技术领域Technical field
本发明涉及通信技术领域，特别是指一种视频质量评价方法、装置及设备。The present invention relates to the field of communication technology, and in particular, to a video quality evaluation method, device and equipment.
背景技术Background technique
目前清晰度更高的视频可以是通过原始较低清晰度视频上变换得到，这样，在分辨率尺度上图像有人为的拉伸痕迹，影响观影体验。Currently, higher-definition videos can be obtained by up-converting the original lower-definition videos. In this way, the image has artificial stretching marks on the resolution scale, which affects the viewing experience.
而现有的视频质量评价方式，是使用不同的比特率以及不同的编码结构对待测视频进行编、解码，测试编解码前后的视频并对其视频质量进行客观评估分析。但是，不同的上变换方法，得到的视频并不相同，采用现有的质量评价方法，对上变换方式不做区分，因此，质量评价的结果不准确。The existing video quality evaluation method is to use different bit rates and different encoding structures to encode and decode the video to be tested, test the video before and after encoding and decoding, and conduct an objective evaluation and analysis of the video quality. However, different up-conversion methods produce different videos. The existing quality evaluation method does not differentiate between up-conversion methods. Therefore, the quality evaluation results are inaccurate.
发明内容Contents of the invention
本发明的目的是提供一种视频质量评价方法、装置及设备，以对视频质量进行更为准确的评价。The purpose of the present invention is to provide a video quality evaluation method, device and equipment to more accurately evaluate video quality.
为达到上述目的，本发明的实施例提供一种视频质量评价方法，包括：To achieve the above objectives, embodiments of the present invention provide a video quality evaluation method, including:
确定与待评价视频对应的参考视频；Determine the reference video corresponding to the video to be evaluated;
根据多个预设上变换方式，对所述参考视频进行上变换处理，得到所述参考视频的多个目标图像；Perform up-conversion processing on the reference video according to multiple preset up-conversion methods to obtain multiple target images of the reference video;
根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式；According to the plurality of target images, determine the up-conversion method of the frame image in the video to be evaluated;
根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果。According to the up-conversion method of frame images in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained.
可选地，所述根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式，包括：Optionally, determining an up-conversion method of frame images in the video to be evaluated based on the plurality of target images includes:
根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数；According to the plurality of target images, obtain the quality parameters of the target frame image in the video to be evaluated;
根据所述质量参数，确定所述目标帧图像的上变换方式；Determine the up-conversion method of the target frame image according to the quality parameter;
其中，所述质量参数包括：峰值信噪比PSNR和结构相似性SSIM。Wherein, the quality parameters include: peak signal-to-noise ratio PSNR and structural similarity SSIM.
可选地，所述根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数，包括：Optionally, obtaining the quality parameters of the target frame image in the video to be evaluated based on the plurality of target images includes:
计算所述待评价视频中目标帧图像与所述多个目标图像的第一组图像中各图像间的PSNR和SSIM；Calculate PSNR and SSIM between the target frame image in the video to be evaluated and each image in the first group of images of the plurality of target images;
将所述计算得到的PSNR中的最大PSNR作为所述目标帧图像的PSNR；Use the maximum PSNR among the calculated PSNRs as the PSNR of the target frame image;
将所述计算得到的SSIM中的最大SSIM作为所述目标帧图像的SSIM；Use the maximum SSIM among the calculated SSIMs as the SSIM of the target frame image;
其中，所述第一组图像是所述多个目标图像中对应所述目标帧图像的一组图像，且各个图像对应不同的预设上变换方式。Wherein, the first group of images is a group of images corresponding to the target frame image among the plurality of target images, and each image corresponds to a different preset up-conversion method.
可选地，所述根据所述质量参数，确定所述目标帧图像的上变换方式，包括：Optionally, determining the up-conversion method of the target frame image according to the quality parameter includes:
将所述目标帧图像的PSNR和/或SSIM对应的预设上变换方式，作为所述目标帧图像的上变换方式；或者，Use the preset up-conversion method corresponding to the PSNR and/or SSIM of the target frame image as the up-conversion method of the target frame image; or,
通过多类别分类模型以及所述目标帧图像与所述第一组图像中各图像间的PSNR和SSIM，得到所述目标帧图像的上变换方式。The up-conversion method of the target frame image is obtained through the multi-category classification model and the PSNR and SSIM between the target frame image and each image in the first group of images.
可选地，所述多类别分类模型是已构建的基于帧图像的PSNR和SSIM，确定帧图像的上变换方式的神经网络模型。Optionally, the multi-category classification model is a neural network model that has been constructed based on the PSNR and SSIM of the frame image and determines the up-conversion method of the frame image.
可选地，所述根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果，包括：Optionally, obtaining the quality evaluation result of the video to be evaluated based on the up-conversion method of the frame image in the video to be evaluated includes:
确定所述待评价视频中，对应帧图像数目最多的目标上变换方式；Determine the target up-conversion method with the largest number of corresponding frame images in the video to be evaluated;
根据所述目标上变换方式所属的上变换方式类别，得到所述待评价视频的质量评价结果；或者，Obtain the quality evaluation result of the video to be evaluated according to the up-conversion method category to which the target up-conversion method belongs; or,
获取所述目标上变换方式对应的帧图像的质量参数的均值，并根据所述均值所属的阈值范围得到所述待评价视频的质量评价结果。Obtain the mean value of the quality parameters of the frame image corresponding to the target up-conversion method, and obtain the quality evaluation result of the video to be evaluated according to the threshold range to which the mean value belongs.
可选地，所述质量评价结果包括：所述待评价视频是否为具有对应清晰度的视频。Optionally, the quality evaluation result includes: whether the video to be evaluated is a video with corresponding definition.
为达到上述目的，本发明的实施例提供一种视频质量评价装置，包括：To achieve the above objectives, embodiments of the present invention provide a video quality evaluation device, including:
第一处理模块，用于确定与待评价视频对应的参考视频；The first processing module is used to determine the reference video corresponding to the video to be evaluated;
第二处理模块，用于根据多个预设上变换方式，对所述参考视频进行上变换处理，得到所述参考视频的多个目标图像；A second processing module, configured to perform up-conversion processing on the reference video according to multiple preset up-conversion methods to obtain multiple target images of the reference video;
第三处理模块，用于根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式；A third processing module, configured to determine the up-conversion method of frame images in the video to be evaluated based on the plurality of target images;
第四处理模块，用于根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果。The fourth processing module is used to obtain the quality evaluation result of the video to be evaluated according to the up-conversion method of the frame image in the video to be evaluated.
为达到上述目的，本发明的实施例提供一种视频质量评价设备，包括收发器、处理器、存储器及存储在所述存储器上并可在所述处理器上运行的程序或指令；所述处理器执行所述程序或指令时实现如上所述的视频质量评价方法。To achieve the above objectives, embodiments of the present invention provide a video quality evaluation device, including a transceiver, a processor, a memory, and a program or instructions stored on the memory and executable on the processor; the processing When the program or instruction is executed by the computer, the video quality evaluation method as described above is implemented.
为达到上述目的，本发明的实施例提供一种可读存储介质，其上存储有程序或指令，所述程序或指令被处理器执行时实现如上所述的视频质量评价方法中的步骤。In order to achieve the above object, embodiments of the present invention provide a readable storage medium on which a program or instructions are stored. When the program or instructions are executed by a processor, the steps in the video quality evaluation method as described above are implemented.
