlated with the target grayscale level.
[0163] In a specific example, different JND value ranges correspond to different target grayscale levels. For example, if the number of pixel channels of the target pixel is 1, Table 2 shows a correspondence between a JND value range and a grayscale level provided in an embodiment of the present application.
[0164] Table 2
[0165] JND value range Grayscale Pixel data width required (0,3] <![CDATA[256(2 8 Level)]]> 8 (3,5] <![CDATA[128(2 7 Level)]]> 7 (5,9] <![CDATA[64(2 6 Level)]]> 6 (9,11] <![CDATA[32(2 5 Level)]]> 5 (11,+∞] <![CDATA[16(2 4 Level)]]> 4
[0166] Exemplarily, if the JND value of the target pixel is 1, the target grayscale level is 256; if the JND value of the target pixel is 4, the target grayscale level of the target pixel is 128.
[0167] It should be noted that the end values of each JND value range are not limited to the end values shown in Table 2, and can be set according to actual conditions and specific scenarios, which will not be elaborated in the embodiments of the present application.
[0168] In an example, Table 3 shows a correspondence between a JNDC value range and a grayscale level provided in an embodiment of the present application.
[0169] Table 3
[0170]
[0171]
[0172] Exemplarily, if the JNDC value of the target pixel is 1, the target grayscale level is 256; if the JNDC value of the target pixel is 9, the target grayscale level of the target pixel is 32.
[0173] It should be noted that the end values of each JNDC value range are not limited to the end values shown in Table 3, and can be set according to actual conditions and specific scenarios, which will not be described in detail in the embodiments of the present application.
[0174] S650, based on the target grayscale level, quantize the grayscale value of each pixel channel of the target pixel to obtain a second image.
[0175] In some embodiments, if the target grayscale level is p, the quantization process of the target pixel (i, j) satisfies formula (18):
[0176]
[0177] Among them, round is the rounding function, M and N are the length and width of the image, and the target grayscale level p = 0, 1, 2, ..., 7. For example, when the grayscale is 4, the grayscale quantization value is 0, 64, 128, 192.
[0178] The image processing method of the embodiment of the present application can determine the target grayscale level of the target pixel according to the correspondence between the preset JND value and the grayscale value after determining the JND value of the pixel. Then, based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is quantized to obtain a second image. Since the fewer the grayscale levels of the pixels in the image, the smaller the amount of data required to store the image, and the JND value can characterize the maximum tolerance of the human eye to image distortion. Determining the target grayscale level of the target pixel by the JND value can reduce the amount of image data while ensuring the subjective perception of the human eye.
[0179] In some embodiments, the grayscale reduction process may cause errors in the pixel values of the surrounding area, thereby affecting the image accuracy. In order to improve the image accuracy and reduce the error caused by the grayscale reduction process, Fig. 9 It is a flowchart of another image processing method provided in an embodiment of the present application. Fig. 9 and Figure 6 The difference is that S650 can be specifically implemented as S651 to S654.
[0180] S651, determine a plurality of adjacent pixels of the target pixel.
[0181] It should be noted that the neighboring pixels of the pixel can be determined according to the size of the error propagation template.
[0182] In some embodiments, if the size of the error transfer template is 2×2, the neighboring pixels of each pixel may include 3 pixels, such as the pixel in the upper left corner, the pixel directly above, and the pixel directly to the left. Exemplarily, if the coordinates of the pixel in the image are (i, j), the coordinates of its neighboring pixels may be (i-1, j-1), (i, j-1), (i-1, j). It should be noted that other neighboring pixels may also be used, such as the pixel directly to the right, the pixel directly below, and the pixel in the lower right corner, and there is no specific limitation on this. Among them, the error transfer template can be implemented as a matrix, and the values in the matrix are the weight values corresponding to each neighboring pixel.
[0183] It should be noted that when the error propagation template uses a 2×2 matrix, compared with other error propagation templates, it can take into account the image display quality, calculation speed and hardware cost. It should be noted that the sum of the values of all elements in the matrix is equal to 1.
