Game quality evaluation method and device, equipment, storage medium and product

Abstract

The invention provides a game quality evaluation method and device, equipment, a storage medium and a product. The method comprises the following steps: acquiring a video corresponding to each of a plurality of running stages of a game; for the video of each stage, performing cross-correlation operation on a target video frame and a reference frame in the video to obtain a row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is a video frame except a reference frame in the video; for each stage, determining a key experience index according to the row and column displacement vector values; and carrying out weighted summation on the key experience indexes of each stage to obtain a quality evaluation index of the game.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a game quality assessment method, device, equipment, storage medium and product. Background Art

[0002] At present, detection and identification methods for game delays are mostly concentrated at the hardware level, such as network performance optimization and game optimization. However, there are relatively few game experience monitoring methods at the software level. Traditional game perception models can only monitor game freezes, but freeze delays, game startup delays, loading delays, etc. cannot be counted. Traditional login delay identification methods mainly rely on monitoring input devices and game engines. This method is often limited by the characteristics of device hardware and drivers, and lacks fine-grained monitoring and feedback of actual game scenarios.

[0003] The prior art relies on the game screen video formed by the game screen, analyzes the game screen video, and generates a video segmentation result based on the analysis result to detect game delay. The focus of the evaluation is on analyzing the running video. The specific abnormal analysis model established is vague in description, and the granularity of the evaluation is large. Relevant data collection for network quality monitoring, such as game startup delay, game login delay, game loading delay, game login success rate and other perception indicators are missing. It relies on the monitoring of input devices and game engines, and is limited by the characteristics of device hardware and drivers. It cannot reflect the quality indicators of different stages, and it is difficult to accurately evaluate the quality of the game. Summary of the invention

[0004] The embodiments of the present application provide a game quality assessment method, apparatus, device, storage medium and product, which can accurately obtain game quality assessment indicators.

[0005] In a first aspect, the present application provides a game quality assessment method, comprising:

[0006] Get the video corresponding to each stage of the game in multiple running stages;

[0007] For each stage of the video, a cross-correlation operation is performed on a target video frame and a reference frame in the video to obtain a row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is a video frame in the video other than the reference frame;

[0008] For each stage, determine the key experience indicators based on the row and column displacement vector values;

[0009] The key experience indicators of each stage are weighted and summed to obtain the quality evaluation index of the game.

[0010] In some possible implementations, for each stage, determining the key experience indicator according to the row and column displacement vector values ​​includes:

[0011] For each stage, perform the following operations:

[0012] Determining the time delay corresponding to the stage according to the row and column displacement vector values;

[0013] According to the preset mapping relationship between latency and service quality value, the key experience index corresponding to the stage is determined, and the key experience index represents the degree to which the corresponding game running process meets the user needs.

[0014] In some possible implementations, obtaining a video corresponding to each stage of the game in multiple running stages includes:

[0015] Get the game running image;

[0016] Performing edge detection on the image in the game running image to determine the text area in the image;

[0017] Performing text recognition on each of the text regions to obtain text information;

[0018] Extracting a plurality of key frames corresponding to each stage in the game running image according to the text information;

[0019] The process is performed for each stage respectively, and a video corresponding to the stage is constructed based on the key frames corresponding to the stage.

[0020] In some possible implementations, obtaining the game running image includes:

[0021] Performing image recognition on the displayed game screen to determine the target position to be clicked in the screen;

[0022] Simulating the user's operation at the target location through a preset script, and recording the displayed game screen to obtain a game test video;

[0023] The game test video is cut to determine the game running image.

[0024] In some possible implementations, cutting the game test video to determine the game running image includes:

[0025] Cutting the game test video according to a preset time range to obtain a target video clip;

[0026] The image in the target video clip is extracted based on a preset frequency to obtain a game running image.

[0027] In some possible implementations, the target video frame is one, and for each stage of the video, performing a cross-correlation operation on the target video frame and a reference frame in the video to obtain row and column displacement vector values ​​of the video includes:

[0028] For each stage of the video, perform the following operations:

[0029] Comparing each pixel in the reference frame and the target video frame respectively to obtain the number of different pixels;

[0030] According to the number of distinguished pixels, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain row and column displacement vector values ​​of the video.

[0031] In some possible implementations, there are at least two target video frames, and for each stage of the video, performing a cross-correlation operation on the target video frame and a reference frame in the video to obtain row and column displacement vector values ​​of the video includes:

[0032] For each stage of the video, perform the following operations:

[0033] Comparing the pixels of the reference frame and each of the target video frames respectively, and obtaining the number of different pixels corresponding to each of the target video frames;

[0034] For each target video frame, respectively performing: performing a cross-correlation operation on the target video frame and the reference frame in the video according to the number of distinguished pixels, to obtain row and column displacement vector values ​​of the video;

[0035] Determine a cross-correlation value curve based on the row and column displacement vector values ​​of each video;

[0036] The peak value of the cross-correlation value curve is used as the row-column displacement vector value.

[0037] In some possible implementations, after obtaining the video corresponding to each stage of the game in the multiple running stages, the method further includes:

[0038] Performing grayscale processing on the videos corresponding to each stage to obtain grayscale processed videos;

[0039] For each stage of the video, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain corresponding row and column displacement vector values, including:

[0040] For the video after the grayscale processing at each stage, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain corresponding row and column displacement vector values.

