CN113516166A - Airborne photoelectric pod electromagnetic performance boundary prediction method and system - Google Patents
Airborne photoelectric pod electromagnetic performance boundary prediction method and system Download PDFInfo
- Publication number
- CN113516166A CN113516166A CN202110514758.7A CN202110514758A CN113516166A CN 113516166 A CN113516166 A CN 113516166A CN 202110514758 A CN202110514758 A CN 202110514758A CN 113516166 A CN113516166 A CN 113516166A
- Authority
- CN
- China
- Prior art keywords
- photoelectric pod
- airborne photoelectric
- interference
- image data
- airborne
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/001—Measuring interference from external sources to, or emission from, the device under test, e.g. EMC, EMI, EMP or ESD testing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Abstract
The invention relates to a method and a system for predicting electromagnetic performance boundary of an airborne photoelectric pod. The method comprises the following steps: acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments; acquiring corresponding image data of the airborne photoelectric pod in an interference-free environment; constructing a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod; training the gradient lifting tree model based on image data of the airborne photoelectric pod under different electromagnetic interference environments and corresponding image data of the airborne photoelectric pod under an interference-free environment to obtain a trained gradient lifting tree model; and predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model. The method can realize the comprehensive prediction of the electromagnetic performance boundary of the airborne photoelectric pod.
Description
Technical Field
The invention relates to the field of electromagnetic performance boundary prediction, in particular to an airborne photoelectric pod electromagnetic performance boundary prediction method and system.
Background
With the development of the technical fields of machinery, optics, electronics and aviation, the pod is widely applied to military and civil fields such as information collection, surveying and mapping, target protection and tracking, civil aviation information collection, aerial photography, high-voltage transmission line maintenance and the like by virtue of the advantages of small volume, light weight, flexibility and convenience in operation, high stability, high-definition shooting and ground return real-time remote control and the like.
Meanwhile, with the rapid development of science and technology, the frequency spectrum of various electronic devices such as radar electronic devices and communication electronic devices is wider and wider, the power is larger and larger, and the types of adopted signal patterns are more and more, so that the electromagnetic environment of the unmanned aerial vehicle is more and more complicated and changeable, and the airborne photoelectric pod is required to be subjected to the strict examination of the complicated electromagnetic environment. In order to clearly understand the influence of complex electromagnetic environment on the unmanned airborne pod, a large number of electromagnetic effect tests are required to be carried out on the unmanned airborne pod. However, in the process of evaluating the electromagnetic environment effect of the airborne photoelectric pod in a complex electromagnetic environment, all electromagnetic signal parameter combinations cannot be exhausted, and the problem of incomplete evaluation of the electromagnetic environment effect exists.
Disclosure of Invention
The invention aims to provide a method and a system for predicting the electromagnetic performance boundary of an airborne photoelectric pod, so as to comprehensively predict the electromagnetic performance boundary of the airborne photoelectric pod.
In order to achieve the purpose, the invention provides the following scheme:
an airborne photoelectric pod electromagnetic performance boundary prediction method comprises the following steps:
acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments;
acquiring corresponding image data of the airborne photoelectric pod in an interference-free environment;
constructing a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod;
training the gradient lifting tree model based on image data of the airborne photoelectric pod under different electromagnetic interference environments and corresponding image data of the airborne photoelectric pod under an interference-free environment to obtain a trained gradient lifting tree model;
and predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model.
Optionally, the acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments specifically includes:
performing an electromagnetic interference experiment on the airborne photoelectric pod to obtain images acquired by the airborne photoelectric pod in different electromagnetic interference environments; the electromagnetic interference environment comprises the type of the interference signal, the frequency of the interference signal, the field strength of the interference signal and the polarization direction of the antenna.
Optionally, the training of the gradient lifting tree model based on the image data of the airborne photoelectric pod in different electromagnetic interference environments and the image data of the airborne photoelectric pod corresponding to the airborne photoelectric pod in an interference-free environment is performed to obtain a trained gradient lifting tree model, which specifically includes:
determining the image quality corresponding to each image in the image data of the airborne photoelectric pod under different electromagnetic interference environments by adopting a full reference method based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the interference-free environment;
determining a performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments;
and taking the performance boundary of the airborne photoelectric pod under each electromagnetic interference environment as a label, taking each electromagnetic interference environment as sample data, taking the sample data and image data corresponding to the airborne photoelectric pod under the interference-free environment as the input of the gradient lifting tree model, and training the gradient lifting tree model to obtain the trained gradient lifting tree model.
