Game screen monitoring method and device

Abstract

The present invention discloses a method and device for monitoring game screens, the method comprising the following steps: starting a test version of a game to be tested, obtaining a first screenshot according to a preset test location and lens posture information; obtaining a second screenshot from the first version of the game to be tested according to the preset test location and lens posture information; comparing the first screenshot with the second screenshot, and determining image quality issues of the game to be tested based on the comparison results. The present invention determines image quality issues of the game to be tested by obtaining screenshots of the test version and the first version at a preset test location with preset lens posture information, and performing comparisons. This allows batch testing of key areas of the game to be tested, effectively reducing the demand for testing manpower and improving the efficiency of screen monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of game testing, and in particular to a game screen monitoring method and device. Background Art

[0002] For gamers, the richer the game's graphics, the more appealing it is. However, for game developers, during version updates, the game's graphics may be altered by developers, artists, or other personnel, resulting in missing elements and a poor visual experience, which can severely impact the user experience. For example, if developers or artists modify a beautiful landscape so that flowers and plants are not visible, the user experience will be poor.

[0003] The quality of a game's graphics directly determines its quality. However, in large-scale games, due to the complex scenes and rapid iteration of game versions, testers cannot conduct comprehensive graphics quality checks before a release. Therefore, how to monitor game graphics to reduce testing manpower and improve monitoring efficiency has become a problem. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a game screen monitoring method that can automatically monitor whether the game screen has abnormalities, reduce the demand for testing manpower, and improve monitoring efficiency.

[0005] The present invention also provides a game screen monitoring system having the above-mentioned game screen monitoring method.

[0006] The present invention also provides a terminal having the above-mentioned game screen monitoring method.

[0007] The present invention also provides a computer-readable storage medium having the above-mentioned game screen monitoring method.

[0008] According to the first aspect of the present invention, the game screen monitoring method includes the following steps: starting the test version of the game to be tested, and obtaining a corresponding first screenshot according to a preset test location and lens posture information; obtaining a second screenshot from the first version of the game to be tested according to the preset test location and lens posture information; comparing the first screenshot and the second screenshot, and determining the image quality problem of the game to be tested based on the comparison result.

[0009] The game screen monitoring method according to the embodiment of the present invention has at least the following beneficial effects: by obtaining screenshots of the version to be tested and the first version at a preset test location with preset lens posture information for comparison, the image quality problems of the game to be tested can be determined, and batch detection can be performed on key areas of the game to be tested, effectively reducing the demand for testing manpower and improving the efficiency of screen monitoring.

[0010] According to some embodiments of the present invention, obtaining corresponding first and second screenshots based on a preset test location and camera pose information includes: using a control command to transport a character to the preset test location; adjusting the orientation of a virtual camera based on the camera pose information; and outputting a screenshot of the current display screen of the game under test in response to a screenshot command. By transporting the character to the target location and adjusting the virtual camera pose, screenshots with the same shooting angle and range are obtained for each test location, increasing the comparability of screenshots between different versions of the game, thereby achieving effective image quality monitoring and judgment.

[0011] According to some embodiments of the present invention, after adjusting the direction of the virtual camera, the method further includes: waiting for a first preset time and then responding to the screenshot instruction. Waiting for a certain period of time can obtain a more stable image, thereby improving comparison accuracy and enhancing monitoring effect.

[0012] According to some embodiments of the present invention, comparing the first screenshot and the second screenshot and determining the image quality issue of the tested game based on the comparison result includes: obtaining image similarity between the first screenshot and the second screenshot; and determining that the image similarity is lower than a first similarity threshold, thereby determining that an image quality risk exists. Comparing based on the similarity facilitates rapid determination of whether an image quality issue exists.

[0013] According to some embodiments of the present invention, comparing the first screenshot with the second screenshot and determining the image quality issues of the game under test based on the comparison results includes: performing target detection on the first screenshot to obtain a first target recognition result; performing target detection on the second screenshot to obtain a second target recognition result; and comparing the first target recognition result with the second target recognition result to determine the image quality issues of the game under test. By identifying scene objects such as flowers, grass, trees, buildings, and people in the screenshots, the accuracy of determining image quality issues can be improved based on the type and number of objects, and detailed and intuitive feedback of the issues can be provided to the relevant personnel.

[0014] According to some embodiments of the present invention, if the first and second target recognition results identify the same target type and quantity, then it is determined that there is no image quality risk; otherwise, it is determined that there is an image quality risk. Only when the target type and quantity are consistent is it considered that there is no image quality risk; if there is a difference in target type or quantity, then it is determined that there is an image quality risk, and the inconsistencies can be output to facilitate relevant personnel to quickly locate the problem.