本发明的上述技术方案的有益效果如下：The beneficial effects of the above technical solutions of the present invention are as follows:
本发明实施例的方法，针对待评价视频，将先确定与其对应的参考视频；然后由多个预设上变换方式，对已确定的参考视频进行上变换处理，得到多个目标图像；之后，再根据所得的多个目标图像，进一步确定该待评价视频中帧图像的上变换方式；最终，结合该待评价视频中帧图像的上变换方式，得到待评价视频的质量评价结果。由于评价过程中考虑了视频的上变换方式，增加了质量评价结果的准确性。According to the method of the embodiment of the present invention, for the video to be evaluated, the corresponding reference video will first be determined; then, the determined reference video will be up-converted using multiple preset up-conversion methods to obtain multiple target images; and then, Then based on the multiple target images obtained, the up-conversion method of the frame image in the video to be evaluated is further determined; finally, combined with the up-conversion method of the frame image in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained. Since the up-conversion method of the video is considered in the evaluation process, the accuracy of the quality evaluation results is increased.
附图说明Description of drawings
图1为本发明实施例的视频质量评价方法的流程图；Figure 1 is a flow chart of a video quality evaluation method according to an embodiment of the present invention;
图2为本发明实施例的视频质量评价装置的结构图；Figure 2 is a structural diagram of a video quality evaluation device according to an embodiment of the present invention;
图3为本发明实施例的视频质量评价设备的结构图；Figure 3 is a structural diagram of a video quality evaluation device according to an embodiment of the present invention;
图4为本发明另一实施例的视频质量评价设备的结构图。Figure 4 is a structural diagram of a video quality evaluation device according to another embodiment of the present invention.
具体实施方式Detailed ways
为使本发明要解决的技术问题、技术方案和优点更加清楚，下面将结合附图及具体实施例进行详细描述。In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.
应理解，说明书通篇中提到的“一个实施例”或“一实施例”意味着与实施例有关的特定特征、结构或特性包括在本发明的至少一个实施例中。因此，在整个说明书各处出现的“在一个实施例中”或“在一实施例中”未必一定指相同的实施例。此外，这些特定的特征、结构或特性可以任意适合的方式结合在一个或多个实施例中。It will be understood that reference throughout this specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic associated with the embodiment is included in at least one embodiment of the invention. Thus, the appearances of "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily referring to the same embodiment. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner in one or more embodiments.
在本发明的各种实施例中，应理解，下述各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本发明实施例的实施过程构成任何限定。In various embodiments of the present invention, it should be understood that the size of the sequence numbers of the following processes does not mean the order of execution. The execution order of each process should be determined by its functions and internal logic, and should not be implemented in the present invention. The implementation of the examples does not constitute any limitations.
另外，本文中术语“系统”和“网络”在本文中常可互换使用。Additionally, the terms "system" and "network" are often used interchangeably in this article.
在本申请所提供的实施例中，应理解，“与A相应的B”表示B与A相关联，根据A可以确定B。但还应理解，根据A确定B并不意味着仅仅根据A确定B，还可以根据A和/或其它信息确定B。In the embodiments provided in this application, it should be understood that "B corresponding to A" means that B is associated with A, and B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B only based on A. B can also be determined based on A and/or other information.
如图1所示，本发明实施例的一种视频质量评价方法，包括：As shown in Figure 1, a video quality evaluation method according to an embodiment of the present invention includes:
步骤101，确定与待评价视频对应的参考视频；Step 101: Determine the reference video corresponding to the video to be evaluated;
步骤102，根据多个预设上变换方式，对所述参考视频进行上变换处理，得到所述参考视频的多个目标图像；Step 102: Perform up-conversion processing on the reference video according to multiple preset up-conversion methods to obtain multiple target images of the reference video;
步骤103，根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式；Step 103: Determine the up-conversion method of the frame image in the video to be evaluated based on the multiple target images;
步骤104，根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果。Step 104: Obtain the quality evaluation result of the video to be evaluated based on the up-conversion method of the frame images in the video to be evaluated.
本发明实施例的方法，按照上述步骤，针对待评价视频，将先确定与其对应的参考视频；然后由多个预设上变换方式，对已确定的参考视频进行上变换处理，得到多个目标图像；之后，再根据所得的多个目标图像，进一步确定该待评价视频中帧图像的上变换方式；最终，结合该待评价视频中帧图像的上变换方式，得到待评价视频的质量评价结果。由于评价过程中考虑了视频的上变换方式，增加了质量评价结果的准确性。According to the method of the embodiment of the present invention, according to the above steps, for the video to be evaluated, the reference video corresponding to the video will be determined first; then, the determined reference video will be up-converted using multiple preset up-conversion methods to obtain multiple targets. image; then, based on the multiple target images obtained, the up-conversion method of the frame image in the video to be evaluated is further determined; finally, combined with the up-conversion method of the frame image in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained . Since the up-conversion method of the video is considered in the evaluation process, the accuracy of the quality evaluation results is increased.
例如，对于高清HP视频的质量评价，通过本发明实施例的方法，因评价过程中考虑了视频的上变换方式，将能够对HP视频是“真/伪4K”视频进行判定。For example, for the quality evaluation of high-definition HP videos, through the method of the embodiment of the present invention, since the up-conversion method of the video is considered in the evaluation process, it will be possible to determine whether the HP video is a "real/fake 4K" video.
可选地，该实施例中，所述参考视频为所述待评价视频的源视频，或者，与所述待评价视频具体相同内容的非源视频。Optionally, in this embodiment, the reference video is the source video of the video to be evaluated, or a non-source video with the same content as the video to be evaluated.
这里，源视频是用于制作待评价视频的基础视频，例如，对于作为待评价视频的HP视频，其源视频是低清LP视频。而与待评价视频具体相同内容的非源视频，与待评价视频的制作无关，例如，对于作为待评价视频的HP视频，与该HP视频具体相同内容的非源视频是landmark高清视频。Here, the source video is the basic video used to make the video to be evaluated. For example, for the HP video as the video to be evaluated, the source video is a low-definition LP video. Non-source videos that have the same content as the video to be evaluated have nothing to do with the production of the video to be evaluated. For example, for the HP video that is the video to be evaluated, the non-source video that has the same content as the HP video is a landmark high-definition video.
应该知道的是，待评价视频与参考视频的映射关系是预先设置的，可通过该映射关系确定与待评价视频对应的参考视频。其中，考虑到源视频用于质量评价对结果的准确性高于非源视频，步骤101具体为：判断评价视频是否有对应的源视频，若有，则将评价视频的源视频作为参考视频；若无，则将与待评价视频具体相同内容的非源视频作为参考视频。It should be noted that the mapping relationship between the video to be evaluated and the reference video is preset, and the reference video corresponding to the video to be evaluated can be determined through this mapping relationship. Among them, considering that the accuracy of the result of the source video used for quality evaluation is higher than that of the non-source video, step 101 is specifically: determine whether the evaluation video has a corresponding source video, and if so, use the source video of the evaluation video as a reference video; If not, a non-source video with the same content as the video to be evaluated will be used as the reference video.
该实施例中，对于确定的参考视频，将通过执行步骤102得到参考视频的多个目标图像，以用于后续对待评价视频中帧图像的上变换方式的确定。这里，对参考视频的上变换处理使用的多个预设上变换方式，可以是多种插值上变换(如最临近插值、双线性插值、双三次插值、区域插值等)，也可以是多种神经网络超分(如基于超分辨率卷积神经网络SRCNN的waifu2x、Meta-Upscale，基于残差密集块RRDB的增强型超分辨率生成对抗网络ESRGAN、超分辨率自然增强库neural-enhance等)。当然，预设上变换方式不限于上述内容，在此不再一一列举。In this embodiment, for the determined reference video, multiple target images of the reference video will be obtained by executing step 102, for subsequent determination of the up-conversion method of the frame images in the video to be evaluated. Here, the multiple preset up-conversion methods used in the up-conversion process of the reference video can be a variety of interpolation up-conversions (such as nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, regional interpolation, etc.), or multiple A variety of neural network super-resolution (such as waifu2x and Meta-Upscale based on the super-resolution convolutional neural network SRCNN, the enhanced super-resolution generative adversarial network ESRGAN based on the residual dense block RRDB, the super-resolution natural enhancement library neural-enhance, etc. ). Of course, the default up-conversion methods are not limited to the above, and will not be listed one by one here.