[0041] In a second aspect, the present application provides a game quality assessment device, the device comprising:

[0042] An acquisition module is used to acquire the video corresponding to each stage of the game in multiple running stages;

[0043] A calculation module, for performing a cross-correlation operation on a target video frame and a reference frame in the video of each stage, to obtain a row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is a video frame in the video other than the reference frame;

[0044] A determination module, used to determine the key experience indicators according to the row and column displacement vector values ​​for each stage;

[0045] The summation module is used to perform weighted summation of the key experience indicators at each stage to obtain the quality evaluation indicators of the game.

[0046] In a third aspect, the present application provides a game quality assessment device, comprising: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the game quality assessment method described above.

[0047] In a fourth aspect, the present application provides a computer-readable storage medium having computer program instructions stored thereon, and the computer program instructions, when executed by a processor, implement the game quality assessment method as described above.

[0048] In a fifth aspect, the present application provides a computer program product. When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the game quality assessment method described above.

[0049] The game quality assessment method, device, equipment, storage medium and product provided in the embodiments of the present application obtain the video corresponding to each stage of the game in multiple running stages, and perform cross-correlation operations on the target video frame and the reference frame in the video of each stage to obtain the corresponding row and column displacement vector values. Then, according to the above row and column displacement vector values ​​of each stage, key experience indicators are determined, and then summed to obtain the quality assessment indicators of the game, so that different stages of the game can be evaluated separately, and then the evaluation results of each stage are comprehensively considered to reduce the granularity of the evaluation, so as to accurately evaluate the quality of the game. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The present application can be better understood from the following description of the specific embodiments of the present application in conjunction with the accompanying drawings, in which:

[0051] Other features, objects and advantages of the present application will become more apparent by reading the following detailed description of non-limiting embodiments with reference to the accompanying drawings, in which the same or similar reference numerals represent the same or similar features.

[0052] Figure 1 is a flowchart of a game quality assessment method provided by an embodiment of the present application;

[0053] Figure 2 This is a flowchart of a simulated user operation process provided by an embodiment of the present application;

[0054] Figure 3 This is a flow chart of cutting a video to obtain an image provided by an embodiment of the present application;

[0055] Figure 4 is a structural diagram of a game quality assessment device provided by an embodiment of the present application;

[0056] Figure 5 It is a schematic diagram of the hardware structure of the game quality assessment device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0058] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0059] In order to solve the problems of the prior art, the embodiments of the present application provide a game quality assessment method, device, equipment, storage medium and product. The game quality assessment method provided by the embodiments of the present application is first introduced below.

[0060] Figure 1 FIG. 1 is a flow chart showing a method for evaluating game quality provided by an embodiment of the present application. Figure 1 As shown, the method includes the following steps S101 to S104.

[0061] Step S101: Obtain the video corresponding to each stage of the game in multiple running stages.

[0062] In the specific implementation, first, the game is recorded at different running stages. Each stage can be a specific event or time period in the game. For example, the entire process of the game from the beginning to the end can be recorded, and specific key parts can be selected from it. The video of each recording stage can be stored as an independent file or clip. These video files will be used in subsequent steps.

[0063] Step S102: For each stage of the video, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is a video frame in the video other than the reference frame.

[0064] In the specific implementation, for each stage of the video, one frame is selected from the video as the reference frame, and the other one or more frames are selected as the target frame. The reference frame is the standard frame used for comparison, and the target frame is the frame to be analyzed. When processing the video, the similarity between the target frame and the reference frame is measured by cross-correlation operation.

[0065] As another example, the specific operation of the cross-correlation operation includes converting the reference frame and the target frame into grayscale images to reduce the computational complexity. The similarity between the two images is then calculated using the cross-correlation function. This operation can reveal the translation of the target frame relative to the reference frame. The maximum value position is extracted from the cross-correlation function to determine the displacement of the target frame in the row and column directions relative to the reference frame, that is, the row and column displacement vector value.

[0066] Step S103: for each stage, determine the key experience index according to the row and column displacement vector values.

[0067] In the specific implementation, for each stage, the key experience index can be determined based on the row and column displacement vector values. For example, the smoothness of the video can be measured by calculating the displacement change between each frame. A large displacement may indicate that there is a freeze or frame loss in the video. Or the stability of the picture can be evaluated by analyzing the change of the displacement vector. Frequent displacement changes may indicate that the picture is unstable. The calculated row and column displacement vector values ​​are converted into specific indicator values. For example, the motion smoothness is determined by calculating the average displacement or the maximum displacement of all frames.

[0068] Step S104: weighted summing up the key experience indicators of each stage to obtain a quality evaluation indicator of the game.

[0069] In the specific implementation, weights are set for the key experience indicators of different stages. The weights can be set based on the impact of different stages on the overall game experience. For example, the key mission stage will be given a higher weight. Weighted calculation is performed on the key experience indicators of each stage. Specifically, the key experience indicator of each stage is multiplied by its corresponding weight. Then all weighted indicator values ​​are added together to obtain the final game quality evaluation indicator.

[0070] The game quality assessment method provided in the embodiment of the present application obtains the video corresponding to each stage of the game in multiple running stages, and performs cross-correlation operation on the target video frame and the reference frame in the video of each stage to obtain the corresponding row and column displacement vector values. Then, according to the above row and column displacement vector values ​​of each stage, the key experience indicators are determined, and then summed to obtain the quality assessment indicators of the game, so that the different stages of the game can be evaluated separately, and then the evaluation results of each stage are comprehensively considered to reduce the granularity of the evaluation, so as to accurately evaluate the quality of the game.