[0184] In one example, the 2×2 error transfer template T can be expressed as equation (19):
[0185]
[0186] For example, the 2×2 error transfer template T can be specifically expressed as formula (20)
[0187]
[0188] It should be noted that the error transfer template T is to obtain the target error value by multiplying the pixel error difference of each pixel by the corresponding weight value, and then disperse the target error value to adjacent pixels. For example, Fig.10 is a logic diagram of an exemplary target error value dispersed to adjacent pixels provided in an embodiment of the present application. Fig.10 As shown, for pixel (i, j), if its pixel value in the first image is f(i, j) and its pixel value in the second image is g(i, j), then its pixel error R(i, j) can be the difference between the pixel value f(i, j) and the pixel value g(i, j). The target error value assigned to the pixel to its right is R(i, j)×w1, the target error value assigned to the pixel directly below is R(i, j)×w2, and the target error value assigned to the pixel in the lower right corner is R(i, j)×w3.
[0189] For example, for pixel (i, j), after its pixel error R(i, j) is dispersed to adjacent pixels, the pixel value of the pixel (i+1, j) to the right of it can satisfy formula (21):
[0190] g′(i+1,j)=g(i+1,j)+w1×R(i,j) (21)
[0191] The pixel value of the pixel (i, j+1) directly below it can satisfy formula (22):
[0192] g′(i,j+1)=g(i,j+1)+w2×R(i,j) (22)
[0193] The pixel value of the lower right corner pixel (i+1, j+1) can satisfy formula (23):
[0194] g′(i+1,j+1)=g(i+1,j+1)+w3×R(i,j) (23)
[0195] It should be noted that the pixel values of each pixel shown in formula (21)-formula (23) are only the pixel values of pixel (i, j) after the pixel error is dispersed. After all pixels are dispersed, the new pixel value of each pixel in the target image satisfies the following formula (25).
[0196] In other embodiments, if the error transfer template is 3×3, the neighboring pixels of each pixel may be its 8 neighboring pixels, the pixel at its upper left corner, the pixel directly above, the pixel at its upper right corner, the pixel directly to the left, the pixel directly to the right, the pixel at its lower left corner, the pixel directly below, and the pixel at its lower right corner.
[0197] It should be noted that other adjacent pixels may be selected based on actual conditions and specific needs, and the embodiments of the present application do not specifically limit this.
[0198] S652, determining grayscale differences of a plurality of adjacent pixels, wherein the grayscale difference of each adjacent pixel refers to the difference between the grayscale value of each channel of each adjacent pixel before quantization and the grayscale value after quantization.
[0199] Among them, for the adjacent pixel (i-1, j-1), the pixel difference R(i-1, j-1) between its pixel value f(i-1, j-1) in the original image and its pixel value g(i-1, j-1) in the target image can be expressed as formula (24):
[0200] R(i-1,j-1)=f(i-1,j-1)-g(i-1,j-1) (24)
[0201] The calculation formulas for the pixel difference values of other pixel points can be found in formula (24), which will not be repeated here.
[0202] S653, obtaining a weighted grayscale difference value of each adjacent pixel according to the product of the grayscale difference value of each adjacent pixel and the weight value of each adjacent pixel.
[0203] S654, updating the weighted grayscale difference values of each of the plurality of adjacent pixels and the sum of the grayscale values of each channel of the target pixel before quantization to the grayscale value of each pixel channel of the target pixel, to obtain a second image.
[0204] For example, taking the pixel value (i, j) as an example, its new pixel value g′(i, j) in the target image can be expressed as formula (25):
[0205] g′(i+1,j+1)=g(i,j)+w3×R(i,j)+w1×R(i+1,j)+w2×R(i,j+1) (25)
[0206] Through this embodiment, the error of each pixel can be dispersed to adjacent pixels, so that the influence of the error on the image is significantly reduced, and the display quality of the image is improved.
[0207] In one example, for each pixel, if its new pixel value in the target graphic is greater than the maximum quantized value of the grayscale level of each pixel, the difference between the new pixel value and the maximum quantized value is determined, and then the difference is dispersed to the adjacent pixels of each pixel.