而通过对参考视频的帧图像进行上变换处理，参考视频的每帧图像使用该多个预设上变换方式生成多个图片，即每个图片对应一种预设上变换方式。具体的，对于参考视频，如待评价HP视频的LP视频(即该待评价HP视频的源视频)，在使用的预设上变换方式为B个的情况下，对该LP视频中的一个帧图像(又可称为LP帧图像)进行上变换处理后，将生成B个HP帧图像：HP1，HP2，…，HPB。这样，该LP视频包括的N个帧图像进行上变换处理后，得到的目标图像为N*B个HP帧图像。这里，LP帧图像上变换处理所得的HP帧图像是与待评价HP视频帧图像同像素尺寸的高清图像。By up-converting the frame images of the reference video, each frame image of the reference video uses the multiple preset up-conversion methods to generate multiple pictures, that is, each picture corresponds to a preset up-conversion method. Specifically, for the reference video, such as the LP video of the HP video to be evaluated (that is, the source video of the HP video to be evaluated), when the preset up-conversion methods used are B, one frame in the LP video After the image (also called LP frame image) is up-converted, B HP frame images will be generated: HP1 , HP2 ,..., HPB . In this way, after the N frame images included in the LP video are up-converted, the obtained target image is N*B HP frame images. Here, the HP frame image obtained by the LP frame image up-conversion process is a high-definition image with the same pixel size as the HP video frame image to be evaluated.
在确定参考视频的多个目标图像之后，即可执行步骤103。可选地，该实施例中，步骤103包括：After determining multiple target images of the reference video, step 103 can be performed. Optionally, in this embodiment, step 103 includes:
根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数；According to the plurality of target images, obtain the quality parameters of the target frame image in the video to be evaluated;
根据所述质量参数，确定所述目标帧图像的上变换方式；Determine the up-conversion method of the target frame image according to the quality parameter;
其中，所述质量参数包括：峰值信噪比PSNR和结构相似性SSIM。Wherein, the quality parameters include: peak signal-to-noise ratio PSNR and structural similarity SSIM.
这里，为了确定待评价视频帧图像的上变换方式，将待评价视频的每帧图像作为目标帧图像，通过获得该目标帧图像的PSNR和SSIM，进一步确定该目标帧图像的上变换方式。Here, in order to determine the up-conversion method of the video frame image to be evaluated, each frame image of the video to be evaluated is used as a target frame image, and the up-conversion method of the target frame image is further determined by obtaining the PSNR and SSIM of the target frame image.
可选地，所述根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数，包括：Optionally, obtaining the quality parameters of the target frame image in the video to be evaluated based on the plurality of target images includes:
计算所述待评价视频中目标帧图像与所述多个目标图像的第一组图像中各图像间的PSNR和SSIM；Calculate PSNR and SSIM between the target frame image in the video to be evaluated and each image in the first group of images of the plurality of target images;
将所述计算得到的PSNR中的最大PSNR作为所述目标帧图像的PSNR；Use the maximum PSNR among the calculated PSNRs as the PSNR of the target frame image;
将所述计算得到的SSIM中的最大SSIM作为所述目标帧图像的SSIM；Use the maximum SSIM among the calculated SSIMs as the SSIM of the target frame image;
其中，所述第一组图像是所述多个目标图像中对应所述目标帧图像的一组图像，且各个图像对应不同的预设上变换方式。Wherein, the first group of images is a group of images corresponding to the target frame image among the plurality of target images, and each image corresponds to a different preset up-conversion method.
这里，第一组图像即目标帧图像进行上变换处理后得到的B个图像，B为预设上变换方式的数量。如此，在计算目标帧图像与第一组图像的PSNR和SSIM后，就能够将所得的最大PSNR作为该目标帧图像的PSNR，将所得的最大SSIM作为该目标帧图像的SSIM。Here, the first group of images, that is, the B images obtained after the up-conversion process is performed on the target frame image, where B is the number of preset up-conversion methods. In this way, after calculating the PSNR and SSIM of the target frame image and the first group of images, the obtained maximum PSNR can be used as the PSNR of the target frame image, and the obtained maximum SSIM can be used as the SSIM of the target frame image.
其中，目标帧图像与当前图像(即第一组图像中的一图像)间的PSNR，可通过PSNR计算公式得到。这里，L为像素点的最大像素值，如255；MSE为均方误差，具体的，/>m为图像宽度上像素点数量，n为图像长度上像素点数量，i为像素点在图像宽度上的编号，j为像素点在图像长度上的编号，X(i,j)表示目标帧图像上一像素点(即图像宽度上的编号i，图像长度上的编号j的像素点)的像素值，Y(i,j)表示第一组图像中当前图像上一像素点(即图像宽度上的编号i，图像长度上的编号j的像素点)的像素值。故通过PSNR计算公式，计算目标帧图像与第一图像中每一图像的PSNR后，其中的最大PSNR即该目标帧图像的PSNR。Among them, the PSNR between the target frame image and the current image (i.e., an image in the first group of images) can be calculated by the PSNR formula get. Here, L is the maximum pixel value of the pixel, such as 255; MSE is the mean square error, specifically, /> m is the number of pixels on the image width, n is the number of pixels on the image length, i is the number of pixels on the image width, j is the number of pixels on the image length, X(i,j) represents the target frame image The pixel value of the previous pixel (that is, the pixel with number i on the image width and the pixel with number j on the image length), Y(i,j) represents the previous pixel of the current image in the first group of images (that is, the image width The number i on, the pixel value of the pixel number j on the image length). Therefore, after calculating the PSNR of each image in the target frame image and the first image through the PSNR calculation formula, the maximum PSNR among them is the PSNR of the target frame image.
其中，目标帧图像与当前图像(即第一组图像中的一图像)间的SSIM可通过目标帧图像与当前图像的结构相似性函数SSIM(x,y)得到，SSIM(x,y)＝[l(x,y)α·c(x,y)β·s(x,y)γ]。这里，x为目标帧图像，y为当前图像，l(x,y)为目标帧图像与当前图像的亮度函数，c(x,y)为目标帧图像与当前图像的对比度函数，s(x,y)为目标帧图像与当前图像的结构函数，α为亮度系数，β为对比度系数，γ为结构系数。Among them, the SSIM between the target frame image and the current image (i.e., an image in the first group of images) can be obtained through the structural similarity function SSIM(x, y) of the target frame image and the current image, SSIM(x, y) =[l(x,y)α ·c(x,y)β ·s(x,y)γ ]. Here, x is the target frame image, y is the current image, l(x,y) is the brightness function of the target frame image and the current image, c(x,y) is the contrast function of the target frame image and the current image, s(x , y) is the structure function of the target frame image and the current image, α is the brightness coefficient, β is the contrast coefficient, and γ is the structure coefficient.
