Interaction terminal of MR virtual reality scene secondary development system

Abstract

The invention discloses an MR virtual reality scene secondary development system interaction terminal, which comprises the steps of recording and shooting a MR virtual reality scene secondary development test process through a virtual camera in a test period to obtain a test image, analyzing the test image and marking an image blurring time period and an image stagnation time period in the test period, carrying out coincidence analysis and data processing on the test image and a simulation static time period and a simulation motion time period of the virtual camera, judging whether the normal use of a user is influenced by the blurring and lagging of the test image or not, if so, generating a problem region analysis signal, and equally dividing a test period into a plurality of region analysis time periods; and analyzing and processing the region analysis time period in combination with the image similarity value to obtain an analysis abnormal value, marking a problem region group in the MR virtual reality scene, analyzing the problem region group to obtain a modification priority value, sorting the modification priority value, generating a repair and optimization report, and performing repair and optimization on the secondary development of the MR virtual reality scene according to the repair and optimization report.

Description

Technical Field

[0001] The present invention relates to the technical field of virtual reality development, and particularly to an interactive terminal for secondary development of an MR virtual reality scene. Background Art

[0002] With the rapid development of technology, MR (Mixed Reality) technology has gradually penetrated into many fields, such as immersive education, virtual design review, remote collaboration, etc., bringing people an unprecedented interactive experience. However, in the process of secondary development of MR virtual reality scenes, many problems have emerged one after another, among which the problems of image quality and smoothness are particularly prominent, seriously hindering its wide application and further development.

[0003] Currently, the existing technologies for such problems have obvious shortcomings. On the one hand, traditional image quality evaluation methods are mostly based on simple comparisons at the pixel level. Indicators such as peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) can only reflect the differences in static images and are completely unable to adapt to the dynamic changes brought about by frequent user interactions and rapid perspective switching in the MR scene. Such problems are difficult to accurately capture with traditional indicators. On the other hand, performance monitoring software focuses on hardware parameters, such as CPU and GPU usage rates. Although it can detect system load, it cannot accurately correlate to the specific image lag, blur positions, and time periods from the user's perspective. It is difficult for developers to locate the root cause of the problem based on this. Moreover, manual testing is highly subjective and inefficient. Different testers have different sensitivities and descriptions of problems, and cannot cover all complex scenarios, easily missing key hidden dangers.

[0004] In view of the above problems, the present invention proposes an interactive terminal for secondary development of an MR virtual reality scene. Summary of the Invention

[0005] The purpose of the present invention is to provide an interactive terminal for secondary development of an MR virtual reality scene to solve at least one of the above-mentioned existing technical problems.

[0006] The present invention provides an interactive terminal for secondary development of an MR virtual reality scene, including the following modules:

[0007] Test analysis module: During the test period, record and shoot the secondary development test process of the MR virtual reality scene through a virtual camera to obtain test images, analyze the test images to obtain a group of blurred comparison images and a group of stagnant comparison images and their image similarity values, and mark the image blur time period and image stagnation time period within the test period according to the image similarity values;

[0008] Time period analysis module: Based on the marked image blurring time period and image stagnation time period, respectively overlap and analyze them with the simulated static time period and simulated motion time period of the virtual camera for data processing, obtain the abnormal blurring time period, abnormal stagnation time period and the problem degree value within the test cycle, and judge whether the blurring and freezing of the test image affect the normal use of users based on the problem degree value. If so, generate a problem area analysis signal;

[0009] Problem area analysis module: Based on the generated problem area analysis signal, evenly divide the test cycle into several area analysis time periods, and combine the image similarity value to analyze and process the area analysis time periods, abnormal blurring time periods and abnormal stagnation time periods to obtain the analysis abnormal value of the area analysis time period, and mark the problem area group in the MR virtual reality scene based on the analysis abnormal value;

[0010] Problem area sorting module: Based on the marked problem area group, analyze the problem area group to obtain the modification priority value, sort the problem area group according to the modification priority value, generate a repair and optimization report based on the sorting, and repair and optimize the MR virtual reality scene according to the repair and optimization report.