[0071] In order to accurately determine the corresponding key experience indicators, in some implementations, the above step S103 may include:

[0072] For each stage, perform steps A1 to B1:

[0073] Step A1: Determine the time delay corresponding to the above stage according to the above row and column displacement vector values.

[0074] In the specific implementation, the stability of the image in the entire stage is measured by calculating the average or sum of the displacement of each frame. Specifically, according to the row and column displacement vector values ​​and the frame rate of the game, the time delay is calculated by dividing the displacement between frames by the frame rate, and the row and column displacement vector values ​​are converted into time delay, thereby determining the delay corresponding to the above stage.

[0075] Step B1: According to the preset mapping relationship between latency and service quality value, the key experience index corresponding to the above stage is determined, and the above key experience index indicates the degree to which the corresponding game running process meets the user's needs.

[0076] In the specific implementation, a mapping relationship or model is predefined to convert the latency value into a service quality value. According to the latency value of each stage, the preset mapping relationship or model is applied to calculate the corresponding service quality value. The service quality value is the key experience indicator, which indicates the degree to which the game operation process at each stage meets the user's needs. For example, if the mapping model determines that a stage with a smaller latency will have a higher service quality score, the experience indicator value of this stage will be higher, and vice versa.

[0077] As another implementation manner, the preset mapping relationship between the delay and the service quality value may be a linear formula, and there is a linear relationship between the delay and the service quality value.

[0078] As another implementation, the preset mapping relationship between the latency and the service quality value may be to set different service quality scores according to different latency ranges. For example, a latency of less than 50 milliseconds is scored as 5, representing "excellent", 50 to 100 milliseconds is scored as 3, representing "good", and more than 100 milliseconds is scored as 0, representing "poor".

[0079] The above implementation method of the embodiment of the present application, after obtaining the row and column displacement vector value, determines the corresponding delay according to the row and column displacement vector value, and thus determines the key experience indicator corresponding to the above stage based on the preset mapping relationship between the delay and the service quality value, and can accurately determine the corresponding key experience indicator according to the delay.

[0080] In order to accurately obtain the video corresponding to each stage, in some implementations, the above step S101 may include steps A2 to E2:

[0081] Step A2: Get the game running image.

[0082] In a specific implementation, during the running of the game, an appropriate time is selected to capture the image, usually at the beginning and end of each stage, or at a fixed time interval. Specifically, the game running image can be obtained by screenshot or screen recording.

[0083] As another implementation method, the game provides an application program interface, which can directly read image data from the game's frame buffer to obtain the game running image.

[0084] As another implementation manner, a stage identifier and a timestamp may be added to each acquired image to facilitate subsequent processing and analysis.

[0085] Step B2: Perform edge detection on the image in the game running image to determine the text area in the image.

[0086] In the specific implementation, a suitable edge detection algorithm is pre-selected, and the edge detection result is used for contour detection to identify the area that may contain text and determine the text area in the image. The coordinate information of the text area is extracted and marked on the original image for subsequent text recognition.

[0087] Step C2: Perform text recognition on each of the above text regions to obtain text information.

[0088] In a specific implementation, optical character recognition is used to identify the marked text area, and the text in the image is converted into a machine-readable text format to obtain text information.

[0089] As another implementation method, the text area can be preprocessed before recognition, such as denoising, binarization, image enhancement, etc., to improve the recognition accuracy of OCR.

[0090] Step D2: extracting a plurality of key frames corresponding to each stage in the game running image according to the text information.

[0091] In the specific implementation, based on the extracted text information, the content of each stage is analyzed, the features related to the key frames are identified, and in the image sequence of each stage, the frames containing important information or events are selected as key frames. These frames usually reflect important moments in the game process. Specifically, key frames can be selected based on criteria such as time interval and image change amount. For example, frames with significant content changes in each stage are selected.

[0092] Step E2: Execute for each stage separately, and construct the video corresponding to the above stage based on the key frames corresponding to the above stage.

[0093] In a specific implementation, a video editing tool is used to arrange key frames in chronological order to generate a video of each stage, thereby constructing a video corresponding to the above stage.

[0094] The above-mentioned implementation method of the embodiment of the present application performs edge detection on the game running image after acquiring the game running image, determines the area containing text in the image, and then performs text recognition on these areas containing text to obtain the text information therein, extracts multiple key frames corresponding to each stage according to the text information, and then constructs the video corresponding to each stage through these key frames, and accurately determines the key frames through text recognition, so as to accurately obtain the video corresponding to each stage.

[0095] In order to quickly obtain the game running image, in some implementations, the above step A2 may include steps A21 to A23:

[0096] Step A21: Perform image recognition on the displayed game screen to determine the target position that needs to be clicked in the above screen.

[0097] In the specific implementation, a set of target images are predefined, which are the key elements that need to be clicked in the game. The images in the target image library are compared with the captured game screen. The position of the target is identified using an image matching algorithm. The coordinates of the target position are calculated based on the matching results. Specifically, the precise position of the target can be determined by comparing the feature points or pixel values ​​of the image, thereby determining the target position that needs to be clicked in the above screen.

