Video quality detection method, device, storage medium and system

Abstract

The invention provides a video quality detection method and device, a storage medium and a system, and is applied to a quality detection device in the video quality detection system, the video quality detection system comprises a source video generation device, a video playing device and a quality detection device, the source video generation device is connected with the video playing device, and the quality detection device is connected with the video playing device. The quality detection device is connected with a source video generation device, and the method comprises the steps that a target recorded image is acquired, a target source image corresponding to the target recorded image is acquired from a source video, the target recorded image is any frame of image in a to-be-detected video, and the to-be-detected video is obtained by recording the source video played by a video playing device through a to-be-evaluated recording system; extracting image areas with the same content from the target recorded image and the target source image to serve as a to-be-tested image corresponding to the target recorded image and a reference image corresponding to the to-be-tested image in the target source image; and determining the recording quality of the target recorded image according to the to-be-tested image and the reference image. According to the technical scheme, the quality of the recording system can be determined.

Description

Technical Field

[0001] This application relates to the field of video quality detection, and particularly to a video quality detection method, device, storage medium, and system. Background Art

[0002] Video quality evaluation is one of the key technologies for ensuring the quality of network video services. Through video quality evaluation, it is possible to monitor the preprocessing, encoding, transmission, and decoding of live videos or recorded videos. It is possible to use reference-based video quality evaluation to evaluate video quality. Among them, reference-based video quality evaluation refers to comparing each pixel in each corresponding frame between an unencoded video used as a reference video and a live video stream or a recorded video stream to obtain the video quality.

[0003] The quality of a recorded video can reflect the quality of the recording system that records the recorded video. In some scenarios, it is necessary to evaluate the quality of recorded videos recorded by the recording systems of multiple manufacturers, so as to evaluate the quality of the recording systems of each manufacturer. Since it is difficult to directly obtain the unencoded videos of each manufacturer, in the case of lacking the reference video corresponding to the recorded video, how to evaluate the quality of the recording system has become a technical problem that needs to be solved urgently. Summary of the Invention

[0004] This application provides a video quality detection method, device, storage medium, and system, aiming to evaluate the quality of a recording system.

[0005] In a first aspect, a video quality detection method is provided, which is applied to a quality detection device in a video quality detection system. The video quality detection system includes a source video generation device, a video playback device, and the quality detection device. The source video generation device is connected to the video playback device, and the source video generation device is used to send a source video to the video playback device for playback. The quality detection device is connected to the source video generation device, and the quality detection device is used to obtain the source video from the source video generation device. The source video includes at least one frame of source image. The method includes:

[0006] Obtain a target recorded image, and obtain a target source image corresponding to the target recorded image from the source video. The target recorded image is any frame image in a to-be-tested video, and the to-be-tested video is obtained by the recording system to be evaluated by recording the source video played by the video playback device;

[0007] Extract an image region with the same content in the target recorded image and the target source image as the to-be-tested image corresponding to the target recorded image and the reference image corresponding to the to-be-tested image in the target source image;

[0008] Determine the recording quality of the target recorded image according to the image to be measured and the reference image.

[0009] In this technical solution, after obtaining the target recorded image and the target source image corresponding to the target recorded image from the source video, extract the image regions with the same content in the target recorded image and the target source image as the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image, and determine the recording quality of the target recorded image according to the image to be measured and the reference image; since the source video corresponding to the target source image is generated by the source video generation device and obtained by the video quality detection device from the source video generation device, and the video to be measured corresponding to the target recorded image is recorded by the recording system to be evaluated for the source video played by the video playback device, the source video is equivalent to the reference video of the video to be measured, and the recording quality of the recorded image in the video to be measured can reflect the quality of the recording system to be evaluated. Therefore, by evaluating the recording quality of the recorded image recorded by the recording system, the quality evaluation of the recording system can be realized; in addition, since the image regions with the same content are extracted from the target recorded image and the target source image as the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image, and the recording quality of the target recorded image is determined according to the image to be measured and the reference image in the target recorded image, it is not necessary for the recording system to completely record the entire image region of the source image, and the perspective of the recording system and the size of the recording site can be less restricted, with fewer detection requirements, and the quality evaluation of the recording system can be completed in any detection environment, with a wide range of applications.

[0010] Combined with the first aspect, in a possible implementation manner, each frame of the source image in the source video includes at least three image markers, and the at least three image markers are distributed at different image positions in the source image; the extracting the image regions with the same content in the target recorded image and the target source image as the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image includes: obtaining first position information and second position information, where the first position information includes the position of each of the at least three image markers in the target recorded image, and the second position information includes the position of each of the at least three image markers in the target source image; intercepting the image region corresponding to the first position information from the target recorded image as the image to be measured corresponding to the target recorded image; and intercepting the image region corresponding to the second position information from the target source image as the reference image corresponding to the image to be measured in the target source image.

[0011] By presetting at least three image markers in the source image and intercepting the image area according to the positions of the image markers in the recorded image and the source image corresponding to the recorded image, the test image and the reference image corresponding to the test image can be obtained, and the extraction of the test image and the reference image corresponding to the test image can be realized; since the image markers are easy to identify, the extraction speed of the test image and the reference image can be increased.

[0012] Combined with the first aspect, in a possible implementation manner, after obtaining the target recorded image, the method further includes: detecting the image marker in the target recorded image; confirming that the image marker cannot be detected in the target recorded image, or the number of the detected image markers in the target recorded image is less than the preset number of markers, and notifying the source video generating device to adjust the positions of at least three image markers in the source image so that the distances between at least three image markers in the source image are reduced, and the preset number of markers is the number of image markers in the source image.