Optionally, the predicting the performance boundary of the airborne photoelectric pod in the interference environment to be predicted based on the trained gradient lifting tree model specifically includes:
and inputting the image data corresponding to the interference environment to be predicted and the airborne photoelectric pod in the interference-free environment into the trained gradient lifting tree model, and outputting the performance boundary of the airborne photoelectric pod in the interference environment to be predicted.
The invention also provides a system for predicting the electromagnetic performance boundary of the airborne photoelectric pod, which comprises the following components:
the image data acquisition module under the electromagnetic interference environment is used for acquiring the image data of the airborne photoelectric pod under different electromagnetic interference environments;
the image data acquisition module in the non-interference environment is used for acquiring the corresponding image data of the airborne photoelectric pod in the non-interference environment;
the gradient lifting tree model building module is used for building a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod;
the training module is used for training the gradient lifting tree model based on image data of the airborne photoelectric pod under different electromagnetic interference environments and corresponding image data of the airborne photoelectric pod under an interference-free environment to obtain a trained gradient lifting tree model;
and the performance boundary prediction module is used for predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model.
Optionally, the image data obtaining module in the electromagnetic interference environment specifically includes:
the electromagnetic interference experiment unit is used for carrying out an electromagnetic interference experiment on the airborne photoelectric pod to obtain images acquired by the airborne photoelectric pod in different electromagnetic interference environments; the electromagnetic interference environment comprises the type of the interference signal, the frequency of the interference signal, the field strength of the interference signal and the polarization direction of the antenna.
Optionally, the training module specifically includes:
the image quality determining unit is used for determining the image quality corresponding to each image in the image data of the airborne photoelectric pod under different electromagnetic interference environments by adopting a full reference method based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the interference-free environment;
the performance boundary determining unit is used for determining the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments;
and the training unit is used for taking the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment as a label, taking each electromagnetic interference environment as sample data, taking the sample data and image data corresponding to the airborne photoelectric pod in an interference-free environment as the input of the gradient lifting tree model, and training the gradient lifting tree model to obtain the trained gradient lifting tree model.
Optionally, the performance boundary prediction module specifically includes:
and the performance boundary prediction unit is used for inputting the image data corresponding to the interference environment to be predicted and the airborne photoelectric pod under the interference-free environment into the trained gradient lifting tree model and outputting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
the method is based on image data under different electromagnetic interference environments, a gradient lifting tree model is constructed and trained, and the performance boundary of the airborne photoelectric pod under any electromagnetic interference environment can be predicted by adopting the gradient lifting tree model, so that comprehensive prediction is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of an airborne photoelectric pod electromagnetic performance boundary prediction method of the present invention;
FIG. 2 is a schematic structural diagram of an airborne photoelectric pod electromagnetic performance boundary prediction system of the present invention;
FIG. 3 is a block flow diagram of an embodiment of the present invention;
FIG. 4 is a block diagram of an EMI experimental system in accordance with an embodiment of the present invention;
FIG. 5 is a graph comparing the predicted performance of GPR and GBDT models in an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
FIG. 1 is a schematic flow chart of the method for predicting the electromagnetic performance boundary of the airborne photoelectric pod. As shown in fig. 1, the method for predicting the electromagnetic performance boundary of the airborne photoelectric pod comprises the following steps:
step 100: and acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments. The invention carries out an electromagnetic interference experiment on the airborne photoelectric pod, stores video information detected by the airborne photoelectric pod in the experiment process in real time, extracts frame images of the video information of the airborne photoelectric pod in different electromagnetic interference environments through a written program, namely converts the video into a plurality of continuous pictures to obtain image data of the airborne photoelectric pod in different electromagnetic interference environments, and stores the image data as gray images. And the electromagnetic interference environment including the type, frequency, field intensity, antenna polarization direction and the like of an interference signal when the airborne photoelectric pod collects images with different qualities is recorded in real time.