[0098] Step A22: Simulate the user's operation at the above-mentioned target position through a preset script, and record the displayed game screen to obtain a game test video.

[0099] In a specific implementation, according to the coordinates recorded in step A21, a preset script is used to simulate clicks or other interactive operations to ensure that these operations can trigger corresponding events in the game. While the script is being executed, a screen recording tool is started to record the process of the simulated operation and the changes in the game screen to obtain a game test video.

[0100] Step A23: Cut the game test video to determine the game running image.

[0101] In a specific implementation, the cutting range in the video is determined according to the game stage or event that needs to be analyzed, and the cutting function in the video editing tool is used to select the start and end time points, extract the relevant game screen fragments, and determine the game running image.

[0102] The above-mentioned implementation method of the embodiment of the present application determines the target position that needs to be clicked in the above-mentioned screen by performing image recognition on the displayed game screen, and then simulates the user's operation at the target position through a preset script, and records the game screen operated by the preset script to obtain a game test video, cuts this game test video, determines the final game running image, and simulates the user operation through the preset script to quickly obtain the game running image.

[0103] In order to reduce the data volume of the game running image, in some implementations, the above step A23 may include steps A231 to A232:

[0104] Step A231: Cut the game test video according to a preset time range to obtain a target video clip.

[0105] In the specific implementation, the start and end time of the video clip is determined according to the test requirements or analysis objectives. The specific time range can be based on the event mark of the test or manually selected. In the video editing tool, mark the start and end time of the cut clip, cut the selected time period from the original video, and obtain the target video clip.

[0106] Step A232: extracting images from the target video clip based on a preset frequency to obtain a game running image.

[0107] In a specific implementation, the frequency of image extraction is defined according to the requirements, for example, a certain number of video frames can be set to be extracted per second. According to this preset frequency, images are extracted from the video clips to obtain game running images.

[0108] The above-mentioned implementation of the embodiment of the present application obtains a target video segment by cutting the game test video based on a preset time range, and then extracts images in the above-mentioned target video segment based on a preset frequency to obtain a game running image. By cutting the game test video and extracting images in the video segment at a preset frequency, the data volume of the game running image can be reduced.

[0109] In order to accurately obtain the row and column displacement vector values, in some implementations, the target video frame is one, and the step S102 may include:

[0110] For each stage of the video, perform the following operations, steps A3 to B3:

[0111] Step A3: Compare the pixels in the reference frame and the target video frame respectively to obtain the number of different pixels.

[0112] In the specific implementation, each pixel in the reference frame and the target video frame is traversed, the color value of each pixel is obtained and compared, and the difference between the pixels is calculated. Specifically, the absolute difference can be used to measure the difference. A difference threshold is set to determine whether the pixel difference exceeds the threshold. If it exceeds, the pixel is considered a "difference pixel". The number of all difference pixels is counted.

[0113] Step B3: According to the above-mentioned number of distinguished pixels, a cross-correlation operation is performed on the target video frame and the reference frame in the above-mentioned video to obtain the row and column displacement vector values ​​of the above-mentioned video.

[0114] In a specific implementation, according to the above-mentioned number of distinguished pixels, the target video frame and the reference frame are calculated by a cross-correlation algorithm, and the result of the cross-correlation operation is used as the row and column displacement vector value of the above-mentioned video.

[0115] The above-mentioned implementation method of the embodiment of the present application obtains the number of distinguishing pixels by respectively comparing each pixel in the above-mentioned reference frame and the above-mentioned target video frame, and then performs cross-correlation operation on the target video frame and the reference frame in the above-mentioned video based on the number of distinguishing pixels to obtain the row and column displacement vector value of the above-mentioned video, thereby accurately obtaining the row and column displacement vector value.

[0116] In order to obtain accurate row and column displacement vector values ​​when there are multiple target video frames, in some implementations, there are at least two target video frames, and the step S102 may include:

[0117] For each stage of the video, perform the following operations, steps A4 to D4:

[0118] Step A4: respectively comparing the pixels of the reference frame and each of the target video frames to obtain the number of different pixels corresponding to each of the target video frames.

[0119] In the specific implementation, each target video frame is extracted from the target video, and each pixel of each target video frame is traversed and compared with the corresponding pixel in the reference frame. The color difference of the pixel is calculated. According to the set threshold, it is determined whether the pixel is considered as a distinguishing pixel. The number of distinguishing pixels in each target video frame is recorded.

[0120] Step B4: for each target video frame, according to the number of distinguished pixels, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector value of the video.

[0121] In a specific implementation, according to the above-mentioned number of distinguished pixels, for each target video frame, a matching degree between the target video frame and the reference frame is calculated by a cross-correlation algorithm to obtain the row and column displacement vector values ​​of the above-mentioned video.

[0122] Step C4: Determine a cross-correlation value curve based on the row and column displacement vector values ​​of each video.

[0123] In a specific implementation, a correlation value curve is generated according to a combination of cross-correlation results of each target video frame.

[0124] Step D4: taking the peak value of the above cross-correlation value curve as the row-column displacement vector value.

[0125] In the specific implementation, the correlation value curve is analyzed to find the peak position of the curve. The peak position corresponds to the displacement of the best match, and the peak position of the curve is used as the final row and column displacement vector value to represent the best alignment displacement between the target video frame and the reference frame.