[0013] When the image marker cannot be detected in the recorded image or the number of detected image markers is less than the number of image markers in the source image, by adjusting the positions of the image markers in the source image so that the distances between the image markers are reduced until the number of detected image markers in the recorded image is equal to the number of image markers in the source image, it can be ensured that a recording system with a small field of view can also record the image markers so as to intercept the test image and the reference image according to the image markers for evaluating the recording quality of the recorded image.

[0014] Combined with the first aspect, in a possible implementation manner, after determining the recording quality of the target recorded image according to the test image and the comparison image, the method further includes: after the source video generating device adjusts the positions of at least three image markers in the source image so that the distances between at least three image markers in the source image change, re-determining the recording quality of the target recorded image; combining the recording qualities of the target recorded image determined multiple times to determine the final recording quality of the target recorded image.

[0015] By adjusting the positions of the image markers in the source image so that the distances between the image markers change, re-determining the recording quality of the target recorded image, and combining the recording qualities of the target recorded image determined multiple times to determine the final recording quality of the target recorded image, the full utilization of the source image can be realized, more quality test results can be obtained, and the reliability of the quality test can be increased.

[0016] In combination with the first aspect, in a possible implementation manner, the extracting, from the target recorded image and the target source image, an image region with the same content as the reference image corresponding to the target recorded image and the reference image corresponding to the target source image in the target source image includes: using the target recorded image as the image to be tested; in the target source image, intercepting the image region corresponding to the target recorded image as the reference image corresponding to the image to be tested in the target source image.

[0017] By using the recorded image as the image to be tested and intercepting the image region corresponding to the recorded image in the source image as the reference image corresponding to the image to be tested, the extraction of the image to be tested and the reference image corresponding to the image to be tested can be realized, and the implementation manner is simple; moreover, by intercepting the image region corresponding to the recorded image in the source image as the reference image, it is not necessary for the recording system to completely record the entire image region of the source image, and the detection requirements are less.

[0018] In combination with the first aspect, in a possible implementation manner, the determining the recording quality of the target recorded image according to the image to be tested and the reference image includes: adjusting the size of the reference image to be the same as the size of the image to be tested to obtain a comparison image corresponding to the image to be tested; determining the recording quality of the target recorded image according to the image to be tested and the comparison image.

[0019] In combination with the first aspect, in a possible implementation manner, each source image in the source video includes an image identifier for uniquely identifying the source image, and different source images include different image identifiers; the obtaining the target source image corresponding to the target recorded image from the source video includes: identifying the image identifier in the target recorded image to obtain a target image identifier; from the source video, determining the source image including the target image identifier as the target source image corresponding to the target recorded image.

[0020] By adding an image identifier in the source image to uniquely identify the source image, frame-by-frame matching of the recorded images in the video to be tested can be realized, preventing the mismatch between the recorded image and the source image caused by frame loss, and ensuring the reliability of the recording quality detection.

[0021] In a second aspect, a video quality detection device is provided, which is applied to a quality detection device in a video quality detection system, wherein the video quality detection system includes a source video generation device, a video playback device, and the quality detection device, wherein the source video generation device is connected to the video playback device, the source video generation device is used to send a source video to the video playback device for playback, the quality detection device is connected to the source video generation device, the quality detection device is used to obtain the source video from the source video generation device, and the source video includes at least one frame of source image; the device includes:

[0022] An image acquisition module, used to acquire a target recorded image, and to acquire a target source image corresponding to the target recorded image from the source video, wherein the target recorded image is any frame of the video to be tested, and the video to be tested is obtained by recording the source video played by the video playback device by the recording system to be evaluated;

[0023] An image extraction module, used for extracting an image area with the same content in the target recorded image and the target source image as an image to be tested corresponding to the target recorded image and a reference image corresponding to the image to be tested in the target source image;

[0024] The quality determination module is used to determine the recording quality of the target recorded image according to the image to be tested and the reference image.

[0025] In a third aspect, a computer device is provided, comprising a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the video quality detection method of the first aspect.

[0026] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor executes the video quality detection method of the first aspect.

[0027] In a fifth aspect, a video quality detection system is provided, comprising a source video generating device, a video playing device and a quality detection device, wherein the source video generating device is connected to the video playing device, and the source video generating device is used to send the source video to the video playing device for playing; the quality detection device is connected to the source video generating device, and the quality detection device obtains the source video from the source video generating device, and the quality detection device is used to execute the video quality detection method of the first aspect mentioned above.

[0028] The present application can achieve the following technical effects: Since the source video corresponding to the target source image is generated by the source video generation device and obtained by the video quality detection device from the source video generation device, and the test video corresponding to the target recorded image is obtained by the recording system to be evaluated by recording the source video played by the video playback device, the source video is equivalent to the reference video of the test video, and the recording quality of the recorded image in the test video can reflect the quality of the recording system to be evaluated. Therefore, by evaluating the recording quality of the recorded image obtained by the recording system, the quality evaluation of the recording system can be achieved; in addition, since the image regions with the same content are extracted from the target recorded image and the target source image as the test image corresponding to the target recorded image and the reference image corresponding to the test image in the target source image, and the recording quality of the target recorded image is determined based on the test image and the reference image in the target recorded image, it is not necessary for the recording system to completely record the entire image region of the source image. The perspective of the recording system and the size of the recording venue can be less restricted, the detection requirements are few, and the quality evaluation of the recording system can be completed in any detection environment, with a wide range of applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments of the present application. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0030] Figure 1 It is a schematic diagram of the system architecture of a video quality detection system provided by an embodiment of the present application;

[0031] Figure 2 It is a schematic flowchart of a video quality detection method provided by an embodiment of the present application;

[0032] Figure 3 It is a schematic diagram of the recorded image and the source image corresponding to the recorded image provided by an embodiment of the present application;

[0033] Figure 4 It is a schematic flowchart of another video quality detection method provided by an embodiment of the present application

[0034] Figure 5 It is a schematic flowchart of yet another video quality detection method provided by an embodiment of the present application;

[0035] Figure 6 It is a schematic diagram of the structure of a video quality detection device provided by an embodiment of the present application;

[0036] Figure 7It is a schematic structural diagram of a computer device provided by an embodiment of the present application. Detailed implementation manners

[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0038] It should be noted that if there is no conflict, the various features in the embodiments of the present application can be combined with each other, and all are within the protection scope of the present application. In addition, although functional module division is performed in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from the module division in the device or the flowchart. Furthermore, the terms "first", "second", "third", etc. used in the present application do not limit the data and execution order, but only distinguish the same items or similar items with basically the same functions and effects.