[0015] According to some embodiments of the present invention, before launching the tested version of the game, the method further includes configuring the display configuration information of the tested version and the first version to be consistent. By aligning the display configurations of different versions of the game, such as image resolution, lens angle, depth of field, and blur, comparison accuracy can be improved, thereby enhancing monitoring quality.

[0016] According to the second aspect of the present invention, the game screen monitoring device includes: a configuration module for receiving and storing preset test locations and lens posture information; a screenshot module for starting the test version of the game to be tested, and obtaining a corresponding first screenshot according to the preset test locations and lens posture information; an acquisition module for obtaining a second screenshot from the first version of the game to be tested according to the preset test locations and lens posture information; and a comparison module for comparing the first screenshot and the second screenshot, and determining the image quality problem of the game to be tested based on the comparison result.

[0017] According to the game screen monitoring device of the embodiment of the present invention, there are at least the following beneficial effects: by obtaining screenshots of the version to be tested and the first version respectively for comparison at a preset test location with preset lens posture information, the image quality problems of the game to be tested can be determined, and batch detection can be carried out in key areas of the game to be tested, which effectively reduces the demand for testing manpower and improves the efficiency of screen monitoring.

[0018] According to some embodiments of the present invention, the screenshot module includes: a character transport module for transporting a character to a preset test location via a control command; a lens adjustment module for adjusting the orientation of a virtual camera based on the lens posture information; and a screenshot output module for outputting a screenshot of the current display screen of the game under test in response to a screenshot command. By transporting the character to the target location and adjusting the virtual camera's posture, screenshots with the same shooting angle and range are obtained for each test location, increasing the comparability of screenshots between different versions of the game and enabling effective image quality monitoring and judgment.

[0019] According to some embodiments of the present invention, the comparison module includes: a similarity comparison module configured to receive the first and second screenshots as input and obtain image similarity between the first and second screenshots; and a first determination module configured to determine if the image similarity is lower than a first similarity threshold and, if so, determine that an image quality risk exists. Comparison based on similarity facilitates rapid determination of image quality issues.

[0020] According to some embodiments of the present invention, the comparison module includes: a target detection module for performing target detection on the first screenshot to obtain a first target recognition result; and a target detection module for performing target detection on the second screenshot to obtain a second target recognition result; and a second judgment module for comparing the first target recognition result with the second target recognition result, and determining image quality issues of the game under test based on the type and number of identified targets. By identifying scene targets such as flowers, grass, trees, buildings, and people in the screenshots, and based on the type and number of targets, the accuracy of image quality problem judgments can be improved, and detailed and intuitive feedback of issues can be provided to relevant personnel.

[0021] According to the third aspect of the present invention, an embodiment includes a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions of the machine-readable storage medium to implement the method steps of the first aspect of the present invention.

[0022] The terminal according to the embodiment of the present invention has at least the same beneficial effects as the method of the embodiment of the first aspect of the present invention.

[0023] According to a fourth aspect of the present invention, a computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the method according to the first aspect of the present invention is implemented.

[0024] The computer-readable storage medium according to the embodiment of the present invention has at least the same beneficial effects as the method of the embodiment of the first aspect of the present invention.

[0025] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments with reference to the accompanying drawings, in which:

[0027] Figure 1 Schematic diagram of the main process of the method of an embodiment of the present invention;

[0028] Figure 2 A schematic diagram for comparing screenshots of an embodiment of the present invention;

[0029] Figure 3 is a schematic flow chart of a method according to a first embodiment of the present invention;

[0030] Figure 4 is a flow chart of a method according to a second embodiment of the present invention;

[0031] Figure 5 is a schematic diagram of the internal modules of the device according to the first embodiment of the present invention;

[0032] Figure 6 is a schematic diagram of the internal modules of a device according to a second embodiment of the present invention;

[0033] Figure 7 Schematic diagram of the internal modules of the screenshot module in the device according to an embodiment of the present invention;

[0034] Figure 8 Schematic diagram of modules of a terminal according to an embodiment of the present invention.

