CN111709889B - Image dynamic real-time analysis system and method based on artificial intelligence - Google Patents
Image dynamic real-time analysis system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN111709889B CN111709889B CN202010518285.3A CN202010518285A CN111709889B CN 111709889 B CN111709889 B CN 111709889B CN 202010518285 A CN202010518285 A CN 202010518285A CN 111709889 B CN111709889 B CN 111709889B
- Authority
- CN
- China
- Prior art keywords
- image
- data
- module
- scene
- artificial intelligence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 42
- 238000010223 real-time analysis Methods 0.000 title claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 54
- 230000008676 import Effects 0.000 claims abstract description 23
- 238000012545 processing Methods 0.000 claims abstract description 21
- 238000000605 extraction Methods 0.000 claims abstract description 18
- 238000012800 visualization Methods 0.000 claims abstract description 15
- 238000010276 construction Methods 0.000 claims abstract description 13
- 238000007781 pre-processing Methods 0.000 claims abstract description 13
- 238000003860 storage Methods 0.000 claims description 24
- 230000009466 transformation Effects 0.000 claims description 15
- 238000013528 artificial neural network Methods 0.000 claims description 12
- 238000005516 engineering process Methods 0.000 claims description 12
- 238000003062 neural network model Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 9
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010191 image analysis Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000001186 cumulative effect Effects 0.000 claims description 5
- 230000002708 enhancing effect Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 3
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 210000002364 input neuron Anatomy 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 2
- 239000000284 extract Substances 0.000 abstract description 2
- 230000008569 process Effects 0.000 abstract description 2
- 238000011496 digital image analysis Methods 0.000 description 3
- 241000195493 Cryptophyta Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of image processing, and discloses an image dynamic real-time analysis system and method based on artificial intelligence, wherein the image dynamic real-time analysis method based on artificial intelligence comprises the following steps: the data acquisition or import module acquires related image data; the data preprocessing module processes the acquired data; the image data dynamic analysis module acquires target, theme, scene, content and other characteristic data of the image; the image classification module is used for classifying images; the scene model construction module is used for constructing a scene model; the data judgment module judges the required image data content; the image feature extraction module extracts corresponding data; the data import module imports the extracted data into the constructed scene model; and performing dynamic analysis on the constructed scene model by using visualization equipment to obtain target, theme, scene, content and other characteristic data. The present invention can contribute to grasping feature points included in an image in all directions.
Description
Technical Field
The invention belongs to the technical field of image processing, and particularly relates to an image dynamic real-time analysis system and method based on artificial intelligence.
Background
At present, digital image analysis techniques play an important role in today's society. As the performance of hardware devices has increased, image resolution has become higher and higher. In the civil field, 1080P resolution has been popularized, and in the industrial and scientific research fields, video data with higher resolution is often adopted to achieve the purpose of acquiring more information.
Digital image processing technology has been around for over 60 years, and as electronic computers have advanced to some level, computers have begun to be used to process graphics and image information. The purpose of early image processing was to improve the quality of images, which was targeted at human subjects with the goal of improving human visual effects. After digital images are applied in various industries, digital image processing technology is used for analyzing video data, extracting semantic information from image information, replacing human eyes for observation to a certain extent, and expanding the application range of digital images and information technology. The digital image analysis technology plays a great role in biomedicine, communication engineering, military police, aerospace, engineering, industrial production and the like.
The traditional digital image analysis processing technology mainly carries out conventional processing such as jade algae and feature extraction on an image, and can only carry out single feature extraction on texture features and the like of the image.
Through the above analysis, the problems and defects of the prior art are as follows: the prior art can not carry out comprehensive dynamic analysis processing on the image and can only carry out single feature extraction aiming at the textural features and the like of the image.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides an image dynamic real-time analysis system and method based on artificial intelligence.
The invention is realized in such a way that an image dynamic real-time analysis method based on artificial intelligence comprises the following steps:
acquiring related image data by using a data interface, scanning equipment or an acquisition device through a data acquisition/import module; and converting the acquired color image into a gray image through a data preprocessing module.
Estimating the brightness background of the image by using an average value method and subtracting the brightness background from the gray image to eliminate the influence of exposure and brightness difference; and performing local rank transformation on the image under different parameter conditions by using a local rank operator to obtain positive local rank transformation and negative local rank transformation of the image.
