State judgment equipment based on probability numerical value analysis

Through the Hofitter neural network model combined with multiple basic data from the image capture mechanism, the subprocess type of the image has been identified, which solves the problem that the image processing type cannot be determined in the prior art and provides accurate recognition and stability of image content.

CN119026021AInactive Publication Date: 2024-11-26NANJING HUANYANHAN TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411250837.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-06
Publication Date
2024-11-26
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The lack of mature mechanisms for numerical analysis in the prior art, and it is impossible to determine which type of image processing the image has undergone, resulting in a lack of a key data basis for the re-identification of image content.

Method used

Using the Hofitter neural network model, multiple learnings are performed by introducing the successively converted devices, combining the parameters of the CMOS sensing component of the image capture mechanism, the visual content of the image and the binary coded values ​​of the processing type, intelligent recognition is performed, and the subprocessing of the output image has been performed. The probability value of the type.

Benefits of technology

Intelligent recognition of the subprocess type of images is realized, valuable reference information of image content is provided, and the reliability and stability of the recognition results are ensured.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026021A_ABST
    Figure CN119026021A_ABST
Patent Text Reader

Abstract

The invention relates to a state judgment device based on probability numerical value analysis, which comprises a content identification component and a probability numerical value analysis component, the method comprises the following steps: sensing a plurality of component parameters of a component based on a CMOS of an image capturing mechanism; each part of probability numerical value corresponding to image processing of each sub-processing type executed by the field captured image is intelligently judged according to each part of visual content of the field captured image and binary coding numerical value of each part of processing type corresponding to each piece of fine classification image processing; and the probability judgment part is used for taking the sub-processing type corresponding to the maximum probability numerical value in the probability numerical values as the executed sub-processing type. The state judgment equipment based on probability numerical value analysis is based on logic, and operation is intelligent. The Hough neural network model can be adopted to intelligently identify which sub-processing type of image processing is performed on an image captured on site based on multiple items of basic data selected in a targeted manner, so that valuable reference information is provided for re-cognition of the image content.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of numerical analysis, and more specifically, to a state judgment device based on probability numerical analysis. Background Art

[0002] With the development of computers, human labor has been liberated, and many mathematicians have been freed from complicated calculations. Computers were originally born for mathematics. Humans are not omnipotent and have their inherent limitations. With the development of science and technology in human society, it is undoubtedly a pipe dream to rely solely on manpower to calculate more and more scientific and technological problems. With computers, people expect to set many mathematical models into a set of algorithms and run them on computers, leaving all the complex mechanized things to computers. Numerical analysis is precisely a very practical subject for computers. Solving scientific and technological problems with computers usually goes through the following steps: establish a mathematical model based on actual problems; give numerical calculation methods based on mathematical models; compile algorithm programs based on calculation methods to obtain results on computers.

[0003] Numerical analysis has a wide range of application prospects. For example, when people obtain a frame of image, they know that the frame of image has undergone a certain type of image processing, for example, image filtering or image enhancement, but they do not know which sub-type of image processing under this type of image processing has been undergone. For example, they do not know whether the image has undergone recursive filtering, median filtering, or wavelet filtering under image filtering, etc., resulting in a lack of key data basis for re-identifying the true content of the frame of image. Obviously, the prior art lacks a mature mechanism for numerical analysis to solve the above technical problems. Summary of the invention

[0004] In order to solve the technical problems in the prior art, the present invention provides a state judgment device based on probabilistic numerical analysis, which can use a Hoffet neural network model to intelligently identify which sub-processing type of image processing the on-site captured image has undergone based on a number of targeted selected basic data when it is known that the on-site captured image output by the image capture mechanism has been processed by a preset large processing type but it is uncertain which sub-processing type of image processing under the preset large processing type has been processed. What is particularly critical is the structural customization of the Hoffet neural network model for intelligent identification, which is manifested in: introducing a sequential conversion device for sequentially performing multiple learning operations on the Hoffet neural network to obtain the Hoffet neural network after completing the multiple learning operations and outputting it as a Hoffet neural network model, thereby providing valuable reference information for the re-cognition of the image content.