Step 200: and acquiring corresponding image data of the airborne photoelectric pod in an interference-free environment.
Step 300: and constructing a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod. According to the gradient lifting tree model, a GBDT (gradient lifting tree) algorithm is used as a classifier, corresponding image data of an electromagnetic interference environment and an airborne photoelectric pod in an interference-free environment is used as model input, and a boundary threshold value of the airborne visible light pod is used as model output.
Step 400: and training the gradient lifting tree model based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the interference-free environment to obtain the trained gradient lifting tree model. The method trains each model of the classifier in the gradient lifting tree model, takes the prediction result of the classifier as a new training set, continuously optimizes and adjusts model parameters, and determines optimal parameters to obtain a posterior model. The specific training process is as follows:
(1) and (4) preprocessing data. The image data of the airborne photoelectric pod under different electromagnetic interference environments is used as sampling data, abnormal values in the sampling data are filtered and removed, and the adopted data are analyzed and processed by using a characteristic engineering technology.
(3) And calculating the image data quality evaluation characteristic value. And taking image data corresponding to the airborne photoelectric pod in an interference-free environment as a Reference picture, and objectively evaluating the image quality of each picture in the sampling data by adopting a Full-Reference (FR) method. The full-reference image quality evaluation refers to comparing the difference between the image to be evaluated and the reference image under the condition of selecting an ideal image as the reference image, and analyzing the distortion degree of the image to be evaluated so as to obtain the quality evaluation of the image to be evaluated. The invention adopts Peak-Signal to Noise Ratio (PSNR) as a quantitative evaluation index.
(4) And constructing training sample data and training. And determining the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments. Performance boundary division: when the image shot by the airborne visible light pod is clear under the action of multi-source electromagnetic interference, the peak signal-to-noise ratio of the image is larger than 25, the peak signal-to-noise ratio of the image shot by the airborne visible light pod is between 15 and 25 when noise points appear on the image shot by the airborne visible light pod due to interference but the outline of an object in the image can be distinguished, the image shot by the airborne visible light pod has full screen horizontal stripes and snowflakes or a large number of blue strip-shaped stripes and white horizontal stripes, and the peak signal-to-noise ratio is smaller than 15 when the outline of the object in the image cannot be distinguished. Therefore, the performance boundaries of the airborne visible light pod under two electromagnetic interferences of the invention are respectively as follows: boundary 1- -performance degradation, corresponding to a peak signal-to-noise ratio of around 25; boundary 2- -imaging error, corresponding to a peak signal-to-noise ratio of around 15.
The performance boundary of the airborne photoelectric pod under different interference signal types and different frequencies has the following characteristics: (1) different interference signal types and frequencies are the same, and the performance boundaries of the airborne photoelectric pod are different. (2) The same interference signal type, different frequencies and different performance boundaries of the airborne photoelectric pod are different, and the two types of interference signals are in nonlinear change. (3) The same interference signal type, different antenna polarization directions and different performance boundaries of the airborne photoelectric pod are different, and the two types of interference signals are in nonlinear change. The reason is that the airborne photoelectric pod is not frequency-used equipment, different devices on the circuit board are interfered when different interference signals act, and due to the fact that the devices are different, the sensitive signals of the different devices are different, and performance boundaries are different.
After the performance boundary is determined, the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment is used as a label, each electromagnetic interference environment is used as sample data, the sample data and image data corresponding to the airborne photoelectric pod in the interference-free environment are used as input of the gradient lifting tree model, the gradient lifting tree model is trained, and the trained gradient lifting tree model is obtained.
Step 500: and predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model. And inputting image data corresponding to the interference environment to be predicted and the airborne photoelectric pod in the interference-free environment into the trained gradient lifting tree model, so that the performance boundary of the airborne photoelectric pod in the interference environment to be predicted can be output.