[0011] Advantages of the present invention:

[0012] 1. The present invention can significantly improve the test efficiency and accuracy of the secondary development of the MR virtual reality scene. This method uses a virtual camera to record the test process, and by analyzing the image pixel values, accurately identifies the blurring and stagnation time periods, effectively avoiding the errors of manual judgment. At the same time, the identified time periods are overlapped and analyzed with the motion state of the virtual camera to further determine the abnormal time periods and problem degree, providing an intuitive optimization basis for developers. In addition, this method can automatically generate a problem area analysis signal to timely remind developers to pay attention to and solve potential problems, thus ensuring the user experience and smoothness of the MR scene.

[0013] 2. The present invention accurately locates and deeply analyzes the problem areas of the MR virtual reality scene. By dividing the test cycle into multiple area analysis time periods and combining the image similarity value, it can accurately calculate the analysis abnormal value of each time period, thus effectively marking the problem areas. Further, sorting the problem areas according to the modification priority value can guide developers to give priority to dealing with key problems, ensuring the efficiency and pertinence of repair and optimization. The finally generated repair and optimization report provides developers with detailed repair suggestions and steps, helping to quickly improve the user experience of the MR virtual reality scene and enhance the overall quality. Description of the drawings

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0015] Figure 1 It is a schematic structural diagram of an interactive terminal of an MR virtual reality scene secondary development system provided by an embodiment of the present invention;

[0016] Figure 2 It is a specific flowchart for obtaining the image similarity value in the interactive terminal of the MR virtual reality scene secondary development system provided in Embodiment 1 of the present invention;

[0017] Figure 3 It is a specific flowchart for obtaining the analysis of outliers in the interactive terminal of the MR virtual reality scene secondary development system provided in Embodiment 2 of the present invention. Specific Embodiments

[0018] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] Embodiment 1

[0020] As Figure 1 shown, the interactive terminal of the MR virtual reality scene secondary development system provided by the embodiment of the present invention specifically includes the following modules:

[0021] Test analysis module: During the test period, record and shoot the MR virtual reality scene secondary development test process through a virtual camera to obtain test images, analyze the test images to obtain a fuzzy comparison image group and a stagnant comparison image group, perform data processing on the image pixel values of the test images to obtain the image similarity values of the fuzzy comparison image group and the stagnant comparison image group, and mark the image fuzzy period and the image stagnant period during the test period according to the image similarity values;

[0022] In some embodiments, a number of monitoring time points are evenly set within a test cycle, where the test cycle represents the time required to test all the modified and added content requirements functions in the secondary development. The time intervals between adjacent monitoring time points are the same. In the secondary development environment of the MR virtual reality scene, a virtual camera associated with the virtual reality device is configured. The virtual camera captures and shoots the virtual reality scene images from the user's perspective at the monitoring time points by simulating the tracking system. The tracking system is a system in the virtual reality device that accurately tracks the head and hand movements of the user through built-in sensors;

[0023] Use MATLAB software to obtain the pixel values of each pixel point in the test image. Specifically, use the imread function to read the test image file, use the size function to obtain the size of the image. The size includes the number of rows, columns, and channels of the image, and then obtain the pixel values of each channel of a specific pixel point by specifying the indexes of the row and column;

[0024] Based on all the modified and added content during the secondary development of the MR virtual reality scene, test its required functions, and analyze the test images taken during the test cycle after the test ends;

[0025] Obtain any test image within the test cycle, and perform Gaussian blur processing on the test image through the Gaussian blur algorithm to obtain the Gaussian blur image of the test image;

