CN116205817A - Data content complexity targeted detection system - Google Patents
Data content complexity targeted detection system Download PDFInfo
- Publication number
- CN116205817A CN116205817A CN202310206982.9A CN202310206982A CN116205817A CN 116205817 A CN116205817 A CN 116205817A CN 202310206982 A CN202310206982 A CN 202310206982A CN 116205817 A CN116205817 A CN 116205817A
- Authority
- CN
- China
- Prior art keywords
- picture
- processed
- complexity
- training
- objects
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 40
- 238000012549 training Methods 0.000 claims abstract description 62
- 230000008859 change Effects 0.000 claims abstract description 47
- 238000012545 processing Methods 0.000 claims abstract description 46
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 42
- 230000007246 mechanism Effects 0.000 claims abstract description 27
- 238000004458 analytical method Methods 0.000 claims abstract description 25
- 238000013075 data extraction Methods 0.000 claims abstract description 11
- 238000005259 measurement Methods 0.000 claims abstract description 11
- 230000000875 corresponding effect Effects 0.000 claims description 33
- 238000000034 method Methods 0.000 claims description 17
- 238000013528 artificial neural network Methods 0.000 claims description 7
- 238000004088 simulation Methods 0.000 claims description 5
- 230000002596 correlated effect Effects 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 3
- 238000009825 accumulation Methods 0.000 claims description 2
- 238000001914 filtration Methods 0.000 description 16
- 238000001228 spectrum Methods 0.000 description 8
- 230000003044 adaptive effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 239000011159 matrix material Substances 0.000 description 6
- 238000010183 spectrum analysis Methods 0.000 description 5
- 230000010287 polarization Effects 0.000 description 3
- 230000004927 fusion Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 230000009466 transformation 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
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- 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
-
- 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]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a data content complexity targeted detection system. The system comprises: the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device; the data extraction mechanism comprises a definition detection device and a contrast measurement device; a prediction processing device, configured to predict the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition, and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by using a convolutional neural network after completing the training for a preset number of times; and the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity based on the number of objects in the picture to be processed. By the system, different prediction models can be provided for each picture to be processed so as to finish targeted judgment of the content complexity of the picture, thereby improving the intelligent level of picture signal processing.
Description
Technical Field
The invention relates to the field of signal processing, in particular to a data content complexity targeted detection system.
Background
The most basic contents of signal processing are transformation, filtering, modulation, demodulation, detection, spectrum analysis, estimation and the like. Transforms such as type fourier transforms, sine transforms, cosine transforms, walsh transforms, etc.; the filtering comprises high-pass filtering, low-pass filtering, band-pass filtering, wiener filtering, kalman filtering, linear filtering, nonlinear filtering, self-adaptive filtering and the like; spectral analysis aspects include analysis of deterministic signals and analysis of random signals, the most common of which is generally studied is analysis of random signals, also known as statistical signal analysis or estimation, which in turn is typically linear and nonlinear spectral estimation; spectrum estimation has cyclic graph estimation, maximum entropy spectrum estimation, etc.