[0208] First, the maximum quantized value of the grayscale level of the pixel is the maximum value among multiple quantized values of the grayscale level of the pixel. Specifically, see Table 1. For example, for a grayscale level of 2 8The maximum quantization value is 255 and the gray level is 2. 7 The maximum quantization value is 254, ..., for gray level 2 2 The maximum quantization value is 192, and the gray level is 2 1 The maximum quantization value is 128.
[0209] For example, taking the pixel value (i, j) as an example, if its gray level is 2 8 If, after steps S531 to S533, the new pixel value g′(i, j) is greater than 255, the difference between the new pixel value g′(i, j) and the maximum quantization value 255 is dispersed to the adjacent pixels of the pixel.
[0210] It should be noted that, through the image processing method provided in the embodiment of the present application, the average compression rates for grayscale images and color images can reach 66.34% and 61.33%, respectively, which can significantly reduce the amount of image data while ensuring image quality.
[0211] Based on the same application concept, in addition to providing an image processing method, the embodiment of the present application also provides a corresponding image processing device.
[0212] The image processing device according to the embodiment of the present application is described in detail below with reference to the accompanying drawings.
[0213] Fig.11 Schematic diagram of the structure of an image processing device provided in an embodiment of the present application. Fig.11 As shown, the image processing apparatus 1100 includes an image acquisition module 1110 , an information determination module 1120 , a JND value determination module 1130 , a grayscale determination module 1140 and an image determination module 1150 .
[0214] An image acquisition module 1110, configured to acquire a first image;
[0215] An information determination module 1120, configured to determine visual characteristic information of a target pixel in the first image;
[0216] A JND value determination module 1130 is used to determine the JND value of the target pixel by using the visual characteristic information and a preset just noticeable difference (JND) model;
[0217] A grayscale level determination module 1140 is used to determine a target grayscale level corresponding to the JND value of the target pixel by using a preset correspondence relationship between the JND value and the grayscale level, wherein the target grayscale level of the target pixel is less than or equal to an original grayscale level of the target pixel in the first image;
[0218] The image determination module 1150 is used to quantify the grayscale value of each pixel channel of the target pixel based on the target grayscale level to obtain a second image.
[0219] In some embodiments, the visual characteristic information includes background brightness and eccentricity angle;
[0220] The JND model is related to background brightness and eccentricity angle.
[0221] In some embodiments, the JND model is expressed as formula (13).
[0222] The specific content of the JND model can be found in the relevant description of formula (13), which will not be repeated here.
[0223] In some embodiments, the visual characteristic information includes background brightness, eccentricity angle, depth, and contrast value;
[0224] The JND model is related to background brightness, eccentricity angle, depth and contrast value.
[0225] In some embodiments, the JND model is expressed as formula (14).
[0226] The specific content of the JND model can be found in the relevant descriptions of formulas (14)-(16), which will not be repeated here.
[0227] In some embodiments, the target pixel comprises a pixel within the slider;
[0228] The information determination module 1120 specifically includes:
[0229] A first determining unit, configured to determine visual characteristic information of reference pixels in the slider during the process of traversing the first image using the slider;
[0230] The second determining unit is used to determine the visual characteristic information of the reference pixel as the visual characteristic information of the target pixel.
[0231] In some embodiments, the number of pixel channels of the target pixel is multiple;
[0232] The JND value determination module 1130 specifically includes:
[0233] a third determining unit, configured to determine a first JND value of each of a plurality of pixel channels of a target pixel by using the visual characteristic information and a preset JND model;
[0234] A fourth determining unit, used to determine the weight value of each of the plurality of pixel channels;
[0235] a fifth determining unit, configured to calculate a weighted JND value by using the first JND values of each of the plurality of pixel channels and the first JND values of each of the plurality of pixel channels of the target pixel;
[0236] The sixth determining unit is configured to determine the weighted JND value as the JND value of the target pixel.
