Application starting time measuring method and system based on image recognition and electronic equipment

Abstract

The invention provides an application starting time measuring method and system based on image recognition and electronic equipment. The method comprises the following steps: processing a starting video of a target application so as to decompose the starting video into a plurality of continuous image frames; analyzing the plurality of image frames by using an image recognition algorithm so as to recognize a starting start frame and a loading end frame for starting the target application from the plurality of image frames; calculating the starting time of the target application according to the starting start frame and the loading end frame; according to the method and the device, the acquired image frame is intelligently analyzed by utilizing an image recognition technology, and the key frame change of the target application in the starting process can be accurately captured, so that the starting time of the target application can be measured based on the key frame; and the method does not depend on log data of an operating system or timestamp information provided by a developer.

Description

Technical Field

[0001] The present invention relates to the field of intelligent devices, relates to the startup of applications, and particularly relates to the startup time of applications. Specifically, it is a method, system, and electronic device for measuring the startup time of applications based on image recognition. Background Art

[0002] With the popularization of intelligent devices, the user experience has become one of the important considerations in application development. The application startup time (i.e., the time from when the user clicks on the application icon to when the application is fully loaded) directly affects the user's perceived experience. A longer startup time often leads to user loss and a poor usage experience. Therefore, how to accurately and seamlessly measure the application startup time has become a key issue in application performance optimization.

[0003] Traditional startup time measurement methods usually rely on system internal timing and log analysis. These methods estimate the startup time by capturing timestamps inside the operating system or application. However, these methods have certain limitations. For example, system logs may not accurately reflect all application interface changes, or are restricted by operating system permissions and cannot obtain the complete startup process. In addition, the log-based measurement method has relatively limited perception ability for the user interface and cannot directly reflect the actual process of application interface loading. Summary of the Invention

[0004] The purpose of the present invention is to provide a method, system, and electronic device for measuring the startup time of applications based on image recognition to solve the problems pointed out in the above background art.

[0005] In a first aspect, the present invention provides a method for measuring the startup time of applications based on image recognition. The method includes: processing the startup video of the target application to decompose the startup video into a plurality of consecutive image frames; each of the image frames corresponds to a moment in the startup video; the target application is installed on an intelligent device; the screen of the intelligent device is used to display the interface changes of the target application during startup; the startup video at least includes the interface changes of the target application during startup; using an image recognition algorithm to analyze the plurality of image frames to identify the startup start frame and the loading end frame of the startup of the target application from the plurality of image frames; calculating the startup time of the target application according to the startup start frame and the loading end frame.

[0006] In the present invention, by using image recognition technology to intelligently analyze the obtained image frames, it is possible to accurately capture the key frame (including the startup start frame and the loading end frame) changes of the target application during startup. Thus, based on this key frame, it is possible to measure the startup time of the target application without relying on the log data of the operating system or the timestamp information provided by the developer.

[0007] In an implementation of the first aspect, the interface changes during the startup process of the target application at least include: the initial interface when the target application starts up and the final interface when the target application finishes loading.

[0008] In an implementation of the first aspect, after the step of decomposing the startup video into multiple consecutive image frames and before the step of calculating the startup time of the target application based on the startup start frame and the loading end frame, the method further includes: numbering the multiple image frames in time sequence; calculating the startup time of the target application based on the startup start frame and the loading end frame includes: calculating the startup time based on the number of the startup start frame, the number of the loading end frame, and the frame interval time; wherein, the frame interval time at least depends on the frame rate of the startup video.

[0009] In an implementation of the first aspect, the analyzing the multiple image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame when the target application starts up includes: comparing the similarity between each image frame and a baseline start frame based on the image recognition algorithm, denoted as the first similarity; determining whether the first similarity meets a first preset condition, and determining the image frame corresponding to the first similarity that meets the first preset condition as the startup start frame; comparing the similarity between each image frame and a baseline end frame based on the image recognition algorithm, denoted as the second similarity; determining whether the second similarity meets a second preset condition, and determining the image frame corresponding to the second similarity that meets the second preset condition as the loading end frame.

[0010] In an implementation of the first aspect, the image recognition algorithm at least includes any one of the following algorithms: histogram comparison algorithm, SIFT feature extraction algorithm, ORB feature extraction algorithm.

[0011] In an implementation of the first aspect, the method further includes: obtaining the startup times corresponding to different test conditions respectively; and / or dividing the startup process into at least two stages according to the image recognition algorithm and the multiple image frames, and determining the time consumption of each stage.