Adding the positive local rank transformation and the negative local rank transformation through a data preprocessing program to obtain a statistical local rank characteristic with continuously changed parameters; and then, on the basis of an image denoising method represented by sparsity, performing primary denoising on the image by taking the statistical local rank characteristic as a constraint condition.
And fourthly, secondarily denoising the image by controlling the difference of the statistical local rank characteristics between the image before denoising and the image after denoising, so as to remove image noise and obtain a denoised clear image.
Step five, constructing a histogram array H' on the basis of the histogram array H of the gray level of the denoised image H; and (5) correspondingly replacing the histogram array H 'with the gray levels in the histogram array H one by one according to the sequence of the gray levels from large to small so as to construct a new histogram array H'.
And step six, performing cumulative summation on the new histogram array H' constructed in the step five to form a new enhanced image through a new gray level, and performing segmentation processing on the enhanced image.
And seventhly, respectively extracting the image data target, the theme, the scene, the content and other multiple different image characteristics by using an image characteristic extraction algorithm based on an artificial intelligence technology through an image data dynamic analysis module and obtaining the segmented image data.
And step eight, processing the data containing the image target, the theme, the scene and the content characteristics and the plurality of segmented image data by using an image processing program to obtain different target, theme, scene and content image processing results.
And step nine, carrying out image fusion on the target, theme, scene and content image processing results obtained in the step eight and the original image data to obtain a plurality of processed sample images.
And step ten, training the artificial neural network model by using the obtained plurality of processed sample image data, and respectively constructing an image target identification analysis model, a theme identification analysis model, a scene identification analysis model and a content identification analysis model.
And step eleven, respectively analyzing the target, the theme, the scene and the content of the image by using the constructed image target identification analysis model, the theme identification analysis model, the scene identification analysis model and the content identification analysis model.
Step twelve, image classification is carried out through an image classification module based on the content characteristics of the images; the image storage module stores the acquired image data and performs classified storage of the processed images according to the image classification result.
Thirteen, constructing a scene model based on the obtained image theme and the scene information through a scene model construction module; and the data judgment module judges the required image data content based on the constructed scene model.
And step fourteen, extracting corresponding data from the acquired image data through an image feature extraction module based on the acquired required image data content.
Step fifteen, importing the extracted data into the constructed scene model through a data importing module; and displaying the constructed scene model by using visualization equipment, and displaying the target, theme, scene, content and other characteristic data obtained by dynamic image analysis.
Further, in the third step, the constraint condition is:
wherein the content of the first and second substances,as a noisy image I i Is counted local rank feature of->After filteringPrimary sharp imageI is an image sequence number, k =0, ± 0.01, ± 0.03.
Further, the statistical local rank characteristic is:
wherein, LRT pk (I i ) For positive local rank transform, LRT nk (I i ) Is a negative local rank transform.
Further, in the fifth step, the constructing a histogram array H' on the basis of the histogram array H of the gray level of the denoised image includes:
counting a histogram array H of the gray level of the denoised image H, and calculating the cumulative sum of the histogram array H; and setting a new gray formula, rounding the new gray to construct an image H1, and counting a histogram array H' of the gray level of the image H1.
Further, in the tenth step, the artificial neural network model comprises an input layer, a hidden layer and an output layer; the method for training the neural network comprises the following steps:
firstly, initializing a neural network model; then, the characteristic vector group data is sent to a neural network model, and calculation is carried out by combining a weight matrix and a bias matrix;
through a back propagation control algorithm, when the accuracy requirement is not met, the number of hidden layers and the network weight are adjusted until the accuracy requirement is met;
and (4) storing neural network parameters including a cyclic neural network model, the number of input neurons and a network weight, and finishing training.
Further, in step thirteen, the method for constructing a scene model based on the obtained image topic and scene information by the scene model construction module includes:
1) Using the image theme and scene data and the raw image data;
2) Carrying out surface support analysis on original image data by using the classification straight lines to obtain a surface to which each pixel belongs in the image, and dividing the image according to the surface to which the pixel belongs to obtain preliminary scene structure information;
3) Constructing an image scene graph based on the obtained image theme and scene data, and generating a final three-dimensional scene structure according to the scene graph;
4) And registering the obtained three-dimensional scene structure into a unified three-dimensional scene by using the characteristics of the single directed line segment.