[0035] Reference numerals:

[0036] Configuration module 100, screenshot module 200, acquisition module 300, comparison module 400;

[0037] Character transmission module 110, lens adjustment module 120, screenshot output module 130;

[0038] Similarity comparison module 410, first determination module 420; target detection module 430, second determination module 440;

[0039] Terminal 500 , processor 510 , and machine-readable storage medium 520 . DETAILED DESCRIPTION

[0040] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0041] In the description of the present invention, the meaning of "several" is one or more, the meaning of "many" is two or more, and "greater than", "less than", "exceed", etc. are understood as excluding the number itself, and "above", "below", "within", etc. are understood as including the number itself. If there is a description of first or second, it is only for the purpose of distinguishing the technical features, and it cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features or implicitly indicating the order of the indicated technical features. In the description of the present invention, the step numbers are only for the convenience of description or citation. The size of the serial number of each step does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0042] Reference Figure 1The method of an embodiment of the present invention includes the following steps: launching a test version of a game to be tested, obtaining a corresponding first screenshot based on a preset test location and camera posture information; obtaining a second screenshot from the first version of the game to be tested based on the preset test location and camera posture information; comparing the first screenshot and the second screenshot, and determining image quality issues in the game to be tested based on the comparison results. That is, for the same game to be tested, obtaining a corresponding first screenshot from the test version and a corresponding second screenshot from the first version; then comparing the first screenshot and the second screenshot to determine image quality issues.

[0043] In an embodiment of the present invention, several preset test locations can be set. For example, based on the player's routine tasks or activity tasks, the activity location that the player may go to can be determined, and the coordinates of the test location can be generated within a certain range of the activity location. Alternatively, several test coordinate points can be generated according to the complexity of the game scene. For example, several coordinate points can be generated according to the type and number of NPCs in the scene area. Scene areas with many NPC types or a large number of NPCs will generate more test coordinate points. The test location can be represented by a map, coordinates, orientation, etc. Taking a three-dimensional multi-scene map game as an example, for example: [MAPID, POSX, POSY, POSZ, lookat], where MapID represents the ID of the map, lookat represents the orientation, and (POSX, POSY, POSZ) represents the three-dimensional coordinates.

[0044] Each test coordinate point can be set to one or more lens postures, and the screenshots and comparisons are shown as follows: Figure 2 According to the same test configuration, that is, the same series of test locations and camera pose information, a plurality of first screenshots and second screenshots are obtained from the version to be tested (eg, version 2) and the first version (eg, version 1) of the game to be tested. Figure 2 In the example of a multi-scene map game, three screenshots of three different test locations are shown for illustration only. Figure 2 In the example shown, the screenshots are named "map_location_orientation_version" and saved in JPG format. The map can be a map symbol or map name, the location can be map coordinates, the orientation represents the virtual camera's posture information, and the version corresponds to the version number of the tested version or the second version. In embodiments of the present invention, the virtual camera's posture information includes, for example, the virtual camera's orientation, angle, and altitude.

[0045] The first screenshot is obtained from the launched version of the game under test. The second screenshot can be obtained from a pre-stored screenshot library of the first version of the game under test. This pre-stored screenshot library can be screenshots retained from a previous test of the first version. When taking screenshots, the screenshot files can be output according to a pre-set naming convention corresponding to the test location and camera pose information.

[0046] In this way, when comparing the first screenshot with the second screenshot, you can find the second screenshot with the same position and angle as the first screenshot according to the above naming rules for comparison. Figure 2 For example, the naming rule for screenshot output is: map_location_direction_version, then the first screenshot and the second screenshot are mapped in the form of "map_location_direction", for example, Figure 1 _Location1_Orientation1_Version2.jpg" and "Location Figure 1 _location1_orientation1_version1.jpg". That is, find screenshots with the same location and orientation from the screenshot library of the first version for comparison.

[0047] It should be understood that Figure 2 The screenshot naming method in the example is only an example and is not limited to this embodiment. As long as the naming method can be used to find screenshots with the same position and angle, for example, the first screenshot is stored in a folder with the corresponding version number (i.e., version 2) and named in the format of "map_location_direction". Alternatively, for example, a code with the same naming rule can be used. Similarly, this embodiment does not limit the storage format of the screenshots.

[0048] The method of the first embodiment of the present invention is as follows: Figure 3 As shown, the following steps are included.

[0049] S100 , for the game to be tested, according to the preset test location and camera posture information, obtain a corresponding first screenshot in the version to be tested, and obtain a corresponding second screenshot in the first version.

[0050] Specifically, the method for obtaining the first screenshot includes:

[0051] Through control commands, such as GM commands, the character is transported to a preset test location;

[0052] Adjust the direction of the virtual camera according to the lens posture information;

[0053] In response to the screenshot instruction, a screenshot of the current display screen of the game to be tested is output.