[0012] In this implementation, by obtaining the startup times of the target application corresponding to different test conditions, and / or by dividing the startup process of the target application into stages and determining the time consumption of each stage, it is convenient for subsequent developers to optimize the target application according to the startup times and / or the time consumption under different test conditions.

[0013] Second aspect, the present invention provides an application startup time measurement system based on image recognition, the system comprising: a video processing module, configured to process a startup video of a target application to decompose the startup video into a plurality of consecutive image frames; each of the image frames corresponds to a moment in the startup video; the target application is installed on a smart device; a screen of the smart device is configured to display interface changes of the target application during startup; the startup video at least includes interface changes of the target application during startup; an image recognition module, configured to analyze the plurality of image frames by using an image recognition algorithm to identify a startup start frame and a loading end frame of the startup of the target application from the plurality of image frames; a time calculation module, configured to calculate a startup time of the target application according to the startup start frame and the loading end frame.

[0014] Third aspect, the present invention provides an electronic device, the electronic device comprising: a processor and a memory; the memory is configured to store a computer program; the processor is configured to execute the computer program stored in the memory so that the electronic device executes the above-mentioned application startup time measurement method based on image recognition.

[0015] As described above, the application startup time measurement method, system and electronic device based on image recognition according to the present invention have the following beneficial effects:

[0016] (1) Compared with the prior art, the present invention aims to monitor interface changes of an application during startup by means of image recognition technology without relying on operating system log data or timestamp information provided by a developer, so as to accurately detect changes in application interface loading, and further accurately calculate the startup time of the application, improving the accuracy of startup time measurement.

[0017] (2) The present invention does not rely on a log mechanism of a specific platform and can measure the startup time of an application across different operating systems (such as Android and iOS), ensuring wide applicability and flexibility.

[0018] (3) The present invention can run independently and does not require a developer to build a time measurement function into the application, is applicable to third-party performance testing, simplifies the testing process, and reduces the workload of development and maintenance. Description of the Drawings

[0019] Figure 1 It shows a flowchart of the application startup time measurement method based on image recognition according to an embodiment of the present invention.

[0020] Figure 2It shows a schematic diagram of using an image recognition algorithm to analyze multiple image frames to identify the start frame and the end frame of loading for the startup of a target application as described in the embodiments of the present invention.

[0021] Figure 3 It shows a schematic diagram of dividing the startup process into stages as described in the embodiments of the present invention.

[0022] Figure 4 It shows a schematic diagram of an application startup time measurement system based on image recognition as described in the embodiments of the present invention. Detailed implementation manners

[0023] The following uses specific specific examples to illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0024] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. The diagrams only show the components related to the present invention, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be an arbitrary change, and the component layout type may also be more complex.

[0025] Refer to Figures 1 to 4 . The following embodiments of the present invention provide an application startup time measurement method, system, and electronic device based on image recognition. Compared with the prior art, the present invention aims to, without relying on the log data of the operating system or the timestamp information provided by the developer, with the help of image recognition technology, monitor the interface changes during the startup process of the application, so as to accurately detect the changes in the application interface loading, and then accurately calculate the startup time of the application, improving the accuracy of startup time measurement; the present invention does not rely on the log mechanism of a specific platform and can measure the startup time of applications across different operating systems (such as Android and iOS), ensuring wide applicability and flexibility; the present invention can run independently and does not require the developer to build a time measurement function into the application, is suitable for third-party performance testing, simplifies the testing process, and reduces the development and maintenance workload.

[0026] It should be noted that an application, namely Application, abbreviated as APP, generally refers to applications for mobile phones and tablets; it is usually classified into personal user applications and enterprise-level applications in terms of the target audience, and mainly includes iOS (such as: Synchronous Push, etc.), Android (such as: AirDroid, Baidu Applications, etc.) and xap and appx for windows phone in terms of mobile operating system classification.

[0027] Next, the technical solutions in the embodiments of the present invention will be described in detail with reference to the accompanying drawings in the embodiments of the present invention.

[0028] As Figure 1 shown, in one embodiment, the present invention provides a method for measuring the application startup time based on image recognition, which is applied to an electronic device.

[0029] Specifically, the method includes:

[0030] Step S1: Process the startup video of the target application to decompose the startup video into a plurality of consecutive image frames.