[0005] According to the present invention, a state judgment device based on probability numerical analysis is provided, the device comprising: A data receiving device, for receiving a live captured image from an image capturing mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capturing mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capturing mechanism, wherein the sub-processing type is a sub-classification of the preset major processing type; A sequential conversion device is used to successively perform multiple learning operations on the Hoffet neural network to obtain the Hoffet neural network after completing the multiple learning operations and output it as a Hoffet neural network model; a content identification component, connected to the data receiving device and the sequential conversion device respectively, for synchronously inputting multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the scene captured image, and binary coded values ​​of various processing types corresponding to various sub-classified image processing into the Hoffet neural network model, and executing the Hoffet neural network model to obtain the probability values ​​of the scene captured image output by the Hoffet neural network model corresponding to the image processing of each sub-processing type executed in the image capture mechanism; A probability judgment component connected to the content identification component, for outputting the sub-processing type corresponding to the probability value with the largest value among the probability values ​​as the sub-processing type executed by the scene captured image in the image capturing mechanism; Wherein, receiving a live captured image from an image capture mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capture mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capture mechanism, wherein the sub-processing type is a subclassification of the preset major processing type, including: the preset major processing type is an enhancement processing type, and each sub-processing type is one of an exponential image enhancement processing, a histogram equalization enhancement processing, a logarithmic image enhancement processing, and an exponential image enhancement processing; Among them, multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the on-site captured image, and binary coded values ​​of various processing types corresponding to various sub-classified image processing are synchronously input into the Hoffet neural network model, and the Hoffet neural network model is executed to obtain the probability values ​​corresponding to the image processing of each sub-processing type executed in the image capture mechanism on the on-site captured image output by the Hoffet neural network model. The multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism.

[0006] Therefore, the present invention has at least the following three beneficial technical effects: Technical effect A: when it is known that the scene-captured image output by the image capture mechanism has been processed by a preset large processing type but it is uncertain which sub-processing type of image processing under the preset large processing type has been processed, a Hoffitt neural network model is used to intelligently identify which sub-processing type of image processing the scene-captured image has been processed by; Technical effect B: In order to ensure the reliability and stability of the intelligent recognition results, multiple basic data are selected in a targeted manner, including multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the on-site captured images, and binary coded values ​​of various processing types corresponding to various sub-processing types, wherein the multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency, and thermal noise amplitude of the CMOS sensing component of the image capture mechanism; Technical effect C: Structural customization of the Hoffitt neural network model for intelligent recognition, which is manifested in: introducing a sequential conversion device for performing multiple learning operations on the Hoffitt neural network in succession to obtain the Hoffitt neural network after completing multiple learning operations and output it as a Hoffitt neural network model.

[0007] The state judgment device based on probability numerical analysis of the present invention has a logical basis and operation intelligence. Since the Hoffitt neural network model can be used to intelligently identify which sub-processing type of image processing the scene captured image has undergone based on multiple targeted basic data, it provides valuable reference information for the re-cognition of the image content. BRIEF DESCRIPTION OF THE DRAWINGS Those skilled in the art may better understand the numerous advantages of the present invention by referring to the accompanying drawings, in which: Figure 1 It is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to the first embodiment of the present invention.

[0009] Figure 2 It is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to the second embodiment of the present invention.

[0010] Figure 3 It is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to the third embodiment of the present invention. DETAILED DESCRIPTION

[0011] The implementation scheme of the state judgment device based on probability numerical analysis of the present invention will be described in detail below with reference to the accompanying drawings.