Along with the complexity of signal types, when the signal to be analyzed cannot meet the conditions of Gaussian distribution, non-minimum phase and the like, a high-order spectrum analysis method is provided. High-order spectral analysis can provide phase information, non-gaussian information and nonlinear information of signals, for example, invention publication CN105137498A is a system and method for detecting and identifying an underground target based on feature fusion, and the method comprises detecting and acquiring echo signals of a ground penetrating radar; preprocessing the collected echo signals; performing enhancement processing on the preprocessed echo signals; extracting time domain features and wavelet packet energy spectrum features of the echo signals subjected to enhancement processing, and carrying out Welch power spectrum estimation and higher-order spectrum analysis on the echo signals subjected to enhancement processing to respectively obtain Welch power spectrum and Gao Jiepu; fusing the time domain characteristics, the wavelet packet energy spectrum characteristics, the Welch power spectrum and the high-order spectrum of the echo signals; performing target recognition on the fused four signal characteristics through a wavelet neural network; outputting a target recognition result; the invention carries out comprehensive detection and identification of the underground target based on multi-feature information fusion, can effectively realize identification of underground targets with different shapes and materials, and has high identification accuracy. The adaptive filtering and equalization are also a large field of application research, for example, the invention publication CN112865754a is an adaptive filtering method based on a Gibbs sampler, and firstly, a process noise mean vector and variance matrix, an observation noise mean vector and variance matrix in a linear state space model are regarded as unknown random variables, and prior distribution is modeled as gaussian-inverse Wishart distribution; for each time epoch, under the framework of the Gibbs sampler, simultaneously performing iterative sampling on the unknown mean vector, the variance matrix, the current time epoch and the system state of the last time epoch; after multiple iterations are carried out, selecting an average value of iteration period sampling after reaching a steady state as a final state estimation value, a noise mean value vector and a variance matrix estimation value; the invention can still obtain better state estimation results when the setting errors of the model noise mean vector and the variance matrix are larger, and can accurately estimate the unknown noise mean vector and the unknown noise variance matrix. The adaptive filtering comprises transversal LMS adaptive filtering, lattice adaptive filtering, adaptive cancellation filtering, adaptive equalization and the like. In addition, for array signals, and array signal processing, for example, the invention publication CN115118320a is a method, apparatus, device, and medium for forming a beam of a nested polarization sensitive antenna array, the method comprising: constructing a virtual polarization sensitive antenna uniform array with higher freedom according to a received signal tensor model of the nested polarization sensitive antenna array, wherein the virtual array has a four-dimensional data structure; carrying out sub-beam formation on the front three-dimension in the virtual array data by means of tensor operation to obtain the optimal weight vector of the corresponding sub-beam former; according to the optimal weight vector, obtaining an optimal weight tensor of virtual array beam forming through tensor outer product, and further obtaining a final beam forming output result; the scheme can effectively solve the problem that the interference signal is difficult to be restrained by the sensor array with the same time scale when the arrival angle of the desired signal is similar to that of the interference signal, and meanwhile, the interference suppression capability is improved due to the higher (virtual) array degree of freedom. .
For the picture signal, when the picture processing is executed, the pertinence of the subsequent picture processing needs to be improved according to the picture content complexity, for example, the higher the picture content complexity is, the higher the adopted contrast improvement algorithm complexity needs to be improved accordingly, however, the picture content complexity needs to be judged manually, and in the prior art, the electronic analysis processing of the picture content complexity cannot be directly completed according to the picture signal data, so that the automation level of the picture signal processing is low, and the existing requirements on the speed and efficiency of the picture signal processing cannot be met.
Disclosure of Invention
In order to solve the technical problems in the related art, the invention provides a data content complexity targeted detection system, which can select a plurality of training data which are the same as the number of pixels of a picture to be processed to perform multiple training processing on a convolutional neural network, wherein the training times are positively correlated with the number of the pixels of the picture to be processed, and the convolutional neural network after the multiple training processing is adopted to predict the number of objects in the picture to be processed based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel in the picture to be processed, so as to provide key information for determining the content complexity of the picture to be processed.
According to an aspect of the present invention, the data content complexity targeted detection system includes:
the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device, wherein the data input device is used for receiving a picture to be processed, the content acquisition device is connected with the data input device and used for acquiring each pixel value corresponding to each pixel point in the picture to be processed, and the shading analysis device is connected with the data input device and used for analyzing the shading change level of the picture to be processed;
the data extraction mechanism is connected with the content acquisition mechanism and comprises a definition detection device and a contrast measurement device, wherein the definition detection device is connected with the data input device and is used for detecting the picture definition of the picture to be processed, and the contrast measurement device is connected with the data input device and is used for measuring the picture contrast of the picture to be processed;
a prediction processing device, which is respectively connected with the content acquisition mechanism and the data extraction mechanism, and is used for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture by adopting a convolution neural network after completing the training of a preset number for multiple times;
the network reconstruction device is connected with the prediction processing device and is used for sending the convolutional neural network after the preset number of training for a plurality of times to the prediction processing device to execute prediction processing;
the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed;
wherein determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed comprises: the picture complexity of the picture to be processed is positively correlated with the number of objects in the picture to be processed;
the smaller the number of each pixel point in the picture to be processed is, the smaller the value of the preset number is.
