Device decoding performance determination method, device, electronic device and storage medium
Through the terminal device's image detection data difference identification and decoding performance judgment, the problem of H.265 playback capability testing of OTT devices is solved, the proportion of H.265 stream playback and user fluency are improved, and bandwidth costs are reduced.
Patent Information
- Application Number
- CN202210907125.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-29
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2042-07-29
AI Technical Summary
The prior art is difficult to effectively determine whether the OTT device supports H.265 code rate, resulting in a large number of long-tail devices being unable to test, and can only play H.264 encoded videos, which consumes high bandwidth and low user playback fluency.
By obtaining the image detection data of the terminal device, performing differences in test video data, identifying the identification object and calculating similarity, judging the decoding performance of the terminal device, and dynamically generating a device whitelist to support H.265 code stream playback.
It has increased the proportion of H.265 stream playback of OTT devices, reduced the company's network bandwidth cost, and improved the playback fluency of users with low access bandwidth.
Smart Images

Figure CN115396661B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of computer technology, and in particular to a method, apparatus, electronic device and storage medium for determining device decoding performance. Background Art
[0002] The number of daily users on the OTT side is very high. On the one hand, due to the high resolution of the TV screen, the playback bit rate is higher than other sides to ensure the viewing clarity. On the other hand, due to the uneven performance of the equipment, in order to ensure playback compatibility, the default playback of H.264 bit rate video leads to a much higher proportion of bandwidth consumption than the number of OTT devices. Compared with H.264, H.265 encoding can reduce the transmission bandwidth by at least half. If the H.265 usage coverage of OTT devices can be improved, on the one hand, the company's bandwidth costs can be saved, and on the other hand, the playback fluency of users with low access bandwidth can be improved.
[0003] However, due to the wide variety of brands and models on the OTT side, it is difficult to determine whether the corresponding device supports the H.265 bit rate through a single dimension (such as chip type, etc.). The current method is manual testing, releasing the H.265 bit rate playback option for each sub-model. This will cause a large number of long-tail devices to be unable to be tested. To ensure playback compatibility, these devices can only play H.264-encoded videos. Summary of the invention
[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present application provides a method and apparatus for determining device decoding performance, an electronic device and a storage medium.
[0005] In a first aspect, the present application provides a method for determining decoding performance of a device, comprising:
[0006] Acquire image detection data from a terminal device, where the image detection data is obtained by the terminal device detecting a playback effect of the test video data;
[0007] Performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result, wherein the encoding format of the detection image is different from the encoding format of the test image;
[0008] The decoding performance of the terminal device is determined based on the identification result.
[0009] Optionally, performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result includes:
[0010] identifying a first identification object in the detection image;
[0011] Determining a test image corresponding to the detection image in the test video data according to the first identification object;
[0012] Determining the similarity between the detection image and the test image;
[0013] When the similarity is greater than or equal to a first threshold, it is determined that the detection image has passed the identification.
[0014] Optionally, determining the similarity between the detection image and the test image includes:
[0015] determining a structural similarity between the detection image and the test image;
[0016] If the structural similarity is greater than a preset similarity threshold, determining that the similarity between the detection image and the test image is greater than or equal to a first threshold;
[0017] Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
[0018] Optionally, determining the similarity between the detection image and the test image includes:
[0019] determining a perceptual hash value between the detection image and the test image;
[0020] The perceptual hash value is less than a preset perceptual hash threshold, and it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold;
[0021] Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
[0022] Optionally, determining the similarity between the detection image and the test image includes:
[0023] determining a structural similarity between the detection image and the test image;
[0024] If the structural similarity is greater than a preset similarity threshold, determining a perceptual hash value between the detection image and the test image;
[0025] The perceptual hash value is less than a preset perceptual hash threshold, and it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold;
[0026] Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
[0027] Optionally, identifying a first identification object in the detection image includes:
[0028] Detecting a target area containing a marked object in the detection image;
[0029] Identifying an identified object in the target area;
[0030] Verifying the text validity of the identified object;
[0031] If the text validity verification of the identification object passes, the identification object is converted into a preset character type to obtain the first identification object.
[0032] Optionally, performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result further includes:
[0033] Acquire the image jitter range of the detection image;
[0034] When the image shaking range is less than or equal to a preset second threshold, it is determined that the detection image has passed the identification.
[0035] Optionally, the method further comprises:
[0036] Acquiring device description data from the terminal device;
[0037] Determine the device category corresponding to each image detection data based on the device description data;
[0038] Based on the image detection data corresponding to each of the device categories, the decoding performance of the terminal device corresponding to each device category is determined.
[0039] Optionally, determining the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category includes:
[0040] For the image detection data corresponding to each device category, determine a detection image sequence according to the image detection data, and determine the device category to which each detection image in the detection image sequence belongs;
[0041] Based on the identification results of the detection images corresponding to each device category, the decoding performance of the terminal device of the device category is determined.