而其中，μx为目标帧图像上像素点的像素值均值，μy为当前图像上像素点的像素值均值，σx2为目标帧图像上像素点的像素值方差，σy2为当前图像上像素点的像素值方差，σxy为目标帧图像上像素点的像素值和当前图像上像素点的像素值的协方差，c1＝(k1L)2，c2＝(k2L)2，c3＝c2/2，k1和k2为常数，如k1＝0.01，k2＝0.03。andAmongthem,μ The pixel value variance of the upper pixel point, σxy is the covariance of the pixel value of the pixel point on the target frame image and the pixel value of the pixel point on the current image, c1 = (k1 L)2 , c2 = (k2 L )2 , c3 = c2 /2, k1 and k2 are constants, such as k1 = 0.01, k2 = 0.03.
可选地，α、β和γ均为1，Optionally, α, β and γ are all 1,
当然，在计算SSIM时，为简化处理，可以在图像上取预设大小的窗口，即x为目标帧图像中窗口对应的图像，y为当前图像中窗口对应的图像，然后通过滑动窗口进行计算，最后取各窗口的SSIM平均值作为全局SSIM。Of course, when calculating SSIM, in order to simplify the process, you can take a preset size window on the image, that is, x is the image corresponding to the window in the target frame image, y is the image corresponding to the window in the current image, and then calculate through the sliding window , and finally take the average SSIM of each window as the global SSIM.
通过选取待评价视频中每一帧图像作为目标帧图像进行PSNR和SSIM计算，最终可得待评价视频的每帧图像的PSNR和SSIM。By selecting each frame of the image in the video to be evaluated as the target frame image to calculate PSNR and SSIM, the PSNR and SSIM of each frame of the video to be evaluated can finally be obtained.
可选地，该实施例中，所述根据所述质量参数，确定所述目标帧图像的上变换方式，包括：Optionally, in this embodiment, determining the up-conversion method of the target frame image according to the quality parameter includes:
将所述目标帧图像的PSNR和/或SSIM对应的预设上变换方式，作为所述目标帧图像的上变换方式；或者，Use the preset up-conversion method corresponding to the PSNR and/or SSIM of the target frame image as the up-conversion method of the target frame image; or,
通过多类别分类模型以及所述目标帧图像与所述第一组图像中各图像间的PSNR和SSIM，得到所述目标帧图像的上变换方式。The up-conversion method of the target frame image is obtained through the multi-category classification model and the PSNR and SSIM between the target frame image and each image in the first group of images.
即，对于参考视频为待评价视频的源视频，将该目标帧图像的PSNR和/或SSIM对应的预设上变换方式，作为目标帧图像的上变换方式。当然，对于参考视频为与待评价视频具体相同内容的非源视频，虽然也可采用上一方式确定目标帧图像的上变换方式，但准确性较差，因此，会通过多类别分类模型来更为准确地确定目标帧图像的上变换方式。其中，多类别分类模型的输入是该目标帧图像与第一组图像中各图像间的PSNR和SSIM，多类别分类模型的输出除该目标帧图像的上变换方式之外，还包括该目标帧图像的PSNR和SSIM。That is, for the source video where the reference video is the video to be evaluated, the preset up-conversion method corresponding to the PSNR and/or SSIM of the target frame image is used as the up-conversion method of the target frame image. Of course, for non-source videos where the reference video has the same content as the video to be evaluated, although the previous method can also be used to determine the up-conversion method of the target frame image, the accuracy is poor. Therefore, a multi-category classification model will be used to update the image. In order to accurately determine the up-conversion method of the target frame image. Among them, the input of the multi-category classification model is the PSNR and SSIM between the target frame image and each image in the first group of images. In addition to the up-conversion method of the target frame image, the output of the multi-category classification model also includes the target PSNR and SSIM of the frame image.
可选地，所述多类别分类模型是已构建的基于帧图像的PSNR和SSIM，确定帧图像的上变换方式的神经网络模型。Optionally, the multi-category classification model is a neural network model that has been constructed based on the PSNR and SSIM of the frame image and determines the up-conversion method of the frame image.
具体地，以待评价的HP视频，参考视频为与该HP视频具体相同内容的非源视频为例，参考视频的每帧图像与对应的第一组图像中各图像间的PSNR和SSIM输入该多类别分类模型，得到的输出为：其中，Z为待评价的HP视频的帧图像数量，labelZ为第Z个帧图像的上变换方式，/>为第Z个帧图像的PSNR，/>为第Z个帧图像的SSIM。Specifically, taking the HP video to be evaluated, the reference video is a non-source video with the same content as the HP video as an example, and the PSNR and SSIM input between each frame of the reference video and each image in the corresponding first group of images are The output of this multi-category classification model is: Among them, Z is the number of frame images of the HP video to be evaluated, labelZ is the up-conversion method of the Z-th frame image, /> is the PSNR of the Z-th frame image,/> is the SSIM of the Z-th frame image.
其中，多类别分类模型是通过多个样本数据训练所得，每组样本中包括待评价视频、该待评价视频的源视频和该待评价视频的每帧图像的上变换方式。这样，训练过程中，通过将样本中待评价视频的源视频的帧图像使用多种预设上变换方式进行上变换处理，并结合待评价视频的对应帧图像计算得PSNR和SSIM后，输入多类别分类模型中，之后得到输出结果与样本中的上变换方式比较，调整模型，直至训练完成。其中，该多类别分类模型训练设置的损失函数为B为预设上变换方式的数量，λcre为预测正确的系数，λnocre为预测错误的系数。而/>在预测正确时值为1，反之则为0；同理/>在预测错误时值为1，反之则为0。The multi-category classification model is trained through multiple sample data, and each group of samples includes the video to be evaluated, the source video of the video to be evaluated, and the up-conversion method of each frame of the video to be evaluated. In this way, during the training process, the frame images of the source video of the video to be evaluated in the sample are up-converted using a variety of preset up-conversion methods, and the PSNR and SSIM are calculated based on the corresponding frame images of the video to be evaluated. After inputting multiple In the category classification model, the output result is then compared with the up-conversion method in the sample, and the model is adjusted until the training is completed. Among them, the loss function set for the multi-category classification model training is B is the number of preset up-conversion methods, λcre is the coefficient of correct prediction, and λnocre is the coefficient of incorrect prediction. And/> When the prediction is correct, the value is 1, otherwise it is 0; similarly/> The value is 1 when the prediction is wrong, and 0 otherwise.
在确定待评价视频中帧图像的上变换方式之后，执行步骤104，得到该待评价视频的质量评价结果。可选地，该实施例中，步骤104包括：After determining the up-conversion method of the frame image in the video to be evaluated, step 104 is performed to obtain the quality evaluation result of the video to be evaluated. Optionally, in this embodiment, step 104 includes:
确定所述待评价视频中，对应帧图像数目最多的目标上变换方式；Determine the target up-conversion method with the largest number of corresponding frame images in the video to be evaluated;
根据所述目标上变换方式所属的上变换方式类别，得到所述待评价视频的质量评价结果；或者，Obtain the quality evaluation result of the video to be evaluated according to the up-conversion method category to which the target up-conversion method belongs; or,
获取所述目标上变换方式对应的帧图像的质量参数的均值，并根据所述均值所属的阈值范围得到所述待评价视频的质量评价结果。Obtain the mean value of the quality parameters of the frame image corresponding to the target up-conversion method, and obtain the quality evaluation result of the video to be evaluated according to the threshold range to which the mean value belongs.