The data content complexity targeted detection system provided by the invention is intelligent in operation and simple and convenient to operate. Because different prediction models can be provided for each picture to be processed to finish the targeted judgment of the content complexity, the intelligent level of picture signal processing is improved.
Brief description of the drawings
Numerous advantages of the present invention may be better understood by those skilled in the art by reference to the accompanying figures.
Fig. 1 is a schematic structural diagram of a data content complexity targeted detection system according to a first embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a data content complexity targeted detection system according to a second embodiment of the present invention.
Fig. 3 is a schematic structural diagram of a data content complexity targeted detection system according to a third embodiment of the present invention.
Detailed Description
First embodiment
Fig. 1 is a schematic structural diagram of a data content complexity targeted detection system according to a first embodiment of the present invention, the system including:
the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device, wherein the data input device is used for receiving a picture to be processed, the content acquisition device is connected with the data input device and used for acquiring each pixel value corresponding to each pixel point in the picture to be processed, and the shading analysis device is connected with the data input device and used for analyzing the shading change level of the picture to be processed;
the shading analysis device is connected with the data input device and is used for analyzing the shading change level of the picture to be processed, wherein the shading change level reflects the distance degree of the signal shading change of one picture;
the data extraction mechanism is connected with the content acquisition mechanism and comprises a definition detection device and a contrast measurement device, wherein the definition detection device is connected with the data input device and is used for detecting the picture definition of the picture to be processed, and the contrast measurement device is connected with the data input device and is used for measuring the picture contrast of the picture to be processed;
a prediction processing device, which is respectively connected with the content acquisition mechanism and the data extraction mechanism, and is used for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture by adopting a convolution neural network after completing the training of a preset number for multiple times;
the network reconstruction device is connected with the prediction processing device and is used for sending the convolutional neural network after the preset number of training for a plurality of times to the prediction processing device to execute prediction processing;
the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed;
wherein determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed comprises: the picture complexity of the picture to be processed is positively correlated with the number of objects in the picture to be processed;
the smaller the number of each pixel point in the picture to be processed is, the smaller the value of the preset number is;
for example, the smaller the number of each pixel point in the to-be-processed picture is, the smaller the value of the preset number is: the value of the preset number is in direct proportion to the number of each pixel point in the picture to be processed;
wherein, sending the convolutional neural network after the completion of the preset number of training for a plurality of times to the prediction processing device to execute the prediction processing comprises: the single training of the convolutional neural network is completed by adopting a certain training picture with the known number of objects and the same number of pixels as the number of pixels of the picture to be processed;
the step of completing the single training of the convolutional neural network by adopting a training picture with the number of known objects and the same number of pixels as the number of pixels of the picture to be processed comprises the following steps: taking the number of the existing objects of a certain training picture as single output information of the convolutional neural network, taking each pixel value corresponding to each pixel point in the certain training picture, the brightness change level of the certain training picture and the picture definition of the certain training picture as each input information of the convolutional neural network, and completing single training of the convolutional neural network;
the method for completing the single training of the convolutional neural network comprises the following steps of taking the number of the existing objects of a certain training picture as single output information of the convolutional neural network, taking each pixel value corresponding to each pixel point in the certain training picture, the brightness change level of the certain training picture and the picture definition of the certain training picture as each input information of the convolutional neural network, wherein the single training of the convolutional neural network comprises the following steps: and completing the preset number of multiple training on the convolutional neural network by adopting the preset number of multiple different training pictures.