[0054] S200 , performing a similarity comparison on the first screenshot and the second screenshot to obtain image similarity.

[0055] Specifically, the similarity can be calculated by one of the following methods, for example:

[0056] (1) Divide the image into blocks using a sliding window, with the total number of blocks being N. Based on the influence of the window shape on the blocks, the mean, variance, and covariance of each window are calculated using Gaussian weighting. The structural similarity (SSIM) of the image blocks corresponding to each window is then calculated. The summed average is used as the structural similarity measure between the first and second screenshots to obtain image similarity. Image similarity can be measured from three aspects: brightness, contrast, and structure.

[0057] (2) Extract key feature points in the image that have a characteristic where pixel values change sharply from light to dark, and calculate the corresponding eigenvectors of the key feature points to represent the intensity pattern around the key feature points; by comparing the key feature points and the corresponding eigenvectors, determine the degree of similarity. This method can be implemented, for example, using the OpenCV image similarity ORB algorithm.

[0058] In the embodiment of the present invention, the method for obtaining the similarity is not limited thereto.

[0059] S300: Determine whether there is an image quality risk based on the image similarity.

[0060] Specifically, if the image similarity between the first and second screenshots is greater than or equal to a first preset threshold, for example, 0.85, then no image quality risk is determined. If the similarity is below the first threshold, then the image quality is determined to be at risk. Information about the risky area (such as a map, coordinates, orientation, etc.) and the corresponding screenshots are communicated to the relevant R&D or art manager via, for example, email or webpage display, so that the relevant personnel can promptly resolve the issue. For example, when comparing image similarity, if two areas in the image are found to have significant differences, they can be highlighted using a more conspicuous color such as red, or a color that significantly differs from the main color of the image, or the corresponding area can be demarcated using a line of the aforementioned color and a screenshot provided to the relevant personnel. Furthermore, the location where the abnormal image quality occurred can be indicated based on the map name or coordinates corresponding to the test location. The degree of image quality abnormality can also be determined based on the amount of difference; that is, the lower the similarity, the higher the degree of image quality abnormality. Different colors can be used to indicate different degrees of image quality abnormality, such as orange, red, and green to indicate severe image quality abnormality, general image quality abnormality, and low-level image quality abnormality, respectively.

[0061] In the embodiments of the present invention, the previous version released to the external network before the version to be tested is generally used as the first version, thereby helping to improve the correctness and accuracy of monitoring. However, the present invention is not limited to this, and does not exclude the possibility of comparing with other versions, such as the first N versions released to the external network.

[0062] The method of the second embodiment of the present invention is as follows: Figure 4 As shown, the following steps are included.

[0063] S100 , for the game to be tested, according to the preset test location and camera posture information, obtain a corresponding first screenshot in the version to be tested, and obtain a corresponding second screenshot in the first version.

[0064] Specifically, the method for obtaining the first screenshot and the second screenshot is the same, including:

[0065] The character is transported to a preset test location through a control command, such as a GM (game master) command;

[0066] Adjust the direction of the virtual camera according to the lens posture information;

[0067] In response to the screenshot instruction, a screenshot of the current display screen of the game to be tested is output.

[0068] S200a, performing target recognition on the first screenshot and the second screenshot respectively, and obtaining corresponding first target recognition results and second target recognition results.

[0069] Specifically, it can be done through a target detection network such as Detectron2 (Detectron2 is a target detection platform launched by Facebook AIResearch, which can be used to identify targets in pictures; Detectron2 is based on a neural network, and after it is trained with a large amount of labeled data, it can perform target recognition. For example, give it 100,000 pictures, mark where there are flowers in the pictures, and then after training, it can recognize the characteristics of the flowers and mark the flowers from the image), input the first screenshot, identify scene targets such as flowers, grass, trees, buildings, and people, and obtain the first target recognition result.

[0070] In the embodiments of the present invention, the level of detail of target recognition will affect the comparison results. Taking the recognition of people and their accessories as an example, the following shows two different levels of target recognition accuracy:

[0071] A. Only the characters are used as scene targets, not the accessories on the characters;

[0072] B. Use the characters and their accessories as scene targets.

[0073] Obviously, the target recognition accuracy of the two is different, and the target recognition accuracy of A is lower than that of B.

[0074] When using method A for recognition, for the first screenshot and the second screenshot with the same coordinates and orientation, if a person is detected but the number of people in the first screenshot is inconsistent with the number of people in the second screenshot, it is confirmed that there is no image quality problem.