Further, in step 2), the method for acquiring scene structure information includes:
analyzing the surface to which the pixel belongs according to the surface support characteristic to further obtain preliminary scene structure information;
after a point A and two groups of coplanar parallel lines in the space are projected to an image plane, if the projection position of the point is in the interior of a quadrangle formed by four straight lines in a projection manner, and other non-coplanar projection straight lines do not exist between the projection point and the projection straight lines to block the visibility of the projection point, the projection point is considered to be supported by a corresponding projection surface and reflected in the space, namely the point A is considered to be on the surface formed by the coplanar parallel lines;
and performing surface support analysis on all pixels in the image, namely dividing the image according to the support to obtain preliminary scene structure information.
Another object of the present invention is to provide an artificial intelligence-based real-time image dynamic analysis system applying the artificial intelligence-based real-time image dynamic analysis method, the artificial intelligence-based real-time image dynamic analysis system comprising:
the data acquisition/import module is connected with the main control module and is used for acquiring related image data by utilizing a data interface, scanning equipment or an acquisition device;
the data preprocessing module is connected with the main control module and used for denoising, enhancing and segmenting the acquired image data through a data preprocessing program;
the main control module is connected with the data acquisition/import module, the data preprocessing module, the image data dynamic analysis module, the image classification module, the image storage module, the scene model construction module, the data judgment module, the image feature extraction module, the data import module and the visualization module and is used for controlling the normal operation of each module of the artificial intelligence based image dynamic real-time analysis system through the main controller;
the image data dynamic analysis module is connected with the main control module and is used for extracting, identifying and analyzing image characteristics based on an artificial intelligence technology to obtain target, theme, scene, content and other characteristic data of the image;
the image classification module is connected with the main control module and is used for classifying the images based on the content characteristics of the images;
the image storage module is connected with the main control module and is used for storing the acquired image data and simultaneously performing classified storage on the processed images according to the image classification result;
the scene model building module is connected with the main control module and used for building a scene model based on the obtained image theme and the scene information;
the data judgment module is connected with the main control module and used for judging the required image data content based on the constructed scene model;
the image characteristic extraction module is connected with the main control module and used for extracting corresponding data from the acquired image data based on the acquired required image data content;
the data import module is connected with the main control module and used for importing the extracted data into the constructed scene model;
and the visualization module is connected with the main control module and used for displaying the constructed scene model by using the visualization equipment and displaying the target, theme, scene, content and other characteristic data obtained by dynamic image analysis.
It is another object of the present invention to provide a computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface to implement the artificial intelligence based image dynamic real-time analysis method when executed on an electronic device.
Another object of the present invention is to provide a computer-readable storage medium storing instructions which, when executed on a computer, cause the computer to execute the artificial intelligence based image dynamic real-time analysis method.
By combining all the technical schemes, the invention has the advantages and positive effects that: according to the method, the data preprocessing module utilizes the statistical local rank characteristic as a constraint condition of noise, and the statistical local rank characteristic is used as the constraint condition to perform primary denoising on the image, so that the removal of noise in a smooth region of the image and the retention of image edge and detail information are realized; subsequently, the image with the retained edge detail information is subjected to secondary denoising by controlling the difference of the statistical local rank characteristics between the image before denoising and the image after denoising, so that a better denoising effect is achieved, a denoised image with higher quality can be obtained, and the reliability of subsequent image processing and analysis is effectively guaranteed. Meanwhile, the image data dynamic analysis module can extract targets, themes, scenes, contents and other features of the acquired image data based on an artificial intelligence technology, and perform classification processing and storage according to the extracted data, so that the method is beneficial to comprehensively mastering the feature points contained in the image, and meanwhile, the method can perform scene construction according to the extracted contents, effectively restore the three-dimensional scene data of the image, simulate various conditions and help to comprehensively master the image scene.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments of the present application will be briefly described below, and it is obvious that the drawings described below are only some embodiments of the present application, and it is obvious for those skilled in the art that other drawings can be obtained from the drawings without creative efforts.
Fig. 1 is a flowchart of an artificial intelligence-based image dynamic real-time analysis method according to an embodiment of the present invention.
FIG. 2 is a schematic structural diagram of an artificial intelligence-based image dynamic real-time analysis system provided by an embodiment of the present invention;
in the figure: 1. a data acquisition/import module; 2. a data preprocessing module; 3. a main control module; 4. an image data dynamic analysis module; 5. an image classification module; 6. an image storage module; 7. a scene model construction module; 8. a data judgment module; 9. an image feature extraction module; 10. a data import module; 11. and a visualization module.