Second embodiment
Fig. 2 is a schematic structural diagram of a data content complexity targeted detection system according to a second embodiment of the present invention, the structure includes:
the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device, wherein the data input device is used for receiving a picture to be processed, the content acquisition device is connected with the data input device and used for acquiring each pixel value corresponding to each pixel point in the picture to be processed, and the shading analysis device is connected with the data input device and used for analyzing the shading change level of the picture to be processed;
the data extraction mechanism is connected with the content acquisition mechanism and comprises a definition detection device and a contrast measurement device, wherein the definition detection device is connected with the data input device and is used for detecting the picture definition of the picture to be processed, and the contrast measurement device is connected with the data input device and is used for measuring the picture contrast of the picture to be processed;
a prediction processing device, which is respectively connected with the content acquisition mechanism and the data extraction mechanism, and is used for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture by adopting a convolution neural network after completing the training of a preset number for multiple times;
the network reconstruction device is connected with the prediction processing device and is used for sending the convolutional neural network after the preset number of training for a plurality of times to the prediction processing device to execute prediction processing;
the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed;
the cloud storage network element is connected with the complexity judging equipment in a network manner and is used for storing the forward association relation between the picture complexity and the number of objects in the picture;
illustratively, a big data storage network element or a blockchain storage network element can be adopted to replace the cloud storage network element;
the forward association relation between the complexity of the storage picture and the number of objects in the picture comprises the following steps: and storing the forward association relation between the picture complexity and the number of objects in the picture by adopting a database.
Third embodiment
Fig. 3 is a schematic structural diagram of a data content complexity targeted detection system according to a third embodiment of the present invention, the structure includes:
the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device, wherein the data input device is used for receiving a picture to be processed, the content acquisition device is connected with the data input device and used for acquiring each pixel value corresponding to each pixel point in the picture to be processed, and the shading analysis device is connected with the data input device and used for analyzing the shading change level of the picture to be processed;
the data extraction mechanism is connected with the content acquisition mechanism and comprises a definition detection device and a contrast measurement device, wherein the definition detection device is connected with the data input device and is used for detecting the picture definition of the picture to be processed, and the contrast measurement device is connected with the data input device and is used for measuring the picture contrast of the picture to be processed;
a prediction processing device, which is respectively connected with the content acquisition mechanism and the data extraction mechanism, and is used for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture by adopting a convolution neural network after completing the training of a preset number for multiple times;
the network reconstruction device is connected with the prediction processing device and is used for sending the convolutional neural network after the preset number of training for a plurality of times to the prediction processing device to execute prediction processing;
the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed;
and the instant display device is connected with the complexity judgment device and is used for receiving and displaying the picture complexity of the picture to be processed.
Next, a further description will be given of the specific structure of the data content complexity targeted detection system of the present invention.
In a data content complexity targeted detection system according to various embodiments of the present invention:
the method for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of times of training comprises the following steps: the predicted objects in the to-be-processed picture are objects occupying the area proportion of the to-be-processed picture to be over-limited and the objects occupying the area proportion of the to-be-processed picture not to be over-limited do not participate in accumulation of the number of the objects in the to-be-processed picture;
the method for predicting the number of the objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of the multiple times of training further comprises: taking each pixel value corresponding to each pixel point in the picture to be processed, the brightness change level of the picture to be processed and the picture definition of the picture to be processed as each input information of the convolutional neural network;
the method for predicting the number of the objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of the multiple times of training further comprises: and taking the number of the objects in the picture to be processed as single output information of the convolutional neural network.