[0026] Specifically, select the size of the Gaussian kernel (also known as the Gaussian filter). The size of the Gaussian kernel determines the spatial range of the blur effect, usually odd × odd (such as 3×3, 5×5, 7×7, etc.) so that the central pixel point can be clearly determined; generate a Gaussian kernel matrix. With the central pixel point as the origin, the values of other points are calculated by the simplified two-dimensional Gaussian function formula where x and y respectively represent the number of rows and columns in the spatial range of the pixel point's blur effect, and σ represents the standard deviation of the Gaussian kernel; apply the constructed Gaussian kernel matrix to the image. The specific steps are as follows: cover the Gaussian kernel matrix on a certain pixel point of the image. Based on any channel of the pixel point, perform a weighted average calculation on all pixel values of the same channel in the covered area. The weights of the weighted average are determined by the element values in the Gaussian kernel matrix, and then the obtained value is used as the new value of this pixel point on this channel. Repeat the operation for each pixel point in the image until the entire image is processed to obtain the Gaussian blur image of the test image;

[0027] Combine the test image and the corresponding Gaussian blur image into a blur comparison image group;

[0028] Combine any two adjacent test images in time series within the test cycle into a stagnation comparison image group;

[0029] As shown in Figure 2 the following, the specific steps for obtaining the image similarity value are as follows:

[0030] Based on any group of fuzzy comparison images or stagnant comparison images, compare the similarity between the two images in the group of fuzzy comparison images or stagnant comparison images;

[0031] Specifically, based on the pixel points with the same number of rows and columns in the two images, obtain the pixel values of each channel of the two pixel points, perform a difference process on the two pixel values with the same channel and take the absolute value to obtain the single-channel pixel absolute difference, and perform a summation and averaging process on the single-channel pixel absolute differences of all channels in the two pixel points to obtain the pixel absolute difference between the two pixel points;

[0032] Compare the obtained pixel absolute difference with the pixel absolute difference threshold;

[0033] If the pixel absolute difference between the two pixel points is greater than or equal to the pixel absolute difference threshold, it indicates that the two pixel points are not similar, and the two pixel points are grouped into a group of non-similar pixels;

[0034] If the pixel absolute difference between the two pixel points is less than the pixel absolute difference threshold, it indicates that the two pixel points are similar, and the two pixel points are grouped into a group of similar pixels;

[0035] Compare all pixel points in the two images, obtain the number of groups of similar pixels in the two images, and perform a ratio process with the sum of the number of groups of similar pixels and non-similar pixels in the two images to obtain the similarity number ratio, denoted as XG;

[0036] Obtain the sum of the pixel absolute differences of all groups of similar pixels and non-similar pixels in the two images and perform an averaging process to obtain the average pixel absolute difference of the two images, denoted as XJ;

[0037] Perform data processing on the obtained similarity number ratio XG and the average pixel absolute difference XJ, and calculate the image similarity value TX through the formula where a1 and a2 are both preset proportional coefficients, a1 = 0.214, a2 = 2.913;

[0038] Compare the obtained image similarity value with the image similarity threshold;

[0039] If the image similarity value of the two images is less than or equal to the image similarity threshold, it indicates that the similarity between the two images is low;

[0040] If the image similarity value of the two images is greater than the image similarity threshold, it indicates that the similarity between the two images is high, and an image similarity signal is generated;

[0041] Based on the fuzzy comparison image groups, all the fuzzy comparison image groups within the test period are compared and analyzed. The time periods when the fuzzy comparison image groups continuously generate image similarity signals are marked as one image blurring period, and several image blurring periods are obtained.

[0042] Based on any stagnant comparison image group, if an image similarity signal is generated, the two images within the stagnant comparison image group are classified into a similar image group. The above-mentioned processing is performed on all the stagnant comparison image groups within the test period, and several similar image groups are obtained.

[0043] Based on any similar image group, the shooting time period of the test image in the similar image group is marked as the image stagnation period.

[0044] Time period analysis module: Based on the marked image blurring periods and image stagnation periods, they are respectively subjected to coincidence analysis and data processing with the simulated stationary period and simulated motion period of the virtual camera to obtain the abnormal blurring period, abnormal stagnation period, and the problem degree value within the test period. Based on the problem degree value, it is judged whether the blurring and freezing of the test images affect the normal use of users. If so, a problem area analysis signal is generated.