[0042] Optionally, determining the decoding performance of a terminal device of each device category based on the identification result of the detection image corresponding to the device category includes:
[0043] Based on the identification results of the detection images corresponding to each device category, determining the identification pass rate corresponding to each device category, wherein the identification pass rate is determined according to the ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0044] If the identification pass rate corresponding to any device category exceeds a preset pass rate threshold, it is determined that the decoding performance of the terminal device of the device category supports the preset encoding method;
[0045] Otherwise, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0046] Optionally, determining the decoding performance of the terminal device of the device category based on the identification result of each of the detection images includes:
[0047] Determine an image detection success rate corresponding to each device category, wherein the image detection success rate is determined according to a ratio between the number of the detection images and the number of the test images;
[0048] If the image detection success rate is greater than a preset detection threshold, determining an identification pass rate corresponding to each device category, wherein the identification pass rate is determined based on a ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0049] If the authentication pass rate is greater than a preset authentication threshold, it is determined that the decoding performance of the terminal device of the device category supports the preset encoding method;
[0050] Otherwise, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0051] In a second aspect, the present application provides a device for determining decoding performance of a device, comprising:
[0052] A first acquisition module is used to acquire image detection data from a terminal device, where the image detection data is obtained by the terminal device detecting a playback effect of the test video data;
[0053] an identification module, used for performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data, and obtaining an identification result, wherein the coding format of the detection image is different from the coding format of the test image;
[0054] The first determination module is used to determine the decoding performance of the terminal device based on the identification result.
[0055] Optionally, the identification module includes:
[0056] A recognition unit, used for recognizing a first identification object in the detection image;
[0057] A first determining unit, configured to determine a test image corresponding to the detection image in the test video data according to the first identification object;
[0058] A second determining unit, configured to determine a similarity between the detection image and the test image;
[0059] The third determination unit is used to determine that the detection image identification has passed when the similarity is greater than or equal to a first threshold.
[0060] Optionally, the second determining unit includes:
[0061] A first determining subunit, configured to determine the structural similarity between the detection image and the test image;
[0062] The second determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the structural similarity is greater than a preset similarity threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0063] Optionally, the second determining unit includes:
[0064] a third determination subunit, configured to determine a perceptual hash value between the detection image and the test image;
[0065] The fourth determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the perceptual hash value is less than a preset perceptual hash threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0066] Optionally, the second determining unit includes:
[0067] a fifth determining subunit, configured to determine a structural similarity between the detection image and the test image;
[0068] a sixth determination subunit, configured to determine a perceptual hash value between the detection image and the test image if the structural similarity is greater than a preset similarity threshold;
[0069] The seventh determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the perceptual hash value is less than a preset perceptual hash threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0070] Optionally, the identification unit includes:
[0071] A detection subunit, configured to detect a target area containing a marked object in the detection image;
[0072] An identification subunit, used to identify the identification object in the target area;
[0073] A verification subunit, used to verify the text validity of the identification object;
[0074] The conversion subunit is used to convert the identification object into a preset character type to obtain the first identification object if the text validity verification of the identification object passes.
[0075] Optionally, the identification module further includes:
[0076] An acquisition unit, used for acquiring an image jitter range of the detection image;
[0077] The fourth determination unit is used to determine that the detection image has passed the identification when the image shaking range is less than or equal to a preset second threshold.
[0078] Optionally, the device further comprises:
[0079] A second acquisition module, used to acquire device description data from the terminal device;
[0080] A second determination module, configured to determine the device category corresponding to each image detection data based on the device description data;
[0081] The third determination module is used to determine the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category.
[0082] Optionally, the third determining module includes:
[0083] a fifth determining unit, configured to determine, for each device category, the image detection data corresponding to the device category, a detection image sequence according to the image detection data, and determine the device category to which each detection image in the detection image sequence belongs;
[0084] The sixth determination unit is used to determine the decoding performance of the terminal device of each device category based on the identification result of the detection image corresponding to the device category.
[0085] Optionally, the third determining module includes:
[0086] a seventh determination unit, configured to determine an identification pass rate corresponding to each device category based on the identification results of the detection images corresponding to each device category, wherein the identification pass rate is determined according to a ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0087] The eighth determination unit is used to determine that the decoding performance of the terminal device of any device category supports the preset encoding method if the identification pass rate corresponding to the device category exceeds the preset pass rate threshold; otherwise, determine that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0088] Optionally, the third determining module includes:
[0089] a ninth determining unit, configured to determine an image detection success rate corresponding to each device category, wherein the image detection success rate is determined according to a ratio between the number of the detection images and the number of the test images;
[0090] a tenth determining unit, configured to determine, if the image detection success rate is greater than a preset detection threshold, an identification pass rate corresponding to each device category, wherein the identification pass rate is determined based on a ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0091] The eleventh determination unit is used to determine that the decoding performance of the terminal device of the device category supports the preset encoding method if the identification pass rate is greater than a preset identification threshold; otherwise, determine that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0092] In a third aspect, the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;
[0093] Memory, used to store computer programs;
[0094] The processor is used to implement the device decoding performance determination method described in any one of the first aspects when executing the program stored in the memory.