如此，将待评价视频中每个帧图像的上变换方式进行统计后，可将对应帧图像数目最多的上变换方式确定为目标上变换方式，作为该待评价视频的上变换方式。这样，一方面，根据该目标上变换方式所属的上变换方式类别，可得到该待评价视频的质量评价结果。当然，此时上变换方式类别是预先设置的，且质量评价结果与上变换方式的映射关系也可预先设置。另一方面，对于目标上变换方式，会获取其对应的帧图像的质量参数的均值，然后根据该均值所述的阈值范围得到该待评价视频的质量评价结果。而此时，会预先设置质量评价结果与不同阈值范围的映射关系。In this way, after counting the up-conversion methods of each frame image in the video to be evaluated, the up-conversion method with the largest number of corresponding frame images can be determined as the target up-conversion method as the up-conversion method of the video to be evaluated. In this way, on the one hand, according to the up-conversion method category to which the target up-conversion method belongs, the quality evaluation result of the video to be evaluated can be obtained. Of course, at this time, the up-conversion method category is preset, and the mapping relationship between the quality evaluation results and the up-conversion method can also be preset. On the other hand, for the target up-conversion method, the mean value of the quality parameters of the corresponding frame images is obtained, and then the quality evaluation result of the video to be evaluated is obtained according to the threshold range described by the mean value. At this time, the mapping relationship between the quality evaluation results and different threshold ranges will be preset.
例如，在待评价视频的帧图像的上变换方式中，上变换方式A对应的帧图像数量PA在待评价视频的P个帧图像中占比最大，大于或等于80％，则该上变换方式A即该待评价视频的上变换方式。之后，对于该待评价视频中对应上变换方式A的PA个帧图像，计算其PSNR和SSIM的均值后，由该均值作为质量评价的指标，通过比较该均值属于第一质量评价结果对应的阈值范围，则得到该第一质量评价结果；若均值属于第二质量评价结果对应的阈值范围，则得到该第二质量评价结果。For example, in the up-conversion method of the frame images of the video to be evaluated, the number of frame images P A corresponding to the up-conversion methodA accounts for the largest proportion among the P frame images of the video to be evaluated. If it is greater than or equal to 80%, then the up-conversion method A is the up-conversion method of the video to be evaluated. After that, for the PA frame images corresponding to the up-conversion method A in the video to be evaluated, after calculating the average value of PSNR and SSIM, the average value is used as an indicator of quality evaluation, and the average value is compared to the corresponding frame of the first quality evaluation result. If the average value belongs to the threshold range corresponding to the second quality evaluation result, the second quality evaluation result is obtained.
该实施例中，可选地，所述质量评价结果包括：所述待评价视频是否为具有对应清晰度的视频。In this embodiment, optionally, the quality evaluation result includes: whether the video to be evaluated is a video with corresponding definition.
这样，对于待评价的HP视频，若延续上例，其对应上变换方式A的PA个帧图像，计算其PSNR和SSIM的均值后，由均值属于“真4K”对应的阈值范围，则可评价该待评价的HP视频为真4K视频；反之，为伪4K视频。In this way, for the HP video to be evaluated, if the above example is continued, it corresponds to PA frame images of the up-conversion method A, and after calculating the average value of its PSNR and SSIM, the average value belongs to the threshold range corresponding to "true 4K", then it can be Evaluate the HP video to be evaluated as a true 4K video; otherwise, it is a pseudo 4K video.
综上所述，本发明实施例的方法，针对待评价视频，通过先确定与其对应的参考视频；然后由多个预设上变换方式，对已确定的参考视频进行上变换处理，得到多个目标图像；之后，再根据所得的多个目标图像，进一步确定该待评价视频中帧图像的上变换方式；最终，结合该待评价视频中帧图像的上变换方式，得到待评价视频的质量评价结果。由于评价过程中考虑了视频的上变换方式，增加了质量评价结果的准确性。To sum up, the method of the embodiment of the present invention first determines the reference video corresponding to the video to be evaluated; and then performs up-conversion processing on the determined reference video using multiple preset up-conversion methods to obtain multiple target image; then, based on the multiple target images obtained, the up-conversion method of the frame image in the video to be evaluated is further determined; finally, combined with the up-conversion method of the frame image in the video to be evaluated, the quality evaluation of the video to be evaluated is obtained result. Since the up-conversion method of the video is considered in the evaluation process, the accuracy of the quality evaluation results is increased.
如图2所示，本发明实施例的一种视频质量评价装置，包括：As shown in Figure 2, a video quality evaluation device according to an embodiment of the present invention includes:
第一处理模块210，用于确定与待评价视频对应的参考视频；The first processing module 210 is used to determine the reference video corresponding to the video to be evaluated;
第二处理模块220，用于根据多个预设上变换方式，对所述参考视频进行上变换处理，得到所述参考视频的多个目标图像；The second processing module 220 is configured to perform up-conversion processing on the reference video according to multiple preset up-conversion methods to obtain multiple target images of the reference video;
第三处理模块230，用于根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式；The third processing module 230 is configured to determine the up-conversion method of the frame image in the video to be evaluated according to the plurality of target images;
第四处理模块240，用于根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果。The fourth processing module 240 is configured to obtain the quality evaluation result of the video to be evaluated according to the up-conversion method of the frame image in the video to be evaluated.
可选地，所述第三处理模块包括：Optionally, the third processing module includes:
第一处理子模块，用于根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数；The first processing submodule is used to obtain the quality parameters of the target frame image in the video to be evaluated based on the multiple target images;
第二处理子模块，用于根据所述质量参数，确定所述目标帧图像的上变换方式；The second processing sub-module is used to determine the up-conversion method of the target frame image according to the quality parameter;
其中，所述质量参数包括：峰值信噪比PSNR和结构相似性SSIM。Wherein, the quality parameters include: peak signal-to-noise ratio PSNR and structural similarity SSIM.
可选地，所述第一处理子模块包括：Optionally, the first processing sub-module includes:
计算单元，用于计算所述待评价视频中目标帧图像与所述多个目标图像的第一组图像中各图像间的PSNR和SSIM；A calculation unit configured to calculate PSNR and SSIM between the target frame image in the video to be evaluated and each image in the first group of images of the plurality of target images;
第一处理单元，用于将所述计算得到的PSNR中的最大PSNR作为所述目标帧图像的PSNR；A first processing unit configured to use the maximum PSNR among the calculated PSNRs as the PSNR of the target frame image;
第二处理单元，用于将所述计算得到的SSIM中的最大SSIM作为所述目标帧图像的SSIM；A second processing unit, configured to use the maximum SSIM among the calculated SSIMs as the SSIM of the target frame image;
其中，所述第一组图像是所述多个目标图像中对应所述目标帧图像的一组图像，且各个图像对应不同的预设上变换方式。Wherein, the first group of images is a group of images corresponding to the target frame image among the plurality of target images, and each image corresponds to a different preset up-conversion method.
可选地，所述第二处理子模块还用于：Optionally, the second processing sub-module is also used to:
将所述目标帧图像的PSNR和/或SSIM对应的预设上变换方式，作为所述目标帧图像的上变换方式；或者，Use the preset up-conversion method corresponding to the PSNR and/or SSIM of the target frame image as the up-conversion method of the target frame image; or,
通过多类别分类模型以及所述目标帧图像与所述第一组图像中各图像间的PSNR和SSIM，得到所述目标帧图像的上变换方式。The up-conversion method of the target frame image is obtained through the multi-category classification model and the PSNR and SSIM between the target frame image and each image in the first group of images.
可选地，所述多类别分类模型是已构建的基于帧图像的PSNR和SSIM，确定帧图像的上变换方式的神经网络模型。Optionally, the multi-category classification model is a neural network model that has been constructed based on the PSNR and SSIM of the frame image and determines the up-conversion method of the frame image.