Fig. 3 is a flowchart of a method for denoising acquired image data by a data preprocessing program according to an embodiment of the present invention.
Fig. 4 is a flowchart of a method for extracting, recognizing and analyzing image features by an image data dynamic analysis module according to an embodiment of the present invention.
Fig. 5 is a flowchart of a method for constructing a scene model based on an obtained image topic and scene information by a scene model construction module according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Aiming at the problems in the prior art, the invention provides an image dynamic real-time analysis system and method based on artificial intelligence, and the invention is described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the method for real-time image dynamic analysis based on artificial intelligence provided in the embodiment of the present invention includes the following steps:
s101, acquiring related image data by using a data interface, scanning equipment or an acquisition device through a data acquisition or import module; and denoising, enhancing and segmenting the acquired data through a data preprocessing module.
S102, extracting, identifying and analyzing image features based on an artificial intelligence technology through an image data dynamic analysis module, and acquiring target, theme, scene, content and other feature data of the image.
S103, classifying the images based on the content characteristics of the images through an image classification module; and storing the acquired image data through an image storage module, and simultaneously performing classified storage on the processed images according to image classification results.
S104, constructing a scene model based on the obtained image theme and scene information through a scene model constructing module; and judging the required image data content based on the constructed scene model through a data judgment module.
And S105, extracting corresponding data from the acquired image data through an image feature extraction module based on the acquired required image data content.
S106, importing the extracted data into the constructed scene model through a data import module; and displaying the constructed scene model by using visualization equipment, and displaying the target, theme, scene, content and other characteristic data obtained by dynamic image analysis.
As shown in fig. 2, the image dynamic real-time analysis system based on artificial intelligence provided by the embodiment of the present invention includes: the system comprises a data acquisition/import module 1, a data preprocessing module 2, a main control module 3, an image data dynamic analysis module 4, an image classification module 5, an image storage module 6, a scene model construction module 7, a data judgment module 8, an image feature extraction module 9, a data import module 10 and a visualization module 11.
The data acquisition/import module 1 is connected with the main control module 3 and is used for acquiring related image data by using a data interface, scanning equipment or an acquisition device;
the data preprocessing module 2 is connected with the main control module 3 and is used for denoising, enhancing and segmenting the acquired image data through a data preprocessing program;
the main control module 3 is connected with the data acquisition/import module 1, the data preprocessing module 2, the image data dynamic analysis module 4, the image classification module 5, the image storage module 6, the scene model construction module 7, the data judgment module 8, the image feature extraction module 9, the data import module 10 and the visualization module 11, and is used for controlling the normal operation of each module of the artificial intelligence-based image dynamic real-time analysis system through the main controller;
the image data dynamic analysis module 4 is connected with the main control module 3 and is used for extracting, identifying and analyzing image features based on an artificial intelligence technology to obtain target, theme, scene, content and other feature data of the image;
the image classification module 5 is connected with the main control module 3 and is used for classifying images based on the content characteristics of the images;
the image storage module 6 is connected with the main control module 3 and used for storing the acquired image data and simultaneously performing classified storage on the processed images according to the image classification result;
the scene model building module 7 is connected with the main control module 3 and used for building a scene model based on the obtained image theme and the scene information;
the data judgment module 8 is connected with the main control module 3 and is used for judging the required image data content based on the constructed scene model;
the image feature extraction module 9 is connected with the main control module 3 and used for extracting corresponding data from the acquired image data based on the acquired required image data content;
the data import module 10 is connected with the main control module 3 and is used for importing the extracted data into the constructed scene model;
and the visualization module 11 is connected with the main control module 3, and is used for displaying the constructed scene model by using visualization equipment and displaying the target, theme, scene, content and other feature data obtained by dynamic image analysis.
The invention is further described with reference to specific examples.
Example 1
The method for dynamically analyzing images in real time based on artificial intelligence provided by the embodiment of the present invention is shown in fig. 1, and as a preferred embodiment, as shown in fig. 3, the method for denoising acquired image data through a data preprocessing program provided by the embodiment of the present invention includes:
s201, converting the color image into a gray image, estimating the brightness background of the image by using an average value method, and subtracting the brightness background from the gray image to eliminate the influence of exposure and brightness difference.
