Data association analysis method for whole machine performance test of AR (Augmented Reality) glasses

Abstract

The invention belongs to the technical field of optical equipment detection, and particularly relates to a data association analysis method for testing the whole machine performance of AR glasses, and the method comprises the steps: S1, extracting geometric position features and luminosity response features from left and right eye images; s2, calculating a luminosity competition load index by using a Weibull perception model in a logarithm form; s3, constructing an anisotropic parallax index by using the anisotropic elliptic field; and S4, based on the anisotropic parallax index, calculating a visual fusion comfort score by using an S-type scoring function, and grading the overall performance of the AR glasses according to a scoring result. According to the method, a deep coupling mechanism of multi-dimensional features is established, a dynamic modulation relation between luminosity data and geometric thresholds is established through an exponential attenuation coupling mechanism, the problem that a user is still dizzy due to the fact that geometric positions are aligned but binocular brightness or definition is uneven is solved, and the user experience is improved. And it is ensured that each piece of leaving-factory AR glasses can provide comfortable visual fusion experience.

Description

Technical Field

[0001] This invention belongs to the technical field of optical equipment testing, specifically relating to a data correlation analysis method for overall performance testing of AR glasses. Background Technology

[0002] Augmented reality (AR) glasses project virtual images onto the user's left and right eyes through two independent optical display modules, which then fuse in the cerebral cortex to form a three-dimensional virtual image. The consistency of the binocular image fusion is a key indicator of user comfort. If the images from the left and right eyes deviate too much in spatial position, or differ significantly in perceptual dimensions such as brightness and clarity, the user's brain will be unable to complete the image fusion, leading to severe dizziness, double vision, or even nausea and other adverse physiological reactions.

[0003] Current binocular fusion detection technology has some limitations in practical applications: Existing technologies typically calculate the linear difference in brightness between the left and right eyes and set a fixed threshold as the pass / fail standard, ignoring the non-linear nature of the human visual system's perception of brightness, which follows the Weber-Fechner law. This means that even small brightness differences in low-light scenes can cause strong discomfort, but are easily overlooked during detection. Furthermore, existing technologies usually use isotropic circular regions as the boundary for geometric position determination, failing to consider the physiological structural characteristics of the human eye—extremely sensitive to vertical parallax but with a relatively strong ability to adjust horizontal parallax—which can easily lead to missed detections of vertical deviations.

[0004] Furthermore, existing detection processes often treat photometric consistency and geometrical positional consistency as two independent parallel steps, lacking a multi-dimensional coupled judgment mechanism. In reality, visual perception is a holistic process. When there is photometric competition between the left and right eyes, the brain significantly reduces its ability to fuse geometrical disparities to suppress interference from low-quality images. In other words, the greater the photometric difference, the more precise the geometrical alignment must be. Current technology lacks a mechanism to dynamically adjust geometrical tolerances based on photometric differences, making it difficult to accurately assess the user's actual visual experience, potentially leading to user discomfort with shipped products. Summary of the Invention

[0005] This invention provides a data correlation analysis method for AR glasses overall performance testing, in order to solve the technical problems in the prior art that ignore the nonlinear perception characteristics of the human eye, lack anisotropic parallax determination, and separate photometric and geometric determination.

[0006] To solve the above problems, the present invention adopts the following technical solution: A data correlation analysis method for AR glasses overall performance testing includes the following steps: S1, control the binocular camera to acquire left and right eye images displayed on the AR glasses, and extract geometric position features and photometric response features from the left and right eye images; S2, Based on the photometric response characteristics, the photometric competition load index is calculated using a logarithmic form of the Weber perceptual model. The photometric competition load index is used to assess the physiological burden on the human eye when processing the difference in photometric intensity between the left and right eyes. S3, Based on the geometric location features and the photometric competition load index, an anisotropic parallax index is constructed using an anisotropic elliptic field, wherein the tolerance threshold for geometric determination decreases exponentially with the increase of the photometric competition load index. S4. Based on the anisotropic parallax index, the visual fusion comfort score is calculated using the S-shaped scoring function, and the overall performance of the AR glasses is graded according to the score results.