[0075] But at this time, the accessories on the character have changed, such as becoming less, and the monitoring and comparison are not detected.

[0076] At this time, when using method B for identification, it is possible not only to compare whether the person has changed, but also to monitor whether the accessories on the person have changed.

[0077] However, improving recognition accuracy will obviously increase the system's data processing requirements and slow down the processing speed. Therefore, in embodiments of the present invention, the target detection and recognition level can be set based on the complexity and importance of the scene, the data processing speed, the required accuracy of the screenshot judgment results, etc. For example, a higher target detection and recognition level can be set for areas where players gather, and a lower target detection and recognition level can be set for simple scenes, and so on.

[0078] For example, a higher object detection and recognition level can be set during the first test. If image quality issues are discovered, the object detection and recognition level can be adjusted in subsequent, modified, iterative tests based on the degree of image quality anomalies detected initially, the content of the modifications, and their size. For example, if a large number of abnormal accessories are found on a person at a certain test point during the first test, the higher object detection and recognition level can be maintained. If the initial test determines that the accessories on a person at a certain test point are normal, the object detection and recognition level can be adjusted.

[0079] The target detection and recognition level can be, for example, the number of recognized objects in a scene. When a preset number is exceeded, recognition is discontinued. The target detection and recognition level can also be, for example, the number of pixels occupied by the recognized objects in an image, that is, objects with a larger number of pixels are prioritized and objects with a smaller number of pixels are discarded. This is merely an example of setting the target detection and recognition level and does not limit the embodiments of the present invention.

[0080] Similarly, for example, through the Detectron2 target detection network, a second target recognition result is obtained from the second screenshot.

[0081] S300a, comparing the first target recognition result and the second target recognition result to determine image quality issues of the game to be tested.

[0082] Specifically, if all target types in the first target recognition result have appeared in the second target recognition result relative to the second target recognition result, and the number of each target type is the same, then the target type and quantity in the first target recognition result are considered to be consistent with those in the second target recognition result, and it is determined that there is no image quality problem at this time.

[0083] If the type and number of identified objects are consistent, some embodiments of the present invention further analyze the similarity threshold of the identified objects. For example, if grass is identified in both the first and second screenshots, a similarity comparison is performed to detect any changes in the area or color of the grass, such as changes from green to yellow or a change in color depth, or whether the area of the grass has shrunk or expanded. If the similarity exceeds a certain threshold, the image quality is considered normal; otherwise, the image quality is considered abnormal.

[0084] By further analyzing the similarity threshold of the identified target, it is possible to locate changes in the target object before the target object is identified, thereby improving detection accuracy and making it easier for relevant personnel to locate problems, thereby avoiding problems such as missing screen elements or abnormal changes in versions released to the external network.

[0085] Obviously, further analyzing the target similarity threshold will also increase the system's data processing requirements and slow down the processing speed. Therefore, some embodiments of the present invention also provide a corresponding switch. The tester can appropriately turn on the switch according to the test needs to obtain the similarity of the identified target.

[0086] If there is a target type in the first target recognition result that does not appear in the second target recognition result, or there is a target type in the second target recognition result that does not appear in the first target recognition result, then the target types of the two are inconsistent, and it is determined that there is a picture quality problem. In addition, if the target types of the first target recognition result and the second target recognition result are the same, but the number of a certain target type is different, then it is also determined that there is a picture quality problem. For example, if two screenshots are taken at a test location facing 1 at position 1. The second target recognition result corresponding to the screenshot obtained from the first version includes five trees, a pond, a house, two lawns and three mountains. The first target recognition result corresponding to the screenshot obtained from the version to be tested also includes: trees, ponds, houses, lawns and mountains, but their numbers are 4, 1, 1, 2 and 2, respectively. Obviously, there is one more tree and one less mountain in the picture. At this time, the embodiment of the present invention will highlight the extra trees in the screenshot obtained from the version to be tested, mark the corresponding area where one mountain is missing with a red line, display the processed screenshot, and annotate: at the test location at position 1 facing 1, there is one missing mountain and one extra tree.

[0087] In an embodiment of the present invention, by automatically taking screenshots and inputting them into a target detection network such as Detectron2 for image target detection and comparison, a large amount of screen content inspection can be carried out covering different coordinates and camera orientations of the game, thereby avoiding the problem of missing screen elements in the version released to the external network and improving the quality of the game.

[0088] In the method of this embodiment of the present invention, before launching the tested version of the game, the display configuration information of the tested version and the first version is configured to be consistent. The display information includes, for example, image resolution, lens angle, depth of field, blur, etc. By using the same display configuration, comparison accuracy can be improved, thereby enhancing monitoring quality.