In a data content complexity targeted detection system according to various embodiments of the present invention:
the content acquisition device is connected with the data input device, and is used for acquiring each pixel value corresponding to each pixel point in the to-be-processed picture, and the method comprises the following steps: the content acquisition equipment is connected with the data input equipment and is used for acquiring each brightness value corresponding to each pixel point in the picture to be processed;
the light and shade analysis device is connected with the data input device, and is used for analyzing the light and shade change level of the to-be-processed picture, and the light and shade change level comprises: obtaining a maximum brightness value and a minimum brightness value of each brightness value corresponding to each pixel point in the picture to be processed, and analyzing the brightness change level of the picture to be processed based on the maximum brightness value and the minimum brightness value;
the obtaining the maximum brightness value and the minimum brightness value of each brightness value corresponding to each pixel point in the to-be-processed picture, and resolving the brightness change level of the to-be-processed picture based on the maximum brightness value and the minimum brightness value comprises the following steps: the larger the absolute value of the difference value between the maximum brightness value and the minimum brightness value is, the higher the brightness change level of the picture to be processed is obtained through analysis;
the larger the absolute value of the difference between the maximum brightness value and the minimum brightness value, the higher the brightness change level of the to-be-processed picture obtained through analysis comprises the following steps: and a numerical simulation formula is adopted for expressing the numerical mapping relation between the brightness change level of the picture to be processed and the absolute value.
In addition, in the data content complexity targeted detection system, the numerical value mapping relationship between the brightness change level of the to-be-processed picture and the absolute value, which is expressed by adopting a numerical value simulation formula, includes: the input parameter of the numerical simulation formula is the absolute value of the difference value between the maximum brightness value and the minimum brightness value, and the output parameter of the numerical simulation formula is the brightness change level of the picture to be processed.
Therefore, the important invention points of the technical scheme of the invention are as follows:
1. for each picture to be processed, adopting a convolution neural network after a preset number of times of training to predict the number of objects in the picture to be processed based on each pixel value corresponding to each pixel point in the picture to be processed, the brightness change level of the picture to be processed, the picture definition of the picture to be processed and the picture contrast of the picture to be processed, and further determining the content complexity of the picture to be processed;
2. for each picture to be processed, selecting a plurality of different training pictures with preset numbers based on the total number of the pixels of the picture to be processed for performing multiple times of training on the convolutional neural network, wherein the total number of the pixels of each training picture is the same as the total number of the pixels of the picture to be processed, and meanwhile, the smaller the total number of the pixels of the picture to be processed is, the smaller the value of the preset number is, so that the reliability and the effectiveness of the convolutional neural network obtained after training are ensured;
3. obtaining a maximum brightness value and a minimum brightness value of each brightness value corresponding to each pixel point in a picture to be processed, analyzing the brightness change level of the picture to be processed based on the maximum brightness value and the minimum brightness value, specifically, the larger the absolute value of the difference value between the maximum brightness value and the minimum brightness value is, the higher the brightness change level of the picture to be processed is obtained through analysis, so that the analysis precision of the brightness change level is ensured.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A data content complexity targeted detection system, the system comprising:
the content acquisition mechanism comprises a data input device, a content acquisition device and a shading analysis device, wherein the data input device is used for receiving a picture to be processed, the content acquisition device is connected with the data input device and used for acquiring each pixel value corresponding to each pixel point in the picture to be processed, and the shading analysis device is connected with the data input device and used for analyzing the shading change level of the picture to be processed;
the data extraction mechanism is connected with the content acquisition mechanism and comprises a definition detection device and a contrast measurement device, wherein the definition detection device is connected with the data input device and is used for detecting the picture definition of the picture to be processed, and the contrast measurement device is connected with the data input device and is used for measuring the picture contrast of the picture to be processed;
a prediction processing device, which is respectively connected with the content acquisition mechanism and the data extraction mechanism, and is used for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of each pixel point in the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture by adopting a convolution neural network after completing the training of a preset number for multiple times;
the network reconstruction device is connected with the prediction processing device and is used for sending the convolutional neural network after the preset number of training for a plurality of times to the prediction processing device to execute prediction processing;
the complexity judging device is connected with the prediction processing device and is used for determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed;
wherein determining the picture complexity of the picture to be processed based on the number of objects in the picture to be processed comprises: the picture complexity of the picture to be processed is positively correlated with the number of objects in the picture to be processed;
the smaller the number of each pixel point in the picture to be processed is, the smaller the value of the preset number is.