[0126] The above-mentioned implementation method of the embodiment of the present application performs cross-correlation operation on the target video frame and the reference frame in the above-mentioned video when there are at least two target video frames, determines the cross-correlation value curve, and finally uses the peak value of the above-mentioned cross-correlation value curve as the row and column displacement vector value, so as to obtain accurate row and column displacement vector values ​​when there are multiple target video frames.

[0127] In order to speed up the processing, in some implementations, after the above step S101, the above method may further include:

[0128] The videos corresponding to each of the above stages are respectively gray-scaled to obtain gray-scaled videos.

[0129] In the specific implementation, each frame of each video is traversed and converted into a grayscale image. Grayscale is usually achieved by converting the RGB value of each pixel into a single grayscale value. Each frame after grayscale processing is saved as a new video file or data stream to obtain a grayscale processed video.

[0130] The above step S102 includes:

[0131] For the video after grayscale processing at each stage, a cross-correlation operation is performed on the target video frame and the reference frame in the above video to obtain the corresponding row and column displacement vector values.

[0132] In a specific implementation, a cross-correlation operation is performed on the target video frame and the reference frame in the video after the grayscale processing at each stage to obtain the corresponding row and column displacement vector values.

[0133] The above implementation of the embodiment of the present application can reduce the amount of data that needs to be processed and speed up the processing by graying the videos corresponding to each of the above stages respectively and then performing cross-correlation operations on the grayed videos.

[0134] In one embodiment of the present application, a computer is used to execute this solution to evaluate the quality of mobile games. First, a relevant test script is created on the computer, and then the mobile terminal to be tested is connected. Figure 2 As shown, the script execution specifically includes steps 201 to 203:

[0135] Step 201: Execute the test script.

[0136] Step 202: Perform image recognition through an image recognition framework.

[0137] The target to be clicked is found on the screen through the preset function, and the path of the target image and the recognition accuracy threshold can be specified.

[0138] Step 203: Simulate user operation.

[0139] According to the found target position, a click operation is simulated at this position, and then the process returns to step 202 to identify the image that has changed after the operation.

[0140] In general, run the test script, use Airtest to perform image recognition on the connected device, find the specified target, and perform simulated click operations. This achieves the simulation of user click behavior through Airtest's image recognition function.

[0141] Afterwards, the video can be cut to obtain image data through the Fast Forward MPEG (ffmpeg) technology, and the test video can be cut to obtain relevant image data, saving computing resources and reducing unnecessary resource consumption. Figure 3 As shown, it includes steps 301 to 303:

[0142] Step 301: Perform frame processing on the video.

[0143] Based on the original video, a video containing many key frames is generated, and the video is cut using the ffmpeg cut command. The cut command specifies the start time and end time, or specifies the range of frames to be cut.

[0144] Step 302: Obtain different frames in the video according to the frame parameters that need to be adjusted.

[0145] Use ffmpeg's screenshot command to obtain image data from the cut video and preprocess the image data to extract the required data. The preprocessing steps include image resizing, graying, filtering, edge detection, segmentation, etc.

[0146] Step 303: Save the acquired key frame image.

[0147] Store the extracted data in appropriate locations and store the data in a relational database.

[0148] Then, the game screen data is extracted by image recognition. The ffmpeg framed images are preprocessed. Since the game screen usually contains interference factors such as noise and lighting changes, the game screen needs to be preprocessed before image recognition. For example, the image processing library can be used to grayscale, binarize, reduce noise, enhance edges, and other operations on the image to improve recognition accuracy. In the game screen, it is necessary to determine the text area containing the required data. Image processing techniques such as edge detection and morphological operations can be used to detect the area where the text is located, and to extract and segment the text area. Text recognition is performed for each extracted text area. Each text area can be used as input for text recognition, and character recognition and text reconstruction are performed for each text area, and the recognition results are output. The output text recognition results are processed and parsed. According to the characteristics of the game screen data, additional processing steps may be required, such as removing spaces, filtering irrelevant characters, and converting data formats. The extracted game frame data is stored in the database. According to specific evaluation requirements, further data processing, analysis and modeling can be performed to evaluate the game perception.

[0149] Experience modeling is performed for game delays, such as game startup delay, game loading delay, game login delay, etc. For example, if the test fails, the corresponding experience service quality (Quality of Service, QOS) value is 0, which means that the game service fails and the user experience is 0. Since users usually do not wait for a service indefinitely, the key experience indicator (Key Experience Indicator, KEI) of the service experience usually also has a threshold value. When the experience KEI is worse than the threshold value, it can be considered that the user has terminated and restarted the service, the service is not completed, and the experience is also 0. Typical situations include game startup failure, long game loading delay or unsuccessful login. Among them, if the delay of the picture during the game is lower than the limit of human visual response, it is uniformly defined as full score, and the domain value is set to 100ms.

[0150] Experience QOS is used to quantify the user's subjective perception corresponding to different service experience KEI results into a unified objective value. The QOS value in this standard divides the subjective feeling of Internet service experience into 5 levels, and the corresponding scores can be shown in Table 1.