[0089] In a third embodiment of the present invention (not shown), the method of the first embodiment and the method of the second embodiment are combined to compare the first screenshot and the second screenshot. That is, the comparison is divided into two stages. In the first stage, the first screenshot and the second screenshot are first compared using the method of the first embodiment. Only when the image similarity between the first screenshot and the second screenshot is lower than the first similarity threshold, the comparison of the second stage is continued. In the second stage, the first screenshot and the second screenshot are compared using the method of the second embodiment to obtain the difference results of the type and quantity of the detected targets, generate a corresponding test report, and send it to the corresponding staff.

[0090] In an embodiment of the present invention, the preset test locations can be based on the reported data related to the player character's map residence time, historical tasks, etc. in the historical version of the game to be tested, and the player's key gathering areas and residence hotspot locations can be extracted through data analysis; or they can be pre-set by the screen monitoring personnel based on the regional usage of the game map scene, the scene boundary range, etc. For certain key test locations, multiple different lens posture information can also be set to obtain screenshots in different directions for screen monitoring. In addition, the lens posture information can also be not limited to the lens direction. For example, different lens wide angles can be set at certain test locations, while the default lens wide angle is used in other places.

[0091] The device of the embodiment of the present invention is used to perform the following Figure 1 The method shown includes the following modules, refer to Figure 5 and Figure 6 .

[0092] The configuration module 100 is used to receive and store preset test locations and lens posture information.

[0093] The screenshot module 200 is used to start the test version of the game to be tested, read the preset test location and camera posture information, and for each test location, obtain the corresponding first screenshot from the test version of the game to be tested according to the corresponding camera posture information.

[0094] Acquisition module 300 is configured to acquire a second screenshot from the first version of the game under test based on the preset test location and camera pose information. The second screenshot may be acquired from a pre-stored screenshot library of the first version of the game under test. This pre-stored screenshot library may be screenshots retained from a previous test of the first version. When taking a screenshot, a screenshot file may be output according to a naming convention corresponding to the preset test location and camera pose information.

[0095] The comparison module 400 is used to compare the first screenshot and the second screenshot, and determine the image quality problem of the game to be tested based on the comparison result.

[0096] Specifically, the screenshot module 200 in the device of the embodiment of the present invention includes the following: Figure 7 The internal modules shown are: The character transfer module 110 is used to set the character's map location information through control instructions, such as GM instructions, and transfer the character to a preset test location. The lens adjustment module 120 is used to adjust the direction of the virtual camera based on the lens posture information so that the virtual camera faces the preset location. The screenshot module 130 is used to output a screenshot of the current display screen of the game under test in response to a screenshot instruction.

[0097] The device of the first embodiment of the present invention, such as Figure 5 As shown. The comparison module 400 includes: a similarity comparison module 410 and a first judgment module 420. The comparison module 400 inputs the received first screenshot and the second screenshot into the similarity comparison module 410 in pairs. The similarity comparison module 410 obtains the image similarity between the first screenshot and the second screenshot, and inputs the image similarity into the first judgment module 420. If the first judgment module 420 detects that the image similarity is lower than the first similarity threshold, it determines that there is an image quality risk. In some embodiments, the first judgment module 420 also generates a test report based on the location information that is lower than the first similarity threshold, such as: map name, map coordinates, etc., and the corresponding screenshot information, and notifies the corresponding staff by means such as sending an email.

[0098] The device of the second embodiment of the present invention, such as Figure 6As shown, the comparison module 400 includes an object detection module 430 and a second determination module 440. The comparison module 400 inputs the received first and second screenshots in pairs into the object detection module 430. Object detection module 430 uses an object detection network, such as Detectron2, to perform object detection on the first screenshot to obtain a first object recognition result; and also performs object detection on the second screenshot to obtain a second object recognition result. The first and second object recognition results are input into the second determination module 440. The second determination module 440 compares the first and second object recognition results and determines the image quality issues of the game under test based on the type and number of identified objects. Specifically, if the target type and number are consistent, it is considered that there are no image quality issues. Otherwise, if the target type is inconsistent, or if the target type is consistent but the number is different, it is considered that there is an image quality risk. In some embodiments, the second determination module also generates a test report based on the location information of the locations where the target type and number are consistent, such as map name and map coordinates, as well as the corresponding screenshot information, and notifies the relevant personnel through methods such as email. The test report will also provide specific information on the target type or quantity inconsistencies.