[0031] In this embodiment, the target application is installed on a smart device; the screen of the smart device is used to display the interface changes during the startup process of the target application; the startup video at least includes the interface changes during the startup process of the target application.

[0032] It should be noted that the smart device is an ordinary smart device, such as a smart phone, a tablet or an emulator, etc.

[0033] In one embodiment, the startup of the target application is achieved through an automated method.

[0034] In one embodiment, the interface changes during the startup process of the target application at least include: the initial interface when the target application starts and the final interface when the target application finishes loading.

[0035] Specifically, the process from the initial interface to the final interface is defined as the startup process of the target application.

[0036] In one embodiment, the startup video meets the preset format requirements.

[0037] In one embodiment, the preset format requirements at least include but are not limited to the following formats: MP4 and / or AVI.

[0038] In one embodiment, the electronic device is a server.

[0039] In this embodiment, each image frame corresponds to a moment in the startup video to capture the screen changes during the startup process of the target application.

[0040] In one embodiment, a startup video is parsed by a video parsing tool to decompose the startup video into a plurality of consecutive image frames (as shown in Figure 2 and Figure 3 , the startup video is decomposed into image frame 1, image frame 2... image frame n... image frame n + x), and the plurality of consecutive image frames are associated with each other before and after.

[0041] In one embodiment, the video parsing tool uses FFmpeg.

[0042] It should be noted that FFmpeg is a free and open-source cross-platform multimedia processing tool. It supports almost all video formats and coding standards. FFmpeg includes a series of sub-projects and tools, such as the ffmpeg command-line tool for transcoding and processing video and audio files.

[0043] In one embodiment, each image frame is marked with a timestamp to ensure that the time information of each image frame is consistent with the recording time of the startup video, thereby providing accurate data for the subsequent calculation of the startup time.

[0044] In one embodiment, the multiple image frames are sorted in order to prepare for subsequent image recognition.

[0045] Step S2: Analyze the multiple image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame of the target application startup from the multiple image frames.

[0046] It should be noted that image recognition technology, as an efficient computer vision technology, has been widely applied to various scenarios; in the present invention, by applying it to the measurement of the application startup time, through the intelligent analysis of the image frames, the subtle changes in the application interface can be accurately captured, especially the changes in the key frames (including the startup start frame and the loading end frame) during the application startup process, making the subsequent startup time calculated based on the key frames more accurate.

[0047] As Figure 2 shown, in one embodiment, the analyzing the multiple image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame of the target application startup from the multiple image frames includes:

[0048] Step S21: Compare the similarity between each image frame and the baseline start frame based on the image recognition algorithm, denoted as the first similarity.

[0049] Step S22: Determine whether the first similarity meets a first preset condition, and determine the image frame corresponding to the first similarity that meets the first preset condition as the startup start frame.

[0050] In one embodiment, the first preset condition is that the first similarity is greater than the first preset threshold.

[0051] It should be noted that what the specific value of the first preset threshold is set to does not limit the conditions of the present invention. In practical applications, it can be set according to specific application scenarios.

[0052] In one embodiment, the first preset threshold is not fixed, that is, it can be adaptively adjusted.

[0053] For example, in order to adapt to different video qualities and recording conditions, the first preset threshold can be automatically adjusted, starting from a higher threshold and gradually decreasing until the starting frame of activation is accurately matched; this adjustment mechanism can improve the accuracy of matching and avoid errors caused by differences in recording quality.

[0054] Specifically, the similarities between each image frame and the baseline starting frame are calculated through the above steps, so as to find the frame most similar to the baseline starting frame from multiple image frames as the starting frame of activation.

[0055] Step S23: Compare the similarity between each image frame and the baseline ending frame based on the image recognition algorithm, denoted as the second similarity.

[0056] Step S24: Determine whether the second similarity meets the second preset condition, and determine the image frame corresponding to the second similarity that meets the second preset condition as the ending frame of loading.

[0057] In one embodiment, the second preset condition is that the second similarity is greater than the second preset threshold.

[0058] It should be noted that what the specific value of the second preset threshold is set to does not limit the conditions of the present invention. In practical applications, it can be set according to specific application scenarios.

[0059] In one embodiment, the second preset threshold is not fixed, that is, it can be adaptively adjusted.

[0060] For example, in order to adapt to different video qualities and recording conditions, the second preset threshold can be automatically adjusted, starting from a higher threshold and gradually decreasing until the ending frame of loading is accurately matched; this adjustment mechanism can improve the accuracy of matching and avoid errors caused by differences in recording quality.