2. The data content complexity targeted detection system of claim 1, wherein:
transmitting the convolutional neural network after the preset number of times of training to the prediction processing equipment to execute the prediction processing comprises the following steps: the single training of the convolutional neural network is completed by adopting a certain training picture with the known number of objects and the same number of pixels as the number of pixels of the picture to be processed;
the step of completing the single training of the convolutional neural network by adopting a training picture with the number of known objects and the same number of pixels as the number of pixels of the picture to be processed comprises the following steps: taking the number of the existing objects of a certain training picture as single output information of the convolutional neural network, taking each pixel value corresponding to each pixel point in the certain training picture, the brightness change level of the certain training picture and the picture definition of the certain training picture as each input information of the convolutional neural network, and completing single training of the convolutional neural network;
the method for completing the single training of the convolutional neural network comprises the following steps of taking the number of the existing objects of a certain training picture as single output information of the convolutional neural network, taking each pixel value corresponding to each pixel point in the certain training picture, the brightness change level of the certain training picture and the picture definition of the certain training picture as each input information of the convolutional neural network, wherein the single training of the convolutional neural network comprises the following steps: and completing the preset number of multiple training on the convolutional neural network by adopting the preset number of multiple different training pictures.
3. The data content complexity targeted detection system of claim 2, wherein the system further comprises:
the cloud storage network element is connected with the complexity judging equipment in a network manner and is used for storing the forward association relation between the picture complexity and the number of objects in the picture;
the forward association relation between the complexity of the storage picture and the number of objects in the picture comprises the following steps: and storing the forward association relation between the picture complexity and the number of objects in the picture by adopting a database.
4. The data content complexity targeted detection system of claim 2, wherein the system further comprises:
and the instant display device is connected with the complexity judgment device and is used for receiving and displaying the picture complexity of the picture to be processed.
5. A data content complexity targeted detection system as claimed in any one of claims 2 to 4, wherein:
the method for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of times of training comprises the following steps: the predicted objects in the to-be-processed picture are objects occupying the area proportion of the to-be-processed picture to be out of limit and the objects occupying the area proportion of the to-be-processed picture not out of limit do not participate in accumulation of the number of the objects in the to-be-processed picture.
6. The data content complexity targeted detection system of claim 5, wherein:
the method for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of the multiple times of training further comprises: and taking each pixel value corresponding to each pixel point in the picture to be processed, the brightness change level of the picture to be processed and the picture definition of the picture to be processed as each input information of the convolutional neural network.
7. The data content complexity targeted detection system of claim 6, wherein:
the method for predicting the number of objects in the to-be-processed picture based on each pixel value, the brightness change level, the picture definition and the picture contrast of the to-be-processed picture, which are respectively corresponding to each pixel point in the to-be-processed picture, by adopting the convolutional neural network after the preset number of the multiple times of training further comprises: and taking the number of the objects in the picture to be processed as single output information of the convolutional neural network.
8. A data content complexity targeted detection system as claimed in any one of claims 2 to 4, wherein:
the content acquisition device is connected with the data input device, and is used for acquiring each pixel value corresponding to each pixel point in the to-be-processed picture, and the method comprises the following steps: the content acquisition equipment is connected with the data input equipment and is used for acquiring each brightness value corresponding to each pixel point in the picture to be processed.
9. The data content complexity targeted detection system of claim 8, wherein:
the light and shade analysis device is connected with the data input device and is used for analyzing the light and shade change level of the picture to be processed, and the light and shade change level comprises: and acquiring a maximum brightness value and a minimum brightness value of each brightness value corresponding to each pixel point in the picture to be processed, and analyzing the brightness change level of the picture to be processed based on the maximum brightness value and the minimum brightness value.