[0095] In a fourth aspect, the present application provides a computer-readable storage medium, on which is stored a program for a method for determining device decoding performance. When the program for the method for determining device decoding performance is executed by a processor, the steps of the method for determining device decoding performance described in any one of the first aspects are implemented.
[0096] The above technical solution provided by the embodiment of the present application has the following advantages compared with the prior art:
[0097] The embodiment of the present application first obtains image detection data from a terminal device, wherein the image detection data is obtained by detecting the playback effect of the test video data by the terminal device; then, a difference identification is performed on a detection image in the image detection data and a corresponding test image in the test video data to obtain an identification result, wherein the encoding format of the detection image is different from the encoding format of the test image; finally, the decoding performance of the terminal device can be determined based on the identification result, thereby improving the efficiency of the decoding performance detection of the terminal device, solving the problem of H.265 playback capability testing of a large number of long-tail OTT devices, and greatly increasing the H.265 code stream playback proportion of OTT devices, thereby significantly reducing the company's network bandwidth cost and effectively improving the playback fluency of low access bandwidth users. BRIEF DESCRIPTION OF THE DRAWINGS
[0098] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0099] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0100] Figure 1 A deployment framework diagram of a data statistical analysis system provided in an embodiment of the present application;
[0101] Figure 2 A flowchart of a method for determining device decoding performance provided in an embodiment of the present application;
[0102] Figure 3 A structural diagram of a device decoding performance determination apparatus provided in an embodiment of the present application;
[0103] Figure 4 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0104] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0105] At present, due to the wide variety of brands and models of OTT terminals, it is difficult to determine whether the corresponding device supports H.265 bitrate through a single dimension (such as chip type, etc.). The current method is manual testing, and the H.265 bitrate playback options are released one by one. This will cause a large number of long-tail devices to be unable to be tested. To ensure playback compatibility, these devices can only play H.264-encoded videos. To this end, the present application embodiment provides a method, device, electronic device and storage medium for determining device decoding performance, which can improve the efficiency of terminal device decoding performance detection, solve the H.265 playback capability test problem of a large number of long-tail OTT devices, and greatly increase the H.265 stream playback ratio of OTT devices, thereby significantly reducing the company's network bandwidth costs and effectively improving the playback fluency of low-access bandwidth users.
[0106] To support the processing of millions of user uploads per day and dynamically adapt to the peak of user uploads on weekends, the detailed deployment framework of the data statistics and analysis system is shown in Figure 1 ,like Figure 1 As shown, the data statistical analysis system includes a front-end proxy server, a user data processing server, a database and a business intelligence (BI) data server.
[0107] Among them, the front-end proxy server is used for load balancing and scheduling distribution, distributing the image sequence data uploaded by each user terminal to different user data processing servers; the user data processing server writes the processing results of the image sequence data into the database; the BI data server will access the database table regularly, classify and count the processing results according to device name / type / version / manufacturer and other information, and dynamically generate a whitelist of devices that support this playback mode.
[0108] like Figure 2 A method for determining device decoding performance is provided in an embodiment of the present application. The method can be applied to a BI data server. The method for determining device decoding performance includes:
[0109] Step S101, acquiring image detection data from a terminal device.
[0110] In the embodiment of the present application, the image detection data is obtained by the terminal device detecting the playback effect of the test video data;
[0111] Since the production standards and production technologies of different manufacturers are different, the decoding performance of the terminal devices produced is also different; decoding performance refers to the ability to process video data; for example: the ability to decode video data encoded using a preset encoding method, the ability to broadcast different video data, etc.
[0112] If the video data is directly sent to the terminal device for playback, a green screen, freezes, or even a failure to play the video may occur during playback due to the terminal device not having the corresponding decoding performance, thereby affecting the user experience. Therefore, in an embodiment of the present invention, before the video data is formally sent to the terminal device, the decoding performance of the terminal device may be tested to determine whether it has the ability to process the video data.
[0113] In the embodiment of the present invention, the terminal device may first play the test video data, and the test video data may be used to detect the decoding performance of the terminal device.
[0114] As an example, test video data can be pre-produced according to actual detection needs; for example, when it is necessary to detect whether the terminal device has the ability to decode video data that has been encoded using a preset encoding method, the preset encoding method can be used to encode the original video data, and then the original video data encoded using the preset encoding method can be sent to the terminal device.
[0115] After receiving the encoded original video data, the terminal device can decode it using a decoder to obtain test video data.
[0116] Since the test video data is used to detect the decoding performance of the terminal device, the content of its video image may not be ornamental; if it is played directly on the screen of the terminal device, it may affect the user's use of the terminal device, for example: when the terminal device is in standby mode, the test video data is suddenly played. Therefore, in an embodiment of the present invention, the test video data can be played off-screen on the terminal device. Playing the test video data off-screen may refer to rendering the test video data off-screen on the terminal device. Off-screen rendering refers to the GPU (Graphics Processing Unit) opening up a new buffer outside the current screen buffer for rendering operations. By playing the test video data off-screen, the entire detection process is imperceptible to the user, thereby avoiding affecting the user's use of the terminal device.