[0045] In some embodiments, within the test period, the virtual camera has and only has a simulated motion state and a simulated stationary state. The time period when the virtual camera is in the simulated motion state within the test period is marked as the simulated motion period, and the time period when the virtual camera is in the simulated stationary state within the test period is marked as the simulated stationary period.

[0046] Obtain the simulated stationary period of the virtual camera within the test period, compare it with the image blurring periods within the test period, obtain the overlapping period between the simulated stationary period and the image blurring periods, mark it as the abnormal blurring period, and perform a ratio process on the duration of the abnormal blurring period and the total duration of the test period to obtain the abnormal blurring ratio, marked as YM.

[0047] Obtain the simulated motion period of the virtual camera within the test period, compare it with the image stagnation periods within the test period, obtain the overlapping period between the simulated motion period and the image stagnation periods, mark it as the abnormal stagnation period, and perform a ratio process on the duration of the abnormal stagnation period and the total duration of the test period to obtain the abnormal stagnation ratio, marked as YT.

[0048] It should be noted that when the virtual camera is in the simulated stationary state but the captured test images are blurred, it indicates that the blurring of the test images at this time has nothing to do with the simulated motion, and there is an abnormal image blurring situation for the test images. Similarly, when the virtual camera is in the simulated motion state but the captured test images show continuous stagnation, it indicates that there is an image freezing situation for the test images.

[0049] Data processing is performed on the obtained abnormal blur ratio YM and abnormal stagnation ratio YT, and the problem degree value YC within the test period is calculated through the formula YC = b1 * YM + b2 * YT, where both b1 and b2 are preset proportionality coefficients, b1 = 3.162, and b2 = 4.118;

[0050] Exemplarily, the period when the virtual camera is in the simulated motion state is 2000 seconds, and the period when it is in the simulated static state is 1600 seconds. During the test, the image blur period is 900 seconds, and the abnormal blur period overlapping with the simulated static period is 600 seconds. Retaining three decimal places, the abnormal blur ratio YM = 0.167. The image stagnation period is 800 seconds, and the abnormal stagnation period overlapping with the simulated motion period is 500 seconds. The abnormal stagnation ratio Y = 0.139. According to the formula, the problem degree value YC = 1.034;

[0051] Compare the obtained problem degree value with the problem degree threshold;

[0052] If the problem degree value within the test period is less than or equal to the problem degree threshold, it indicates that there are few image blur and freezing situations during the test period and it does not affect the normal use of users;

[0053] If the problem degree value within the test period is greater than the problem degree threshold, it indicates that there are many image blur and freezing situations during the test period, seriously affecting the normal use of users, and a problem area analysis signal is generated;

[0054] The technical solution of the embodiment of the present invention is as follows: During the test period, the test process of the secondary development of the MR virtual reality scene is recorded and photographed by a virtual camera to obtain test images. The test images are analyzed to obtain a blurred comparison image group and a stagnant comparison image group. Data processing is performed on the image pixel values of the test images to obtain the image similarity values of the blurred comparison image group and the stagnant comparison image group. The image blur period and the image stagnation period within the test period are marked according to the image similarity values; based on the marked image blur period and image stagnation period, they are respectively subjected to coincidence analysis and data processing with the simulated static period and the simulated motion period of the virtual camera to obtain the abnormal blur period, the abnormal stagnation period, and the problem degree value within the test period. Based on the problem degree value, it is judged whether the test image blur and freezing affect the normal use of users. If so, a problem area analysis signal is generated.