[0061] Specifically, the similarities between each image frame and the baseline ending frame are calculated through the above steps, so as to find the frame most similar to the baseline ending frame from multiple image frames as the ending frame of loading.

[0062] Generally, among the multiple image frames obtained in step S1, there may be multiple frames that match the baseline start frame or the baseline end frame (i.e., meet the above first preset condition or second preset condition).

[0063] In one embodiment, the last frame that meets the first preset condition (the last frame among all the image frames that meet the first preset condition) is taken as the start start frame.

[0064] In one embodiment, the first frame that meets the second preset condition (the first frame among all the image frames that meet the second preset condition) is taken as the loading end frame.

[0065] It should be noted that when initially using the electronic device to calculate the startup time of the target application, the above baseline start frame and baseline end frame are both manually marked by the user; specifically, the user manually marks the baseline start frame (usually the initial interface when the application starts, such as the display of the application icon or the start of the startup animation) and the baseline end frame (usually the final interface when the application loading is completed, marking the end of the startup process) on the startup video; after the user first marks the baseline start frame and baseline end frame, the electronic device will save them; in subsequent tests, the electronic device will automatically perform image recognition based on the manually marked baseline start frame and baseline end frame.

[0066] It should be noted that the above step S21 and step S22, as a whole, are used to determine the start start frame; step S23 and step S24, as a whole, are used to determine the loading end frame; among them, the steps for determining the start start frame (step S21 and step S22) and the steps for determining the loading end frame (step S23 and step S24) can be executed simultaneously or in a sequential order, that is, first execute the steps for determining the start start frame, and then execute the steps for determining the loading end frame, or first execute the steps for determining the loading end frame, and then execute the steps for determining the start start frame.

[0067] In one embodiment, the image recognition algorithm includes at least, but is not limited to, any one of the following algorithms: histogram comparison algorithm, SIFT feature extraction algorithm, ORB feature extraction algorithm, to ensure the accuracy of matching the start start frame and the loading end frame even under different devices or different recording conditions.

[0068] Among them, the histogram comparison algorithm: is used for cross-sections with obvious color changes. By adopting the method of comparing color histograms, it calculates the color distribution difference between each frame image and the baseline image, and matches the frames with higher similarity, so as to quickly determine the start start frame and the loading end frame.

[0069] SIFT (Scale-Invariant Feature Transform) feature extraction algorithm: A description used in the field of image processing. This description has scale invariance and can detect key points in an image. It is a local feature descriptor. Specifically, when there are significant rotations, scalings, or lighting changes in the interface of the target application, SIFT provides strong robustness by extracting local feature points and descriptors in the image, adapting to complex backgrounds and a startup process with large variations.

[0070] ORB (Oriented FAST and Rotated BRIEF) feature extraction algorithm: Can be used to quickly create feature vectors for key points in an image, and these feature vectors can be used to identify objects in the image. In real-time scenarios requiring high computational efficiency, ORB provides a high computational speed by combining FAST corner detection and BRIEF descriptors, while ensuring the accuracy of feature matching, and is applicable to the measurement of fast startup times in different devices and environments.

[0071] It should be noted that both the SIFT and ORB algorithms extract local feature points of an image and perform matching through feature point matching. These two algorithms can still maintain a good matching effect in the case of image rotation, scaling, or lighting changes.

[0072] It should be noted that in practical applications, the most suitable image recognition algorithm can be dynamically selected according to specific application scenarios (video quality, recording environment) and device conditions. For example: when the quality of the recorded video is low or there are lighting changes, the SIFT or ORB algorithm can be selected to ensure accuracy; when the video quality is high and the interface color changes significantly, the histogram comparison algorithm can be preferentially used to improve computational efficiency; at the same time, the thresholds for similarity calculation (including the above-mentioned first preset threshold and second preset threshold) can also be dynamically adjusted according to the actual situation to ensure the accuracy of the recognition result under different conditions.

[0073] In one embodiment, after step S1 and before step S2, the method further includes: preprocessing multiple image frames to enable, in step S2, the use of an image recognition algorithm to analyze the multiple preprocessed image frames.

[0074] Specifically, the analysis of the preprocessed image frames using an image recognition algorithm has the same principle as the above direct analysis of the image frames, so it will not be elaborated in detail here.

[0075] In one embodiment, the preprocessing includes at least but is not limited to the following processing methods: denoising and / or grayscaling.