10. The data content complexity targeted detection system of claim 9, wherein:
obtaining a maximum brightness value and a minimum brightness value of brightness values respectively corresponding to each pixel point in the picture to be processed, and analyzing the brightness change level of the picture to be processed based on the maximum brightness value and the minimum brightness value comprises the following steps: the larger the absolute value of the difference value between the maximum brightness value and the minimum brightness value is, the higher the brightness change level of the picture to be processed is obtained through analysis;
the larger the absolute value of the difference between the maximum brightness value and the minimum brightness value, the higher the brightness change level of the to-be-processed picture obtained through analysis comprises the following steps: and a numerical simulation formula is adopted for expressing the numerical mapping relation between the brightness change level of the picture to be processed and the absolute value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310206982.9A CN116205817A (en) | 2023-03-07 | 2023-03-07 | Data content complexity targeted detection system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310206982.9A CN116205817A (en) | 2023-03-07 | 2023-03-07 | Data content complexity targeted detection system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116205817A true CN116205817A (en) | 2023-06-02 |
Family
ID=86507528
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310206982.9A Pending CN116205817A (en) | 2023-03-07 | 2023-03-07 | Data content complexity targeted detection system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116205817A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662206A (en) * | 2023-07-24 | 2023-08-29 | 泰山学院 | Computer software online real-time visual debugging method and device |
-
2023
- 2023-03-07 CN CN202310206982.9A patent/CN116205817A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116662206A (en) * | 2023-07-24 | 2023-08-29 | 泰山学院 | Computer software online real-time visual debugging method and device |
| CN116662206B (en) * | 2023-07-24 | 2024-02-13 | 泰山学院 | Computer software online real-time visual debugging method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rocchini et al. | Satellite remote sensing to monitor species diversity: Potential and pitfalls | |
| CN103020478A (en) | Method for checking reality of ocean color remote sensing product | |
| CN109472393B (en) | Spatial downscaling precipitation data detection method and device and electronic equipment | |
| CN105469079A (en) | Object material identification method based on multi-sensor information fusion | |
| CN111191673B (en) | Ground surface temperature downscaling method and system | |
| KR102319145B1 (en) | Method and device for generating high-resolution ocean data | |
| CN116205817A (en) | Data content complexity targeted detection system | |
| CN110706208A (en) | Infrared dim target detection method based on tensor mean square minimum error | |
| CN115965862A (en) | SAR ship target detection method based on mask network fusion image characteristics | |
| CN114330163B (en) | Typhoon-ionosphere disturbance dynamics model modeling method for high-frequency ground wave beyond-visual-range radar | |
| CN117423013B (en) | Ore potential prediction method based on geological big data | |
| CN112668615B (en) | Satellite cloud picture prediction method based on depth cross-scale extrapolation fusion | |
| CN111624606B (en) | Radar image rainfall identification method | |
| CN110363863B (en) | Input data generation method and system of neural network | |
| CN117407710A (en) | Hyperspectral remote sensing water quality parameter inversion method and device, electronic equipment and storage medium | |
| CN116719002A (en) | Quantitative precipitation estimation method, quantitative precipitation estimation device, electronic equipment and computer storage medium | |
| CN111105417A (en) | Image noise positioning method and system | |
| Sah et al. | Acoustic signal-based indigenous real-time rainfall monitoring system for sustainable environment | |
| Ucan et al. | A New Approach to Gravity Anomaly Separation: Differential Markov Random Field (DMRF) | |
| Raisuddin et al. | 3D-OutDet: A Fast and Memory Efficient Outlier Detector for 3D LiDAR Point Clouds in Adverse Weather | |
| CN114018773B (en) | PM 2.5 Method, device and equipment for acquiring concentration spatial distribution data and storage medium | |
| CN105069775B (en) | SAR image ship detection method based on Fourier space | |
| CN115630492B (en) | Tunnel lining defect change characteristic intelligent inversion method, system and storage medium | |
| Mitchell | Multiscale Modeling and Simulation of Clutter in ISAR Imaging | |
| CN107607931A (en) | A kind of laser radar echo image processing method |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20231127 Address after: No. 66 Jingu East Road, Xincheng, Dantu District, Zhenjiang City, Jiangsu Province, 212100 Applicant after: Zhenjiang Anjian Imaging Co.,Ltd. Address before: No. 161 Shidai Road, Shifeng District, Zhuzhou City, Hunan Province, 412001 Applicant before: Chen Chunlan |
|
| TA01 | Transfer of patent application right |