[0117] As an example, the terminal device may refer to a device for playing video data, such as a high-definition player, a high-definition box, a smart TV, a smart phone, a tablet computer, and a smart projector, etc., which is not limited in this embodiment of the present invention.
[0118] In the embodiment of the present invention, if the terminal device does not have the decoding performance to be detected by the test video data, the test video data cannot be played normally; if the terminal device has the decoding performance to be detected by the test video data, the test video data can be played normally.
[0119] Therefore, when the terminal device plays the test video data off-screen, it can generate image detection data for the test video data played off-screen. The image detection data may refer to the image of the test video data played off-screen by the terminal device when the terminal device plays the test video data off-screen, and the collected data, such as: a screenshot of the image of the test video data played off-screen, etc., which is not limited in the embodiment of the present invention; the image detection data can be used to analyze whether an abnormality occurs when the terminal device plays the test video data (such as: black screen, flower screen, playback freeze, change of rendering area before and after ABS switching, etc.).
[0120] Step S102, performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result;
[0121] In an embodiment of the present application, the coding format of the detection image is different from the coding format of the test image. In the present application, the detection image adopts a coding format in the whitelist supported by the device to test a coding format with unknown support. For example, H.265 is one of the coding formats with unknown support.
[0122] In an embodiment of the present application, the image detection data includes a detection image sequence generated by periodic screenshots during the playing of test image data by a group of terminal devices, the detection image sequence includes at least one detection image, the test video data includes a test image sequence, the test image sequence includes at least one test image, and the detection images in the image detection data correspond one-to-one to the test images in the test image sequence.
[0123] In this step, the difference between the detection image and the corresponding test image can be determined, and the identification result can be determined based on the difference.
[0124] Specifically, the similarity between the detection image and the test image may be determined, and when the similarity is greater than or equal to a first threshold, it is determined that the detection image has passed the identification.
[0125] Furthermore, when the similarity is greater than or equal to a first threshold, the image jitter range of the detection image can be determined, and when the image jitter range is less than or equal to a preset second threshold, it can be determined that the detection image identification has passed.
[0126] Step S103: determining the decoding performance of the terminal device based on the identification result.
[0127] Since the image detection data includes a detection image sequence, which includes at least one detection image, and the test video data includes a test image sequence, which includes at least one test image, the decoding performance of the terminal device can be determined based on the identification results of each detection image in the terminal device.
[0128] Specifically, the identification pass rate corresponding to the terminal device can be calculated. The identification pass rate is the ratio between the number of the detection images that have passed the identification and the number of all detection images. When the identification pass rate exceeds a first preset threshold, it is determined that the decoding performance of the terminal device supports the preset encoding method. Otherwise, it is determined that the decoding performance of the terminal device does not support the preset encoding method.
[0129] Alternatively, the image detection success rate corresponding to the terminal device can be calculated first, and the image detection success rate is the ratio of the number of detection images in the detection image sequence to the number of test images in the test image sequence. When the image detection success rate is greater than a second preset threshold, the identification pass rate corresponding to the terminal device is calculated, and the identification pass rate is the ratio of the number of the detection images that have passed the identification to the number of all detection images. When the identification pass rate exceeds the first preset threshold, it is determined that the decoding performance of the terminal device supports the preset encoding method; otherwise, it is determined that the decoding performance of the terminal device does not support the preset encoding method.
[0130] The embodiment of the present application first obtains image detection data from a terminal device, wherein the image detection data is obtained by detecting the playback effect of the test video data by the terminal device; then, a difference identification is performed on a detection image in the image detection data and a corresponding test image in the test video data to obtain an identification result, wherein the encoding format of the detection image is different from the encoding format of the test image; finally, the decoding performance of the terminal device can be determined based on the identification result, thereby improving the efficiency of the decoding performance detection of the terminal device, solving the problem of H.265 playback capability testing of a large number of long-tail OTT devices, and greatly increasing the H.265 code stream playback proportion of OTT devices, thereby significantly reducing the company's network bandwidth cost and effectively improving the playback fluency of low access bandwidth users.
[0131] In another embodiment of the present application, step S102, performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result includes:
[0132] Step S201, identifying a first identification object in the detection image;
[0133] In this step, a target area containing an identification object can be detected in the detection image; the identification object in the target area can be identified; the text validity of the identification object can be verified; if the text validity verification of the identification object passes, the identification object can be converted into a preset character type to obtain the first identification object.
[0134] Specifically, in order to improve the success rate of digital recognition, only text in a specific area can be recognized, and the first identification object can be detected in the detection image to obtain an identification object image. Then, the object image can be processed using a preset image processing method (such as using the Leptonica algorithm), such as decontamination and color enhancement, etc. The Tesseract algorithm can be used to perform optical character recognition (such as OCR optical character recognition) on the processed object image. The identified identification object can verify the text validity based on preset rules. When the text validity verification passes, the identification object is converted into a preset character type. Specifically, the preset character type can be a digital type to obtain a first identification object.