[0055] Embodiment 2

[0056] As Figure 1 shown, the interactive terminal of the MR virtual reality scene secondary development system provided by the embodiment of the present invention further includes the following modules:

[0057] Problem area analysis module: Based on the generated problem area analysis signals, evenly divide the test cycle into several area analysis time periods, and analyze and process the area analysis time periods, abnormal blur time periods, and abnormal stagnation time periods in combination with the image similarity values to obtain the analysis abnormal values of the area analysis time periods. Mark the problem area groups in the MR virtual reality scene based on the analysis abnormal values;

[0058] As Figure 3 shown, the specific steps for obtaining the analysis abnormal values are as follows:

[0059] In some embodiments, based on the generated problem area analysis signals, evenly divide the test cycle into several area analysis time periods, and the durations of the area analysis time periods are the same;

[0060] Based on any area analysis time period, obtain the overlapping time period between the area analysis time period and the abnormal blur time period, process the ratio of the duration of the overlapping time period to the duration of the area analysis time period to obtain the blur overlap ratio. Obtain the image similarity values of all blurred comparison image groups within the area analysis time period, sum and average them, and multiply the result by the obtained blur overlap ratio to obtain the blur abnormal degree value of the area analysis time period, marked as MC;

[0061] Similarly, obtain the overlapping time period between the area analysis time period and the abnormal stagnation time period, process the ratio of the duration of the overlapping time period to the duration of the area analysis time period to obtain the stagnation overlap ratio. Obtain the image similarity values of all stagnation comparison image groups within the area analysis time period, sum and average them, and multiply the result by the obtained stagnation overlap ratio to obtain the stagnation abnormal degree value of the area analysis time period, marked as TC;

[0062] Sum the obtained blur abnormal degree value MC and stagnation abnormal degree value TC to obtain the analysis abnormal value of the area analysis time period, and compare the obtained analysis abnormal value with the analysis abnormal threshold;

[0063] If the analysis abnormal value of the area analysis time period is less than or equal to the analysis abnormal threshold, it indicates that there is no or a slight degree of abnormal blur and stuttering in the images at the positions passed by the virtual camera during this area analysis time period;

[0064] If the analysis abnormal value of the area analysis time period is greater than the analysis abnormal threshold, it indicates that there is a serious abnormal blur and stuttering in the images at the positions passed by the virtual camera during this area analysis time period. Obtain the three-dimensional coordinates of all positions passed by the virtual camera within the area analysis time period and classify them into a problem area group;

[0065] Analyze and process all area analysis time periods within the test cycle to obtain several problem area groups, and perform position analysis on any two problem area groups;

[0066] Specifically, randomly select a three-dimensional coordinate from each of the two problem area groups, calculate the distance between the two three-dimensional coordinates, and compare the distance with the distance threshold;

[0067] If the distance between the two three-dimensional coordinates is greater than or equal to the distance threshold, it indicates that the two three-dimensional coordinates are far apart;

[0068] If the distance between the two three-dimensional coordinates is less than the distance threshold, it indicates that the two three-dimensional coordinates are close, and a short-distance area signal is generated;

[0069] Compare the distances of all three-dimensional coordinates within the two problem area groups. If there are three-dimensional coordinates within two different problem area groups that generate short-distance area signals, then merge the two problem area groups into one problem area group;

[0070] Compare and merge the distances of all problem area groups in the MR virtual reality scene until there are no three-dimensional coordinates within two different problem area groups that generate short-distance area signals in the MR virtual reality scene;

[0071] Problem area sorting module: Based on the marked problem area groups, analyze the problem area groups to obtain the modification priority value, sort the problem area groups according to the modification priority value, generate a repair and optimization report based on the sorting, and repair and optimize the MR virtual reality scene according to the repair and optimization report;

[0072] In some embodiments, based on any problem area group, obtain the number of three-dimensional coordinates in the problem area group, and perform a ratio process with the total number of three-dimensional coordinates in the MR virtual reality scene to obtain the problem volume ratio, marked as WT;

[0073] Obtain the analysis outliers of all area analysis time periods corresponding to the problem area group, sum and average them to obtain the analysis anomaly average value of the problem area group, marked as FJ;

[0074] Perform data processing on the obtained problem volume ratio WT and analysis anomaly average value FJ, through the formula Obtain the modification priority value XY of the problem area group, where s1 is a preset proportionality coefficient, s1 = 10;