S202, local rank transformation is carried out on the image under different parameter conditions by using a local rank operator, and positive local rank transformation and negative local rank transformation of the image are obtained.
S203, adding the positive local rank transformation and the negative local rank transformation to obtain a statistical local rank characteristic with continuously changed parameters; and then, on the basis of an image denoising method of sparse representation, performing primary denoising on the image by taking the statistical local rank characteristic as a constraint condition.
S204, secondarily denoising the image by controlling the difference of the statistical local rank characteristics between the image before and after denoising, so as to remove image noise and obtain a denoised clear image.
The constraint conditions provided by the embodiment of the invention are as follows:
wherein the content of the first and second substances,as a noisy image I i Is counted local rank feature of->For filtered primary sharp imagesI is an image sequence number, k =0, ± 0.01, ± 0.03.
The statistical local rank characteristic provided by the embodiment of the invention is as follows:
wherein, LRT pk (I i ) Is a positive local rankConversion, LRT nk (I i ) Is a negative local rank transform.
The image enhancement method provided by the embodiment of the invention comprises the following steps:
constructing a histogram array H' on the basis of a histogram array H of the gray level of the denoised image H;
replacing the histogram array H' and each gray level in the histogram array H in a one-to-one correspondence mode according to the sequence of the gray levels from large to small so as to construct a new histogram array H ";
the new histogram array H "is cumulatively summed to form a new enhanced image with the new gray levels.
The method for constructing the histogram array H' on the basis of the histogram array H of the gray level of the denoised image comprises the following steps:
counting a histogram array H of the gray level of the denoised image H, and calculating the cumulative sum of the histogram array H; and setting a new gray formula, rounding the new gray to construct an image H1, and counting a histogram array H' of the gray level of the image H1.
Example 2
As shown in fig. 1 and fig. 4, which are preferred embodiments of the method for dynamically analyzing an image based on artificial intelligence according to an embodiment of the present invention, a method for extracting, identifying, and analyzing image features by using an image data dynamic analysis module according to an embodiment of the present invention includes:
s301, a plurality of different image characteristics of an image data target, a theme, a scene and content are respectively extracted through an image characteristic extraction algorithm by an image data dynamic analysis module, and the segmented image data is obtained.
S302, processing data containing image target, theme, scene and content characteristics and a plurality of segmented image data by using an image processing program respectively to obtain different target, theme, scene and content image processing results.
And S303, carrying out image fusion on the obtained target, theme, scene and content image processing results and the original image data to obtain a plurality of processed sample images.
S304, training the artificial neural network model by using the obtained processed sample image data, and respectively constructing an image target identification analysis model, a theme identification analysis model, a scene identification analysis model and a content identification analysis model.
S305, respectively analyzing the target, the theme, the scene and the content of the image by using the constructed image target identification analysis model, the theme identification analysis model, the scene identification analysis model and the content identification analysis model.
The artificial neural network model provided by the embodiment of the invention comprises an input layer, a hidden layer and an output layer; the method for training the neural network comprises the following steps:
firstly, initializing a neural network model; then, the characteristic vector group data is sent to a neural network model, and calculation is carried out by combining a weight matrix and a bias matrix;
through a back propagation control algorithm, when the accuracy requirement is not met, the number of hidden layers and the network weight are adjusted until the accuracy requirement is met;
and (4) storing neural network parameters including a cyclic neural network model, the number of input neurons and a network weight, and finishing training.
Example 3
The method for dynamically analyzing images in real time based on artificial intelligence provided by the embodiment of the invention is shown in fig. 1, and as a preferred embodiment, as shown in fig. 5, the method for constructing a scene model based on the obtained image subjects and scene information through a scene model construction module provided by the embodiment of the invention comprises the following steps:
s401, using the image theme and scene data and the raw image data.
S402, carrying out surface support analysis on the original image data by using the classification straight lines to obtain the surface to which each pixel belongs in the image, and dividing the image according to the surface to which the pixel belongs to obtain preliminary scene structure information.
And S403, constructing an image scene graph based on the obtained image theme and the scene data, and generating a final three-dimensional scene structure according to the scene graph.
And S404, registering the obtained three-dimensional scene structure into a unified three-dimensional scene by using the characteristics of the single directed line segment.