[0012] Figure 1 1 is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to a first embodiment of the present invention, the device comprising: A data receiving device, for receiving a live captured image from an image capturing mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capturing mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capturing mechanism, wherein the sub-processing type is a sub-classification of the preset major processing type; For example, a data receiving device is used to receive a live captured image from an image capturing mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capturing mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capturing mechanism, wherein the sub-processing type is a sub-classification of the preset major processing type, and includes: the data receiving device includes a synchronous driving interface, which is used to control the timing of receiving the live captured image from the image capturing mechanism; A sequential conversion device is used to successively perform multiple learning operations on the Hoffet neural network to obtain the Hoffet neural network after completing the multiple learning operations and output it as a Hoffet neural network model; a content identification component, connected to the data receiving device and the sequential conversion device respectively, for synchronously inputting multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the scene captured image, and binary coded values ​​of various processing types corresponding to various sub-classified image processing into the Hoffet neural network model, and executing the Hoffet neural network model to obtain the probability values ​​of the scene captured image output by the Hoffet neural network model corresponding to the image processing of each sub-processing type executed in the image capture mechanism; A probability judgment component connected to the content identification component, for outputting the sub-processing type corresponding to the probability value with the largest value among the probability values ​​as the sub-processing type executed by the scene captured image in the image capturing mechanism; Wherein, receiving a live captured image from an image capture mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capture mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capture mechanism, wherein the sub-processing type is a subclassification of the preset major processing type, including: the preset major processing type is an enhancement processing type, and each sub-processing type is one of an exponential image enhancement processing, a histogram equalization enhancement processing, a logarithmic image enhancement processing, and an exponential image enhancement processing; Wherein, multiple component parameters of the CMOS sensing component of the image capture mechanism, each visual content of the scene captured image, and binary coded values ​​of each processing type corresponding to various sub-classified image processing are synchronously input into the Hoffet neural network model, and the Hoffet neural network model is executed to obtain the probability values ​​of each sub-processing type of the scene captured image output by the Hoffet neural network model in the image capture mechanism. The multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism; And wherein, the multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism, including: using multiple different parameter acquisition components to respectively acquire the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism.

[0013] Figure 2 It is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to the second embodiment of the present invention.

[0014] and Figure 1 different, Figure 2 The state judgment device based on probability numerical analysis may also include the following components: A radiation amount detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequential conversion device and is connected to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively; Among them, the radiation detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequence conversion device and is respectively connected to the probability judgment component, the content identification component, the data receiving device and the sequence conversion device, including: the radiation detection device is used to respectively realize the on-site measurement of the current radiation amount of the probability judgment component, the content identification component, the data receiving device and the sequence conversion device.

[0015] Figure 3 It is a schematic diagram of the internal structure of a state judgment device based on probability numerical analysis according to the third embodiment of the present invention.

[0016] and Figure 1 different, Figure 3 The state judgment device based on probability numerical analysis may also include the following components: A height detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequential conversion device and is connected to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively; Among them, the height detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequence conversion device and is respectively connected to the probability judgment component, the content identification component, the data receiving device and the sequence conversion device, including: the height detection device is used to respectively realize on-site measurement of the current height of the probability judgment component, the content identification component, the data receiving device and the sequence conversion device.

[0017] Next, the specific structure of the state judgment device based on probability numerical analysis of the present invention will be further described.

[0018] In a state judgment device based on probability numerical analysis according to any embodiment of the present invention: A GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving component and the sequential conversion component to obtain output processing data corresponding to the probability judgment component, the content identification component, the data receiving component and the sequential conversion component respectively.

[0019] In a state judgment device based on probability numerical analysis according to any embodiment of the present invention: The GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively, including: performing non-scaling transformation fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

[0020] In a state judgment device based on probability numerical analysis according to any embodiment of the present invention: Using a GAL device to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively includes: performing median fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

[0021] In a state judgment device based on probability numerical analysis according to any embodiment of the present invention: Using a GAL device to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively includes: performing Gaussian blur processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

[0022] And in the state judgment device based on probability numerical analysis according to any embodiment of the present invention: The GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively, including: performing bilateral filtering fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

[0023] In addition, in the state judgment device based on probability numerical analysis, multiple component parameters of the CMOS sensing component of the image capture mechanism, each visualization content of the on-site captured image, and binary coded numerical values ​​of each processing type corresponding to various sub-classified image processing are synchronously input into the Hoffitt neural network model, and the Hoffitt neural network model is executed to obtain the probability numerical values ​​corresponding to each sub-processing type of image processing performed in the image capture mechanism on the on-site captured image output by the Hoffitt neural network model, including: the probability numerical values ​​corresponding to each sub-processing type of image processing performed in the image capture mechanism on the on-site captured image output by the Hoffitt neural network model are all in binary numerical representation.

[0024] Those skilled in the art should understand that various improvements can be made to the device disclosed in the above invention without departing from the content of the invention. Therefore, the protection scope of the present invention should be determined by the content of the attached claims.