[0076] It should be noted that by preprocessing the image frames, the contrast and detail features of the images are enhanced, thereby improving the recognition accuracy of subsequent image recognition.

[0077] Step S3: Calculate the startup time of the target application according to the startup start frame and the loading end frame.

[0078] In one embodiment, the unit of the startup time is seconds.

[0079] In one embodiment, the startup time is calculated according to the timestamps of the startup start frame and the loading end frame.

[0080] Specifically, the startup time is the timestamp of the loading end frame minus the timestamp of the startup start frame.

[0081] In one embodiment, after the step of decomposing the startup video into multiple consecutive image frames and before the step of calculating the startup time of the target application according to the startup start frame and the loading end frame, the method further includes: numbering the multiple image frames in time sequence.

[0082] In this embodiment, calculating the startup time of the target application according to the startup start frame and the loading end frame includes: calculating the startup time according to the number of the startup start frame, the number of the loading end frame and the frame interval time (that is, the time required for the final interface to be fully displayed from the initial interface at the start of the application to the completion of loading).

[0083] Specifically, startup time = (number of loading end frame - number of startup start frame) × frame interval time.

[0084] In one embodiment, the frame interval time depends at least on the frame rate of the startup video.

[0085] For example, in one embodiment, the frame rate of the startup video is 30 frames per second, then the frame interval time is 1 / 30 second.

[0086] It should be noted that the higher the frame rate, the more accurate the calculation of the startup time.

[0087] In one embodiment, the method further includes: generating a data report based on the startup time.

[0088] In one embodiment, the data report includes at least but is not limited to the startup time.

[0089] In one embodiment, the method further includes: obtaining the startup times corresponding to different test conditions respectively.

[0090] Specifically, under different test conditions, by repeating the above method, the startup times of the corresponding target applications are calculated respectively.

[0091] In one embodiment, the different test conditions include at least any one or two or more combinations of the following: different smart devices, the same smart device with different test environments, and the same smart device with different operating systems; wherein, the test environment includes but is not limited to the network environment.

[0092] In one embodiment, the data report includes at least but is not limited to: the startup times corresponding to different test conditions respectively.

[0093] It should be noted that by measuring the startup time of the target application under different test conditions multiple times, performance evaluation is carried out, and a corresponding data report is generated to help developers analyze the application startup performance and optimize the application startup speed.

[0094] In one embodiment, the data report is presented at least but not limited to in the following forms: tables and / or text.

[0095] For example, as shown in Table 1, in one embodiment, by repeating the above method, the startup times of the target application corresponding to smart devices with different performances are calculated respectively.

[0096] Table 1

[0097]

[0098]

[0099] In one embodiment, the data report also includes: the maximum value, minimum value and average value of the startup times corresponding to different test conditions respectively.

[0100] For example, in the embodiment corresponding to Table 1 above, the data report also includes:

[0101] (1) The maximum value of the startup time: 5.2 seconds;

[0102] (2) The minimum value of the startup time: 2.3 seconds;

[0103] (3) The average value of the startup time: 3.67 seconds.

[0104] In one embodiment, the data report also includes: optimization suggestions for the target application.

[0105] Specifically, by comparing the startup times of the target application under different test conditions, performance bottlenecks are analyzed to help developers find the optimization direction for the target application, and possible optimization suggestions are provided based on the startup time.

[0106] For example, in one embodiment, the data report also includes:

[0107] Optimization suggestions:

[0108] (1) Optimize the resources loaded on the startup page;

[0109] (2) Compress image and icon resources.

[0110] In one embodiment, after the step of generating the data report, the method further includes: displaying the data report.

[0111] In one embodiment, the electronic device displays the data report through a browser.

[0112] In one embodiment, the user views the data report through the web interface of the electronic device.

[0113] In one embodiment, after the step of generating the data report, the method further includes: generating a link to the data report for the user to download the data report by clicking the link.

[0114] In one embodiment, the method further includes: dividing the startup process into at least two stages according to the image recognition algorithm and the multiple image frames, and determining the time consumption of each stage.

[0115] Specifically, according to the calculation of the above similarity (including the first similarity and / or the second similarity), the startup process is divided into at least two stages (as Figure 3 shown, the startup process includes: stage 1, stage 2, stage 3, stage 4...), and each stage is composed of a group of similar (i.e., similar in similarity) image frames. Then, according to the frame interval time, the time consumption of each stage is calculated (the principle is the same as the above calculation of the startup time, so it will not be elaborated here).