[0099] In a third embodiment of the present invention (not shown), the apparatus of the first embodiment and the method of the second embodiment are installed to compare the first screenshot and the second screenshot. That is, the comparison module 400 includes: a similarity comparison module 410, a first judgment module 420, a target detection module 430 and a second judgment module 440. The comparison is divided into two stages. In the first stage, the first screenshot and the second screenshot are compared using the similarity comparison module 410 and the first judgment module 420. Only when the image similarity between the first screenshot and the second screenshot is lower than the first similarity threshold, the comparison of the second stage is continued. In the second stage, the target detection module 430 and the second judgment module 440 are used to compare the first screenshot and the second screenshot, obtain the difference results of the type and quantity of the detected targets, generate a corresponding test report, and send it to the corresponding staff.

[0100] The present invention also provides a terminal 500, such as Figure 8 As shown, the system includes a processor 510 and a machine-readable storage medium 520. The machine-readable storage medium 520 stores machine-executable instructions that can be executed by the processor 510. The processor 510 executes the machine-executable instructions of the machine-readable storage medium 520 to implement the above method steps in the embodiment of the present invention.

[0101] Although specific embodiments are described herein, those skilled in the art will recognize that many other modifications or alternative embodiments are also within the scope of this disclosure. For example, any of the functions and / or processing capabilities described in conjunction with a particular device or component may be performed by any other device or component. In addition, although various exemplary implementations and architectures have been described in accordance with embodiments of the present disclosure, those skilled in the art will recognize that many other modifications to the exemplary implementations and architectures described herein are also within the scope of this disclosure.

[0102] Some aspects of the present disclosure have been described above with reference to the block diagrams and flow charts of the systems, methods, systems and / or computer program products according to the exemplary embodiments. It should be understood that the combination of one or more blocks in the block diagram and the flow chart and the blocks in the block diagram and the flow chart can be realized by executing computer executable program instructions respectively. Equally, according to some embodiments, some blocks in the block diagram and the flow chart may not need to be executed in the order shown, or may not need to be executed in full. In addition, additional components and / or operations beyond those components and / or operations shown in the blocks in the block diagram and the flow chart may be present in certain embodiments.

[0103] Therefore, the blocks in the block diagrams and flow charts support combinations of means for performing the specified functions, combinations of elements or steps for performing the specified functions, and program instruction means for performing the specified functions. It should also be understood that each block in the block diagrams and flow charts, and combinations of blocks in the block diagrams and flow charts, can be implemented by a dedicated hardware computer system that performs the specific functions, elements, or steps, or a combination of dedicated hardware and computer instructions.

[0104] The program modules, applications, etc. described herein may include one or more software components, including, for example, software objects, methods, data structures, etc. Each such software component may include computer-executable instructions that, in response to execution, cause at least a portion of the functionality described herein (e.g., one or more operations of the illustrative methods described herein) to be performed.

[0105] Software component can be encoded with any one in various programming languages.A kind of exemplary programming language can be low-level programming language, such as the assembly language associated with specific hardware architecture and / or operating system platform.Comprise that the software component of assembly language instruction may need to be converted to executable machine code by assembler before being executed by hardware architecture and / or platform.Another exemplary programming language can be a more advanced programming language, and it can be transplanted across multiple architectures.Comprise that the software component of more advanced programming language may need to be converted to intermediate representation by interpreter or compiler before execution.Other examples of programming language include but are not limited to macro language, shell or command language, job control language, script language, database query or search language or report writing language.In one or more exemplary embodiments, the software component that comprises the instruction of one in the above-mentioned programming language example can be directly executed by operating system or other software component, without first being converted into another form.

[0106] Software components can be stored as files or other data storage structures. Software components of similar types or related functions can be stored together, such as in a specific directory, folder, or library. Software components can be static (e.g., preset or fixed) or dynamic (e.g., created or modified at execution time).

[0107] The embodiments of the present invention are described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Various changes can be made within the knowledge of ordinary technicians in the relevant technical field without departing from the scope of the present invention.