[0007] Further, geometric position features are extracted from the left and right eye images, including: The left and right eye images are preprocessed, and the pixel coordinates of the central target are extracted using a subpixel-level centroid algorithm and converted into angular coordinates as geometric position features.

[0008] Further, photometric response features are extracted from the left and right eye images, including: The average brightness value is calculated within the effective display area of ​​the left and right eye images. The modulation transfer function is calculated using the tilted edge method, and its MTF value at a preset spatial frequency is taken as the sharpness index. The average brightness value and the extracted MTF value are used as the photometric response features.

[0009] Furthermore, assuming a spatial frequency of 20 cycles / degree, the MTF50 value corresponding to the modulation transfer function at 20 cycles / degree is used as the sharpness index.

[0010] Furthermore, the photometric competition load index satisfies the following relationship: ; In the formula, The photometric competition load index. It is the natural logarithm function. and These are the average brightness values ​​for the left and right eyes, respectively. It is the arithmetic mean of the average brightness values ​​of the left and right eyes. This is the dark vision compensation constant. For clarity weighting coefficients, and These are the acuity indicators for the left and right eyes, respectively. The arithmetic mean of the acuity index values ​​for both eyes. This is the fuzzy compensation constant.

[0011] Furthermore, the anisotropic parallax index satisfies the following relationship: ; In the formula, The anisotropic parallax index, and These are the horizontal coordinates of the center target in the left and right eye images, respectively. and These are the vertical coordinates of the center target in the left and right eye images, respectively. This is the horizontal baseline tolerance threshold. The vertical reference tolerance threshold, It is a natural exponential function. and These are the coupling sensitivity coefficients in the horizontal and vertical directions, respectively. This is the photometric competition load index.

[0012] Furthermore, the vertical reference tolerance threshold is smaller than the horizontal reference tolerance threshold.

[0013] Furthermore, the coupling sensitivity coefficient in the vertical direction is greater than that in the horizontal direction.

[0014] Furthermore, the visual fusion comfort score satisfies the following relationship: ; In the formula, Score the visual integration comfort level. The anisotropic parallax index, This is the steepness factor of the judgment.

[0015] Furthermore, the binocular camera consists of left and right human-eye-like camera modules of the testing equipment.

[0016] The beneficial effects are: 1. This invention establishes a deep coupling mechanism for multi-dimensional features. Through an exponential decay coupling mechanism, it establishes a dynamic modulation relationship between photometric data and geometric thresholds. This mechanism solves the problem of user dizziness caused by uneven brightness or clarity of the binoculars even when the geometric position is aligned, ensuring that every AR glasses manufactured can provide a comfortable visual fusion experience.

[0017] 2. Through sub-pixel-level algorithms and tilted edge methods, geometric and photometric data can be accurately extracted, eliminating image noise interference and providing a high-quality input source for subsequent calculations.

[0018] 3. By introducing the log-Weber model, this invention simulates the nonlinear perception characteristics of human eyes for brightness in industrial automated inspection, avoiding the oversensitivity and misjudgment of small brightness differences by traditional linear algorithms in high-brightness environments, while capturing the differences that human eyes are truly sensitive to in low-brightness environments.

[0019] 4. By constructing a dynamic elliptical judgment field, it strictly follows the physiological principle that the human eye has high horizontal tolerance and low vertical tolerance. Compared with the coarse circular judgment area of ​​existing technologies, it avoids the risk of missed detection due to diplopia caused by slight deviations in the vertical direction, and also avoids misjudging good products due to excessive stringency in the horizontal direction.

[0020] 5. By converting the binary judgment result into a continuous numerical score, it can more accurately identify products in the critical risk zone compared to the traditional hard threshold. Attached Figure Description

[0021] Figure 1 A flowchart illustrating the data correlation analysis method used for overall performance testing of AR glasses; Figure 2 A schematic diagram of the dynamic determination field for anisotropy of binocular image fusion; Figure 3 This is a comparison chart of the detection performance of existing technologies and the present invention. Detailed Implementation

[0022] An embodiment of the data correlation analysis method for AR glasses overall performance testing provided by this invention: like Figure 1 As shown, a data correlation analysis method for AR glasses overall performance testing includes the following steps: S1, control the binocular camera to acquire left and right eye images displayed on the AR glasses, and extract geometric position features and photometric response features from the left and right eye images.