[0135] In one embodiment of the present application, the frame mark in each image frame in the test video data is pre-generated and set on the image frame. In order to maximize the success rate of optical character recognition, the area where the frame mark is located in the test image can be screenshotted as a training sample for optical character recognition. Because these digital fonts are fixed and limited in number, the accuracy of OCR recognition can be greatly improved. The practical result is that the recognition error rate can be less than 10 -6 .
[0136] Step S202, determining a test image corresponding to the detection image in the test video data according to the first identification object;
[0137] In an embodiment of the present application, a frame identifier is set at a designated position of each image frame in the test video data to uniquely identify the image frame, and the detection image is generated by detecting the playback effect of the test video data (such as taking a screenshot), so a normal detection image must contain a first identification object that is the same as the frame identifier.
[0138] That is to say, the frame identifier identical to the first identifier object can be determined, and then a test image having the frame identifier can be searched in the test image data as the test image corresponding to the detection image.
[0139] Step S203, determining the similarity between the detection image and the test image;
[0140] In one embodiment of the present application, the structural similarity between the detection image and the test image can be determined; if the structural similarity is greater than a preset similarity threshold, it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold; otherwise, it is determined that the similarity between the detection image and the test image is less than the first threshold.
[0141] Specifically, the structural similarity between the detection image and the test image may be determined by a structural similarity index measure (SSIM) algorithm.
[0142] In another embodiment of the present application, a perceptual hash value between the detection image and the test image can be determined; if the perceptual hash value is less than a preset perceptual hash threshold, it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold; otherwise, it is determined that the similarity between the detection image and the test image is less than the first threshold.
[0143] Specifically, a perceptual hash value between the detection image and the test image is determined by a perceptual hash algorithm (PH).
[0144] In another embodiment of the present application, the structural similarity between the detection image and the test image is determined; if the structural similarity is greater than a preset similarity threshold, the perceptual hash value between the detection image and the test image is determined; if the perceptual hash value is less than the preset perceptual hash threshold, it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold; otherwise, it is determined that the similarity between the detection image and the test image is less than the first threshold.
[0145] Specifically, the structural similarity between the detection image and the test image can be determined by a structural similarity index measure (SSIM) algorithm, and the perceptual hash value between the detection image and the test image can be determined by a perceptual hash algorithm (PH).
[0146] Step S204: when the similarity is greater than or equal to a first threshold, determining that the detection image identification has passed.
[0147] When the similarity is less than a first threshold, it is determined that the detection image identification fails.
[0148] The embodiment of the present application can automatically determine whether the detection image has passed the identification based on the similarity, and then determine the decoding performance of the terminal device. By judging the image similarity, the detection precision and accuracy can be improved.
[0149] In another embodiment of the present application, performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result further includes:
[0150] Step S301, obtaining the image jitter range of the detection image;
[0151] In this step, when the similarity between the detection image and the test image is greater than the first threshold, the image jitter range of the detection image can be determined. Specifically, the image jitter range of the first identification object in the detection image can be determined. Exemplarily, the image jitter range can be determined by the following formula:
[0152] K = abs (first identification object * sampling time interval (seconds) * frame rate - frame identification)
[0153] Step S302: When the image shaking range is less than or equal to a preset second threshold, it is determined that the detection image identification has passed.
[0154] The embodiment of the present application can automatically determine whether the detection image has passed the identification based on the image jitter range, and then determine the decoding performance of the terminal device. By judging the image jitter range, the detection precision and accuracy can be improved.
[0155] In one embodiment of the present application, when the similarity is less than a first threshold, the image jitter range of the detection image can be obtained, and when the image jitter range is less than or equal to a preset second threshold, it is determined that the detection image identification has passed.
[0156] The embodiment of the present application determines whether the detected image has passed the identification by first judging the similarity and then judging the image jitter range, and then determines the decoding performance of the terminal device. Through multiple threshold judgments, the detection precision and accuracy are improved.
[0157] In another embodiment of the present application, the method further includes:
[0158] Step S401, obtaining device description data from the terminal device;
[0159] In the embodiment of the present application, the device description data may refer to relevant information of the terminal device itself, such as: device manufacturer / brand / type / version and other information, which is not limited in the embodiment of the present invention.
[0160] Step S402, determining the device category corresponding to each image detection data based on the device description data;
[0161] Since the device description data is relevant information of the terminal device itself, different terminal devices have different device description data. Each terminal device can be classified according to the device description data. Then, each image detection data can be classified according to the device category to which each terminal device belongs, so as to obtain the device category corresponding to each image detection data.
[0162] Step S403: Determine the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category.
[0163] After obtaining the image detection data corresponding to each device category, the BI data server can determine the decoding performance of the terminal devices corresponding to each device category based on the image detection data corresponding to each device category, so as to determine whether the terminal devices of each device category can play the test video data normally, and then determine whether the terminal devices of each device category have the decoding performance to be detected by the test video data.