[0237] In some embodiments, the image determination module 1150 specifically includes:
[0238] a seventh determining unit, configured to determine a plurality of adjacent pixels of a target pixel;
[0239] an eighth determining unit, configured to determine grayscale differences of each of a plurality of adjacent pixels, wherein the grayscale difference of each adjacent pixel refers to a difference between a grayscale value of each channel of each adjacent pixel before quantization and a grayscale value after quantization;
[0240] A ninth determining unit, configured to obtain a weighted grayscale difference value of each adjacent pixel according to the product of the grayscale difference value of each adjacent pixel and the weight value of each adjacent pixel;
[0241] The grayscale value updating unit is used to update the weighted grayscale difference values of multiple adjacent pixels and the sum of the grayscale values of each channel of the target pixel before quantization to the grayscale value of each pixel channel of the target pixel to obtain a second image.
[0242] The image processing device of the embodiment of the present application can determine the target grayscale level of the target pixel according to the correspondence between the preset JND value and the grayscale value after determining the JND value of the pixel. Then, based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is quantized to obtain a second image. Since the fewer the grayscale levels of the pixels in the image, the smaller the amount of data required to store the image, and the JND value can characterize the maximum tolerance of the human eye to image distortion. By determining the target grayscale level of the target pixel through the JND value, the amount of image data can be reduced while ensuring the subjective perception of the human eye.
[0243] According to other details of the image processing device of the embodiment of the present application, in combination with the above Figures 6 to 10 The image processing method described in the example shown is similar and can achieve its corresponding technical effect. For the sake of brevity, it will not be repeated here.
[0244] Based on the same application concept, in addition to providing an image processing method and an image processing device, the embodiment of the present application also provides a corresponding micro display system.
[0245] Fig.12 is a system architecture diagram of a micro display system provided in an embodiment of the present application. Fig.12 As shown, the micro display system 1200 includes a micro display controller 1210 and a micro display 1220 .
[0246] The micro display controller 1210 is used to execute the embodiment of the present application in combination with Figures 6 to 9 The image processing method shown generates a control signal of a target image and outputs the control signal to a micro display.
[0247] The micro display 1220 is used to display a target image based on a control signal.
[0248] In some embodiments, information can be transmitted between the micro display controller 1210 and the micro display 1220 via wired communication or wireless communication.
[0249] In some embodiments, the micro display 1220 may be a silicon-based display, for example, a digital micromirror device (DMD) display, a liquid crystal on silicon (LCOS) display, a light emitting diode (LED) display, and an organic light emitting diode (OLED) display.
[0250] The image processing system of the embodiment of the present application can determine the target grayscale level of the target pixel according to the correspondence between the preset JND value and the grayscale value after determining the JND value of the pixel. Then, based on the target grayscale level, the grayscale value of each pixel channel of the target pixel is quantized to obtain a second image. Since the fewer the grayscale levels of the pixels in the image, the smaller the amount of data required to store the image, and the JND value can characterize the maximum tolerance of the human eye to image distortion. By determining the target grayscale level of the target pixel through the JND value, the amount of image data can be reduced while ensuring the subjective perception of the human eye.
[0251] According to other details of the micro display system of the embodiment of the present application, combined with the above Figures 6 to 10 The image processing method described in the example shown is similar and can achieve its corresponding technical effect. For the sake of brevity, it will not be repeated here.
[0252] Fig.13 A schematic diagram of the hardware structure of an image processing device provided in an embodiment of the present application is shown.
[0253] The image processing device may include a processor 1301 and a memory 1302 storing computer program instructions.
[0254] Specifically, the processor 1301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.
[0255] The memory 1302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 1302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In some instances, the memory 1302 may include a removable or non-removable (or fixed) medium, or the memory 1302 is a non-volatile solid-state memory. In some embodiments, the memory 1302 may be inside or outside the image processing device.
[0256] In some examples, the memory 1302 may be a read-only memory (ROM). In one example, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.
[0257] The memory 1302 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, in general, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to an aspect of the present disclosure.
[0258] The processor 1301 reads and executes the computer program instructions stored in the memory 1302 to implement Figure 6-Figure 10 The method in the illustrated embodiment and the corresponding technical effect are achieved, which will not be described in detail here for the sake of brevity.