In conclusion, the factors causing the instability of the camera shooting of the airborne photoelectric pod comprise interference signal patterns, frequency and antenna polarization direction, because the relationship between the three influencing factors and the performance boundary of the airborne photoelectric pod is nonlinear, nonlinear predictive modeling is difficult to realize by adopting the traditional deterministic analysis method, and the effect threshold of the airborne photoelectric pod under all multi-source electromagnetic interference is obtained through experimental tests, the workload is huge, the realization is difficult, therefore, the method can be used for training and modeling by constructing an original test data sample and utilizing a machine learning method, the performance boundary of the airborne photoelectric pod can be predicted under different electromagnetic sensitivity parameters (including interference signal type combination, interference signal frequency and antenna polarization direction) through the optimization model, and therefore the workload of determining the effect threshold of the airborne photoelectric pod through a large number of tests is reduced.
Based on the scheme, the invention also provides an airborne photoelectric pod electromagnetic performance boundary prediction system, and fig. 2 is a schematic structural diagram of the airborne photoelectric pod electromagnetic performance boundary prediction system. As shown in fig. 2, the system for predicting the electromagnetic performance boundary of the airborne photoelectric pod of the invention comprises the following structures:
the image data acquisition module 201 in the electromagnetic interference environment is used for acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments.
And the image data acquisition module 202 in the non-interference environment is used for acquiring the corresponding image data of the airborne photoelectric pod in the non-interference environment.
The gradient lifting tree model building module 203 is used for building a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod.
The training module 204 is configured to train the gradient lifting tree model based on image data of the airborne photoelectric pod in different electromagnetic interference environments and image data corresponding to the airborne photoelectric pod in an interference-free environment, so as to obtain a trained gradient lifting tree model.
And the performance boundary predicting module 205 is configured to predict a performance boundary of the airborne photoelectric pod in the interference environment to be predicted based on the trained gradient lifting tree model.
As a specific embodiment, in the system for predicting the electromagnetic performance boundary of the airborne photoelectric pod, the image data obtaining module 201 in the electromagnetic interference environment specifically includes:
the electromagnetic interference experiment unit is used for carrying out an electromagnetic interference experiment on the airborne photoelectric pod to obtain images acquired by the airborne photoelectric pod in different electromagnetic interference environments; the electromagnetic interference environment comprises the type of the interference signal, the frequency of the interference signal, the field strength of the interference signal and the polarization direction of the antenna.
As a specific embodiment, in the system for predicting electromagnetic performance boundary of airborne optoelectronic pod of the present invention, the training module 204 specifically includes:
the image quality determining unit is used for determining the image quality corresponding to each image in the image data of the airborne photoelectric pod under different electromagnetic interference environments by adopting a full reference method based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the non-interference environment.
And the performance boundary determining unit is used for determining the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments.
And the training unit is used for taking the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment as a label, taking each electromagnetic interference environment as sample data, taking the sample data and image data corresponding to the airborne photoelectric pod in an interference-free environment as the input of the gradient lifting tree model, and training the gradient lifting tree model to obtain the trained gradient lifting tree model.
As a specific embodiment, in the system for predicting electromagnetic performance boundary of airborne optoelectronic pod of the present invention, the performance boundary predicting module 205 specifically includes:
and the performance boundary prediction unit is used for inputting the image data corresponding to the interference environment to be predicted and the airborne photoelectric pod under the interference-free environment into the trained gradient lifting tree model and outputting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted.
The following provides a detailed description of the above-described aspects of the invention. In the embodiment, the photoelectric pod of the unmanned aerial vehicle is taken as a tested object, and a background interference signal, a communication interference signal and a radar interference signal are taken as typical interference type combinations. Fig. 3 is a flow chart of an embodiment of the present invention, as shown in fig. 3, including the following processes:
step 1: the dynamic multi-source electromagnetic environment effect test system is connected according to the structure shown in fig. 4, the tested airborne photoelectric pod is placed 1 meter in front of the antenna, and the tested airborne photoelectric pod is electrified to work normally.
Step 2: setting parameters of a background electromagnetic environment generation module, programming and setting signals (including signal patterns, amplitudes, frequencies and the like) generated by any signal generator, microwave signal sources (including signals), signals (including signal patterns, amplitudes, frequencies and the like) generated by the microwave signal sources, and background interference signals generated by gains of power amplifiers.