[0164] In one embodiment of the present invention, when the BI data server determines that a terminal device of any device category has (or does not have) the decoding performance to be detected for testing video data, a white list of terminal devices of each device category with the decoding performance to be detected can be dynamically generated, so that when the video data is subsequently formally sent to each terminal device, it can be sent based on the decoding performance of each terminal device in the white list, thereby avoiding sending video data that cannot be played normally to the terminal device.
[0165] In another embodiment of the present application, step S403 determines the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category, including:
[0166] Step S501, determining a detection image sequence according to the image detection data, and determining the device category to which each detection image in the detection image sequence belongs;
[0167] Step S502: Determine the decoding performance of the terminal device of each device category based on the identification result of the detection image corresponding to each device category.
[0168] In one implementation of the present application, step S502 determines the decoding performance of the terminal device of the device category based on the identification results of each of the detection images, including:
[0169] Step S601, determining the identification pass rate corresponding to each device category based on the identification result of the detection image corresponding to each device category;
[0170] In an embodiment of the present application, the identification pass rate is determined based on the ratio between the number of the detection images that have passed the identification and the number of all detection images, that is, the identification pass rate is equal to the ratio between the number of detection images that have passed the identification and the number of all detection images.
[0171] Step S602, if the identification pass rate of the detection image in the image detection data corresponding to any device category exceeds a preset pass rate threshold, it is determined that the decoding performance of the terminal device of the device category supports the preset encoding method;
[0172] Otherwise, in step S603, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0173] In the embodiment of the present application, the identification pass rate may refer to the proportion of detection images that have passed the identification to all detection images. For example, the identification pass rate is 99%.
[0174] That is, among all the image detection data corresponding to any device category, if the identification pass rate of the detection images in all the image detection data exceeds the preset pass rate threshold, it is determined that the decoding performance of the terminal device of this device category supports the preset encoding method; otherwise, it is determined that the decoding performance of the terminal device of this device category does not support the preset encoding method.
[0175] In the embodiment of the present application, the decoding performance of the terminal devices in each device category can be determined by the identification pass rate. The method is simple and improves the efficiency of the terminal device decoding performance detection.
[0176] In one implementation of the present application, step S502 determines the decoding performance of the terminal device of the device category based on the identification results of each of the detection images, including:
[0177] Step S701, determining the image detection success rate corresponding to each device category.
[0178] In the embodiment of the present application, the image detection success rate is determined based on the ratio between the number of detection images and the number of test images, that is, image detection success rate = number of detection images / number of test images.
[0179] Step S702: If the image detection success rate is greater than a preset detection threshold, determine the identification pass rate corresponding to each device category.
[0180] In the embodiment of the present application, the identification pass rate is determined based on the ratio between the number of the detection images that have passed the identification and the number of all detection images, that is, the identification pass rate = the number of detection images that have passed the identification / the number of all detection images.
[0181] Step S703, if the identification pass rate is greater than a preset identification threshold, it is determined that the decoding performance of the terminal device of the device category supports a preset encoding method;
[0182] Otherwise, in step S704, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0183] The embodiment of the present application determines the decoding performance of each terminal device in each device category through the image detection success rate and identification success rate corresponding to each device category, and improves the detection precision and accuracy through multiple threshold judgments.
[0184] In another embodiment of the present application, Figure 3 As shown, a device decoding performance determination apparatus is also provided, comprising:
[0185] A first acquisition module 11 is used to acquire image detection data from a terminal device, where the image detection data is obtained by the terminal device detecting a playback effect of the test video data;
[0186] An identification module 12, used for performing a difference identification on the detection image in the image detection data and the corresponding test image in the test video data, and obtaining an identification result that the coding format of the detection image is different from the coding format of the test image;
[0187] The first determination module 13 is configured to determine the decoding performance of the terminal device based on the identification result.
[0188] Optionally, the identification module includes:
[0189] A recognition unit, used for recognizing a first identification object in the detection image;
[0190] A first determining unit, configured to determine a test image corresponding to the detection image in the test video data according to the first identification object;
[0191] A second determining unit, configured to determine a similarity between the detection image and the test image;
[0192] The third determination unit is used to determine that the detection image identification has passed when the similarity is greater than or equal to a first threshold.
[0193] Optionally, the second determining unit includes:
[0194] A first determining subunit, configured to determine the structural similarity between the detection image and the test image;
[0195] The second determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the structural similarity is greater than a preset similarity threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0196] Optionally, the second determining unit includes:
[0197] a third determination subunit, configured to determine a perceptual hash value between the detection image and the test image;
[0198] The fourth determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the perceptual hash value is less than a preset perceptual hash threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0199] Optionally, the second determining unit includes:
[0200] a fifth determining subunit, configured to determine a structural similarity between the detection image and the test image;
[0201] a sixth determination subunit, configured to determine a perceptual hash value between the detection image and the test image if the structural similarity is greater than a preset similarity threshold;
[0202] The seventh determination subunit is used to determine that the similarity between the detection image and the test image is greater than or equal to a first threshold if the perceptual hash value is less than a preset perceptual hash threshold; otherwise, determine that the similarity between the detection image and the test image is less than the first threshold.