[0039] The technical solution of the present application is applicable to a video quality evaluation scenario, which can evaluate the quality of a recorded video obtained by a recording system to evaluate the quality of the recording system.

[0040] To implement the quality evaluation of the recorded video, it is necessary to obtain the unencoded source video corresponding to the recorded video. However, the unencoded source video corresponding to the recorded video is in the hands of the device manufacturer corresponding to the recording system that records the recorded video. It is difficult to directly obtain the unencoded source video corresponding to the recorded video. Therefore, it is necessary to evaluate the quality of the recording system by other means.

[0041] In view of this, the present application proposes an evaluation scheme for the quality of a recording system. By setting up a video quality detection system, an unencoded source video is generated by a source video generation device in the video quality detection system, and the unencoded source video is played by a video playback device in the video quality detection system. The source video played by the video playback device is recorded by the recording system to be evaluated to obtain a test video, and the recording quality of the recorded images in the test video is detected. The recording quality of the recorded images in the test video can reflect the quality of the recording system, so the quality evaluation of the recording system can be realized. Moreover, by extracting the image regions with the same content in the recorded image and the corresponding source image as the test image and the reference image corresponding to the test image, and determining the recording quality of the recorded image according to the test image and the reference image corresponding to the test image, it is not necessary for the recording system to completely record the entire image region of the source image. The perspective of the recording system and the size of the recording site can be less restricted, the detection requirements are few, and the quality evaluation of the recording system can be completed in any detection environment, with a wide range of applications.

[0042] The technical solution of the present application will be specifically introduced below.

[0043] For ease of understanding, the video quality detection system of the present application is first introduced. Refer to Figure 1 , Figure 1 which is a schematic diagram of the system architecture of a video quality detection system provided by an embodiment of the present application. As Figure 1 shown, the video quality detection system 10 includes a source video generation device 101, a video playback device 102, and a quality detection device 103. The source video generation device 101 is connected to the video playback device 102, and the quality detection device 103 is connected to the source video generation device 101, where:

[0044] The source video generation device 101 is used to generate a source video, and the source video includes at least one frame of source image. Each frame of source image in the source video is an unencoded source image. The source video generation device 101 can obtain a material video obtained by encoding, decode each frame of the video in the material video one by one to obtain the color encoding data corresponding to each frame of the video. The color encoding data can be YUV format data or RGB format data. An unencoded source image is generated according to the color encoding data, and each frame of unencoded source image generated is sent to the video playback device 102. The multiple frames of unencoded source images sent by the source video generation device 101 form a source video stream.

[0045] In some possible cases, during the process of generating an unencoded source image based on color-coded data, the source video generating device 101 may also add an image identifier to each frame of the source image. The image identifier is used to uniquely identify the source image, and different source images contain different image identifiers. For example, the source video generating device 101 may add a QR code to each frame of the source image, and different identification information is carried in the QR codes of different source images. After adding the image identifier to each frame of the source image, the source video generating device 101 may also record and save the correspondence between the image identifier and the source image. For example, the source video generating device 101 may record and save the correspondence between the frame number of the source image in the source video and the added image identifier. The frame number of the source image in the source video is used to represent the order of the source image in the source video.

[0046] In some possible cases, during the process of generating an unencoded source image based on color-coded data, the source video generating device 101 may also add image markers to each frame of the source image. Each frame of the source image contains at least three image markers. Different image markers are distributed at different image positions in the source image. The image markers can be at any position in the source image, and the positions of the image markers in each frame of the source image are the same. The image markers can be used as positioning markers in the source image to locate an image area in the source image; the image markers can also be used as identification markers in the source image to be quickly identified in the source image. The image markers can be, for example, solid black boxes, positioning codes, etc. After adding the image markers to each frame of the source image, the source video generating device 101 may also record and save the number of image markers in the source image, and record and save the positions of the image markers in the source image.

[0047] In some specific embodiments, during the process of generating an unencoded source image based on color-coded data, the source video generating device 101 may add at least three ArUco codes to each frame of the source image. The ArUco codes can not only be used as image markers but also as image identifiers. Using the ArUco codes as image markers and image identifiers can combine the image markers and image identifiers into one, reducing the types of additional markers added to the source image.

[0048] The source video generating device 101 may also adjust the number and positions of the image markers in the source image according to detection requirements. For example, the source video generating device 101 may also adjust the positions of the image markers in the source image according to actual needs, so that the distance between the image markers increases, thereby expanding the image area located by the image markers in the source image; or, the source video generating device 101 may also adjust the positions of the image markers in the source image according to actual needs, so that the distance between the image markers decreases, thereby reducing the image area located by the image markers in the source image.

[0049] The video playback device 102 may be connected to the source video generation device 101 via a high definition multimedia interface (HDMI). The video playback device 102 is used to play each frame of unencoded source image sent by the source video generation device, thereby playing the source video.