[0151] Table 1

[0152] QOS Value Experience Degree of experience damage 4.5～5 excellent Ideally no experience loss 3.5～4.5 good Slightly noticeable impairment 2.5～3.5 middle Acceptable experience loss 1.5～2.5 Difference Severe impairment of user experience 0～1.5 inferior Unbearable experience damage

[0153] Specifically, a frame image is selected as the starting time Ts, and the current frame image and the reference frame image (the reference frame image is the starting node image of the definition index, such as the reference frame image of the startup, and the reference frame image of the startup completion) are grayscale projected as Te, and the difference i between the two is calculated as the number of samples, and the peak of the cross-correlation value curve is selected as the row and column displacement vector value of the two frames of images. The relative motion degree of the two frames of images is judged according to the absolute value of this vector, and a threshold is set. Whether jitter occurs or not is judged according to the threshold. This method directly processes the grayscale image, with a small amount of calculation and a more accurate diagnosis effect.

[0154] The game startup delay algorithm is divided into two steps: image grayscale mapping and cross-correlation operation. The specific calculation formula is as follows:

[0155]

[0156] Where Ts represents the reference frame image, Te represents the current frame image, i represents the difference in the number of pixels between the reference frame image and the previous frame image, and the game startup delay is: The key reference points for the experience QOS of the game startup delay mapping are: within 1.5s, with a full score of 5 points; 2.5s passing score, 3 points; above 2.5s, the minimum score is 0 points.

[0157] The game loading delay algorithm is divided into two steps: image grayscale mapping and cross-correlation operation. The specific calculation formula is as follows:

[0158]

[0159] Where Ts represents the reference frame image, Te represents the current frame image, k represents the difference in the number of pixels between the reference frame image and the previous frame image, and the game loading delay is: The key reference points for the experience QOS of the game loading delay mapping are: within 3.5s, with a full score of 5 points; 5.5s, a passing score of 3 points; 8s to the lowest score of 0 points.

[0160] The game login delay algorithm is divided into two steps: image grayscale mapping and cross-correlation operation.

[0161]

[0162] Where Ts represents the reference frame image, Te represents the current frame image, j represents the difference in the number of pixels between the reference frame image and the previous frame image, and the game login delay is:

[0163] The key reference points for the experience QOS of the game login delay mapping are: within 1s, with a full score of 5 points; 1.5s, a passing score of 3 points; 5s, the lowest score of 0 points.

[0164] The game login success rate algorithm is divided into one step: image grayscale mapping association operation. The specific calculation formula is as follows:

[0165]

[0166] The key reference points of the experience QOS mapped by the game login success rate are: 100%, full score, and 0 points for failure.

[0167] Each KEI indicator in the game experience monitoring indicator set is selected as the feature input. The feature coefficient ε needs to be trained for the above KEI indicators. Specifically:

[0168]

[0169] The gaming experience QoS value is obtained by calculating each KEI, specifically:

[0170]

[0171] The mobile game experience is evaluated based on the QOS experience value. The experience is excellent between 4.5 and 5 points, good between 3.5 and 4.5 points, medium between 2.5 and 3.5 points, poor between 1.5 and 2.5 points, and poor between 0 and 1.5 points.

[0172] Based on the above method, the network experience of mobile games can be monitored. Through a large number of tests and analyses every day, game problems can be discovered before users do. Comparative analysis of network experience with competitors can be conducted to promptly discover and resolve network problems of mobile games. Game network experience monitoring and optimization can be continuously carried out to enhance user gaming experience and satisfaction.

[0173] Based on the game quality assessment method provided in the above embodiment, the present application also provides a specific implementation of a game quality assessment device. Please refer to the following embodiment.

[0174] See first Figure 4 The game quality assessment device 400 provided in the embodiment of the present application includes the following modules:

[0175] The acquisition module 401 is used to acquire the video corresponding to each stage of the game in multiple running stages.

[0176] The operation module 402 is used to perform cross-correlation operation on the target video frame and the reference frame in the video of each stage to obtain the row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is the video frame in the video except the reference frame.

[0177] The determination module 403 is used to determine the key experience index according to the row and column displacement vector values ​​for each stage.

[0178] The summing module 404 is used to perform weighted summing of the key experience indicators of each stage to obtain a quality evaluation indicator of the game.

[0179] The game quality assessment device provided in the embodiment of the present application obtains the video corresponding to each stage of the game in multiple running stages, and performs cross-correlation operation on the target video frame and the reference frame in the video of each stage to obtain the corresponding row and column displacement vector values. Then, according to the above row and column displacement vector values ​​of each stage, the key experience indicators are determined, and then summed up to obtain the quality assessment indicators of the game, so that the different stages of the game can be evaluated separately, and then the evaluation results of each stage are comprehensively considered to reduce the granularity of the evaluation, so as to accurately evaluate the quality of the game.

[0180] As an implementation of the present application, the determination module 403 may include:

[0181] The determination unit is used to determine the time delay corresponding to the above stage according to the above row and column displacement vector value.

[0182] The determination unit is also used to determine the key experience indicators corresponding to the above stages according to the preset mapping relationship between the delay and the service quality value, and the above key experience indicators represent the degree to which the corresponding game running process meets the user needs.

[0183] The above implementation method of the embodiment of the present application, after obtaining the row and column displacement vector value, determines the corresponding delay according to the row and column displacement vector value, and thus determines the key experience indicator corresponding to the above stage based on the preset mapping relationship between the delay and the service quality value, and can accurately determine the corresponding key experience indicator according to the delay.

[0184] As an implementation of the present application, the acquisition module 401 may include:

[0185] The acquisition unit is used to acquire the game running image.

