CN117893877A

CN117893877A - Method and system for detecting quasi-periodic radiation phenomenon of Zhangheng first satellite based on DETR

Info

Publication number: CN117893877A
Application number: CN202410061814.XA
Authority: CN
Inventors: 冉子霖; 杨德贺; 泽仁志玛; 杨艳艳; 黄建平; 刘大鹏; 林剑
Original assignee: National Institute of Natural Hazards
Current assignee: National Institute of Natural Hazards
Priority date: 2024-01-16
Filing date: 2024-01-16
Publication date: 2024-04-16

Abstract

The invention discloses a method and a system for detecting quasi-periodic radiation phenomenon of a Zhangheng first satellite based on DETR, wherein the method comprises the following steps: converting an input image into high-level semantic features through a convolutional neural network to obtain a feature map; multiplying the channel weight with the corresponding element of the feature map to obtain a final output feature map; mapping the final output feature map into a sequence form, then inputting the sequence form into an encoder, and adding position codes; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; the features are mapped to the prediction space through a plurality of fully connected layers to predict the class and location of the target. By the processing scheme, accuracy and detection speed are improved, and the method can be deployed in satellite equipment.

Description

Method and system for detecting quasi-periodic radiation phenomenon of Zhangheng first satellite based on DETR

Technical Field

The invention relates to the technical field of detection, in particular to a method and a system for detecting quasi-periodic radiation phenomenon of a Zhangheng first satellite based on DETR.

Background

Ionospheric quasi-periodic radiation is a typical very low frequency/very low frequency electromagnetic wave phenomenon propagating in whistle wave mode, has important scientific research value, and has become a research hotspot in the academic world for more than thirty years. At present, related studies have preliminary knowledge of ionospheric quasi-periodic radiation: the source region may be located near the geomagnetic equatorial plane of the magnetic layer, with a frequency range of approximately 500Hz-4kHz, and with significant periodic variations in the fluctuation intensity, lasting for a period of 10-80 a, which occurs mainly during the day, and which typically occurs during periods of relatively calm of geomagnetism, is most easily observed from a spatial distribution at locations with L-shell values greater than 3, with periods approximately coinciding with the Pc3-Pc5 geomagnetic pulses.

Electromagnetic field observation is an important means for comprehensively understanding the ionospheric quasi-periodic radiation phenomenon in near-earth space. Some knowledge of the periodic radiation is known, but none is known in depth due to the limitations of processing large amounts of electromagnetic field data from satellites. Before overcoming these problems, extensive and intensive research into a large number of ionospheric quasi-periodic emissions is required. Currently, ionospheric electromagnetic field observations, CSES, have accumulated for five years. However, in the face of such a large amount of electromagnetic field data, the conventional means for selecting the electromagnetic wave phenomenon relies on a manual identification method, and has problems of low detection efficiency and easiness in omission. Along with the accumulation of satellite observation data, a large amount of space phenomena hidden in the observation data are urgently needed to be extracted quickly, accurately and intelligently.

Since emission in 2018, 2 nd month, CSES has been running steadily on orbit for more than five years, global space electromagnetic fields, ionospheric plasma environments, etc. have been monitored comprehensively, and a large amount of electromagnetic field waveform data has been obtained. The massive electromagnetic information carries electromagnetic disturbance information from the sun to the rock ring, and has important scientific research and practical application values. However, in the face of these large amounts of electromagnetic field data, conventional data processing and research methods are inefficient and have very limited information mining capabilities, and thus have not been fully studied so far. Therefore, we apply advanced artificial intelligence technology cross fusion to this field to achieve the ability to extract target information from electromagnetic big data quickly and accurately. Therefore, there is a need for artificial intelligence methods that explore quasi-periodic radiation phenomena to solve the problem of automatic detection of a large number of phenomena in databases, as well as the problem of near real-time detection of subsequent orbiting satellite observations.

DETR (Detection Transformer) is an end-to-end object detector based on a transducer architecture. Compared with the traditional target detection method, the DETR adopts completely different ideas to solve the target detection problem, and the position and the category of the target can be directly predicted from the input image under the condition that the traditional target detection components such as an anchor frame, non-maximum inhibition and the like which are manually designed are not required.

The following is a general flow of DETR to achieve target detection:

Encoder (Encoder): the input image is passed through a pre-trained convolutional neural network (typically ResNet or other CNN model) as an encoder for extracting image features. The output of the encoder is a series of feature vectors that contain semantic information for each location in the image.

Decoder (Decoder): the decoder is a module based on a transducer structure and comprises a Self-Attention layer (Self-Attention) and a plurality of multi-head Attention layers. The input to the decoder is the output feature vector of the encoder, which is processed to generate the result of the object detection.

Object Queries (objects): the decoder may initialize specific object query vectors for directing the decoder to find the target at a specific location. These object query vectors typically contain predefined location and category information.