[0116] It should be noted that by determining the time consumption of each stage, it is possible to further determine in which stage the target application takes a long time, thus facilitating the developer to identify the bottleneck of the target application during the startup process.

[0117] In one embodiment, the optimization suggestions provided by the above data report are for the stage with the longest time consumption; for example: if a certain stage in the startup process (such as loading the interface, transition animation, etc.) takes a long time, it will be reminded that this stage needs further optimization.

[0118] In one embodiment, the electronic device is a smart device.

[0119] Specifically, the smart device obtains the startup video of the target application installed on it through its recording function, and then measures the startup time of the target application based on the startup video.

[0120] It should be noted that the intelligent device can be any high-performance device, any medium-performance device, or any low-performance device; in practical applications, if the selected device performance is different, the corresponding measured startup time will also be different; specifically, the higher the selected device performance, the shorter the corresponding measured startup time.

[0121] In one embodiment, the same target application is tested on multiple different intelligent devices (such as Android phones, iOS devices, emulators, etc.) and different operating system versions; during each startup process of the target application, the electronic device performs video parsing, image recognition, and calculation of the startup time according to the above steps.

[0122] Through this method, developers can comprehensively understand the startup performance of the target application in different device and operating system environments, identify potential performance issues, and optimize them.

[0123] In one embodiment, to ensure data security, encryption transmission technology is adopted during data processing; at the same time, all data will be anonymously stored according to the privacy policy after processing to protect user privacy.

[0124] It should be noted that the present invention provides a method for measuring the startup time of an application based on image recognition. This method accurately obtains startup time data by analyzing the changes in the screen image during the application startup process, thereby providing a more intuitive and reliable measurement means; through this method, the startup time measurement is not only more accurate, but also has higher flexibility and cost-effectiveness.

[0125] The protection scope of the method for measuring the startup time of an application based on image recognition described in the embodiments of the present invention is not limited to the execution order of the steps listed in this embodiment. Any solution achieved by adding or subtracting steps of the prior art and replacing steps according to the principle of the present invention is included in the protection scope of the present invention.

[0126] The embodiments of the present invention also provide an electronic device, which includes: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the computer program stored in the memory so that the electronic device executes the above-mentioned method for measuring the startup time of an application based on image recognition.

[0127] In one embodiment, the electronic device is an intelligent device.

[0128] In one embodiment, the electronic device is a server.

[0129] In this embodiment, the intelligent device serves as the client, and the server serves as the server side.

[0130] In one embodiment, the client uploads the recorded startup video through a browser, and the server calculates the startup time based on the startup video and displays it through the browser.

[0131] An embodiment of the present invention also provides an application startup time measurement system based on image recognition. The application startup time measurement system based on image recognition can implement the application startup time measurement method based on image recognition of the present invention. However, the implementation devices of the application startup time measurement method based on image recognition of the present invention include but are not limited to the structures of the application startup time measurement system based on image recognition listed in this embodiment. Any structural deformation and substitution of the prior art made according to the principles of the present invention are included in the protection scope of the present invention.

[0132] As Figure 4 shown, in one embodiment, the present invention provides an application startup time measurement system based on image recognition. The system includes:

[0133] A video processing module 41, configured to process the startup video of the target application to decompose the startup video into a plurality of consecutive image frames; each image frame corresponds to a moment in the startup video; the target application is installed on the intelligent device; the screen of the intelligent device is used to display the interface changes during the startup process of the target application; the startup video at least includes the interface changes during the startup process of the target application.

[0134] An image recognition module 42, configured to analyze the plurality of image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame of the startup of the target application from the plurality of image frames.

[0135] A time calculation module 43, configured to calculate the startup time of the target application according to the startup start frame and the loading end frame.

[0136] In one embodiment, the system further includes: a report generation module 44, configured to generate a data report based on the startup time.

[0137] It should be noted that this data report is the same as the data report described above when introducing the application startup time measurement method based on image recognition, so it will not be elaborated here in detail.

[0138] It should be noted that the structures and principles of the video processing module 41, the image recognition module 42, and the time calculation module 43 correspond one by one to the steps (step S1 to step S3) in the above application startup time measurement method based on image recognition. The specific working principles can also refer to the introduction of the application startup time measurement method based on image recognition in the foregoing embodiments, so they will not be elaborated here.