The method for acquiring the scene structure information provided by the embodiment of the invention comprises the following steps:
analyzing the surface to which the pixel belongs according to the surface support characteristic to further obtain preliminary scene structure information;
after a point A and two groups of coplanar parallel lines in the space are projected to an image plane, if the projection position of the point is in the interior of a quadrangle formed by four straight lines in a projection manner, and other non-coplanar projection straight lines do not exist between the projection point and the projection straight lines to block the visibility of the projection point, the projection point is considered to be supported by a corresponding projection surface and reflected in the space, namely the point A is considered to be on the surface formed by the coplanar parallel lines;
and performing surface support analysis on all pixels in the image, namely dividing the image according to the support to obtain preliminary scene structure information.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When used in whole or in part, can be implemented in a computer program product that includes one or more computer instructions. When loaded or executed on a computer, cause the flow or functions according to embodiments of the invention to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wire (e.g., coaxial cable, fiber optic, digital Subscriber Line (DSL), or wireless (e.g., infrared, wireless, microwave, etc.)). The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that includes one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), among others.
The above description is only for the purpose of illustrating the embodiments of the present invention, and the scope of the present invention should not be limited thereto, and any modifications, equivalents and improvements made by those skilled in the art within the technical scope of the present invention as disclosed in the present invention should be covered by the scope of the present invention.
Claims (10)
1. An image dynamic real-time analysis method based on artificial intelligence is characterized by comprising the following steps:
acquiring related image data by using a data interface, scanning equipment or an acquisition device through a data acquisition/import module; converting the obtained color image into a gray image through a data preprocessing module;
estimating the brightness background of the image by using an average value method and subtracting the brightness background from the gray image to eliminate the influence of exposure and brightness difference; performing local rank transformation on the image under different parameter conditions by using a local rank operator to obtain positive local rank transformation and negative local rank transformation of the image;
thirdly, adding the positive local rank transformation and the negative local rank transformation through a data preprocessing program to obtain a statistical local rank characteristic with continuously changed parameters; then, on the basis of an image denoising method of sparse representation, performing primary denoising on the image by taking the statistical local rank characteristic as a constraint condition;
fourthly, secondarily denoising the image by controlling the difference of the statistical local rank characteristics between the image before denoising and the image after denoising, so as to remove image noise and obtain a denoised clear image;
step five, constructing a histogram array H' on the basis of the histogram array H of the gray level of the denoised image H; replacing the histogram array H' and each gray level in the histogram array H in a one-to-one correspondence mode according to the sequence of the gray levels from large to small so as to construct a new histogram array H ";
step six, carrying out cumulative summation on the new histogram array H' constructed in the step five so as to form a new enhanced image through a new gray level, and carrying out segmentation processing on the enhanced image;
step seven, respectively extracting image data targets, themes, scenes, contents and other multiple different image characteristics by an image characteristic extraction algorithm based on an artificial intelligence technology through an image data dynamic analysis module and obtaining segmented image data;
processing data containing image target, theme, scene and content characteristics and a plurality of segmented image data by using an image processing program to obtain different target, theme, scene and content image processing results;
step nine, carrying out image fusion on the target, theme, scene and content image processing results obtained in the step eight and the original image data to obtain a plurality of processed sample images;
step ten, training an artificial neural network model by using the obtained plurality of processed sample image data, and respectively constructing an image target identification analysis model, a theme identification analysis model, a scene identification analysis model and a content identification analysis model;
eleven, respectively analyzing the target, the theme, the scene and the content of the image by using the constructed image target identification analysis model, the theme identification analysis model, the scene identification analysis model and the content identification analysis model;
step twelve, image classification is carried out through an image classification module based on the content characteristics of the images; the image storage module stores the acquired image data and performs classified storage of the processed images according to image classification results;
thirteen, constructing a scene model based on the obtained image theme and the scene information through a scene model construction module; the data judgment module judges the required image data content based on the constructed scene model;
fourteen, extracting corresponding data from the acquired image data through an image feature extraction module based on the acquired required image data content;
step fifteen, importing the extracted data into the constructed scene model through a data importing module; and displaying the constructed scene model by using a visualization device, and displaying the target, theme, scene, content and other characteristic data obtained by dynamic image analysis.