And step 3: the method comprises the steps of setting parameters of a radar interference generation module, setting signals (including signal patterns, amplitudes, frequencies and the like) generated by any signal generator and radio frequency signal sources (including signals) through programming, setting signals (including signal patterns, amplitudes, frequencies and the like) generated by the radio frequency signal sources and setting radar interference signals generated by gains of power amplifiers.
And 4, step 4: the method comprises the steps of setting parameters of a communication interference generation module, setting signals (including signal patterns, amplitudes, frequencies and the like) generated by any signal generator and radio frequency signal sources (including signals) through programming, setting signals (including signal patterns, amplitudes, frequencies and the like) generated by the radio frequency signal sources and setting communication interference signals generated by gains of power amplifiers.
And 5: and the antenna polarization direction, the background signal, the radar signal and the communication signal are all set to be output to form a multi-source radiation field.
Step 6: and adjusting the type, frequency and field intensity of a radiation field signal, radiating the airborne photoelectric pod, recording the peak signal-to-noise ratio of images of the airborne photoelectric pod in different interference environments, and determining a corresponding performance boundary.
And 7: and (5) repeating the steps 2-6, and recording the performance boundary of the airborne photoelectric pod under each interference signal.
And 8: finally 27590 groups of data samples are collected, after data averaging processing, 275 groups of data samples are constructed, 192 groups of the data samples are used as training samples, and 83 groups of the data samples are used as testing samples.
And step 9: and determining an iterative model and a regression model, and setting initial model parameters. Selecting a tolerant early-stop iteration model and an objective function model based on a least square method, setting initial model parameters on the basis, and determining a prior model.
Step 10: and (4) training the GBDT model. The method comprises the steps of taking pictures of an interference signal in a training sample, taking the pictures of an airborne photoelectric pod as 4 parameters as model input, taking a performance boundary of the airborne visible light pod as model output, taking a GBDT (gradient lifting tree) algorithm as a classifier, training each model of the classifier based on an iterative model and a regression model, taking a prediction result of the classifier as a new training set, continuously optimizing and adjusting model parameters, determining optimal parameters and obtaining a posterior model.
Step 11: and (5) testing the prediction accuracy of the model. 4 parameters in the test sample are used as input of the posterior model, output of the model can be obtained through prediction, and the model is compared with an effect threshold value of the airborne visible light pod in the test sample for analysis, so that error of a predicted value is obtained. The algorithm randomly divides the characteristic value data into a training set (70%) and a testing set (30%) according to a certain proportion. And carrying out model training by using the training set data, and evaluating the trained model by using the test set.
The GBDT model training is carried out through the process, in order to test the effect of the GBDT model in the aspect of prediction, a test sample is utilized to compare the prediction performances of the GPR model and the GBDT model, the result is shown in FIG. 5, part (a) in FIG. 5 is the prediction precision of the GPR model regressed in the Gaussian process, and part (b) in FIG. 5 is the prediction precision of the GBDT model of the gradient lifting tree, so that the prediction error of the GPR model is large, and the maximum position exceeds 6 dB; the GBDT model of the invention has smaller prediction error, the maximum error is only 0.67dB, which is superior to the prediction result of the GPR model and meets the tolerance of +/-3 dB specified by the national military standard.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (8)
1. The method for predicting the electromagnetic performance boundary of the airborne photoelectric pod is characterized by comprising the following steps:
acquiring image data of the airborne photoelectric pod in different electromagnetic interference environments;
acquiring corresponding image data of the airborne photoelectric pod in an interference-free environment;
constructing a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod;
training the gradient lifting tree model based on image data of the airborne photoelectric pod under different electromagnetic interference environments and corresponding image data of the airborne photoelectric pod under an interference-free environment to obtain a trained gradient lifting tree model;
and predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model.
2. The method for predicting the electromagnetic performance boundary of the airborne photoelectric pod according to claim 1, wherein the acquiring of the image data of the airborne photoelectric pod in different electromagnetic interference environments specifically comprises:
performing an electromagnetic interference experiment on the airborne photoelectric pod to obtain images acquired by the airborne photoelectric pod in different electromagnetic interference environments; the electromagnetic interference environment comprises the type of the interference signal, the frequency of the interference signal, the field strength of the interference signal and the polarization direction of the antenna.
