CN-120640067-B - Object attribute dynamic configuration method for digital media multi-platform

Abstract

The invention relates to the technical field of digital media processing, in particular to a dynamic configuration method for object attributes of a digital media multi-platform. Aiming at the configuration difficulty brought by the technical specification difference of digital media when different platforms are displayed, platform feature vectors containing composite features are generated by collecting and analyzing the technical specifications of each platform in real time, a nonlinear conversion model from object basic attributes to platform specific attributes is established by using a multi-layer perceptron, a subscription-publishing architecture containing subscription management, message publishing and message distributing modules is constructed, so that digital media objects can subscribe to platform policy change notification, configuration recalculation is triggered after the notification is received, and new attributes are applied to the objects and configuration information is updated. The invention can dynamically adapt to the standard change of the platform, improves the management efficiency and the display effect of the digital media content, has the advantages of strong universality, high configuration accuracy and the like, and effectively meets the propagation requirement of the multi-platform digital media.

Inventors

• TAO YUHENG

Assignees

• 中国传媒大学

Dates

Publication Date

20260512

Application Date

20250423

Claims (2)

1. 1. A method for dynamically configuring object properties for a digital media multi-platform, comprising: S1, collecting technical specifications and generating a feature vector, namely collecting the technical specifications of each platform in terms of color space, file format, compression format, file size, encryption method, frame rate and resolution in real time through web crawlers, API calls or accessing platform official documents, and analyzing the information to generate a platform feature vector; The feature vector generation process comprises the steps of combining and constructing new features according to the collected features, and specifically comprises the following steps: constructing frame rate and resolution comprehensive features and file size and compression ratio comprehensive features based on feature calculation; The specific method for constructing the frame rate and resolution integrated feature based on feature calculation is to calculate the product of the frame rate and the resolution, and then perform standardization processing, wherein the obtained result is the frame rate and resolution integrated feature, and is defined as the information density of the video content in unit time, and a specific calculation formula is as follows: ; In the formula, For the frame rate to be the same, Is the number of horizontal pixels and, Is the number of vertical pixels and, 、 And The maximum values of the number of resolution vertical pixels, the number of horizontal pixels and the frame rate in the considered platform are respectively; The specific method for constructing the file size and compression ratio comprehensive features based on feature calculation is to calculate the product of the file size and the compression ratio, then perform standardization processing, and the obtained result is the file size and compression ratio comprehensive features and is used for reflecting the situation after the files are compressed under different platforms, wherein the specific calculation formula is as follows: ; In the formula, In order to be of the size of the file, In order to achieve a file compression ratio, For the maximum file size allowed in the platform, Is the maximum compression ratio in the platform; the integrated feature of the file size and the compression ratio reflects the situation after the file compression under different platforms through the combined feature of the file size and the compression ratio; constructing color space, encryption method and file format cross features based on feature cross, frame rate, resolution and compression format cross features; the specific method for constructing the color space, the encryption method and the file format cross features based on the feature cross comprises the following steps: For setting the color space and the color gamut Representation, encryption method is used after label coding Representing, the vector of the file format after being subjected to one-hot coding is that The combined color space, encryption method and file format cross feature vector is: ; the specific method for constructing the frame rate, resolution and compression format cross features based on the feature cross comprises the following steps: let the frame rate be Number of horizontal pixels for resolution And vertical pixel count Representation, compressed format, tag encoded and then used Representing, then combining the frame rate, resolution and compressed format cross feature vectors as ; The frame rate, resolution and compression format cross features are combined with the frame rate, resolution and compression format to reflect the characteristics of the compression format which are commonly used under different frame rates and resolutions; s2, establishing a nonlinear conversion model, namely, based on the generated platform feature vector, selecting a multi-layer perceptron model framework, and establishing a nonlinear conversion model from the object basic attribute to the platform specific attribute; the multi-layer perceptron model framework consists of an input layer, a hidden layer and an output layer, wherein the input layer receives basic attributes and platform characteristic vectors of a digital media object, the hidden layer comprises a plurality of neurons, nonlinear transformation is carried out on input information through an activation function, the output layer outputs predicted platform specific attributes, a large amount of sample data is used during training, error gradients are calculated through a back propagation algorithm, and the weights and the biases of the network are updated, so that a model prediction result approximates to the actual platform specific attributes; s3, implementing a subscription-release mode, namely subscribing the digital media object to the policy change notification of each platform by adopting the subscription-release mode; And S4, triggering configuration recalculation, namely after the digital media object receives the platform policy change notification, analyzing change information in the notification to generate a new platform feature vector.
2. 2. The method of claim 1, wherein after the recalculation of the object attributes is completed, the newly calculated specific attributes are applied to the digital media object and the configuration information of the object stored on the platform in the digital media object management system database is updated, including the new specific attributes, the attached platform and the update time, to ensure that the presentation and distribution of the object on the platform meets the latest specifications.