Claims

1. A state judgment device based on probability numerical analysis, characterized in that: The device comprises: A data receiving device, for receiving a live captured image from an image capturing mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capturing mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capturing mechanism, wherein the sub-processing type is a sub-classification of the preset major processing type; A sequential conversion device is used to successively perform multiple learning operations on the Hoffet neural network to obtain the Hoffet neural network after completing the multiple learning operations and output it as a Hoffet neural network model; a content identification component, connected to the data receiving device and the sequential conversion device respectively, for synchronously inputting multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the scene captured image, and binary coded values ​​of various processing types corresponding to various sub-classified image processing into the Hoffet neural network model, and executing the Hoffet neural network model to obtain the probability values ​​of the scene captured image output by the Hoffet neural network model corresponding to the image processing of each sub-processing type executed in the image capture mechanism; A probability judgment component connected to the content identification component, for outputting the sub-processing type corresponding to the probability value with the largest value among the probability values ​​as the sub-processing type executed by the scene captured image in the image capturing mechanism; Wherein, receiving a live captured image from an image capture mechanism, wherein the live captured image is determined to have been subjected to image processing of a preset major processing type in the image capture mechanism but is not determined to have been subjected to image processing of a sub-processing type in the image capture mechanism, wherein the sub-processing type is a subclassification of the preset major processing type, including: the preset major processing type is an enhancement processing type, and each sub-processing type is one of an exponential image enhancement processing, a histogram equalization enhancement processing, a logarithmic image enhancement processing, and an exponential image enhancement processing; Among them, multiple component parameters of the CMOS sensing component of the image capture mechanism, various visual contents of the on-site captured image, and binary coded values ​​of various processing types corresponding to various sub-classified image processing are synchronously input into the Hoffet neural network model, and the Hoffet neural network model is executed to obtain the probability values ​​corresponding to the image processing of each sub-processing type executed in the image capture mechanism on the on-site captured image output by the Hoffet neural network model. The multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism.

2. The state judgment device based on probability numerical analysis according to claim 1, characterized in that: The multiple component parameters of the CMOS sensing component of the image capture mechanism are the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism, including: using multiple different parameter acquisition components to respectively acquire the sensitivity, quantum efficiency and thermal noise amplitude of the CMOS sensing component of the image capture mechanism.

3. The state judgment device based on probability numerical analysis according to claim 2, characterized in that: The device also includes: A radiation amount detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequential conversion device and is connected to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively; Among them, the radiation detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequence conversion device and is respectively connected to the probability judgment component, the content identification component, the data receiving device and the sequence conversion device, including: the radiation detection device is used to respectively realize the on-site measurement of the current radiation amount of the probability judgment component, the content identification component, the data receiving device and the sequence conversion device.

4. The state judgment device based on probability numerical analysis according to claim 2, characterized in that: The device also includes: A height detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequential conversion device and is connected to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively; Among them, the height detection device is arranged near the probability judgment component, the content identification component, the data receiving device and the sequence conversion device and is respectively connected to the probability judgment component, the content identification component, the data receiving device and the sequence conversion device, including: the height detection device is used to respectively realize on-site measurement of the current height of the probability judgment component, the content identification component, the data receiving device and the sequence conversion device.

5. The state judgment device based on probability numerical analysis according to any one of claims 2 to 4, characterized in that: A GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving component and the sequential conversion component to obtain output processing data corresponding to the probability judgment component, the content identification component, the data receiving component and the sequential conversion component respectively.

6. The state judgment device based on probability numerical analysis according to claim 5, characterized in that: The GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively, including: performing non-scaling transformation fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

7. The state judgment device based on probability numerical analysis according to claim 5, characterized in that: Using a GAL device to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively includes: performing median fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

8. The state judgment device based on probability numerical analysis according to claim 5, characterized in that: Using a GAL device to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively includes: performing Gaussian blur processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.

9. The state judgment device based on probability numerical analysis according to claim 5, characterized in that: The GAL device is used to perform image data processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively, including: performing bilateral filtering fuzzy processing on the output data of the probability judgment component, the content identification component, the data receiving device and the sequential conversion device to obtain the output processing data corresponding to the probability judgment component, the content identification component, the data receiving device and the sequential conversion device respectively.