[0023] Specifically, the binocular camera here consists of left and right human-eye camera modules of the testing equipment. In operation, firstly, the left and right human-eye camera modules of the testing equipment are moved along a linear motor guide rail to adjust the entrance pupil distance to the standard interpupillary distance (IPD), for example, 64mm, ensuring that the camera optical axis is aligned with the design visual axis of the AR glasses. Then, the AR glasses are driven to display a dedicated comprehensive test chart, which includes a central crosshair target, a checkerboard pattern distributed around it, and color blocks of different gray levels. The crosshair target is used for geometric positioning, and the color blocks of different gray levels are used for photometric measurements. Afterward, the binocular cameras are triggered to simultaneously expose, acquiring the left-eye and right-eye images respectively.

[0024] When extracting geometric position features, the left and right eye images are first preprocessed by grayscale conversion and Gaussian filtering to remove noise. Next, a sub-pixel centroid algorithm is used to extract the pixel coordinates of the central crosshair target in the left and right eye images. Finally, based on the camera's focal length and pixel size parameters, the obtained pixel coordinates are converted into angular coordinates for standardized calculations conforming to the human eye's field of view.

[0025] When extracting photometric response features, the average luminance value is calculated within the effective display area of ​​the left-eye and right-eye images, respectively. The unit of the average luminance value is nit. For example, the effective display area is the center of the field of view. Range. At the same time, using the black and white edge regions in the image, the modulation transfer function (MTF) of the left and right eyes is calculated separately using the tilted edge method. The MTF value at a preset spatial frequency is taken as the sharpness index. For example, the MTF50 value at a spatial frequency of 20 cycles / degree is taken as the sharpness index. The average brightness value and the sharpness index constitute the luminous response characteristics.

[0026] Step S1 can extract the basic feature data reflecting the optical performance of AR glasses, eliminate the interference of environmental noise and system errors, provide accurate and reliable input basis for subsequent multidimensional data correlation analysis, and ensure the authenticity of the final evaluation results.

[0027] S2, based on the photometric response characteristics, calculate the photometric competition load index using a logarithmic form of the Weber perceptual model. The photometric competition load index is used to assess the physiological burden on the human eye when processing the difference in photometric intensity between the left and right eyes.

[0028] Specifically, the photometric competition load index satisfies the following relationship:

[0029] In the formula, The photometric competition load index is a dimensionless value. , , , These are the average brightness values ​​for the left and right eyes, respectively. , These are the clarity indicators for the left and right eyes, respectively. This is the dark vision compensation constant, which, for example, can be 0.1 nit, is used to simulate the dark noise level of the retina in a completely dark environment and to prevent the denominator from being zero. This is the fuzzy compensation constant; for example, it can be 0.01. This is the sharpness weighting coefficient, which assigns a higher weight to the sharpness difference item; for example, it can be 1.5.

[0030] In one example, suppose the average brightness of the left eye is measured to be... nit, the average brightness value of the right eye is nit, then The clarity index for the left eye is nit. The clarity index for the left eye is ,but .

[0031] First, calculate the brightness difference term: Then calculate the natural logarithm of this term: .

[0032] Next, calculate the sharpness difference: Squaring it yields... Then multiply by the clarity weighting factor to calculate the result. .

[0033] Finally, the total load is calculated as follows: .

[0034] By employing a logarithmic Weber perception model, the physiological stress experienced by the human eye when faced with inconsistent brightness in binocular images can be assessed. This solves the problem of poor adaptability of traditional linear judgment under different brightness backgrounds, making the assessment results more consistent with the real visual experience of the human eye.

[0035] S3, based on the geometric location features and the photometric competition load index, an anisotropic parallax index is constructed using an anisotropic elliptic field, wherein the tolerance threshold for geometric determination decreases exponentially with the increase of the photometric competition load index.