[0075] Sort the problem area groups from largest to smallest according to the modification priority value, write the problem area group number and three-dimensional coordinates into the repair and optimization report, and generate a repair and optimization report;

[0076] Repair and optimize the problem area groups according to the sorting in the repair and optimization report, reduce the scene complexity, reduce the number of unnecessary polygons and texture maps, adjust rendering effects such as lighting, shadows, and reflections, improve the clarity and smoothness of the MR virtual reality scene, and enhance the user's interaction experience;

[0077] It should be noted that the purpose of repairing and optimizing the problem area group according to the sorting is to ensure that developers first focus on the areas that have the greatest impact on the user experience and the most serious problems, efficiently and reasonably allocate repair resources, and improve the stability and fluency of the secondary development system of the entire MR virtual reality scene at the fastest speed;

[0078] The technical solution of the embodiment of the present invention is as follows: based on the generated problem area analysis signal, the test cycle is evenly divided into several area analysis time periods, and the area analysis time periods, abnormal blur time periods and abnormal stagnation time periods are analyzed and processed in combination with the image similarity value to obtain the analysis abnormal value of the area analysis time period, and the problem area group in the MR virtual reality scene is marked based on the analysis abnormal value; based on the marked problem area group, the problem area group is analyzed to obtain the modification priority value, the problem area group is sorted according to the modification priority value, a repair and optimization report is generated based on the sorting, and the MR virtual reality scene is repaired and optimized according to the repair and optimization report.

[0079] The above formulas are all dimensionless and take their numerical calculations. The formula is a formula obtained by collecting a large amount of data for software simulation to get the closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0080] The above has described a detailed description of an embodiment of the present invention, but the content described above is only the preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equal changes and improvements made according to the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. MR virtual reality scene secondary development system interactive terminal, characterized by: include: Test analysis module: During the test cycle, the virtual camera is used to record and shoot the secondary development test process of the MR virtual reality scene to obtain test images, and the test images are analyzed to obtain the fuzzy contrast image group and the stagnant contrast image group and their image similarity values, and the image fuzzy period and image stagnant period in the test cycle are marked according to the image similarity values; Period analysis module: Based on the marked image blur period and image stagnation period, overlap analysis and data processing are performed with the simulated static period and simulated motion period of the virtual camera to obtain the abnormal blur period, abnormal stagnation period and the problem degree value within the test cycle. Based on the problem degree value, it is determined whether the test image blur and freeze affect the normal use of the user. If so, a problem area analysis signal is generated; Problem area analysis module: Based on the generated problem area analysis signal, the test cycle is divided into several area analysis periods. The area analysis period, abnormal fuzzy period and abnormal stagnation period are analyzed and processed in combination with the image similarity value to obtain the analysis abnormal value of the area analysis period. The problem area group in the MR virtual reality scene is marked based on the analysis abnormal value. Problem area sorting module: Based on the marked problem area groups, the problem area groups are analyzed to obtain modification priority values, the problem area groups are sorted according to the modification priority values, a repair optimization report is generated based on the sorting, and the MR virtual reality scene is repaired and optimized according to the repair optimization report.

2. The MR virtual reality scene secondary development system interactive terminal according to claim 1, characterized in that: The specific method of obtaining the image blur period and the image stagnant period is: Based on the fuzzy contrast image group, all fuzzy contrast image groups within the test period are compared and analyzed, and the period in which the fuzzy contrast image group continuously generates image similarity signals is marked as an image blur period, thereby obtaining several image blur periods; Based on any stagnation contrast image group, if an image similarity signal is generated, two images in the stagnation contrast image group are classified into one similar image group, and the above processing is performed on all stagnation contrast image groups in the test period to obtain a plurality of similar image groups; Based on any one of the similar image groups, a shooting period of a test image in the similar image group is marked as an image stagnation period.