[0139] In several embodiments provided by the present invention, it should be understood that the disclosed system, apparatus, or method can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For example, the division of modules / units is only a logical function division. In actual implementation, there can be other division methods. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of devices or modules or units can be in electrical, mechanical, or other forms.

[0140] The modules / units described as separate components may or may not be physically separated. The components shown as modules / units may or may not be physical modules, that is, they can be located in one place or distributed to multiple network units. Some or all of the modules / units can be selected according to actual needs to achieve the purpose of the embodiments of the present invention. For example, in each embodiment of the present invention, the functional modules / units can be integrated in a processing module, or each module / unit can exist physically alone, or two or more modules / units can be integrated in one module / unit.

[0141] Those of ordinary skill in the art should also be further aware that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0142] The descriptions of the processes or structures corresponding to the above-mentioned respective drawings each have their own focuses. For parts not detailed in a certain process or structure, reference can be made to the relevant descriptions of other processes or structures.

[0143] The above embodiments only illustratively explain the principles and effects of the present invention and are not used to limit the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. A method for measuring the application startup time based on image recognition, characterized in that, The method includes: Processing the startup video of the target application to decompose the startup video into a plurality of consecutive image frames; each of the image frames corresponds to a moment in the startup video; the target application is installed on a smart device; the screen of the smart device is used to display the interface changes during the startup process of the target application; the startup video at least includes the interface changes during the startup process of the target application; Analyzing the plurality of image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame of the startup of the target application from the plurality of image frames; Calculating the startup time of the target application according to the startup start frame and the loading end frame.

2. The method for measuring the application startup time based on image recognition according to claim 1, wherein The interface changes during the startup process of the target application at least include: the initial interface when the target application starts and the final interface when the target application finishes loading.

3. The method for measuring the application startup time based on image recognition according to claim 1, wherein After the step of decomposing the startup video into a plurality of consecutive image frames and before the step of calculating the startup time of the target application according to the startup start frame and the loading end frame, the method further includes: numbering the plurality of image frames in chronological order; Calculating the startup time of the target application according to the startup start frame and the loading end frame includes: calculating the startup time according to the number of the startup start frame, the number of the loading end frame and the frame interval time; wherein, the frame interval time at least depends on the frame rate of the startup video.

4. The method for measuring the application startup time based on image recognition according to claim 1, wherein The analyzing the plurality of image frames by using an image recognition algorithm to identify the startup start frame and the loading end frame of the startup of the target application from the plurality of image frames includes: Comparing the similarity of each of the image frames with a baseline start frame based on the image recognition algorithm, denoted as the first similarity; Judging whether the first similarity meets a first preset condition, and determining the image frame corresponding to the first similarity that meets the first preset condition as the startup start frame; Comparing the similarity of each of the image frames with a baseline end frame based on the image recognition algorithm, denoted as the second similarity; Judging whether the second similarity meets a second preset condition, and determining the image frame corresponding to the second similarity that meets the second preset condition as the loading end frame.

5. The method for measuring the application startup time based on image recognition according to claim 1, characterized in that, The image recognition algorithm at least includes any one of the following algorithms: histogram comparison algorithm, SIFT feature extraction algorithm, ORB feature extraction algorithm.

6. The method for measuring the application startup time based on image recognition according to any one of claims 1 to 5, characterized in that, The method further includes: Obtaining the startup times respectively corresponding to different test conditions; and / or Dividing the startup process into at least two stages according to the image recognition algorithm and the plurality of image frames, and determining the time consumption of each stage.

7. An application startup time measurement system based on image recognition, characterized in that, The system includes: A video processing module, configured to process the startup video of the target application to decompose the startup video into a plurality of consecutive image frames; each of the image frames corresponds to a moment in the startup video; the target application is installed on a smart device; the screen of the smart device is used to display the interface changes during the startup process of the target application; the startup video at least includes the interface changes during the startup process of the target application; An image recognition module, configured to analyze the multiple image frames by using an image recognition algorithm, so as to identify a start start frame and a loading end frame at which the target application is started from the multiple image frames; A time calculation module, configured to calculate a startup time of the target application according to the start start frame and the loading end frame.

8. An electronic device, characterized in that, The electronic device includes: a processor and a memory; The memory is used for storing a computer program; The processor is configured to execute the computer program stored in the memory, so that the electronic device executes the method for measuring an application startup time based on image recognition according to any one of claims 1 to 6.

Images