[0036] Specifically, the anisotropic parallax index satisfies the following relationship:

[0037] In the formula, The anisotropic parallax index, and These are the horizontal coordinates of the center target in the left and right eye images, respectively. and These are the vertical coordinates of the center target in the left and right eye images, respectively. This is the horizontal reference tolerance threshold; for example, the horizontal reference tolerance threshold is 1.0°. The vertical reference tolerance threshold is, for example, 0.3°. It is a natural exponential function; and These are the coupling sensitivity coefficients in the horizontal and vertical directions, respectively. For example, =0.5、 =0.8.

[0038] Following the example in S1, the photometric competition load index is known to be... Assuming the horizontal parallax is measured Vertical parallax .

[0039] First, calculate the horizontal dynamic threshold: ,in, This is the horizontal dynamic threshold.

[0040] Next, calculate the vertical dynamic threshold: ,in, This is the vertical dynamic threshold.

[0041] Substitute the formula for the anisotropic parallax index to calculate the anisotropic parallax index: Horizontal component square: .

[0042] Square of vertical components: .

[0043] .

[0044] because Even after considering the photometric load tightening threshold, it is still judged as qualified. It can be seen that the existence of photometric load compresses the original tolerance threshold. For example, the horizontal reference tolerance threshold of 1.0 is compressed to 0.806, and the vertical reference tolerance threshold of 0.3 is compressed to 0.212, realizing the dynamic tightening of the standard; if the photometric difference is greater, the threshold will be further reduced.

[0045] By introducing anisotropic parallax index and exponential decay coupling mechanism, not only is the physiological characteristic that the human eye is more sensitive to vertical parallax reflected, preventing missed detections in the vertical direction, but also an adaptive judgment logic is realized that the worse the photometric quality, the higher the geometric alignment requirement, thereby intercepting products that may appear to meet the geometric position but cause dizziness due to photometric differences.

[0046] S4. Based on the anisotropic parallax index, the visual fusion comfort score is calculated using the S-shaped scoring function, and the overall performance of the AR glasses is graded according to the score results.

[0047] Specifically, in this embodiment, to provide the production line with more refined quality management methods, the binary judgment result is transformed into a continuous numerical score. The visual fusion comfort score satisfies the following relationship: ; In the formula, The decision steepness factor is, for example, set to 5.0, to simulate the critical abrupt change in the human eye from comfort to dizziness.

[0048] Following the example in S3, it is known that... .

[0049] First, calculate the internal structure of the exponential term: .

[0050] Next, calculate the denominator: .

[0051] Final score calculation: .

[0052] A score of 87.7 is close to 88, indicating good visual comfort, which can be classified as excellent, such as Grade A.

[0053] In other embodiments, if the anisotropic parallax index is slightly greater than 1, for example 1.1, then The exponent term is With a denominator of 3.85, the score dropped to approximately 26 points, reflecting the severe punishment for products that exceeded the standards.

[0054] In an optional embodiment, refer to Figure 2 , Figure 2 A schematic diagram of the dynamic determination field for binocular image fusion anisotropy is shown. In the diagram, the horizontal axis represents horizontal geometric parallax, and the vertical axis represents vertical geometric parallax. The gray dashed circle represents the fixed threshold determination boundary of the prior art, which is circular and does not distinguish between horizontal and vertical differences. The colored solid ellipse represents the dynamic determination boundary of this invention; the boundary is a flattened ellipse, visually demonstrating that the anisotropy characteristic with a much smaller tolerance in the vertical direction than in the horizontal direction. Figure 2 The image shows three ellipses of different colors (green, orange, and red), corresponding to low, medium, and high photometric competition loads, respectively. As the photometric load increases, the area of ​​the ellipses shrinks significantly, intuitively revealing the logic that the worse the photometric consistency, the tighter the geometric alignment standard automatically becomes.

[0055] In one embodiment, refer to Figure 3 , Figure 3 A comparison chart of the detection performance of existing technologies and this invention is presented. The horizontal axis represents the photometric competition load index, and the vertical axis represents the comprehensive amplitude of geometric parallax. Green dots represent high-quality products, located below the judgment curve of this invention. Gray dots represent obviously defective products, located above the judgment line of the existing technology. Red crosses represent latent defects; these dots are located below the gray dashed line but above the blue solid line of this invention. This means that although these products may appear to meet the geometric error standards, the excessive photometric differences can actually cause dizziness in users.