3. The method for predicting the boundary of the electromagnetic performance of the airborne photoelectric pod of claim 1, wherein the training of the gradient lifting tree model based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the image data corresponding to the airborne photoelectric pod under an interference-free environment is performed to obtain the trained gradient lifting tree model, and specifically comprises:
determining the image quality corresponding to each image in the image data of the airborne photoelectric pod under different electromagnetic interference environments by adopting a full reference method based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the interference-free environment;
determining a performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments;
and taking the performance boundary of the airborne photoelectric pod under each electromagnetic interference environment as a label, taking each electromagnetic interference environment as sample data, taking the sample data and image data corresponding to the airborne photoelectric pod under the interference-free environment as the input of the gradient lifting tree model, and training the gradient lifting tree model to obtain the trained gradient lifting tree model.
4. The method for predicting the electromagnetic performance boundary of the airborne photoelectric pod according to claim 1, wherein the predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model specifically comprises:
and inputting the image data corresponding to the interference environment to be predicted and the airborne photoelectric pod in the interference-free environment into the trained gradient lifting tree model, and outputting the performance boundary of the airborne photoelectric pod in the interference environment to be predicted.
5. An airborne optoelectronic pod electromagnetic performance boundary prediction system, comprising:
the image data acquisition module under the electromagnetic interference environment is used for acquiring the image data of the airborne photoelectric pod under different electromagnetic interference environments;
the image data acquisition module in the non-interference environment is used for acquiring the corresponding image data of the airborne photoelectric pod in the non-interference environment;
the gradient lifting tree model building module is used for building a gradient lifting tree model for predicting the electromagnetic performance boundary of the airborne photoelectric pod;
the training module is used for training the gradient lifting tree model based on image data of the airborne photoelectric pod under different electromagnetic interference environments and corresponding image data of the airborne photoelectric pod under an interference-free environment to obtain a trained gradient lifting tree model;
and the performance boundary prediction module is used for predicting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted based on the trained gradient lifting tree model.
6. The system for predicting the electromagnetic performance boundary of the airborne optoelectronic pod according to claim 5, wherein the image data acquisition module in the electromagnetic interference environment specifically comprises:
the electromagnetic interference experiment unit is used for carrying out an electromagnetic interference experiment on the airborne photoelectric pod to obtain images acquired by the airborne photoelectric pod in different electromagnetic interference environments; the electromagnetic interference environment comprises the type of the interference signal, the frequency of the interference signal, the field strength of the interference signal and the polarization direction of the antenna.
7. The system for predicting the electromagnetic performance boundary of the airborne optoelectronic pod as set forth in claim 5, wherein the training module specifically comprises:
the image quality determining unit is used for determining the image quality corresponding to each image in the image data of the airborne photoelectric pod under different electromagnetic interference environments by adopting a full reference method based on the image data of the airborne photoelectric pod under different electromagnetic interference environments and the corresponding image data of the airborne photoelectric pod under the interference-free environment;
the performance boundary determining unit is used for determining the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment based on the image quality corresponding to each image in the image data of the airborne photoelectric pod in different electromagnetic interference environments;
and the training unit is used for taking the performance boundary of the airborne photoelectric pod in each electromagnetic interference environment as a label, taking each electromagnetic interference environment as sample data, taking the sample data and image data corresponding to the airborne photoelectric pod in an interference-free environment as the input of the gradient lifting tree model, and training the gradient lifting tree model to obtain the trained gradient lifting tree model.