Description

Object attribute dynamic configuration method for digital media multi-platform Technical Field The invention relates to the technical field of digital media processing, in particular to a dynamic configuration method for object attributes of a digital media multi-platform. Background With the rapid development of digital media technology, digital media content needs to be displayed and spread on a number of different platforms, such as televisions, computers, cell phones, tablets, etc. And different platforms have significant technical specification differences in terms of color space, file format, compression format, file size, encryption method, frame rate, resolution, and the like. Conventional digital media object attribute configuration methods are often static, and once configuration is performed according to the requirements of a specific platform after the content is produced. When the technical specifications of the platforms change, the digital media object attributes are required to be manufactured and configured again, so that a great deal of time and resources are consumed, the display effects of the content on different platforms are inconsistent, and the requirements of users on high-quality digital media experience cannot be met. Disclosure of Invention The invention provides an object attribute dynamic configuration method for digital media multi-platform, which is characterized in that technical specifications of all the platforms are analyzed in real time to generate platform feature vectors, a nonlinear conversion model from object basic attributes to platform specific attributes is established by using a multi-layer perceptron, platform policy change notification is received by adopting a subscription-release mode, configuration recalculation is triggered, dynamic configuration of the digital media object attributes is realized, and display effects and management efficiency of digital media content on different platforms are improved. A method for dynamic configuration of object properties for a digital media multi-platform, comprising: s1, collecting technical specifications and generating feature vectors, namely collecting the technical specifications of each platform in terms of color space, file format, compression format, file size, encryption method, frame rate and resolution in real time through web crawlers, API calls or accessing platform official documents, and analyzing the information to generate the platform feature vectors. Preferably, for color space, detailed parameters such as standard name, color gamut range, white point coordinate, gamma correction curve and the like are collected, for file format, coding mode, extension, supported metadata and the like of the file are collected, for compression format, compression algorithm (such as JPEG and PNG compression algorithm) and compression ratio range adopted are known, for file size, maximum and minimum allowable values of a single file are obtained, for encryption method, encryption algorithm (such as AES and RSA) and encryption key management mode which are definitely used are obtained, for frame rate, frame number per second of video playing is determined, and for resolution, horizontal and vertical pixel number of an image or video is obtained. Preferably, the collected original information is subjected to sorting and normalization processing, and the data represented by different formats are converted into a unified mathematical model or quantization index. And then constructing a platform feature vector according to the analyzed data, wherein each dimension of the vector corresponds to a key technical feature parameter. Preferably, the generating process of the feature vector includes constructing new features according to the combination of the collected features, and specifically includes: constructing frame rate and resolution comprehensive features and file size and compression ratio comprehensive features based on feature calculation; the integrated feature of the file size and the compression ratio reflects the situation after the file compression under different platforms through the combined feature of the file size and the compression ratio; constructing color space, encryption method and file format cross features based on feature cross, frame rate, resolution and compression format cross features; The frame rate, resolution and compression format cross features are combined with the frame rate, resolution and compression format, so that the characteristics of the compression format which are commonly used under different frame rates and resolutions are reflected. Preferably, the specific method for constructing the frame rate and resolution integrated feature based on feature calculation is to calculate the product of the frame rate and the resolution, and then perform standardization processing, where the obtained result is the frame rate and resolution integrated feature, and is defined as the informati