[0050] The quality detection device 103 is used to obtain the source video from the source video generating device 101. The source image in the source video obtained by the quality detection device 103 from the source video generating device 101 is an uncoded source image. The quality detection device 103 is used to connect with the recording system 20 to be evaluated. The quality detection device 103 obtains the video to be tested obtained by recording the source video played by the video playing device 102 by the recording system 20 to be evaluated through the connection with the recording system 20 to be evaluated. The recording system 20 to be evaluated refers to the recording system whose recording quality needs to be determined. The recording system may include a camera and an encoder corresponding to the camera, etc.

[0051] Among them, the source video generating device 101, the video playing device 102 and the quality detection device 103 can be independent devices, or they can be integrated into one device as related functional modules. Regarding the specific forms of the source video generating device 101, the video playing device 102 and the quality detection device 103, this application does not limit this.

[0052] based on Figure 1 The video quality detection system shown can implement the technical solution of the present application, and the technical solution of the present application can be applied to a video quality detection device in a video quality detection system.

[0053] See also Figure 2 , Figure 2 A flow chart of a video quality detection method provided in an embodiment of the present application is shown as follows: Figure 2 As shown, the method comprises the following steps:

[0054] S201, obtaining a target recorded image.

[0055] Here, the target recorded image is any frame image in the video to be tested. The video to be tested is obtained by recording the source video played by the video playback device by the recording system to be evaluated. For an introduction to the recording system to be evaluated, the video playback device and the source video, please refer to the above description.

[0056] The video quality detection device may obtain the video to be tested from the recording system to be evaluated, and obtain a frame of image from the video to be tested as a target recorded image.

[0057] S202, obtaining a target source image corresponding to the target recorded image from the source video.

[0058] In a feasible implementation, the frame number of the target recorded image in the video to be tested can be determined to obtain the target frame number, and the target frame number is used to represent the order of the target recorded image in the video to be tested; in the source video, the source image with the frame number being the target frame number is determined as the target source image corresponding to the target recorded image. This implementation is applicable to the situation where while the video playback device plays the source video, the recording device to be evaluated synchronously records the source video.

[0059] In another feasible implementation, when each source image in the source video contains an image identifier for uniquely identifying the source image, the target source image corresponding to the target recorded image can be obtained from the source video through the following steps A1 - A2:

[0060] A1. Identify the image identifier in the target recorded image to obtain the target image identifier.

[0061] A2. From the source video, the source image containing the target image identifier is determined as the target source image corresponding to the target recorded image.

[0062] Among them, the correspondence between the image identifier and the source image can be obtained from the source video generation device, and according to the correspondence between the image identifier and the source image, the source image containing the target image identifier is determined.

[0063] By adding an image identifier to the source image to uniquely identify the source image, frame - by - frame matching of the recorded images in the video to be tested can be achieved, preventing the mismatch between the recorded image and the source image caused by dropped frames, and ensuring the reliability of the recording quality detection.

[0064] S203. Extract the image regions with the same content in the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image.

[0065] Here, the image to be tested corresponding to the target recorded image is obtained based on the image region extracted from the target recorded image and is used to reflect the image content in the target recorded image; the reference image corresponding to the image to be tested in the target source image is obtained based on the image region extracted from the target source image and is used to reflect the image content in the target source image.

[0066] In some possible cases, the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image can be determined through the following steps B1 - B2:

[0067] B1. Take the target recorded image as the image to be tested.

[0068] B2. In the target source image, intercept the image area corresponding to the target recorded image as the reference image corresponding to the image to be measured in the target source image.

[0069] Among them, the image area corresponding to the target recorded image refers to the area formed by the four matching points corresponding to the four vertices of the target recorded image in the target source image. For example, if the target source image is as shown by P1 in Figure 3 and the target recorded image is as shown by P2 in Figure 3 , then the image area corresponding to the target recorded image is as shown by Q1 in Figure 3 .

[0070] In a feasible implementation manner, the reference image corresponding to the image to be measured can be determined through the following steps B21 - B24:

[0071] B21. Determine at least four groups of matching feature points according to the target source image and the target recorded image.

[0072] Here, each group of matching feature points includes a first feature point and a second feature point that matches the first feature point. The first feature point is a feature point in the target recorded image, and the second feature point is a feature point in the target source image.

[0073] When the source image contains at least three image markers, at least the three image markers can be respectively identified from the target source image and the target recorded image, and the positions of the at least three image markers in the target source image and the target recorded image are respectively used as the first feature point and the second feature point.

[0074] Taking the target source image and the target recorded image as shown by P1 and P2 in Figure 3 as an example, both the target source image and the target recorded image contain four ArUco codes, and the ArUco code is an image marker. Then, four ArUco codes can be identified in the target recorded image, the positions of the four ArUco codes in the target recorded image are determined as the first feature points, and four ArUco codes are identified in the target source image, and the positions of the four ArUco codes in the target source image are determined as the second feature points.

[0075] Optionally, feature points can also be extracted from the target recorded image to obtain the feature points in the target recorded image, and feature points can be extracted from the target source image to obtain the feature points in the target source image. The feature points in the target recorded image are matched with the feature points in the target source image to obtain the first feature point and the second feature point.

[0076] Among them, any one or more feature point extraction algorithms can be used to extract feature points from the target recorded image and the target source image, so as to obtain the feature points in the target recorded image and the feature points in the target source image. The feature point extraction algorithms include, but are not limited to, the scale-invariant feature transform (SIFT) algorithm, the oriented FAST and rotated BRIEF (ORB) algorithm, etc.

[0077] Among them, based on any one feature point matching algorithm, the feature points in the target recorded image can be matched with the feature points in the target source image to obtain the first feature point and the second feature point.