可选地，所述第四处理模块包括：Optionally, the fourth processing module includes:
确定子模块，用于确定所述待评价视频中，对应帧图像数目最多的目标上变换方式；The determination submodule is used to determine the target up-conversion method with the largest number of corresponding frame images in the video to be evaluated;
第三处理子模块，用于根据所述目标上变换方式所属的上变换方式类别，得到所述待评价视频的质量评价结果；或者，The third processing submodule is used to obtain the quality evaluation result of the video to be evaluated according to the up-conversion method category to which the target up-conversion method belongs; or,
获取所述目标上变换方式对应的帧图像的质量参数的均值，并根据所述均值所属的阈值范围得到所述待评价视频的质量评价结果。Obtain the mean value of the quality parameters of the frame image corresponding to the target up-conversion method, and obtain the quality evaluation result of the video to be evaluated according to the threshold range to which the mean value belongs.
可选地，所述质量评价结果包括：所述待评价视频是否为具有对应清晰度的视频。Optionally, the quality evaluation result includes: whether the video to be evaluated is a video with corresponding definition.
可选地，所述参考视频为所述待评价视频的源视频，或者，与所述待评价视频具体相同内容的非源视频。Optionally, the reference video is a source video of the video to be evaluated, or a non-source video with the same content as the video to be evaluated.
该装置针对待评价视频，通过先确定与其对应的参考视频；然后由多个预设上变换方式，对已确定的参考视频进行上变换处理，得到多个目标图像；之后，再根据所得的多个目标图像，进一步确定该待评价视频中帧图像的上变换方式；最终，结合该待评价视频中帧图像的上变换方式，得到待评价视频的质量评价结果。由于评价过程中考虑了视频的上变换方式，增加了质量评价结果的准确性。For the video to be evaluated, the device first determines the reference video corresponding to it; then uses multiple preset up-conversion methods to up-convert the determined reference video to obtain multiple target images; and then, based on the obtained multiple target image, and further determine the up-conversion method of the frame image in the video to be evaluated; finally, combined with the up-conversion method of the frame image in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained. Since the up-conversion method of the video is considered in the evaluation process, the accuracy of the quality evaluation results is increased.
需要说明的是，该装置是应用了上述视频质量评价方法的装置，上述方法实施例的实现方式适用于该装置，也能达到相同的技术效果，在此不再赘述。It should be noted that this device is a device that applies the above video quality evaluation method. The implementation of the above method embodiment is applicable to this device and can achieve the same technical effect, which will not be described again here.
如图3所示，本发明实施例的一种视频质量评价设备300，包括处理器310，所述处理器310用于：As shown in Figure 3, a video quality evaluation device 300 according to the embodiment of the present invention includes a processor 310, and the processor 310 is used for:
确定与待评价视频对应的参考视频；Determine the reference video corresponding to the video to be evaluated;
根据多个预设上变换方式，对所述参考视频进行上变换处理，得到所述参考视频的多个目标图像；Perform up-conversion processing on the reference video according to multiple preset up-conversion methods to obtain multiple target images of the reference video;
根据所述多个目标图像，确定所述待评价视频中帧图像的上变换方式；According to the plurality of target images, determine the up-conversion method of the frame image in the video to be evaluated;
根据所述待评价视频中帧图像的上变换方式，得到所述待评价视频的质量评价结果。According to the up-conversion method of frame images in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained.
可选地，所述处理器还用于：Optionally, the processor is also used to:
根据所述多个目标图像，得到所述待评价视频中目标帧图像的质量参数；According to the plurality of target images, obtain the quality parameters of the target frame image in the video to be evaluated;
根据所述质量参数，确定所述目标帧图像的上变换方式；Determine the up-conversion method of the target frame image according to the quality parameter;
其中，所述质量参数包括：峰值信噪比PSNR和结构相似性SSIM。Wherein, the quality parameters include: peak signal-to-noise ratio PSNR and structural similarity SSIM.
可选地，所述处理器还用于：Optionally, the processor is also used to:
计算所述待评价视频中目标帧图像与所述多个目标图像的第一组图像中各图像间的PSNR和SSIM；Calculate PSNR and SSIM between the target frame image in the video to be evaluated and each image in the first group of images of the plurality of target images;
将所述计算得到的PSNR中的最大PSNR作为所述目标帧图像的PSNR；Use the maximum PSNR among the calculated PSNRs as the PSNR of the target frame image;
将所述计算得到的SSIM中的最大SSIM作为所述目标帧图像的SSIM；Use the maximum SSIM among the calculated SSIMs as the SSIM of the target frame image;
其中，所述第一组图像是所述多个目标图像中对应所述目标帧图像的一组图像，且各个图像对应不同的预设上变换方式。Wherein, the first group of images is a group of images corresponding to the target frame image among the plurality of target images, and each image corresponds to a different preset up-conversion method.
可选地，所述处理器还用于：Optionally, the processor is also used to:
将所述目标帧图像的PSNR和/或SSIM对应的预设上变换方式，作为所述目标帧图像的上变换方式；或者，Use the preset up-conversion method corresponding to the PSNR and/or SSIM of the target frame image as the up-conversion method of the target frame image; or,
通过多类别分类模型以及所述目标帧图像与所述第一组图像中各图像间的PSNR和SSIM，得到所述目标帧图像的上变换方式。An up-conversion method of the target frame image is obtained through a multi-category classification model and the PSNR and SSIM between the target frame image and each image in the first group of images.
可选地，所述多类别分类模型是已构建的基于帧图像的PSNR和SSIM，确定帧图像的上变换方式的神经网络模型。Optionally, the multi-category classification model is a neural network model that has been constructed based on the PSNR and SSIM of the frame image and determines the up-conversion method of the frame image.
可选地，所述处理器还用于：Optionally, the processor is also used to:
确定所述待评价视频中，对应帧图像数目最多的目标上变换方式；Determine the target up-conversion method with the largest number of corresponding frame images in the video to be evaluated;
根据所述目标上变换方式所属的上变换方式类别，得到所述待评价视频的质量评价结果；或者，Obtain the quality evaluation result of the video to be evaluated according to the up-conversion method category to which the target up-conversion method belongs; or,
获取所述目标上变换方式对应的帧图像的质量参数的均值，并根据所述均值所属的阈值范围得到所述待评价视频的质量评价结果。Obtain the mean value of the quality parameters of the frame image corresponding to the target up-conversion method, and obtain the quality evaluation result of the video to be evaluated according to the threshold range to which the mean value belongs.
可选地，所述质量评价结果包括：所述待评价视频是否为具有对应清晰度的视频。Optionally, the quality evaluation result includes: whether the video to be evaluated is a video with corresponding definition.
可选地，所述参考视频为所述待评价视频的源视频，或者，与所述待评价视频具体相同内容的非源视频。Optionally, the reference video is a source video of the video to be evaluated, or a non-source video with the same content as the video to be evaluated.
本发明实施例的视频质量评价设备300，还可以包括收发器320，用于在处理器310的控制下收发数据。The video quality evaluation device 300 in the embodiment of the present invention may also include a transceiver 320 for sending and receiving data under the control of the processor 310.
该实施例的视频质量评价设备，针对待评价视频，通过先确定与其对应的参考视频；然后由多个预设上变换方式，对已确定的参考视频进行上变换处理，得到多个目标图像；之后，再根据所得的多个目标图像，进一步确定该待评价视频中帧图像的上变换方式；最终，结合该待评价视频中帧图像的上变换方式，得到待评价视频的质量评价结果。由于评价过程中考虑了视频的上变换方式，增加了质量评价结果的准确性。The video quality evaluation device of this embodiment first determines the reference video corresponding to the video to be evaluated; and then performs up-conversion processing on the determined reference video using multiple preset up-conversion methods to obtain multiple target images; After that, based on the multiple target images obtained, the up-conversion method of the frame image in the video to be evaluated is further determined; finally, combined with the up-conversion method of the frame image in the video to be evaluated, the quality evaluation result of the video to be evaluated is obtained. Since the up-conversion method of the video is considered in the evaluation process, the accuracy of the quality evaluation results is increased.