3. The MR virtual reality scene secondary development system interactive terminal according to claim 2 is characterized in that: The specific method of obtaining the fuzzy contrast image group and the stagnant contrast image group is as follows: Several monitoring time points are evenly set within the test cycle. The test cycle represents the time required to test all modifications and added content requirements in the secondary development. The intervals between adjacent monitoring time points are the same. In the secondary development environment of the MR virtual reality scene, a virtual camera associated with the virtual reality device is configured. The virtual camera captures the virtual reality scene image from the user's perspective at the monitoring time point through the simulation of the tracking system to obtain the test image. The tracking system is a system in the virtual reality device that accurately tracks the user's head and hand movements through built-in sensors. Based on all modifications and additions during the secondary development of the MR virtual reality scene, test its required functions and analyze the test images taken during the test cycle after the test; Obtain any test image within the test period, perform Gaussian blur processing on the test image by using a Gaussian blur algorithm to obtain a Gaussian blurred image of the test image, and combine the test image and the corresponding Gaussian blurred image into a fuzzy contrast image group; Any two temporally adjacent test image sets within the test period are grouped as a stagnation comparison image group.

4. The MR virtual reality scene secondary development system interactive terminal according to claim 2 is characterized in that: The specific method of generating the image similarity signal is: Based on any fuzzy contrast image group or stagnant contrast image group, compare the similarity of two images in the fuzzy contrast image group or stagnant contrast image group, compare all pixels in the two images, obtain the number of similar pixel groups in the two images, and perform ratio processing on the sum of the number of similar pixel groups and the number of dissimilar pixel groups in the two images to obtain a similar number ratio, which is marked as XG; Obtain the absolute differences of all similar pixel groups and dissimilar pixel groups in the two images, sum them up and take the average, and obtain the average of the absolute differences of the pixels of the two images, marked as XJ; The obtained similarity number ratio XG and pixel absolute difference mean XJ are processed, and the image similarity value TX is calculated by the formula; Compare the obtained image similarity value with the image similarity threshold; If the image similarity value of the two images is greater than the image similarity threshold, it means that the two images are highly similar, and an image similarity signal is generated.

5. The MR virtual reality scene secondary development system interactive terminal according to claim 4 is characterized in that: The specific method of obtaining the similar pixel group and the dissimilar pixel group is: Based on any fuzzy contrast image group or stagnant contrast image group, comparing the similarity between two images in the fuzzy contrast image group or the stagnant contrast image group; Based on any pixel point with the same number of rows and columns in the two images, the pixel values ​​of each channel of the two pixel points are obtained, the difference processing is performed on the two pixel values ​​with the same channel and the absolute value is taken to obtain the single-channel pixel absolute difference, and the single-channel pixel absolute differences of all channels in the two pixel points are summed and averaged to obtain the pixel absolute difference of the two pixel points; Compare the obtained pixel absolute difference with the pixel absolute difference threshold; If the absolute pixel difference between two pixels is greater than or equal to the absolute pixel difference threshold, it means that the two pixels are not similar, and the two pixels are grouped into a dissimilar pixel group; If the absolute pixel difference between two pixels is less than the absolute pixel difference threshold, it means that the two pixels are similar, and the two pixels are grouped into a similar pixel group.

6. The MR virtual reality scene secondary development system interactive terminal according to claim 1, characterized in that: The specific method of obtaining the problem area analysis signal is as follows: The obtained abnormal fuzzy ratio YM and abnormal stagnation ratio YT are processed, and the problem degree value YC within the test cycle is calculated by the formula; Compare the obtained problem degree value with the problem degree threshold; If the problem degree value during the test cycle is greater than the problem degree threshold, it means that there are many cases of image blur and freeze during the test cycle, which seriously affects the normal use of users and generates a problem area analysis signal.