[0078] In a specific implementation manner, for any feature point (hereinafter referred to as the to-be-matched feature point) in the target recorded image obtained by extraction, the feature descriptor of the to-be-matched feature point can be obtained. The feature descriptor is an attribute parameter used to describe the feature point, usually presented as a multi-dimensional position vector, and the feature descriptor is obtained by the feature point extraction algorithm; calculate the Euclidean distance between the feature descriptor of the to-be-matched feature point and the feature descriptors of the feature points in the target source image. If the Euclidean distance between the feature descriptor of the to-be-matched feature point and the feature descriptor of the target feature point in the target source image is less than the preset distance threshold, then the to-be-matched feature point is determined as the first feature point, and the target feature point is determined as the second feature point; if the Euclidean distance between the feature descriptor of the to-be-matched feature point and the feature descriptors of each feature point in the target source image is greater than or equal to the preset distance threshold, it is determined that there is no feature point in the target source image that matches the to-be-matched feature point. For each feature point in the target recorded image, the feature point matching is performed with the feature points in the target source image in the same manner, then all the first feature points and the second feature points can be determined, that is, all the matching feature points in the target recorded image and the target source image are determined.

[0079] B22. Determine the homography matrix between the target recorded image and the target source image according to at least four groups of matching feature points.

[0080] Here, the homography matrix between the target recorded image and the target source image is used to describe the affine transformation between the target recorded image and the target source image.

[0081] Among them, the homography matrix between the target recorded image and the target source image can be expressed as:

[0082]

[0083] The first feature point and the second feature point satisfy the following corresponding relationship:

[0084]

[0085] (x 1 ,y 1 ) represents the position coordinates of the first feature point, and (x 2 ,y 2 ) represents the position coordinates of the second feature point. According to the position coordinates of at least four groups of matching feature points, the homography matrix between the target recorded image and the target source image can be solved.

[0086] B23. Determine the matching points of the four vertices of the image to be measured in the target source image according to the homography matrix between the target recorded image and the target source image, and obtain four vertex matching points.

[0087] Among them, according to the corresponding relationship between the above-mentioned first feature point and the second feature point, the matching points of the four vertices of the image to be measured in the target source image can be determined, and four vertex matching points can be obtained.

[0088] B24. In the target source image, intercept the image area corresponding to the four vertex matching points as the reference image corresponding to the image to be measured.

[0089] In the above steps B1 - B2, by using the recorded image as the image to be measured and intercepting the image area corresponding to the recorded image in the source image as the reference image corresponding to the image to be measured, the extraction of the image to be measured and the reference image corresponding to the image to be measured can be realized, and the implementation method is simple; moreover, by intercepting the image area corresponding to the recorded image in the source image as the reference image, it is not necessary for the recording system to completely record the entire image area of the source image, and the detection requirements are less.

[0090] In another feasible implementation manner, the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image can also be determined through subsequent steps S306 - S308, and the present application does not limit this.

[0091] S204. Determine the recording quality of the target recorded image according to the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image.

[0092] Among them, the recording quality of the target recorded image can be determined through the following steps C1 - C2:

[0093] C1. Adjust the size of the reference image corresponding to the image to be measured to be the same as the size of the image to be measured, and obtain the comparison image corresponding to the image to be measured.

[0094] C2. Determine the recording quality of the target recorded image according to the image to be measured and the comparison image corresponding to the image to be measured.

[0095] By adjusting the size of the reference to be the same as the size of the image to be tested, a comparison image corresponding to the image to be tested is obtained, and by comparing the image to be tested and the comparison image, the recording quality of the recorded image can be detected; by determining the recording quality of the recorded image after adjusting the size of the reference image to be the same as the size of the image to be tested, a one-by-one comparison of the pixels in the image to be tested and the comparison image can be achieved, ensuring the reliability of the recording quality.

[0096] In a feasible implementation, the image to be tested and the comparison image corresponding to the image to be tested can be input into a video multimethod assessment fusion (VMAF) model for VMAF scoring, and the score output by the VMAF model can be used as the recording quality score of the target recorded image. The higher the score output by the VMAF model, the better the recording quality of the target recorded image, and the lower the score output by the VMAF model, the worse the recording quality of the target recorded image.

[0097] According to the above steps S201 to S204, the recording quality of each frame of the recorded image in the video to be tested can be determined, and the recording quality of each frame of the recorded image in the video to be tested can be obtained. Furthermore, the quality of the recording system to be evaluated can be determined based on the recording quality of each frame of the recorded image in the video to be tested. Exemplarily, the quality score of the recording system to be evaluated can be the mean or sum of the recording quality scores of each frame of the recorded image in the video to be tested.

[0098] In the above Figure 2In the corresponding technical solution, after the target recorded image is acquired and the target source image corresponding to the target recorded image is acquired from the source video, an image area with the same content is extracted from the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image, and the recording quality of the target recorded image is determined according to the image to be tested and the reference image; since the source video corresponding to the target source image is generated by a source video generating device and acquired by a video quality detection device from the source video generating device, the video to be tested corresponding to the target recorded image is obtained by recording the source video played by a video playback device by a recording system to be evaluated, the source video is equivalent to the reference video of the video to be tested, and the recorded image in the video to be tested is The recording quality can reflect the quality of the recording system to be evaluated. Therefore, by evaluating the recording quality of the recorded images obtained by the recording system, the quality evaluation of the recording system can be achieved. In addition, since the image area with the same content is extracted from the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image, and the recording quality of the target recorded image is determined based on the image to be tested in the target recorded image and the reference image, the recording system does not need to fully record the entire image area of ​​the source image, the viewing angle of the recording system and the size of the recording site can be free from excessive restrictions, the detection requirements are few, and the quality evaluation of the recording system can be completed in any detection environment, with a wide range of applications.

[0099] See also Figure 4 , Figure 4 A flowchart of another video quality detection method provided in an embodiment of the present application is shown as follows: Figure 4 As shown, the method comprises the following steps:

[0100] S301, obtaining a target recorded image.

[0101] Here, the specific implementation of step S301 can refer to the description of the aforementioned step S201, which will not be repeated here.

[0102] S302: Determine whether an image mark is detected in the target recorded image.

[0103] Here, for the definition of image tags, please refer to the above description.

[0104] If the image mark is detected in the target recorded image, step S303 is executed; if the image mark is not detected in the target recorded image, step S304 is executed.

[0105] S303: Determine whether the number of image marks in the target recorded image is less than a preset number of marks.

[0106] Here, the preset number of markers is the number of image markers included in the source image.

[0107] If the number of image markers in the target recorded image is less than the preset number of markers, step S304 is executed; if the number of image markers in the target recorded image is equal to the preset number of markers, step S305 is executed.

[0108] S304, Notify the source video generating device to adjust the positions of at least three image markers in the source image of the source video so that the distances between at least three image markers in the source image of the source video are reduced, and then execute step S301.

[0109] S305, Obtain the target source image corresponding to the target recorded image from the source video.

[0110] Here, for the specific implementation manner of step S305, reference can be made to the description of the foregoing step S202, which will not be elaborated here.

[0111] S306, Obtain the first position information and the second position information.

[0112] Here, the first position information includes the position of each of at least three image markers in the target recorded image, and the second position information includes the position of each of at least three image markers in the target source image. For example, if the source icon includes 4 image markers, the first position information includes the position of each of the 4 image markers in the target recorded image, and the second position information includes the position of each of the 4 image markers in the target recorded image.

[0113] If the target recorded image is the first frame image in the video to be measured, the positions of at least three image markers in the target recorded image can be recognized to obtain the first position information; if the target recorded image is not the first frame image in the video to be measured, the positions of at least three image markers in the target recorded image can be recognized to obtain the first position information, or the position information obtained by directly recognizing the positions of at least three image markers in the first frame image of the video to be measured can be used as the first position information.

[0114] Among them, the second position information can be obtained from the source video generating device.

[0115] S307, In the target recorded image, intercept the image area corresponding to the first position information as the image to be measured in the target recorded image.

[0116] Here, the image region corresponding to the first position information refers to the image region corresponding to the first circumscribed rectangle determined based on the image markers in the target recorded image. The first circumscribed rectangle can enclose all the image markers in the target recorded image. For example, if the target recorded image is as shown in Figure 3 P2, the first circumscribed rectangle can be as shown in Figure 3 J2 in P2.

[0117] Among them, the maximum coordinate values and minimum coordinate values of the image markers in the target recorded image in two directions of the target recorded image can be determined according to the first position information. Based on the maximum coordinate values and minimum coordinate values of the image markers in the target recorded image in two directions of the target recorded image, the first circumscribed rectangle is determined; the four vertex coordinates of the first circumscribed rectangle are expressed as (x min1 , y min1 ), (x min1 , y max1 ), (x max1 , y min1 ), (x max1 , y max1 ), x min1 and x max1 represent the minimum and maximum values of the image markers in the target recorded image in the x-axis direction of the target recorded image, and y min1 and y max1 represent the minimum and maximum values of the image markers in the target recorded image in the y-axis direction of the target recorded image.

[0118] Taking the first circumscribed rectangle as shown in Figure 3 J2 in P2 as an example, then x min1 is the x coordinate value of the image marker j21, x max1 is the x coordinate value of the image marker j22, y min1 is the y coordinate value of the image marker j23, and y max1 is the y coordinate value of the image marker j24.

[0119] S308. From the target source image, intercept the image region corresponding to the second position information as the reference image corresponding to the image to be tested in the target source image.

[0120] Here, the image region corresponding to the second position information refers to the image region corresponding to the second circumscribed rectangle determined based on the image markers in the target source image. The second circumscribed rectangle can enclose all the image markers in the target source image. For example, if the target source image is as shown in Figure 3 P1, the second circumscribed rectangle can be as shown in Figure 3 J1 in P1.

[0121] Among them, according to the second position information, the maximum coordinate values and minimum coordinate values of the image markers in the target source image in two directions can be determined, and according to the maximum coordinate values and minimum coordinate values of the image markers in the target source image in two directions, a second circumscribed rectangle is determined; the four vertex coordinates of the second circumscribed rectangle are expressed as (x min2 , y min2 ), (x min2 , y max2 ), (x max2 , y min2 ), (x max2 , y max2 ), x min2 and x max2 represent the minimum value and maximum value of the image markers in the target source image in the x-axis direction of the target source image, and y min2 and y max2 represent the minimum value and maximum value of the image markers in the target source image in the y-axis direction of the target source image.

[0122] Taking the example of J1 in P1 in the second circumscribed rectangle as Figure 3 shown, then x min2 is the x coordinate value of the image marker j11, x max2 is the x coordinate value of the image marker j12, y min2 is the y coordinate value of the image marker j13, and y max2 is the y coordinate value of the image marker j14.

[0123] Through the above steps S306 - S308, by presetting at least three image markers in the source image and intercepting the image area according to the positions of the image markers in the recorded image and the source image corresponding to the recorded image, the image to be measured and the reference image corresponding to the image to be measured can be obtained, and the extraction of the image to be measured and the reference image corresponding to the image to be measured can be realized; since the image markers are easy to identify, the extraction speed of the image to be measured and the reference image can be accelerated.

[0124] S309. Determine the recording quality of the target recorded image according to the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image.

[0125] Here, for the specific implementation manner of step S309, reference can be made to the description of the foregoing step S204, which will not be elaborated here.