[0259] In one example, the image processing device may further include a communication interface 1303 and a bus 1310. Fig.13As shown, the processor 1301, the memory 1302, and the communication interface 1303 are connected via a bus 1310 and communicate with each other.
[0260] The communication interface 1303 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiment of the present invention.
[0261] Bus 1310 includes hardware, software or both, and the components of online data flow billing equipment are coupled to each other. For example, but not limitation, the bus may include Accelerated Graphics Port (AGP) or other graphics bus, Enhanced Industry Standard Architecture (EISA) bus, Front Side Bus (FSB), Hyper Transport (HT) interconnection, Industry Standard Architecture (ISA) bus, InfiniBand interconnection, Low Pin Count (LPC) bus, Memory bus, Micro Channel Architecture (MCA) bus, Peripheral Component Interconnect (PCI) bus, PCI-Express (PCI-X) bus, Serial Advanced Technology Attachment (SATA) bus, Video Electronics Standards Association Local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 1310 may include one or more buses. Although the embodiment of the present invention describes and shows a specific bus, the present invention considers any suitable bus or interconnection.
[0262] The image processing device can execute the image processing method in the embodiment of the present invention, thereby realizing the combination Figures 6 to 11 The image processing method and apparatus are described.
[0263] In addition, in combination with the image processing method in the above embodiment, the embodiment of the present invention can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the image processing methods in the above embodiment is implemented.
[0264] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.
[0265] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), appropriate firmware, plug-in, function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (Radio Frequency, RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.
[0266] It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.
[0267] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device, equipment and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.
[0268] The above is only a specific implementation of the present invention. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be covered within the protection scope of the present invention.
Claims
1. An image processing method, It is characterized in that include: acquiring a first image; Determining visual characteristic information of a target pixel in the first image; Determine the JND value of the target pixel by using the visual characteristic information and a preset just noticeable difference (JND) model; Determine a target grayscale level corresponding to the JND value of the target pixel by using a preset correspondence relationship between the JND value and the grayscale level, wherein the target grayscale level of the target pixel is less than or equal to an original grayscale level of the target pixel in the first image; Based on the target grayscale level, grayscale values of each pixel channel of the target pixel are quantized by grayscale reduction to obtain a second image; Quantifying the grayscale value of each pixel channel of the target pixel to obtain the second image specifically includes: Determining a plurality of neighboring pixels of the target pixel; Determine the grayscale difference of each of the plurality of adjacent pixels, wherein the grayscale difference of each adjacent pixel refers to the difference between the grayscale value of each channel of each adjacent pixel before quantization and the grayscale value after quantization; Obtaining a weighted grayscale difference value of each adjacent pixel according to the product of the grayscale difference value of each adjacent pixel and the weight value of each adjacent pixel; The weighted grayscale difference values of each of the plurality of adjacent pixels and the sum of the grayscale values of each channel of the target pixel before quantization are updated to the grayscale value of each pixel channel of the target pixel to obtain a second image.
2. The method according to claim 1, It is characterized in that The visual characteristic information includes background brightness and eccentricity angle; The JND model is related to the background brightness and the eccentricity angle.
3. The method according to claim 2, It is characterized in that The JND model is expressed as: Among them, FD-JND L is the JND model related to the background brightness and the eccentricity angle, L 1 is a first related parameter, which is related to the background brightness of the target pixel, E 1 is a second related parameter, which is related to the eccentricity angle of the target pixel, D 1 is a third related parameter, and the third related parameter is related to the depth value of the target pixel, E 1 (0) is the reference background brightness value, D 1 (1.0) is the base depth value, α 1 is the first correction parameter, α 2 is the second calibration parameter.
4. The method according to claim 1, It is characterized in that The visual characteristic information includes background brightness, eccentricity angle, depth and contrast value; The JND model is related to the background brightness, the eccentricity angle, the depth, and the contrast value.