本发明另一实施例的一种视频质量评价设备，如图4所示，包括收发器410、处理器400、存储器420及存储在所述存储器420上并可在所述处理器400上运行的程序或指令；所述处理器400执行所述程序或指令时实现上述应用于视频质量评价方法。A video quality evaluation device according to another embodiment of the present invention, as shown in Figure 4, includes a transceiver 410, a processor 400, a memory 420, and a program stored on the memory 420 and capable of running on the processor 400. Program or instruction; when the processor 400 executes the program or instruction, the above-mentioned video quality evaluation method is implemented.
所述收发器410，用于在处理器400的控制下接收和发送数据。The transceiver 410 is used to receive and send data under the control of the processor 400.
其中，在图4中，总线架构可以包括任意数量的互联的总线和桥，具体由处理器400代表的一个或多个处理器和存储器420代表的存储器的各种电路链接在一起。总线架构还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起，这些都是本领域所公知的，因此，本文不再对其进行进一步描述。总线接口提供接口。收发器410可以是多个元件，即包括发送机和接收机，提供用于在传输介质上与各种其他装置通信的单元。处理器400负责管理总线架构和通常的处理，存储器420可以存储处理器400在执行操作时所使用的数据。In FIG. 4 , the bus architecture may include any number of interconnected buses and bridges, specifically one or more processors represented by processor 400 and various circuits of the memory represented by memory 420 are linked together. The bus architecture can also link together various other circuits such as peripherals, voltage regulators, and power management circuits, which are all well known in the art and therefore will not be described further herein. The bus interface provides the interface. Transceiver 410 may be a plurality of elements, including a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium. The processor 400 is responsible for managing the bus architecture and general processing, and the memory 420 can store data used by the processor 400 when performing operations.
本发明实施例的一种可读存储介质，其上存储有程序或指令，所述程序或指令被处理器执行时实现如上所述的视频质量评价方法中的步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。A readable storage medium according to an embodiment of the present invention has a program or instruction stored thereon. When the program or instruction is executed by a processor, the steps in the video quality evaluation method as described above are implemented and the same technical effect can be achieved. , to avoid repetition, we will not go into details here.
其中，所述处理器为上述实施例中所述的视频质量评价设备中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器(Read-Only Memory，简称ROM)、随机存取存储器(Random Access Memory，简称RAM)、磁碟或者光盘等。Wherein, the processor is the processor in the video quality evaluation device described in the above embodiment. The readable storage media includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks.
进一步需要说明的是，此说明书中所描述的许多功能部件都被称为模块，以便更加特别地强调其实现方式的独立性。It should be further noted that many of the functional components described in this specification are called modules to more specifically emphasize the independence of their implementation.
本发明实施例中，模块可以用软件实现，以便由各种类型的处理器执行。举例来说，一个标识的可执行代码模块可以包括计算机指令的一个或多个物理或者逻辑块，举例来说，其可以被构建为对象、过程或函数。尽管如此，所标识模块的可执行代码无需物理地位于一起，而是可以包括存储在不同位里上的不同的指令，当这些指令逻辑上结合在一起时，其构成模块并且实现该模块的规定目的。In the embodiment of the present invention, the module can be implemented in software so as to be executed by various types of processors. For example, an identified module of executable code may include one or more physical or logical blocks of computer instructions, which may be structured, for example, as an object, procedure, or function. Nonetheless, the executable code of an identified module need not be physically located together, but may include different instructions stored on different bits that, when logically combined, constitute the module and implement the provisions of the module Purpose.
实际上，可执行代码模块可以是单条指令或者是许多条指令，并且甚至可以分布在多个不同的代码段上，分布在不同程序当中，以及跨越多个存储器设备分布。同样地，操作数据可以在模块内被识别，并且可以依照任何适当的形式实现并且被组织在任何适当类型的数据结构内。所述操作数据可以作为单个数据集被收集，或者可以分布在不同位置上(包括在不同存储设备上)，并且至少部分地可以仅作为电子信号存在于系统或网络上。In fact, an executable code module can be a single instruction or many instructions, and can even be distributed over multiple different code segments, distributed among different programs, and distributed across multiple memory devices. Likewise, operational data may be identified within modules and may be implemented in any suitable form and organized within any suitable type of data structure. The operational data may be collected as a single data set, or may be distributed over different locations (including on different storage devices), and may exist, at least in part, solely as electronic signals on a system or network.
在模块可以利用软件实现时，考虑到现有硬件工艺的水平，所以可以以软件实现的模块，在不考虑成本的情况下，本领域技术人员都可以搭建对应的硬件电路来实现对应的功能，所述硬件电路包括常规的超大规模集成(VLSI)电路或者门阵列以及诸如逻辑芯片、晶体管之类的现有半导体或者是其它分立的元件。模块还可以用可编程硬件设备，诸如现场可编程门阵列、可编程阵列逻辑、可编程逻辑设备等实现。When the module can be implemented using software, taking into account the level of existing hardware technology, those skilled in the art can build corresponding hardware circuits to implement the corresponding functions without considering the cost. The hardware circuits include conventional very large scale integration (VLSI) circuits or gate arrays as well as existing semiconductors such as logic chips, transistors, or other discrete components. Modules can also be implemented using programmable hardware devices, such as field programmable gate arrays, programmable array logic, programmable logic devices, etc.
上述范例性实施例是参考该些附图来描述的，许多不同的形式和实施例是可行而不偏离本发明精神及教示，因此，本发明不应被建构成为在此所提出范例性实施例的限制。更确切地说，这些范例性实施例被提供以使得本发明会是完善又完整，且会将本发明范围传达给那些熟知此项技术的人士。在该些图式中，组件尺寸及相对尺寸也许基于清晰起见而被夸大。在此所使用的术语只是基于描述特定范例性实施例目的，并无意成为限制用。如在此所使用地，除非该内文清楚地另有所指，否则该单数形式“一”、“一个”和“该”是意欲将该些多个形式也纳入。会进一步了解到该些术语“包含”及/或“包括”在使用于本说明书时，表示所述特征、整数、步骤、操作、构件及/或组件的存在，但不排除一或更多其它特征、整数、步骤、操作、构件、组件及/或其族群的存在或增加。除非另有所示，陈述时，一值范围包含该范围的上下限及其间的任何子范围。The above exemplary embodiments have been described with reference to the accompanying drawings. Many different forms and embodiments are possible without departing from the spirit and teachings of the invention. Therefore, the invention should not be construed as the exemplary embodiments set forth herein. limits. Rather, these exemplary embodiments are provided so that this disclosure will be thorough and complete, and will convey the scope of the invention to those skilled in the art. In the drawings, component sizes and relative sizes may be exaggerated for clarity. The terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It will be further understood that the terms "comprising" and/or "including" when used in this specification indicate the presence of stated features, integers, steps, operations, components and/or components, but do not exclude the presence of one or more other The existence or addition of features, integers, steps, operations, components, components and/or families thereof. Unless otherwise indicated, when stated, a range of values includes the upper and lower limits of the range and any subranges therebetween.