[0126] In the above Figure 4In the corresponding technical solution, when no image marker is detected in the recorded image or the number of detected image markers is less than the number of image markers in the source image, the positions of the image markers in the source image are adjusted to reduce the distances between the image markers until the number of detected image markers in the recorded image is equal to the number of image markers in the source image, so as to ensure that a recording system with a small field of view can also record the image markers, and the recorded image and the reference image can be intercepted according to the image markers for evaluating the recording quality of the recorded image.

[0127] See Figure 5 , Figure 5 which is a schematic flowchart of another video quality detection method provided by an embodiment of the present application. As Figure 5 shown, the method includes the following steps:

[0128] S401, Obtain a target recorded image.

[0129] S402, Determine whether an image marker is detected in the target recorded image.

[0130] If an image marker is detected in the target recorded image, execute step S403; if no image marker is detected in the target recorded image, it means that the image marker in the source image is outside the viewing angle of the recording device to be evaluated, and execute step S411.

[0131] S403, Determine whether the number of image markers in the target recorded image is less than a preset marker number.

[0132] If the number of image markers in the target recorded image is less than the preset marker number, execute step S411; if the number of image markers in the target recorded image is equal to the preset marker number, execute step S404.

[0133] S404, Obtain a target source image corresponding to the target recorded image from the source video.

[0134] S405, Obtain first position information and second position information.

[0135] S406, Intercept the image area corresponding to the first position information from the target recorded image as the image to be measured in the target recorded image.

[0136] S407, Intercept the image area corresponding to the second position information from the target source image as the reference image corresponding to the image to be measured in the target source image.

[0137] S408, Determine the recording quality of the target recorded image according to the image to be measured corresponding to the target recorded image and the reference image corresponding to the image to be measured in the target source image.

[0138] Here, for the specific implementation manners of steps S404 to S408, reference may be made to the descriptions of the foregoing steps S305 to S309, which will not be elaborated herein.

[0139] S409, determine whether the quantity of the recording quality of the target recorded image reaches the quality quantity threshold.

[0140] Here, the quantity of the recording quality of the target recorded image is used to reflect the number of times of determining the recording quality of the current target recorded image.

[0141] If the quantity of the recording quality of the target recorded image reaches the quality quantity threshold, execute step S412; if the quantity of the recording quality of the target recorded image does not reach the quality quantity threshold, execute step S410.

[0142] S410, the source video generating device adjusts the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video change, and then execute step S401.

[0143] Wherein, the source video generating device can adjust the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video increase, or can also adjust the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video decrease. It can be that after the video playing device finishes playing the source video, the source video generating device automatically adjusts the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video change; or it can be that after the video quality detection device detects the recording quality of each frame image in the to-be-tested video, it notifies the source video generating device to adjust the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video change. The present application does not limit the triggering manner for the source video generating device to adjust the positions of at least three image markers in the source image in the source video, so that the distances between at least three image markers in the source image in the source video change.

[0144] S411, determine whether the quantity of the recording quality of the target recorded image reaches the quality quantity threshold.

[0145] If the quantity of the recording quality of the target recorded image reaches the quality quantity threshold, execute step S412; if the quantity of the recording quality of the target recorded image does not reach the quality quantity threshold, execute step S401.

[0146] S412, combine the recording qualities of the target recorded image determined multiple times to determine the final recording quality of the target recorded image.

[0147] In a feasible implementation manner, an average of the recording quality scores of the target recorded image determined multiple times may be determined as the final recording quality score of the target recorded image.

[0148] It can be understood that, when the final recording quality of the target recorded image is determined in combination with the recording qualities of the target recorded images determined multiple times, the final recording quality of each frame of the recorded image in the video to be tested can be determined in accordance with the above steps S401 to S412, and the quality of the recording system to be evaluated can be determined based on the final recording quality of each frame of the recorded image in the video to be tested.

[0149] In the above Figure 5 In the corresponding technical solution, by adjusting the positions of image marks in the source image so that the distances between the image marks change, and re-determining the recording quality of the target recorded image, and combining the recording qualities of the target recorded image determined multiple times, the final recording quality of the target recorded image is determined, thereby making full use of the source image, obtaining more quality test results, and increasing the reliability of the quality test.

[0150] The method of the present application is introduced above, and the device of the present application is introduced below.

[0151] See also Figure 6 , Figure 6 is a structural schematic diagram of a video quality detection device provided in an embodiment of the present application, wherein the video quality detection device is applied to a quality detection device in a video quality detection system, wherein the video quality detection system comprises a source video generation device, a video playback device and the quality detection device, wherein the source video generation device is connected to the video playback device, wherein the source video generation device is used to send a source video to the video playback device for playback, wherein the quality detection device is connected to the source video generation device, wherein the quality detection device is used to obtain the source video from the source video generation device, wherein the source video comprises at least one frame of a source image; Figure 6 As shown, the video quality detection device 50 includes:

[0152] An image acquisition module 501 is used to acquire a target recorded image and acquire a target source image corresponding to the target recorded image from the source video, wherein the target recorded image is any frame of the video to be tested, and the video to be tested is obtained by recording the source video played by the video playback device by the recording system to be evaluated;

[0153] An image extraction module 502 is used to extract an image region with the same content in the target recorded image and the target source image as an image to be tested corresponding to the target recorded image and a reference image corresponding to the image to be tested in the target source image;

[0154] A quality determination module 503 is configured to determine the recording quality of the target recorded image according to the image to be measured and the reference image.

[0155] It should be noted that the above video quality detection device 50 can execute the above video quality detection method provided by the embodiments of the present application, and has corresponding functional modules and beneficial effects for executing the method. For technical details not described in detail in the embodiments of the video quality detection device 50, reference can be made to the above video quality detection method provided by the embodiments of the present application.