[0203] Optionally, the identification unit includes:
[0204] A detection subunit, configured to detect a target area containing a marked object in the detection image;
[0205] An identification subunit, used to identify the identification object in the target area;
[0206] A verification subunit, used to verify the text validity of the identification object;
[0207] The conversion subunit is used to convert the identification object into a preset character type to obtain the first identification object if the text validity verification of the identification object passes.
[0208] Optionally, the identification module further includes:
[0209] An acquisition unit, used for acquiring an image jitter range of the detection image;
[0210] The fourth determination unit is used to determine that the detection image has passed the identification when the image shaking range is less than or equal to a preset second threshold.
[0211] Optionally, the device further comprises:
[0212] A second acquisition module, used to acquire device description data from the terminal device;
[0213] A second determination module, configured to determine the device category corresponding to each image detection data based on the device description data;
[0214] The third determination module is used to determine the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category.
[0215] Optionally, the third determining module includes:
[0216] a fifth determining unit, configured to determine, for each device category, the image detection data corresponding to the device category, a detection image sequence according to the image detection data, and determine the device category to which each detection image in the detection image sequence belongs;
[0217] The sixth determination unit is used to determine the decoding performance of the terminal device of each device category based on the identification result of the detection image corresponding to the device category.
[0218] Optionally, the third determining module includes:
[0219] a seventh determination unit, configured to determine an identification pass rate corresponding to each device category based on the identification results of the detection images corresponding to each device category, wherein the identification pass rate is determined according to a ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0220] The eighth determination unit is used to determine that the decoding performance of the terminal device of any device category supports the preset encoding method if the identification pass rate corresponding to the device category exceeds the preset pass rate threshold; otherwise, determine that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0221] Optionally, the third determining module includes:
[0222] a ninth determining unit, configured to determine an image detection success rate corresponding to each device category, wherein the image detection success rate is determined according to a ratio between the number of the detection images and the number of the test images;
[0223] a tenth determining unit, configured to determine, if the image detection success rate is greater than a preset detection threshold, an identification pass rate corresponding to each device category, wherein the identification pass rate is determined based on a ratio between the number of the detection images that have passed the identification and the number of all detection images;
[0224] The eleventh determination unit is used to determine that the decoding performance of the terminal device of the device category supports the preset encoding method if the identification pass rate is greater than a preset identification threshold; otherwise, determine that the decoding performance of the terminal device of the device category does not support the preset encoding method.
[0225] In another embodiment of the present application, an electronic device is provided, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus;
[0226] Memory, used to store computer programs;
[0227] The processor is used to implement the device decoding performance determination method described in any of the above method embodiments when executing the program stored in the memory.
[0228] In the electronic device provided by the embodiment of the present invention, the processor executes the program stored in the memory to first obtain the image detection data and the device description data from each terminal device, then determines the device category corresponding to each image detection data based on the device description data, and finally automatically determines the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category, thereby improving the efficiency of the terminal device decoding performance detection, solving the H.265 playback capability testing problem of a large number of long-tail OTT devices, and greatly increasing the H.265 code stream playback proportion of OTT devices, thereby significantly reducing the company's network bandwidth cost and effectively improving the playback fluency of low access bandwidth users.
[0229] The communication bus 1140 mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus 1140 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.
[0230] The communication interface 1120 is used for communication between the above electronic device and other devices.
[0231] The memory 1130 may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.
[0232] The above-mentioned processor 1110 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.
[0233] In another embodiment of the present application, a computer-readable storage medium is further provided, on which a program of a method for determining device decoding performance is stored. When the program of the method for determining device decoding performance is executed by a processor, the steps of the method for determining device decoding performance described in any of the aforementioned method embodiments are implemented.
[0234] It should be noted that, in this article, relational terms such as "first" and "second" 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 sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0235] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.
Claims
1. A method for determining decoding performance of a device, characterized in that: include: Acquire image detection data from a terminal device, where the image detection data is obtained by the terminal device detecting a playback effect of the test video data; Performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result, wherein the encoding format of the detection image is different from the encoding format of the test image; Determining the decoding performance of the terminal device based on the identification result; The method further comprises: Acquiring device description data from the terminal device; Determine the device category corresponding to each image detection data based on the device description data; Determining decoding performance of a terminal device corresponding to each device category based on the image detection data corresponding to each device category; Determining the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category includes: For the image detection data corresponding to each device category, determine a detection image sequence according to the image detection data, and determine the device category to which each detection image in the detection image sequence belongs; Determining the decoding performance of the terminal device of each device category based on the identification result of the detection image corresponding to each device category; Determining the decoding performance of the terminal device of the device category based on the identification result of the detection image corresponding to each device category includes: Determine an image detection success rate corresponding to each device category, wherein the image detection success rate is determined according to a ratio between the number of the detection images and the number of the test images; If the image detection success rate is greater than a preset detection threshold, determining an identification pass rate corresponding to each device category, wherein the identification pass rate is determined based on a ratio between the number of the detection images that have passed the identification and the number of all detection images; If the authentication pass rate is greater than a preset authentication threshold, it is determined that the decoding performance of the terminal device of the device category supports the preset encoding method; Otherwise, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
2. The method according to claim 1, characterized in that: Performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result, including: identifying a first identification object in the detection image; Determining a test image corresponding to the detection image in the test video data according to the first identification object; Determining the similarity between the detection image and the test image; When the similarity is greater than or equal to a first threshold, it is determined that the detection image has passed the identification.
3. The method according to claim 2, characterized in that Determining the similarity between the detection image and the test image includes: determining a structural similarity between the detection image and the test image; If the structural similarity is greater than a preset similarity threshold, determining that the similarity between the detection image and the test image is greater than or equal to a first threshold; Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
4. The method according to claim 2, characterized in that: Determining the similarity between the detection image and the test image includes: determining a perceptual hash value between the detection image and the test image; If the perceptual hash value is less than a preset perceptual hash threshold, determining that the similarity between the detection image and the test image is greater than or equal to a first threshold; Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
5. The method according to claim 2, characterized in that: Determining the similarity between the detection image and the test image includes: determining a structural similarity between the detection image and the test image; If the structural similarity is greater than a preset similarity threshold, determining a perceptual hash value between the detection image and the test image; The perceptual hash value is less than a preset perceptual hash threshold, and it is determined that the similarity between the detection image and the test image is greater than or equal to a first threshold; Otherwise, it is determined that the similarity between the detection image and the test image is less than a first threshold.
6. The method according to claim 2, characterized in that Identifying a first identification object in the detection image includes: Detecting a target area containing a marked object in the detection image; Identifying an identified object in the target area; Verifying the text validity of the identified object; If the text validity verification of the identification object passes, the identification object is converted into a preset character type to obtain the first identification object.
7. The method according to claim 1, characterized in that Performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data to obtain an identification result, further comprising: Acquire the image jitter range of the detection image; When the image shaking range is less than or equal to a preset second threshold, it is determined that the detection image identification has passed.
8. The method according to claim 1, characterized in that Determining the decoding performance of the terminal device of the device category based on the identification result of the detection image corresponding to each device category includes: Based on the identification results of the detection images corresponding to each device category, determining the identification pass rate corresponding to each device category, wherein the identification pass rate is determined according to the ratio between the number of the detection images that have passed the identification and the number of all detection images; If the authentication pass rate corresponding to any device category exceeds a preset pass rate threshold, it is determined that the decoding performance of the terminal device of the device category supports the preset encoding method; Otherwise, it is determined that the decoding performance of the terminal device of the device category does not support the preset encoding method.
9. A device for determining decoding performance of a device, characterized in that: include: A first acquisition module is used to acquire image detection data from a terminal device, where the image detection data is obtained by the terminal device detecting a playback effect of the test video data; an identification module, used for performing difference identification on the detection image in the image detection data and the corresponding test image in the test video data, and obtaining an identification result, wherein the coding format of the detection image is different from the coding format of the test image; A first determination module, configured to determine the decoding performance of the terminal device based on the identification result; The device also includes: A second acquisition module, used to acquire device description data from the terminal device; A second determination module, used to determine the device category corresponding to each image detection data based on the device description data; A third determination module, configured to determine the decoding performance of the terminal device corresponding to each device category based on the image detection data corresponding to each device category; The third determination module comprises: a fifth determining unit, configured to determine, for each device category, the image detection data corresponding to the device category, a detection image sequence according to the image detection data, and determine the device category to which each detection image in the detection image sequence belongs; a sixth determining unit, configured to determine the decoding performance of a terminal device of each device category based on the identification result of the detection image corresponding to the device category; The third determination module comprises: a ninth determining unit, configured to determine an image detection success rate corresponding to each device category, wherein the image detection success rate is determined according to a ratio between the number of the detection images and the number of the test images; a tenth determining unit, configured to determine, if the image detection success rate is greater than a preset detection threshold, an identification pass rate corresponding to each device category, wherein the identification pass rate is determined based on a ratio between the number of the detection images that have passed the identification and the number of all detection images; The eleventh determination unit is used to determine that the decoding performance of the terminal device of the device category supports the preset encoding method if the identification pass rate is greater than a preset identification threshold; otherwise, determine that the decoding performance of the terminal device of the device category does not support the preset encoding method.
10. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, used to implement the device decoding performance determination method described in any one of claims 1 to 8 when executing a program stored in a memory.
11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program of a method for determining device decoding performance, and when the program of the method for determining device decoding performance is executed by a processor, the steps of the method for determining device decoding performance described in any one of claims 1-8 are implemented.