Claims

1. A method for monitoring a game screen, characterized in that: The following steps are involved: Launch the test version of the game to be tested, and obtain the first screenshot based on the preset test location and camera posture information; Obtaining a second screenshot from the first version of the game to be tested according to the preset test location and camera posture information; Comparing the first screenshot and the second screenshot, and determining a picture quality problem of the game to be tested based on the comparison result; The step of comparing the first screenshot and the second screenshot and determining the image quality problem of the game to be tested according to the comparison result includes: Determining an object detection and recognition level according to the scenes in which the first screenshot and the second screenshot are located, wherein the object detection and recognition level is used to represent the recognition accuracy of the scene object; performing target recognition on the first screenshot and the second screenshot respectively based on the target detection recognition level through the target detection network, and obtaining corresponding first target recognition results and second target recognition results, wherein the first target recognition results and the second target recognition results record the type and quantity of each of the scene targets; The image quality problem of the game to be tested is determined by comparing the first target recognition result and the second target recognition result.

2. The game screen monitoring method according to claim 1, characterized in that: The obtaining of the corresponding first screenshot according to the preset test location and lens posture information includes: The character is transported to the preset test location through control instructions; Adjusting the orientation of the virtual camera according to the lens posture information; In response to the screenshot instruction, a screenshot of the current display screen of the game to be tested is output.

3. The game screen monitoring method according to claim 2, characterized in that: After adjusting the direction of the virtual camera, the method further includes: waiting for a first preset time and responding to the screenshot instruction.

4. The game screen monitoring method according to claim 1, characterized in that: The comparing the first screenshot and the second screenshot and determining the image quality problem of the game to be tested according to the comparison result includes: Obtaining image similarity between the first screenshot and the second screenshot; If it is determined that the image similarity is lower than a first similarity threshold, it is determined that there is an image quality risk.

5. The game screen monitoring method according to claim 1, characterized in that: The comparing the first screenshot and the second screenshot and determining the image quality problem of the game to be tested according to the comparison result includes: Performing target detection on the first screenshot to obtain a first target recognition result; Performing target detection on the second screenshot to obtain a second target recognition result; The first target recognition result and the second target recognition result are compared to determine image quality issues of the game to be tested.

6. The game screen monitoring method according to claim 5, characterized in that: If the type and quantity of the identified targets in the first target recognition result and the second target recognition result are consistent, it is determined that there is no image quality risk; otherwise, it is determined that there is an image quality risk.

7. The game screen monitoring method according to claim 1, characterized in that: Before launching the test version of the game under test, it also includes: The display configuration information of the version to be tested and the first version are configured to be consistent.

8. A game screen monitoring device, characterized in that: include: Configuration module, used to receive and store preset test location and lens posture information; A screenshot module is used to start the test version of the game to be tested and obtain the corresponding first screenshot according to the preset test location and camera posture information; an acquisition module, configured to acquire a second screenshot from the first version of the game to be tested according to a preset test location and camera posture information; a comparison module, configured to compare the first screenshot and the second screenshot, and determine image quality issues of the game to be tested based on the comparison result; The step of comparing the first screenshot and the second screenshot and determining the image quality problem of the game to be tested according to the comparison result includes: Determining an object detection and recognition level according to the scenes in which the first screenshot and the second screenshot are located, wherein the object detection and recognition level is used to represent the recognition accuracy of the scene object; performing target recognition on the first screenshot and the second screenshot respectively based on the target detection recognition level through the target detection network, and obtaining corresponding first target recognition results and second target recognition results, wherein the first target recognition results and the second target recognition results record the type and quantity of each of the scene targets; The image quality problem of the game to be tested is determined by comparing the first target recognition result and the second target recognition result.

9. The game screen monitoring device according to claim 8, characterized in that: The screenshot module includes: A character transmission module, used to transmit the character to the preset test location through control instructions; A lens adjustment module, configured to adjust the orientation of the virtual camera according to the lens posture information; The screenshot output module is used to output a screenshot of the current display screen of the game to be tested in response to a screenshot instruction.

10. The game screen monitoring device according to claim 8, characterized in that: The comparison module includes: a similarity comparison module, configured to receive the first screenshot and the second screenshot as input, and obtain image similarity between the first screenshot and the second screenshot; The first determination module is configured to determine that if the image similarity is lower than a first similarity threshold, then determine that there is an image quality risk.

11. The game screen monitoring device according to claim 8, characterized in that: The comparison module includes: an object detection module, configured to perform object detection on the first screenshot to obtain a first object recognition result; and to perform object detection on the second screenshot to obtain a second object recognition result; The second judgment module is used to compare the first target recognition result and the second target recognition result, and determine the image quality problem of the game to be tested based on the type and number of recognized targets.

12. A terminal, characterized in that: The method comprises a processor and a machine-readable storage medium, wherein the machine-readable storage medium stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions of the machine-readable storage medium to implement the method steps described in any one of claims 1 to 7.

13. A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images