[0186] The detection unit is used to perform edge detection on the image in the game running image to determine the text area in the image.

[0187] The recognition unit is used to perform text recognition on each of the above text areas to obtain text information.

[0188] The extraction unit is used to extract multiple key frames corresponding to each stage in the game running image according to the text information.

[0189] The construction unit is used to execute for each stage respectively, and construct the video corresponding to the above stage based on the key frames corresponding to the above stage.

[0190] The above-mentioned implementation method of the embodiment of the present application performs edge detection on the game running image after acquiring the game running image, determines the area containing text in the image, and then performs text recognition on these areas containing text to obtain the text information therein, extracts multiple key frames corresponding to each stage according to the text information, and then constructs the video corresponding to each stage through these key frames, and accurately determines the key frames through text recognition, so as to accurately obtain the video corresponding to each stage.

[0191] As an implementation of the present application, the acquisition unit may include:

[0192] The recognition subunit is used to perform image recognition on the displayed game screen and determine the target position that needs to be clicked in the above screen.

[0193] The recording subunit is used to simulate the user's operation at the above-mentioned target position through a preset script, and simultaneously record the displayed game screen to obtain a game test video.

[0194] The cutting subunit is used to cut the above-mentioned game test video to determine the game running image.

[0195] The above-mentioned implementation method of the embodiment of the present application determines the target position that needs to be clicked in the above-mentioned screen by performing image recognition on the displayed game screen, and then simulates the user's operation at the target position through a preset script, and records the game screen operated by the preset script to obtain a game test video, cuts this game test video, determines the final game running image, and simulates the user operation through the preset script to quickly obtain the game running image.

[0196] As an implementation of the present application, the shearing subunit may include:

[0197] The cutting subunit is used to cut the game test video according to a preset time range to obtain a target video clip.

[0198] The extraction subunit is used to extract the image in the above-mentioned target video clip based on a preset frequency to obtain a game running image.

[0199] The above-mentioned implementation of the embodiment of the present application obtains a target video segment by cutting the game test video based on a preset time range, and then extracts images in the above-mentioned target video segment based on a preset frequency to obtain a game running image. By cutting the game test video and extracting images in the video segment at a preset frequency, the data volume of the game running image can be reduced.

[0200] As an implementation of the present application, the operation module 402 may include:

[0201] The comparison unit is used to compare each pixel in the reference frame and the target video frame respectively to obtain the number of different pixels.

[0202] The operation unit is used to perform a cross-correlation operation on the target video frame and the reference frame in the above video according to the above-mentioned number of distinguished pixels, so as to obtain the row and column displacement vector value of the above video.

[0203] The above-mentioned implementation method of the embodiment of the present application obtains the number of distinguishing pixels by respectively comparing each pixel in the above-mentioned reference frame and the above-mentioned target video frame, and then performs cross-correlation operation on the target video frame and the reference frame in the above-mentioned video based on the number of distinguishing pixels to obtain the row and column displacement vector value of the above-mentioned video, thereby accurately obtaining the row and column displacement vector value.

[0204] As an implementation of the present application, the operation module 402 may include:

[0205] The comparison unit is used to compare the pixels of the reference frame and each of the target video frames respectively, and obtain the number of different pixels corresponding to each of the target video frames.

[0206] The operation unit is used to perform, for each target video frame, the following operations: performing cross-correlation operation on the target video frame and the reference frame in the video according to the number of distinguished pixels, to obtain row and column displacement vector values ​​of the video.

[0207] The determination unit is used to determine the cross-correlation value curve based on the row and column displacement vector values ​​of each video.

[0208] The determination unit is further configured to use the peak value of the cross-correlation value curve as the row-column displacement vector value.

[0209] The above-mentioned implementation method of the embodiment of the present application performs cross-correlation operation on the target video frame and the reference frame in the above-mentioned video when there are at least two target video frames, determines the cross-correlation value curve, and finally uses the peak value of the above-mentioned cross-correlation value curve as the row and column displacement vector value, so as to obtain accurate row and column displacement vector values ​​when there are multiple target video frames.

[0210] As an implementation of the present application, the game quality evaluation device 400 may further include:

[0211] The processing module is used to perform grayscale processing on the videos corresponding to each of the above stages to obtain grayscale processed videos.

[0212] The operation module is used for performing cross-correlation operation on the target video frame and the reference frame in the video after the grayscale processing in each stage, so as to obtain the corresponding row and column displacement vector values.

[0213] The above implementation of the embodiment of the present application can reduce the amount of data that needs to be processed and speed up the processing by graying the videos corresponding to each of the above stages respectively and then performing cross-correlation operations on the grayed videos.

[0214] Each module in the game quality assessment device provided in the embodiment of the present application can implement each step in the above-mentioned game quality assessment method and achieve corresponding effects. For the sake of concise description, it will not be repeated here.

[0215] Figure 5 A schematic diagram of the structure of the game quality assessment hardware provided in an embodiment of the present application is shown.

[0216] The game quality assessment device may include a processor 501 and a memory 502 storing computer program instructions.

[0217] Specifically, the processor 501 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0218] The memory 502 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 502 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 502 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 502 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 502 is a non-volatile solid-state memory.

[0219] The memory may include a read-only memory (ROM), a random access memory (RAM), a disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Therefore, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the game quality assessment method according to any one embodiment of the present disclosure.