2. The method for dynamically analyzing images in real time based on artificial intelligence as claimed in claim 1, wherein in step three, the constraint conditions are:
4. The method for real-time analysis of image dynamics based on artificial intelligence as claimed in claim 1, wherein in the fifth step, said constructing the histogram array H' based on the histogram array H of the gray levels of the de-noised image comprises:
counting a histogram array H of the gray level of the denoised image H, and calculating the cumulative sum of the histogram array H; and setting a new gray formula, rounding the new gray to construct an image H1, and counting a histogram array H' of the gray level of the image H1.
5. The method for dynamically analyzing images in real time based on artificial intelligence as claimed in claim 1, wherein in step ten, the artificial neural network model comprises an input layer, a hidden layer and an output layer; the method for training the neural network comprises the following steps:
firstly, initializing a neural network model; then, the characteristic vector group data is sent to a neural network model, and calculation is carried out by combining a weight matrix and a bias matrix;
through a back propagation control algorithm, when the accuracy requirement is not met, the number of hidden layers and the network weight are adjusted until the accuracy requirement is met;
and (4) storing neural network parameters including a cyclic neural network model, the number of input neurons and a network weight, and finishing training.
6. The method for dynamically analyzing images in real time based on artificial intelligence of claim 1, wherein in step thirteen, the method for constructing the scene model based on the obtained image subjects and scene information by the scene model construction module comprises:
1) Using the image theme and scene data and the raw image data;
2) Carrying out surface support analysis on original image data by using the classification straight lines to obtain a surface to which each pixel belongs in the image, and dividing the image according to the surface to which the pixel belongs to obtain preliminary scene structure information;
3) Constructing an image scene graph based on the obtained image theme and scene data, and generating a final three-dimensional scene structure according to the scene graph;
4) And registering the obtained three-dimensional scene structure into a unified three-dimensional scene by using the characteristics of the single directed line segment.
7. The method for dynamically analyzing images in real time based on artificial intelligence as claimed in claim 6, wherein in step 2), the method for obtaining scene structure information comprises:
analyzing the surface to which the pixel belongs according to the surface support characteristic to further obtain preliminary scene structure information;
after a point A and two groups of coplanar parallel lines in the space are projected to an image plane, if the projection position of the point is in the interior of a quadrangle formed by four straight lines in a projection manner, and other non-coplanar projection straight lines do not exist between the projection point and the projection straight lines to block the visibility of the projection point, the projection point is considered to be supported by a corresponding projection surface and reflected in the space, namely the point A is considered to be on the surface formed by the coplanar parallel lines;
and performing surface support analysis on all pixels in the image, namely dividing the image according to the support to obtain preliminary scene structure information.
8. An artificial intelligence based image dynamic real-time analysis system applying the artificial intelligence based image dynamic real-time analysis method according to any one of claims 1 to 7, wherein the artificial intelligence based image dynamic real-time analysis system comprises:
the data acquisition/import module is connected with the main control module and is used for acquiring related image data by utilizing a data interface, scanning equipment or an acquisition device;
the data preprocessing module is connected with the main control module and used for denoising, enhancing and segmenting the acquired image data through a data preprocessing program;
the main control module is connected with the data acquisition/import module, the data preprocessing module, the image data dynamic analysis module, the image classification module, the image storage module, the scene model construction module, the data judgment module, the image feature extraction module, the data import module and the visualization module and is used for controlling the normal operation of each module of the artificial intelligence based image dynamic real-time analysis system through the main controller;
the image data dynamic analysis module is connected with the main control module and is used for extracting, identifying and analyzing image characteristics based on an artificial intelligence technology to obtain target, theme, scene, content and other characteristic data of the image;
the image classification module is connected with the main control module and is used for classifying the images based on the content characteristics of the images;
the image storage module is connected with the main control module and used for storing the acquired image data and simultaneously performing classified storage on the processed images according to the image classification result;
the scene model building module is connected with the main control module and used for building a scene model based on the obtained image theme and the scene information;
the data judgment module is connected with the main control module and used for judging the required image data content based on the constructed scene model;
the image characteristic extraction module is connected with the main control module and used for extracting corresponding data from the acquired image data based on the acquired required image data content;
the data import module is connected with the main control module and used for importing the extracted data into the constructed scene model;
and the visualization module is connected with the main control module and used for displaying the constructed scene model by utilizing visualization equipment and displaying the target, theme, scene, content and other characteristic data obtained by dynamic image analysis.