5. The method according to claim 4, It is characterized in that The JND model is expressed as: in, Among them, FD-JND C =FD-JND L +CM Among them, L 1 is a first related parameter, which is related to the background brightness of the target pixel, E 1 is a second related parameter, which is related to the eccentricity angle of the target pixel, D 1 is a third related parameter, and the third related parameter is related to the depth value of the target pixel, C(i,j) is the contrast of the target pixel, and E 1 (0) is the reference background brightness value, D 1 (1.0) is the base depth value, α 1 is the first correction parameter, α 2 is the second correction parameter, CM is the contrast mask, μ 1 is the first preset parameter, μ 2 is the second preset parameter, μ 3 is the third preset parameter, μ 4 is the fourth preset parameter, v 1 is the fifth preset parameter, ν 2 is the sixth preset parameter, ν 3 is the seventh preset parameter, ν 4 is the eighth preset parameter, ω 1 is the ninth preset parameter, ω 2 is the tenth preset parameter, ω 3 is the eleventh preset parameter, ω 4 It is the twelfth preset parameter, FD-JND L is the JND model related to the background brightness and the eccentricity angle, FD-JND C is the JND model related to the background brightness, the eccentricity angle, and the contrast masking, L bg (i, j) is the background brightness of the target pixel, and E(i, j) is the eccentricity angle of the target pixel.
6. The method according to claim 1, It is characterized in that The target pixel includes a pixel within the slider; The determining of visual characteristic information of a target pixel in the first image specifically includes: In the process of traversing the first image using the slider, determining visual characteristic information of reference pixels in the slider; The visual characteristic information of the reference pixel is determined as the visual characteristic information of the target pixel.
7. The method according to claim 1, It is characterized in that The number of pixel channels of the target pixel is multiple; The determining the JND value of the target pixel by using the visual characteristic information and a preset just noticeable difference (JND) model specifically includes: Determine a first JND value of each of a plurality of pixel channels of the target pixel by using the visual characteristic information and a preset JND model; Determining weight values of each of the plurality of pixel channels; Calculate a weighted JND value using the first JND values of each of the plurality of pixel channels and the first JND values of each of the plurality of pixel channels of the target pixel; The weighted JND value is determined as the JND value of the target pixel.
8. An image processing device, It is characterized in that The device comprises: An image acquisition module, used for acquiring a first image; An information determination module, used to determine visual characteristic information of a target pixel in the first image; A JND value determination module, used to determine the JND value of the target pixel by using the visual characteristic information and a preset just noticeable difference (JND) model; a grayscale determination module, configured to determine a target grayscale corresponding to the JND value of the target pixel by using a preset correspondence relationship between the JND value and the grayscale, wherein the target grayscale of the target pixel is less than or equal to an original grayscale of the target pixel in the first image; An image determination module, configured to quantify the grayscale value of each pixel channel of the target pixel based on the target grayscale level to obtain a second image; The image determination module specifically includes: a seventh determining unit, configured to determine a plurality of adjacent pixels of the target pixel; an eighth determining unit, configured to determine the grayscale difference of each of the plurality of adjacent pixels, wherein the grayscale difference of each adjacent pixel refers to the difference between the grayscale value of each channel of each adjacent pixel before quantization and the grayscale value after quantization; A ninth determining unit, configured to obtain a weighted grayscale difference value of each adjacent pixel according to a product of the grayscale difference value of each adjacent pixel and a weight value of each adjacent pixel; The grayscale value updating unit is used to update the weighted grayscale difference values of each of the plurality of adjacent pixels and the sum of the grayscale values of each channel of the target pixel before quantization to the grayscale value of each pixel channel of the target pixel to obtain a second image.
9. A micro display system, the micro display system include: A micro display controller, configured to execute the image processing method according to any one of claims 1 to 7, generate a control signal of a target image, and output the control signal to a micro display; The micro display is used to display the target image based on the control signal.
10. An image processing device, It is characterized in that The device comprises: a processor and a memory storing computer program instructions; When the processor executes the computer program instructions, the image processing method according to any one of claims 1 to 7 is implemented.
11. A computer storage medium, It is characterized in that The computer storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the image processing method according to any one of claims 1 to 7 is implemented.