以上所述是本发明的优选实施方式，应当指出，对于本技术领域的普通技术人员来说，在不脱离本发明所述原理的前提下，还可以作出若干改进和润饰，这些改进和润饰也应视为本发明的保护范围。The above is the preferred embodiment of the present invention. It should be pointed out that for those of ordinary skill in the art, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications can also be made. should be regarded as the protection scope of the present invention.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011624696.7ACN112767310B (en) | 2020-12-31 | 2020-12-31 | Video quality evaluation method, device and equipment |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011624696.7ACN112767310B (en) | 2020-12-31 | 2020-12-31 | Video quality evaluation method, device and equipment |
| Publication Number | Publication Date |
|---|---|
| CN112767310A CN112767310A (en) | 2021-05-07 |
| CN112767310Btrue CN112767310B (en) | 2024-03-22 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011624696.7AActiveCN112767310B (en) | 2020-12-31 | 2020-12-31 | Video quality evaluation method, device and equipment |
| Country | Link |
|---|---|
| CN (1) | CN112767310B (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114841960B (en)* | 2022-05-05 | 2025-04-29 | 咪咕视讯科技有限公司 | Three-dimensional video quality evaluation method, device, equipment and storage medium |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101156451A (en)* | 2005-04-12 | 2008-04-02 | 皇家飞利浦电子股份有限公司 | Video processing with region-based multiple-pass motion estimation and update of temporal motion vector candidates |
| CN101156450A (en)* | 2005-04-12 | 2008-04-02 | 皇家飞利浦电子股份有限公司 | Region- based 3drs motion estimation using dynamic asoect ratio of region |
| JP2011109176A (en)* | 2009-11-12 | 2011-06-02 | Nippon Telegr & Teleph Corp <Ntt> | Device and method for multiplexing video, and program |
| CN102982508A (en)* | 2011-06-17 | 2013-03-20 | 索尼公司 | Image processing apparatus and method, program, and recording medium |
| CN103414915A (en)* | 2013-08-22 | 2013-11-27 | 合一网络技术(北京)有限公司 | Quality evaluation method and device for uploaded videos of websites |
| EP2736261A1 (en)* | 2012-11-27 | 2014-05-28 | Alcatel Lucent | Method For Assessing The Quality Of A Video Stream |
| CN103856775A (en)* | 2014-03-18 | 2014-06-11 | 天津大学 | Processing method for subjective evaluation result of stereo video quality |
| CN106210767A (en)* | 2016-08-11 | 2016-12-07 | 上海交通大学 | A kind of video frame rate upconversion method and system of Intelligent lifting fluidity of motion |
| CN109068174A (en)* | 2018-09-12 | 2018-12-21 | 上海交通大学 | Video frame rate upconversion method and system based on cyclic convolution neural network |
| CN109379550A (en)* | 2018-09-12 | 2019-02-22 | 上海交通大学 | Video frame rate up-conversion method and system based on convolutional neural network |
| CN109587474A (en)* | 2018-12-14 | 2019-04-05 | 央视国际网络无锡有限公司 | No-reference video quality evaluating method and device based on distortion restoring degree |
| WO2020080698A1 (en)* | 2018-10-19 | 2020-04-23 | 삼성전자 주식회사 | Method and device for evaluating subjective quality of video |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8848061B2 (en)* | 2012-06-27 | 2014-09-30 | Apple Inc. | Image and video quality assessment |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101156451A (en)* | 2005-04-12 | 2008-04-02 | 皇家飞利浦电子股份有限公司 | Video processing with region-based multiple-pass motion estimation and update of temporal motion vector candidates |
| CN101156450A (en)* | 2005-04-12 | 2008-04-02 | 皇家飞利浦电子股份有限公司 | Region- based 3drs motion estimation using dynamic asoect ratio of region |
| JP2011109176A (en)* | 2009-11-12 | 2011-06-02 | Nippon Telegr & Teleph Corp <Ntt> | Device and method for multiplexing video, and program |
| CN102982508A (en)* | 2011-06-17 | 2013-03-20 | 索尼公司 | Image processing apparatus and method, program, and recording medium |
| EP2736261A1 (en)* | 2012-11-27 | 2014-05-28 | Alcatel Lucent | Method For Assessing The Quality Of A Video Stream |
| CN103414915A (en)* | 2013-08-22 | 2013-11-27 | 合一网络技术(北京)有限公司 | Quality evaluation method and device for uploaded videos of websites |
| CN103856775A (en)* | 2014-03-18 | 2014-06-11 | 天津大学 | Processing method for subjective evaluation result of stereo video quality |
| CN106210767A (en)* | 2016-08-11 | 2016-12-07 | 上海交通大学 | A kind of video frame rate upconversion method and system of Intelligent lifting fluidity of motion |
| CN109068174A (en)* | 2018-09-12 | 2018-12-21 | 上海交通大学 | Video frame rate upconversion method and system based on cyclic convolution neural network |
| CN109379550A (en)* | 2018-09-12 | 2019-02-22 | 上海交通大学 | Video frame rate up-conversion method and system based on convolutional neural network |
| WO2020080698A1 (en)* | 2018-10-19 | 2020-04-23 | 삼성전자 주식회사 | Method and device for evaluating subjective quality of video |
| CN109587474A (en)* | 2018-12-14 | 2019-04-05 | 央视国际网络无锡有限公司 | No-reference video quality evaluating method and device based on distortion restoring degree |
| Title |
|---|
| "面向 HDTV 高刷新率的视频帧速率变化算 法研究";陈学伟等;《新型工业化》;第第1卷卷(第第7期期);第58-70页* |
| Publication number | Publication date |
|---|---|
| CN112767310A (en) | 2021-05-07 |
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | Dd-cyclegan: Unpaired image dehazing via double-discriminator cycle-consistent generative adversarial network | |
| CN114339219B (en) | Inter-frame prediction method, inter-frame prediction device, encoding and decoding method, encoding and decoding device and electronic equipment | |
| Liu et al. | Perceptual reduced-reference visual quality assessment for contrast alteration | |
| CN107636690A (en) | Full-reference image quality assessment based on convolutional neural network | |
| Hou et al. | A perceptual quality metric for video frame interpolation | |
| CN110738235A (en) | Pulmonary tuberculosis determination method, pulmonary tuberculosis determination device, computer device, and storage medium | |
| CN111046893B (en) | Image similarity determining method and device, image processing method and device | |
| Kim et al. | Deep blind image quality assessment by employing FR-IQA | |
| Yang et al. | Blind VQA on 360° video via progressively learning from pixels, frames, and video | |
| CN117726542A (en) | Controllable noise removal method and system based on diffusion model | |
| CN112767310B (en) | Video quality evaluation method, device and equipment | |
| CN112950579B (en) | Image quality evaluation method and device and electronic equipment | |
| Ceruso et al. | Relative multiscale deep depth from focus | |
| Zhou et al. | Stereoscopic video quality prediction based on end-to-end dual stream deep neural networks | |
| CN109871790A (en) | A Video Decolorization Method Based on Hybrid Neural Network Model | |
| CN111612766B (en) | Image quality evaluation method and device and electronic equipment | |
| CN112399177B (en) | Video coding method, device, computer equipment and storage medium | |
| Cemiloglu et al. | Blind video quality assessment via spatiotemporal statistical analysis of adaptive cube size 3D‐DCT coefficients | |
| US20240177277A1 (en) | System, devices and/or processes for application of an intensity derivative for temporal image stability | |
| Hu et al. | No-reference image quality assessment based on global awareness | |
| CN105959685A (en) | A Compression Bit Rate Prediction Method Based on Video Content and Cluster Analysis | |
| CN115760603A (en) | Interference array broadband imaging method based on big data technology | |
| Dhondea et al. | DFTS2: Simulating deep feature transmission over packet loss channels | |
| CN113936135A (en) | A detection image generation method and system | |
| CN115623204A (en) | Encoding processing method and device, electronic equipment and storage medium |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |