Distributed type number storage application system
Through the customized structure-designed AI analysis model, the problem of lack of high-precision analysis mechanism in the existing technology is solved, the monitoring terminal's awareness of the content of the target image is improved, and intelligent special effects processing is realized Type identification and processing.
Patent Information
- Application Number
- CN202411191276.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-28
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art lacks a high-precision analysis mechanism to process the special effects processing type that the target image undergoes in the target imaging mechanism, resulting in a low level of awareness of the target image content at the monitoring end.
Using a customized structure design AI analysis model, the target image and Bayer image are analyzed by setting the number of trained BP neural networks in proportion to the number of pixel circuits of the target camera mechanism to analyze the image signal of the target image and Bayer image to obtain clarity and background The area proportion, noise type and noise amplitude are used to determine the special effect processing type and perform corresponding processing.
It provides an intelligent analysis foundation, helps the monitoring end to improve its awareness of the target image content, is logically reliable and has intelligent design, and can filter the basic data in a targeted manner for intelligent analysis of special effects processing types.
Smart Images
Figure CN119094906A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data analysis, and in particular to a distributed type number storage application system. Background Art
[0002] The main activities of the data analysis process consist of identifying information needs, collecting data, analyzing data, and evaluating and improving the effectiveness of data analysis.
[0003] Identifying information needs is the first condition to ensure the effectiveness of the data analysis process, which can provide clear goals for collecting and analyzing data. Identifying information needs is the responsibility of managers. Managers should put forward information needs based on the needs of decision-making and process control. In terms of process control, managers should identify the information needed to support the review of process inputs, process outputs, rationality of resource allocation, optimization plans for process activities, and discovery of abnormal process variations. Collecting data is a purposeful collection of data, which is the basis for ensuring the effectiveness of the data analysis process. Organizations need to plan the content, channels, and methods of collecting data. Analyzing data is to convert the collected data into information through processing, organization, and analysis.
[0004] Currently, it is necessary to use a data analysis process to analyze the types of special effects processing that the received target image has undergone, so as to help the monitoring end improve its understanding of the content of the target image. For example, it is necessary to perform data analysis on the types of special effects processing that have been performed on target images captured by mobile terminals such as smartphones that may have built-in special effects functions. Obviously, the existing technology lacks a corresponding targeted and high-precision analysis mechanism. Summary of the invention
[0005] In order to solve the technical problems in the prior art, the present invention provides a distributed type number storage application system, which intelligently analyzes the special effect processing type of the target image in the target camera mechanism by adopting an AI analysis model with a customized structure design. The structural customization of the AI analysis model is that the AI analysis model is a BP neural network after each training of a set number and the value of the set number is proportional to the number of pixel circuits of the photoelectric sensing device of the target camera mechanism. The target image is subjected to image signal analysis to obtain the clarity, background area ratio, various noise type numbers and the noise amplitudes corresponding to the various noise types corresponding to the target image and output them as multiple auxiliary data of the target image, and the photoelectric sensing device of the target camera mechanism is directly sensed and output. The Bayer image is analyzed for image signals to obtain the clarity, background area ratio, various noise type numbers and noise amplitudes corresponding to various noise types corresponding to the target image and output them as multiple auxiliary data of the Bayer image, thereby providing reliable basic content for the intelligent analysis of the special effect processing type that the target image has undergone in the target camera mechanism, and using an AI analysis model to judge the type number of the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image, wherein the Bayer image corresponding to the target image is subjected to special effect processing corresponding to the special effect processing type of the type number obtained by judgment, to obtain the target image, thereby helping the monitoring end to improve its content recognition of the target image.
[0006] According to the present invention, a distributed type number storage application system is provided, the system comprising: A number storage device is provided at the big data service network element, and is used to pre-store type numbers of various special effect processings corresponding to different types of special effect processings, wherein different special effect processings correspond to different type numbers and each type number is represented by a binary value; A signal extraction device, arranged at the big data service network element, is used to obtain a target image to be subjected to special effect analysis, wherein the target image to be subjected to special effect analysis originates from a target camera mechanism; A model application mechanism is provided at a big data service network element and is electrically connected to the signal extraction device network and the number storage device, respectively, and is used to perform image signal analysis on a target image to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the target image; perform image signal analysis on a Bayer image directly sensed and output by a photoelectric sensing device of a target camera mechanism to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the Bayer image; and use an AI analysis model to determine the type number of the special effect processing type of the target image based on the resolution of the target image, the multiple auxiliary data of the target image, and the multiple auxiliary data of the Bayer image; Among them, the type number of the special effect processing type of the target image is determined by using the AI analysis model based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image, including: the Bayer image corresponding to the target image is subjected to special effect processing corresponding to the special effect processing type of the type number obtained by the determination, thereby obtaining the target image; Among them, the type number of using the AI analysis model to determine the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the AI analysis model is a set number of BP neural networks after each training and the value of the set number is proportional to the number of pixel circuits of the photoelectric sensing device of the target camera mechanism; The number storage device is arranged at the big data service network element and is used to pre-store the type numbers of each copy corresponding to various special effect processing, the type numbers corresponding to different special effect processing are different and each type number is represented by a binary value, including: the number storage device uses different storage spaces to store each type number respectively; Among them, the type number of using AI analysis model to judge the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the special effect processing type of the target image is one of animation special effect, skin smoothing special effect, transition special effect and lens flare special effect.
[0007] It can be seen that the present invention has at least the following significant substantive features: Substantive feature 1: The AI analysis model with customized structure design is used to intelligently analyze the special effects processing type of the target image in the target camera mechanism. The structure customization of the AI analysis model is that the AI analysis model is a set number of BP neural networks after each training, and the value of the set number is proportional to the number of pixel circuits of the photoelectric sensing device of the target camera mechanism: Substantive feature 2: Perform image signal analysis on the target image to obtain the clarity, background area ratio, various noise type numbers and noise amplitudes corresponding to the target image and output them as multiple auxiliary data of the target image; perform image signal analysis on the Bayer image directly sensed and output by the photoelectric sensing device of the target camera mechanism to obtain the clarity, background area ratio, various noise type numbers and noise amplitudes corresponding to the target image and output them as multiple auxiliary data of the Bayer image, thereby providing reliable basic content for the intelligent analysis of the special effect processing type of the target image in the target camera mechanism; Substantive feature three: An AI parsing model is used to determine the type number of the special effects processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image, wherein the Bayer image corresponding to the target image is processed by the special effects processing corresponding to the special effects processing type of the type number obtained by the determination, thereby helping the monitoring end to improve its understanding of the content of the target image.
[0008] The distributed type number storage application system of the present invention is logically reliable and intelligently designed. Since the AI analysis model with customized structure design can be used to intelligently analyze the special effect processing type of the target image in the target camera based on the targeted selected basic data, it helps the monitoring end to improve the content cognition of the target image. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 It is a diagram showing the internal structure of a distributed type number storage application system according to the first embodiment of the present invention.
[0010] Figure 2 It is a diagram showing the internal structure of a distributed type number storage application system according to a second embodiment of the present invention.
[0011] Figure 3 It is a diagram showing the internal structure of a distributed type number storage application system according to the third embodiment of the present invention. DETAILED DESCRIPTION
[0012] The implementation scheme of the distributed type number storage application system of the present invention will be described in detail below with reference to the accompanying drawings.
[0013] Figure 1 The internal structure diagram of the distributed type number storage application system according to the first embodiment of the present invention is shown, and the system includes: A number storage device is provided at the big data service network element, and is used to pre-store type numbers of various special effect processings corresponding to different types of special effect processings, wherein different special effect processings correspond to different type numbers and each type number is represented by a binary value; Specifically, the number storage device is arranged at the big data service network element, and is used to pre-store the type numbers of each copy corresponding to various special effect processing, wherein the type numbers corresponding to different special effect processing are different and each type number is represented by a binary value, including: selecting and using different physical storage addresses to pre-store the type numbers of each copy corresponding to various special effect processing; A signal extraction device, arranged at the big data service network element, is used to obtain a target image to be subjected to special effect analysis, wherein the target image to be subjected to special effect analysis originates from a target camera mechanism; A model application mechanism is provided at a big data service network element and is electrically connected to the signal extraction device network and the number storage device, respectively, and is used to perform image signal analysis on a target image to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the target image; perform image signal analysis on a Bayer image directly sensed and output by a photoelectric sensing device of a target camera mechanism to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the Bayer image; and use an AI analysis model to determine the type number of the special effect processing type of the target image based on the resolution of the target image, the multiple auxiliary data of the target image, and the multiple auxiliary data of the Bayer image; Among them, the type number of the special effect processing type of the target image is determined by using the AI analysis model based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image, including: the Bayer image corresponding to the target image is subjected to special effect processing corresponding to the special effect processing type of the type number obtained by the determination, thereby obtaining the target image; Among them, the type number of using the AI analysis model to determine the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the AI analysis model is a set number of BP neural networks after each training and the value of the set number is proportional to the number of pixel circuits of the photoelectric sensing device of the target camera mechanism; The number storage device is arranged at the big data service network element and is used to pre-store the type numbers of each copy corresponding to various special effect processing, the type numbers corresponding to different special effect processing are different and each type number is represented by a binary value, including: the number storage device uses different storage spaces to store each type number respectively; Among them, the type number of using the AI parsing model to judge the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the special effect processing type of the target image is one of animation special effect, skin smoothing special effect, transition special effect and lens flare special effect; And wherein, using the AI analytical model to determine the type number of the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image also includes: using a numerical simulation mode to implement the simulation and testing of the data processing process of using the AI analytical model to determine the type number of the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image; And wherein, the type number of using the AI analysis model to determine the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: performing binary value conversion processing on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image respectively and then synchronously inputting them into the AI analysis model.
[0014] Figure 2 It is a diagram showing the internal structure of a distributed type number storage application system according to a second embodiment of the present invention.
[0015] exist Figure 2 In, with Figure 1 Differently, the distributed type number storage application system shown in the second embodiment of the present invention may also include: A bandwidth measurement mechanism, connected to the number storage device, the signal extraction device and the model application mechanism, respectively, for measuring the instantaneous bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively; The bandwidth measurement mechanism is connected to the number storage device, the signal extraction device and the model application mechanism respectively, and is used to measure the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively, including: the bandwidth measurement mechanism includes a plurality of bandwidth measurement units, which are used to be connected to the number storage device, the signal extraction device and the model application mechanism respectively, so as to complete the measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively; Wherein, the bandwidth measurement mechanism comprises a plurality of bandwidth measurement units, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to complete the respective measurement of the instantaneous bandwidth data of the number storage device, the signal extraction device and the model application mechanism, including: the plurality of bandwidth measurement units are a plurality of bandwidth sensing circuits, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to complete the respective measurement of the instantaneous bandwidth data of the number storage device, the signal extraction device and the model application mechanism; The plurality of bandwidth measurement units are a plurality of bandwidth sensing circuits, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to respectively measure the instantaneous bandwidth data of the number storage device, the signal extraction device and the model application mechanism, including: the plurality of bandwidth sensing circuits have the same structure; And wherein, the multiple bandwidth measurement units are multiple bandwidth sensing circuits, which are used to be connected to the number storage device, the signal extraction device and the model application mechanism respectively to complete the separate measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively, and also include: the multiple bandwidth sensing circuits have the same bandwidth measurement upper limit value and bandwidth measurement lower limit value.
[0016] Figure 3 It is a diagram showing the internal structure of a distributed type number storage application system according to the third embodiment of the present invention.
[0017] exist Figure 3 In, with Figure 1 Differently, the distributed type number storage application system shown in the third embodiment of the present invention may also include: a wired configuration interface, arranged near the number storage device, the signal extraction device and the model application mechanism and connected to the number storage device, the signal extraction device and the model application mechanism respectively, for providing data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively; Wherein, the wired configuration interface is arranged near the number storage device, the signal extraction device and the model application mechanism and is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to provide data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively, including: the wired configuration interface is connected to the number storage device, the signal extraction device and the model application mechanism through different wired configuration units; And wherein, a wired configuration interface is arranged near the number storage device, the signal extraction device and the model application mechanism and is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to provide data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively, and also includes: the internal structures of different wired configuration units connecting the number storage device, the signal extraction device and the model application mechanism are the same.
[0018] In addition, in the distributed type number storage application system, the type number of the special effect processing type of the target image is determined based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image using the AI analysis model, and also includes: the AI analysis model is stored in a model storage device, and the model storage device completes the model storage of the AI analysis model by storing various model parameters of the AI analysis model.
[0019] Although the present invention has been disclosed in detail with various preferred embodiments as above, the present invention should not be interpreted as only covering the above embodiments, which are only exemplary embodiments. Any person skilled in the art may make some changes and improvements without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the claims and their equivalents.
Claims
1. A distributed type number storage application system, characterized in that: The system comprises: A number storage device is provided at the big data service network element, and is used to pre-store type numbers of various special effect processings corresponding to different types of special effect processings, wherein different special effect processings correspond to different type numbers and each type number is represented by a binary value; A signal extraction device, arranged at the big data service network element, is used to obtain a target image to be subjected to special effect analysis, wherein the target image to be subjected to special effect analysis originates from a target camera mechanism; A model application mechanism is provided at a big data service network element and is electrically connected to the signal extraction device network and the number storage device, respectively, and is used to perform image signal analysis on a target image to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the target image; perform image signal analysis on a Bayer image directly sensed and output by a photoelectric sensing device of a target camera mechanism to obtain the clarity, background area ratio, various noise type numbers, and noise amplitudes corresponding to the target image, and output them as multiple auxiliary data of the Bayer image; and use an AI analysis model to determine the type number of the special effect processing type of the target image based on the resolution of the target image, the multiple auxiliary data of the target image, and the multiple auxiliary data of the Bayer image; Among them, the type number of the special effect processing type of the target image is determined by using the AI analysis model based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image, including: the Bayer image corresponding to the target image is subjected to special effect processing corresponding to the special effect processing type of the type number obtained by the determination, thereby obtaining the target image; Among them, the type number of using the AI analysis model to determine the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the AI analysis model is a set number of BP neural networks after each training and the value of the set number is proportional to the number of pixel circuits of the photoelectric sensing device of the target camera mechanism; The number storage device is arranged at the big data service network element and is used to pre-store the type numbers of each copy corresponding to various special effect processing, the type numbers corresponding to different special effect processing are different and each type number is represented by a binary value, including: the number storage device uses different storage spaces to store each type number respectively; Among them, the type number of using AI analysis model to judge the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: the special effect processing type of the target image is one of animation special effect, skin smoothing special effect, transition special effect and lens flare special effect.
2. The distributed type number storage application system according to claim 1, characterized in that: The method of using an AI analytical model to determine the type number of special effect processing type of a target image based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image also includes: using a numerical simulation mode to implement the simulation and testing of the data processing process of using an AI analytical model to determine the type number of special effect processing type of a target image based on the resolution of the target image, multiple auxiliary data of the target image, and multiple auxiliary data of the Bayer image; Among them, the type number of using the AI analysis model to determine the special effect processing type of the target image based on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image also includes: performing binary value conversion processing on the resolution of the target image, multiple auxiliary data of the target image and multiple auxiliary data of the Bayer image respectively, and then synchronously inputting them into the AI analysis model.
3. The distributed type number storage application system according to claim 2, characterized in that: The system further comprises: A bandwidth measurement mechanism, connected to the number storage device, the signal extraction device and the model application mechanism, respectively, for measuring the instantaneous bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively; Among them, the bandwidth measurement mechanism is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to respectively measure the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism, including: the bandwidth measurement mechanism includes multiple bandwidth measurement units, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to complete the respective measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism.
4. The distributed type number storage application system according to claim 3, characterized in that: The bandwidth measurement mechanism includes a plurality of bandwidth measurement units, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to complete the respective measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism, including: the plurality of bandwidth measurement units are a plurality of bandwidth sensing circuits, which are respectively connected to the number storage device, the signal extraction device and the model application mechanism to complete the respective measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism.
5. The distributed type number storage application system according to claim 4, characterized in that: The multiple bandwidth measurement units are multiple bandwidth sensing circuits, which are used to be connected to the number storage device, the signal extraction device and the model application mechanism respectively to complete the separate measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively, including: the structures of the multiple bandwidth sensing circuits are the same.
6. The distributed type number storage application system according to claim 5, characterized in that: The multiple bandwidth measurement units are multiple bandwidth sensing circuits, which are used to be connected to the number storage device, the signal extraction device and the model application mechanism respectively to complete the separate measurement of the real-time bandwidth data of the number storage device, the signal extraction device and the model application mechanism respectively, and also include: the multiple bandwidth sensing circuits have the same bandwidth measurement upper limit value and bandwidth measurement lower limit value.
7. The distributed type number storage application system according to claim 2, characterized in that: The system further comprises: A wired configuration interface is arranged near the number storage device, the signal extraction device and the model application mechanism and is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to provide data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively.
8. The distributed type number storage application system according to claim 7, characterized in that: A wired configuration interface is arranged near the number storage device, the signal extraction device and the model application mechanism and is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to provide data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively. The wired configuration interface includes: the wired configuration interface connects the number storage device, the signal extraction device and the model application mechanism through different wired configuration units.
9. The distributed type number storage application system according to claim 8, characterized in that: A wired configuration interface is arranged near the number storage device, the signal extraction device and the model application mechanism and is respectively connected to the number storage device, the signal extraction device and the model application mechanism, and is used to provide data configuration services based on wired transmission for the number storage device, the signal extraction device and the model application mechanism respectively. It also includes: different wired configuration units connecting the number storage device, the signal extraction device and the model application mechanism have the same internal structure.