9. A computer program product stored on a computer readable medium, comprising a computer readable program for providing a user input interface for implementing the artificial intelligence based method for dynamic real-time analysis of images as claimed in any one of claims 1 to 7 when executed on an electronic device.
10. A computer-readable storage medium storing instructions which, when executed on a computer, cause the computer to perform the artificial intelligence based image dynamic real-time analysis method according to any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010518285.3A CN111709889B (en) | 2020-06-09 | 2020-06-09 | Image dynamic real-time analysis system and method based on artificial intelligence |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010518285.3A CN111709889B (en) | 2020-06-09 | 2020-06-09 | Image dynamic real-time analysis system and method based on artificial intelligence |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111709889A CN111709889A (en) | 2020-09-25 |
CN111709889B true CN111709889B (en) | 2023-04-07 |
Family
ID=72539669
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010518285.3A Active CN111709889B (en) | 2020-06-09 | 2020-06-09 | Image dynamic real-time analysis system and method based on artificial intelligence |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111709889B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106096613A (en) * | 2016-05-31 | 2016-11-09 | 哈尔滨工业大学深圳研究生院 | Image multi-target detection method and device based on corner feature |
CN107808132A (en) * | 2017-10-23 | 2018-03-16 | 重庆邮电大学 | A kind of scene image classification method for merging topic model |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20190138786A1 (en) * | 2017-06-06 | 2019-05-09 | Sightline Innovation Inc. | System and method for identification and classification of objects |
-
2020
- 2020-06-09 CN CN202010518285.3A patent/CN111709889B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106096613A (en) * | 2016-05-31 | 2016-11-09 | 哈尔滨工业大学深圳研究生院 | Image multi-target detection method and device based on corner feature |
CN107808132A (en) * | 2017-10-23 | 2018-03-16 | 重庆邮电大学 | A kind of scene image classification method for merging topic model |
Non-Patent Citations (1)
Title |
---|
"一种基于增强模糊分类器的快速图像分类方法";徐小龙 等;《电子技术》;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111709889A (en) | 2020-09-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yang et al. | Underwater image enhancement based on conditional generative adversarial network | |
US11830230B2 (en) | Living body detection method based on facial recognition, and electronic device and storage medium | |
CN109035172B (en) | Non-local mean ultrasonic image denoising method based on deep learning | |
CN109657612B (en) | Quality sorting system based on facial image features and application method thereof | |
Khan et al. | Localization of radiance transformation for image dehazing in wavelet domain | |
CN110097522B (en) | Single outdoor image defogging method based on multi-scale convolution neural network | |
Feng et al. | URNet: A U-Net based residual network for image dehazing | |
CN115082966B (en) | Pedestrian re-recognition model training method, pedestrian re-recognition method, device and equipment | |
Gao et al. | Background subtraction via 3D convolutional neural networks | |
Zhao et al. | A multi-scale U-shaped attention network-based GAN method for single image dehazing | |
Yan et al. | Tongue image texture classification based on image inpainting and convolutional neural network | |
CN113158860B (en) | Deep learning-based multi-dimensional output face quality evaluation method and electronic equipment | |
Qu et al. | Umle: Unsupervised multi-discriminator network for low light enhancement | |
CN111881803B (en) | Face recognition method based on improved YOLOv3 | |
CN110136164B (en) | Method for removing dynamic background based on online transmission transformation and low-rank sparse matrix decomposition | |
CN111709889B (en) | Image dynamic real-time analysis system and method based on artificial intelligence | |
CN115294162B (en) | Target identification method, device, equipment and storage medium | |
Andrei et al. | Unsupervised Machine Learning Algorithms Used in Deforested Areas Monitoring | |
CN113627342B (en) | Method, system, equipment and storage medium for video depth feature extraction optimization | |
CN116977683A (en) | Object recognition method, apparatus, computer device, storage medium, and program product | |
Avazov et al. | Automatic moving shadow detection and removal method for smart city environments | |
Zhang | (Retracted) Computer-aided three-dimensional animation design and enhancement based on spatial variation and convolution algorithm | |
CN114283087A (en) | Image denoising method and related equipment | |
CN113177879A (en) | Image processing method, image processing apparatus, electronic device, and storage medium | |
Zhang et al. | A modified image processing method for deblurring based on GAN networks |
Legal Events
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 |