8. The system for predicting the electromagnetic performance boundary of the airborne optoelectronic pod of claim 5, wherein the performance boundary prediction module specifically comprises:
and the performance boundary prediction unit is used for inputting the image data corresponding to the interference environment to be predicted and the airborne photoelectric pod under the interference-free environment into the trained gradient lifting tree model and outputting the performance boundary of the airborne photoelectric pod under the interference environment to be predicted.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110514758.7A CN113516166B (en) | 2021-05-12 | 2021-05-12 | Airborne photoelectric pod electromagnetic performance boundary prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110514758.7A CN113516166B (en) | 2021-05-12 | 2021-05-12 | Airborne photoelectric pod electromagnetic performance boundary prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113516166A true CN113516166A (en) | 2021-10-19 |
| CN113516166B CN113516166B (en) | 2022-03-15 |
Family
ID=78064381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110514758.7A Active CN113516166B (en) | 2021-05-12 | 2021-05-12 | Airborne photoelectric pod electromagnetic performance boundary prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113516166B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102722618A (en) * | 2012-06-04 | 2012-10-10 | 北京航空航天大学 | Method for building and paralleling quasi three-dimensional electromagnetic environmental model on basis of parabolic equation |
| US8316555B2 (en) * | 2009-12-09 | 2012-11-27 | Honeywell International Inc. | Non-contact data transfer from moving systems |
| CN108227687A (en) * | 2016-12-09 | 2018-06-29 | 广东德豪润达电气股份有限公司 | Intelligent robot identification virtual boundary method, mode of progression and beacon |
| CN111740796A (en) * | 2020-06-05 | 2020-10-02 | 中国人民解放军陆军工程大学 | Unmanned aerial vehicle data link electromagnetic interference situation prediction method and device |
-
2021
- 2021-05-12 CN CN202110514758.7A patent/CN113516166B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8316555B2 (en) * | 2009-12-09 | 2012-11-27 | Honeywell International Inc. | Non-contact data transfer from moving systems |
| CN102722618A (en) * | 2012-06-04 | 2012-10-10 | 北京航空航天大学 | Method for building and paralleling quasi three-dimensional electromagnetic environmental model on basis of parabolic equation |
| CN108227687A (en) * | 2016-12-09 | 2018-06-29 | 广东德豪润达电气股份有限公司 | Intelligent robot identification virtual boundary method, mode of progression and beacon |
| CN111740796A (en) * | 2020-06-05 | 2020-10-02 | 中国人民解放军陆军工程大学 | Unmanned aerial vehicle data link electromagnetic interference situation prediction method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113516166B (en) | 2022-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102749539B (en) | Fast electromagnetic compatibility test and diagnosis system with quantization electromagnetic interference | |
| CN110191335B (en) | Remote sensing camera imaging quality evaluation and simulation system | |
| CN115187527B (en) | Separation and identification method for multi-source mixed ultrahigh frequency partial discharge spectrum | |
| CN103246556A (en) | General performance simulation method and framework of air surveillance radar | |
| CN105389591A (en) | Method for verifying simulation of typical effect of infrared imaging sensor | |
| CN105163106A (en) | Multi-data-processing video quality evaluation system | |
| JPWO2019106850A1 (en) | SAR image analysis system, image processing equipment, image processing method and image processing program | |
| CN111858340A (en) | Deep neural network test data generation method based on stability transformation | |
| CN113159447B (en) | Laser radar electromagnetic environment effect prediction method and system | |
| CN104980734A (en) | Device for detecting image sensor performance and use method thereof | |
| KR102174424B1 (en) | Method for Inspecting Component basesd Server and system and apparatus therefor | |
| CN113516166B (en) | Airborne photoelectric pod electromagnetic performance boundary prediction method and system | |
| Xia et al. | Quality assessment for remote sensing images: approaches and applications | |
| CN105046706B (en) | SAR image ship detection method based on rational polynominal Function Fitting sea clutter | |
| US9083418B2 (en) | Versatile antenna received signal strength measurement system not affecting antenna pattern and receiver performance | |
| CN115598993A (en) | Evaluation method and system for physical effect simulation effect of semi-physical simulation infrared scene | |
| CN115941934A (en) | Multi-frame Fusion Video Noise Evaluation Method Based on Deep Convolutional Neural Network | |
| CN111062168A (en) | Design method suitable for vibration reduction structure of airborne wireless laser communication equipment | |
| CN116205817A (en) | Data content complexity targeted detection system | |
| CN115201766A (en) | Radar mixed interference sensing method based on deep target detection network | |
| Li et al. | Machine learning based tool chain solution for free space optical communication (FSOC) propagation modeling | |
| US11507803B2 (en) | System for generating synthetic digital data for data multiplication | |
| CN115856881B (en) | Millimeter wave radar behavior sensing method based on dynamic lightweight network | |
| CN113591323B (en) | Embedded image dynamic scene simulation imaging system and method | |
| US11551061B2 (en) | System for generating synthetic digital data of multiple sources |
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 |