[0056] The soft edge detection mechanism, built using the S-shaped scoring function, transforms complex detection data into an intuitive percentage score. This not only sensitively identifies products in critical risk zones but also provides quantifiable quality grading criteria for the production line, helping manufacturers optimize yield control strategies and improve overall shipment quality.

[0057] In addition, in the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

Claims

1. A data correlation analysis method for AR glasses system performance testing, characterized in that, The method comprises the following steps: S1, controlling a binocular camera to collect left and right eye images displayed by AR glasses, and extracting geometric position features and luminosity response features from the left and right eye images; S2, calculating a luminosity competition load index based on the luminosity response features and using a logarithmic form of a Weber perception model, the luminosity competition load index being used to evaluate a physiological burden of human eyes when processing luminosity differences between the left and right eyes; S3, constructing an anisotropic disparity index using an anisotropic ellipse field according to the geometric position features and the luminosity competition load index, wherein a tolerance threshold of geometric determination exponentially decays with an increase of the luminosity competition load index; S4, calculating a visual fusion comfort score using an S-shaped scoring function based on the anisotropic disparity index, and grading overall performance of the AR glasses according to the scoring result.

2. The data correlation analysis method for AR glasses system performance test according to claim 1, characterized in that, The geometric position features are extracted from the left and right eye images, comprising: performing preprocessing on the left and right eye images, extracting pixel coordinates of a central target using a sub-pixel level barycenter algorithm, and converting the pixel coordinates into angle coordinates as the geometric position features.

3. The data correlation analysis method for AR glasses system performance testing according to claim 2, characterized in that, The luminosity response features are extracted from the left and right eye images, comprising: calculating an average luminance value in an effective display area of the left and right eye images, calculating a modulation transfer function using a tilted edge method, and taking an MTF value of the modulation transfer function at a preset spatial frequency as a definition index, and taking the average luminance value and the extracted MTF value as the luminosity response features.

4. The data correlation analysis method for AR glasses system performance testing according to claim 3, characterized in that, The preset spatial frequency is 20 cycle / degree, and an MTF50 value corresponding to the modulation transfer function at 20 cycle / degree is taken as the definition index.

5. The data correlation analysis method for AR glasses system performance test according to claim 3, characterized in that, The luminosity competition load index satisfies the following relationship: ； wherein is the photometric competition load index, is the natural logarithm function, and are the average luminance values for the left and right eyes, respectively, is the arithmetic mean of the average luminance values for the left and right eyes, is the dark vision compensation constant, is the sharpness weight coefficient, and are the sharpness indicators for the left and right eyes, respectively, is the arithmetic mean of the sharpness indicator values for the left and right eyes, is the blur compensation constant.

6. The data correlation analysis method for AR glasses system performance testing according to claim 5, characterized in that, The anisotropic disparity index satisfies the following relationship: ； wherein is the anisotropic disparity index, and are the coordinates of the center target of the left-eye image and the right-eye image in the horizontal direction, respectively, and are the coordinates of the center target of the left-eye image and the right-eye image in the vertical direction, respectively, is the horizontal reference tolerance threshold, is the vertical reference tolerance threshold, is the natural exponential function, and are the coupling sensitivity coefficients in the horizontal and vertical directions, respectively, is the photometric competition load index.

7. The data association analysis method for AR glasses system performance test according to claim 6, characterized in that, The vertical reference tolerance threshold is less than the horizontal reference tolerance threshold.

8. The data association analysis method for AR glasses system performance test according to claim 7, characterized in that, The coupling sensitivity coefficient in the vertical direction is greater than the coupling sensitivity coefficient in the horizontal direction.

9. The data association analysis method for AR glasses system performance testing according to any one of claims 1-8, characterized in that, The visual fusion comfort score satisfies the following relationship: ； In the formula, is a visual fusion comfort score, is an anisotropic disparity index, is a decision steepness factor.

10. The data association analysis method for AR glasses system performance testing according to any one of claims 1-8, characterized in that, The binocular camera is composed of left and right artificial eye camera modules of a test device.

Images