[0220] The processor 501 implements any one of the game quality assessment methods in the above embodiments by reading and executing computer program instructions stored in the memory 502 .

[0221] In one example, the game quality assessment device may further include a communication interface 503 and a bus 510. Figure 5 As shown, the processor 501, the memory 502, and the communication interface 503 are connected via a bus 510 and communicate with each other.

[0222] The communication interface 503 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0223] Bus 510 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 510 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0224] In addition, in combination with the method for evaluating game quality in the above embodiments, the present application embodiment may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any one of the game quality evaluation methods in the above embodiments is implemented.

[0225] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed, implements any one of the game quality assessment methods in the above embodiments.

[0226] It should be clear that the present application is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present application is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present application.

[0227] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0228] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0229] Aspects of the present disclosure are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field programmable logic circuit. It can also be understood that each box in the block diagram and / or flowchart and the combination of boxes in the block diagram and / or flowchart can also be implemented by dedicated hardware that performs a specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions.

[0230] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, modules and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this application, and these modifications or replacements should be included in the protection scope of this application.

Claims

1. A game quality assessment method, It is characterized in that include: Get the video corresponding to each stage of the game in multiple running stages; For each stage of the video, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector value of the video; the reference frame is any video frame in the video; The target video frame is a video frame in the video other than the reference frame; For each stage, determine the key experience indicators based on the row and column displacement vector values; The key experience indicators of each stage are weighted and summed to obtain the quality evaluation index of the game.

2. The game quality assessment method according to claim 1, It is characterized in that For each stage, the key experience indicators are determined according to the row and column displacement vector values, including: For each stage, perform the following operations: Determining the time delay corresponding to the stage according to the row and column displacement vector values; According to the preset mapping relationship between latency and service quality value, the key experience index corresponding to the stage is determined, and the key experience index represents the degree to which the corresponding game running process meets the user needs.

3. The game quality assessment method according to claim 1, It is characterized in that The obtaining of the video corresponding to each stage of the game in the multiple running stages includes: Get the game running image; Performing edge detection on the image in the game running image to determine the text area in the image; Performing text recognition on each of the text regions to obtain text information; Extracting a plurality of key frames corresponding to each stage in the game running image according to the text information; The process is performed for each stage respectively, and a video corresponding to the stage is constructed based on the key frames corresponding to the stage.

4. The game quality assessment method according to claim 3, It is characterized in that The obtaining of the game running image comprises: Performing image recognition on the displayed game screen to determine the target position to be clicked in the screen; Simulating the user's operation at the target location through a preset script, and recording the displayed game screen to obtain a game test video; The game test video is cut to determine the game running image.

5. The game quality assessment method according to claim 4, It is characterized in that The step of cutting the game test video to determine the game running image includes: Cutting the game test video according to a preset time range to obtain a target video clip; The image in the target video clip is extracted based on a preset frequency to obtain a game running image.

6. The game quality assessment method according to claim 1, It is characterized in that The target video frame is one, and for each stage of the video, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector values ​​of the video, including: For each stage of the video, perform the following operations: Comparing each pixel in the reference frame and the target video frame respectively to obtain the number of different pixels; According to the number of distinguished pixels, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain row and column displacement vector values ​​of the video.

7. The game quality assessment method according to claim 1, It is characterized in that There are at least two target video frames, and for each stage of the video, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector values ​​of the video, including: For each stage of the video, perform the following operations: Comparing the pixels of the reference frame and each of the target video frames respectively, and obtaining the number of different pixels corresponding to each of the target video frames; For each target video frame, respectively performing: performing a cross-correlation operation on the target video frame and the reference frame in the video according to the number of distinguished pixels, to obtain row and column displacement vector values ​​of the video; Determine a cross-correlation value curve based on the row and column displacement vector values ​​of each video; The peak value of the cross-correlation value curve is used as the row-column displacement vector value.

8. The game quality assessment method according to any one of claims 1 to 7, It is characterized in that After obtaining the video corresponding to each stage of the game in the multiple running stages, the method further includes: Performing grayscale processing on the videos corresponding to each stage to obtain grayscale processed videos; For each stage of the video, cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain the row and column displacement vector values ​​of the video, including: For the video after the grayscale processing at each stage, a cross-correlation operation is performed on the target video frame and the reference frame in the video to obtain corresponding row and column displacement vector values.

9. A game quality assessment device, It is characterized in that The device comprises: An acquisition module is used to acquire the video corresponding to each stage of the game in multiple running stages; A calculation module, for performing a cross-correlation operation on a target video frame and a reference frame in the video of each stage, to obtain a row and column displacement vector value of the video; the reference frame is any video frame in the video; the target video frame is a video frame in the video other than the reference frame; A determination module is used to determine the key experience indicators according to the row and column displacement vector values ​​for each stage; The summation module is used to perform weighted summation of the key experience indicators at each stage to obtain the quality evaluation indicators of the game.

10. A game quality assessment device, It is characterized in that The device includes: a processor, and a memory storing computer program instructions; the processor reads and executes the computer program instructions to implement the game quality assessment method as described in any one of claims 1-8.

11. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the game quality assessment method according to any one of claims 1 to 8 is implemented.

12. A computer program product, It is characterized in that When the instructions in the computer program product are executed by a processor of an electronic device, the electronic device executes the game quality assessment method as described in any one of claims 1-8.

Images