7. The MR virtual reality scene secondary development system interactive terminal according to claim 6, characterized in that: The specific method of obtaining the abnormal fuzzy ratio YM and the abnormal stagnation ratio YT is: Marking the period during which the virtual camera is in a simulated motion state during the test period as a simulated motion period, and marking the period during which the virtual camera is in a simulated static state during the test period as a simulated static period; Obtain the simulated static period of the virtual camera in the test cycle, compare it with the image blur period in the test cycle, obtain the overlapping period between the simulated static period and the image blur period, mark it as the abnormal blur period, and perform ratio processing on the duration of the abnormal blur period and the total duration of the test cycle to obtain the abnormal blur ratio, marked as YM; The simulated motion period of the virtual camera within the test cycle is obtained, and compared with the image stagnation period within the test cycle. The overlapping period between the simulated motion period and the image stagnation period is obtained, marked as the abnormal stagnation period, and the length of the abnormal stagnation period is ratioed with the total length of the test cycle to obtain the abnormal stagnation ratio, marked as YT.

8. The MR virtual reality scene secondary development system interactive terminal according to claim 1, characterized in that: The specific method of obtaining the problem area group is as follows: Compare the obtained analysis anomaly value with the analysis anomaly threshold; If the analysis anomaly value of the regional analysis period is greater than the analysis anomaly threshold, the three-dimensional coordinates of all the positions passed by the virtual camera in the regional analysis period are obtained and classified into a problem area group; Analyze and process all regional analysis periods within the test cycle to obtain several problem regional groups, and perform position analysis on any two problem regional groups; Randomly select a three-dimensional coordinate in each of the two problem area groups, calculate the distance between the two three-dimensional coordinates, and compare the distance with the distance threshold; If the distance between the two three-dimensional coordinates is less than the distance threshold, a close-range area signal is generated; The distance comparison is performed on the three-dimensional coordinates in the two problem area groups. If there are three-dimensional coordinates in two different problem area groups that generate close-range area signals, the two problem area groups are merged into one problem area group.

9. The MR virtual reality scene secondary development system interactive terminal according to claim 8, characterized in that: The specific method for obtaining the analysis outlier value is as follows: Based on the generated problem area analysis signal, the test cycle is evenly divided into a number of area analysis periods, and the duration of the area analysis periods is the same; Based on any regional analysis period, the overlapping period of the regional analysis period and the abnormal fuzzy period is obtained, and the duration of the overlapping period is compared with the duration of the regional analysis period to obtain the fuzzy overlap ratio, and the image similarity values ​​of all fuzzy contrast image groups in the regional analysis period are obtained, and the sum and average are taken and multiplied with the obtained fuzzy overlap ratio to obtain the fuzzy abnormality degree value of the regional analysis period, which is marked as MC; Obtain the overlapping period of the regional analysis period and the abnormal stagnation period, perform ratio processing on the length of the overlapping period and the length of the regional analysis period to obtain the stagnation overlap ratio, obtain the image similarity values ​​of all stagnation comparison image groups in the regional analysis period, sum and average them, and perform product processing with the obtained stagnation overlap ratio to obtain the stagnation abnormality degree value of the regional analysis period, marked as TC; The obtained fuzzy abnormality degree value MC and the stagnation abnormality degree value TC are summed to obtain the analysis abnormality value of the regional analysis period.

10. The MR virtual reality scene secondary development system interactive terminal according to claim 1, characterized in that: The specific method of generating the repair optimization report is as follows: Based on any problem area group, the number of three-dimensional coordinates in the problem area group is obtained, and the number is compared with the total number of three-dimensional coordinates in the MR virtual reality scene to obtain the problem volume ratio, which is marked as WT; Obtain the analysis anomaly values ​​of all regional analysis periods corresponding to the problem regional group, sum and average them, and obtain the analysis anomaly mean of the problem regional group, marked as FJ; The obtained problem volume ratio WT and the analysis abnormal mean FJ are processed, and the modification priority value XY of the problem area group is obtained through the formula; The problem area groups are sorted from large to small according to the modification priority value, and the problem area group sequence number and three-dimensional coordinates are written into the repair optimization report to generate the repair optimization report.

Images