[0156] See Figure 7 , Figure 7 FIG. is a schematic structural diagram of a computer device provided by an embodiment of the present application. The computer device 60 includes a processor 601 and a memory 602. The memory 602 is connected to the processor 601. The memory 602 is connected to the processor 601 through a bus, for example.

[0157] The processor 601 is configured to support the computer device 60 to execute the corresponding functions in the method in the above method embodiments. The processor 601 may be a central processing unit (CPU), a network processor (NP), a hardware chip, or any combination thereof. The above hardware chip may be an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The above PLD may be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0158] The memory 602 is used to store program codes and the like. The memory 602 may include volatile memory (VM), such as random access memory (RAM); the memory 602 may also include non-volatile memory (NVM), such as read-only memory (ROM), flash memory, hard disk drive (HDD) or solid-state drive (SSD); the memory 602 may further include a combination of the above types of memories.

[0159] The memory 602 is used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as the program instructions / modules corresponding to the video quality detection method in the embodiments of the present application. The processor executes various functional applications and data processing of the video quality detection method by running the non-volatile software programs, instructions, and modules stored in the memory, that is, implements the functions of the video quality detection method provided in the above method embodiments.

[0160] The memory 602 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system and application programs required for at least one function. The data storage area may store data created according to the use of the video quality detection device. In some embodiments, the memory may include a memory remotely set relative to the processor, and these remote memories may be connected to the video quality detection device through a network. Examples of the above network include but are not limited to the Internet, enterprise intranet, local area network, mobile communication network, and their combinations.

[0161] The one or more modules are stored in the memory and, when executed by the one or more processors, execute the video quality detection method in any of the above method embodiments. For example, execute the method steps described in the above method embodiments to implement the functions of the modules described in the above device embodiments.

[0162] The embodiments of the present application further provide a computer-readable storage medium. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the method as described in the foregoing embodiments.

[0163] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above various methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0164] The above-disclosed are only the preferred embodiments of the present application. Of course, the scope of rights of the present application cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A video quality detection method, characterized in that: A quality detection device applied to a video quality detection system, the video quality detection system comprising a source video generation device, a video playback device and the quality detection device, the source video generation device is connected to the video playback device, the source video generation device is used to send a source video to the video playback device for playback, the quality detection device is connected to the source video generation device, the quality detection device is used to obtain the source video from the source video generation device, the source video comprises at least one frame of source image; the method comprises: Acquire a target recorded image, and acquire a target source image corresponding to the target recorded image from the source video, wherein the target recorded image is any frame of the video to be tested, and the video to be tested is obtained by recording the source video played by the video playback device by the recording system to be evaluated; Extracting image regions with the same content in the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image; The recording quality of the target recorded image is determined according to the image to be tested and the reference image.

2. The method according to claim 1, characterized in that Each source image frame in the source video includes at least three image markers, and the at least three image markers are distributed at different image positions in the source image; The step of extracting an image region with the same content in the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image comprises: Acquire first position information and second position information, wherein the first position information includes the position of each of the at least three image marks in the target recorded image, and the second position information includes the position of each of the at least three image marks in the target source image; From the target recorded image, intercepting an image area corresponding to the first position information as an image to be tested corresponding to the target recorded image; An image region corresponding to the second position information is intercepted from the target source image as a reference image corresponding to the image to be measured in the target source image.

3. The method according to claim 2, characterized in that After acquiring the target recorded image, the method further includes: detecting the image marker in the target recorded image; Confirm that the image mark cannot be detected in the target recorded image, or that the number of the image marks detected in the target recorded image is less than a preset number of marks, and notify the source video generating device to adjust the positions of at least three image marks in the source image so that the distances between the at least three image marks in the source image are reduced, and the preset number of marks is the number of image marks in the source image.

4. The method according to claim 2, characterized in that: After determining the recording quality of the target recorded image according to the image to be tested and the reference image, the method further includes: After the source video generation device adjusts the positions of at least three image marks in the source image so that the distances between the at least three image marks in the source image change, re-determining the recording quality of the target recorded image; The final recording quality of the target recorded image is determined by combining the recording qualities of the target recorded image determined multiple times.

5. The method according to claim 1, characterized in that The step of extracting an image region with the same content in the target recorded image and the target source image as the image to be tested corresponding to the target recorded image and the reference image corresponding to the image to be tested in the target source image comprises: Using the target recorded image as the image to be tested; In the target source image, an image area corresponding to the target recorded image is intercepted as a reference image corresponding to the image to be tested in the target source image.

6. The method according to any one of claims 1 to 5, characterized in that: The step of determining the recording quality of the target recorded image according to the image to be tested and the reference image includes: Adjusting the size of the reference image to be the same as the size of the image to be tested, to obtain a comparison image corresponding to the image to be tested; The recording quality of the target recorded image is determined according to the image to be tested and the comparison image.

7. The method according to any one of claims 1 to 5, characterized in that: Each source image frame in the source video contains an image identifier, and the image identifier is used to uniquely identify the source image, and different source images contain different image identifiers; The step of acquiring a target source image corresponding to the target recorded image from the source video includes: Identify the image identifier in the target recorded image to obtain the target image identifier; From the source video, a source image containing the target image identifier is determined as a target source image corresponding to the target recorded image.

8. A computer device, characterized in that: The method comprises a memory and a processor, wherein the memory is connected to the processor, and the processor is used to execute one or more computer programs stored in the memory. When the processor executes the one or more computer programs, the computer device implements the method according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

10. A video quality detection system, characterized in that: It includes a source video generating device, a video playing device and a quality detection device, wherein the source video generating device is connected to the video playing device, and the source video generating device is used to send the source video to the video playing device for playing; The quality detection device is connected to the source video generating device, the quality detection device is used to obtain the source video from the source video generating device, and the quality detection device is used to execute the method according to any one of claims 1-7.

Images