Self-Attention mechanism (Self-Attention): the self-attention mechanism of the decoders allows global association between the decoders so that they can see the entire input feature map when generating the results. The self-attention mechanism maps the input features to a new representation, enabling the encoding of semantic relationships between different locations.

Target classification and location prediction: the decoder aligns the locations in the feature map with the object query vector through a self-attention mechanism to predict the class and location of the target. This can be achieved by classifying and regressing the feature map.

Matching process (Bipartite Matching): the predicted target is matched with the real target through a Hungary algorithm and other methods, and a loss function is calculated in the training process so as to monitor the prediction accuracy of the model.

Directly outputting a result: finally, DETR directly outputs information such as predicted target boxes and categories without requiring further screening and optimization of the results by post-processing steps (e.g., non-maximal suppression).

However, the quasi-periodic radiation phenomenon has large density difference, complex background, targets with different dimensions, and DETR cannot meet the requirements on precision and speed when facing the problems, and needs a lightweight model when deployed in satellite equipment,

Therefore, the above-mentioned existing detection method is still inconvenient and disadvantageous, and needs to be further improved. How to create a new detection method becomes the aim of improvement in the current industry.

Disclosure of Invention

In view of this, the embodiments of the present disclosure provide a DETR-based method for detecting quasi-periodic radiation of a first satellite in a tensor manner, which at least partially solves the problems in the prior art.

In a first aspect, embodiments of the present disclosure provide a DETR-based method for detecting quasi-periodic radiation phenomenon of a first satellite in tensor, the method including the steps of:

Converting an input image into high-level semantic features through a convolutional neural network to obtain a feature map;

carrying out global average pooling on an input feature map based on a high-efficiency channel attention module, then carrying out convolution operation, obtaining channel weights through an activation function, and multiplying the channel weights with corresponding elements of the feature map to obtain a finally output feature map;

Mapping the final output feature map into a sequence form, then inputting into an encoder, and adding position codes; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; and interacting the position in the finally output feature map with the object query vector based on a self-attention mechanism in the decoder to obtain the representation information of the feature, mapping the feature to a prediction space through a plurality of fully connected layers, and predicting the category and the position of the target.

According to a specific implementation of an embodiment of the disclosure, the method further includes: physical constraints of proton cyclotron frequency are added at the time of prediction.

According to a specific implementation of an embodiment of the disclosure, the adding physical constraint of proton cyclotron frequency in prediction includes:

Calculating proton cyclotron frequency of predicted orbit by observed magnetic field strength based on the following formula _：

Wherein ω is proton convolution rate; q is proton charge; b is the magnetic field intensity; m is proton mass;

interpolating the latitude value and the frequency value to the number of the pixel points to obtain the position of the image where the pixel points are located; wherein, the abscissa is latitude, and the ordinate is frequency;

Judging whether the value of the ordinate is larger than the pixel value of the proton convolution frequency when verifying the position of the abscissa at the center of the prediction frame when the position of the image is at the center coordinate of the prediction frame; wherein when less than, the prediction box is discarded.

According to a specific implementation of an embodiment of the disclosure, the convolutional neural network is a EFFICIECTNETV network.

According to a specific implementation manner of the embodiment of the disclosure, the converting, by a convolutional neural network, an input image into high-level semantic features includes:

Acquiring global features through deformable convolution in an encoder; wherein the encoder is further provided with a multi-layer deformable sparse attention fusion mechanism;

the obvious individual features are parsed by a decoder.

According to a specific implementation of an embodiment of the disclosure, the obtaining the global feature by a deformable convolution in the encoder includes:

and calculating K offsets, and adding the K offsets and the reference points to obtain coordinates of K attention points.

In a second aspect, embodiments of the present disclosure provide a DETR-based tensor first satellite quasi-periodic radiation phenomenon detection system, the system comprising:

The feature extraction module is configured to convert the input image into high-level semantic features through a convolutional neural network to obtain a feature map; and

A prediction module configured to map the final output feature map into a sequence form, and then input to an encoder, and add a position code; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; based on a self-attention mechanism in a decoder, interacting the position in the finally output feature map with an object query vector to obtain the representation information of the feature, mapping the feature to a prediction space through a plurality of full-connection layers, and predicting the category and position of a target;

According to a specific implementation of an embodiment of the disclosure, the system further includes:

a constraint module configured to add physical constraints of proton cyclotron frequency at the time of prediction, comprising:

In a third aspect, embodiments of the present disclosure further provide an electronic device, including:

At least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, which when executed by the at least one processor, cause the at least one processor to implement the DETR-based tensor first satellite quasi-periodic radiation phenomenon detection method of any one of the preceding first aspect or any implementation of the first aspect.

In a fourth aspect, the presently disclosed embodiments also provide a non-transitory computer-readable storage medium storing computer instructions that, when executed by at least one processor, cause the at least one processor to perform the DETR-based method of detecting the quasi-periodic radiation phenomenon of the tensor satellite number one in any of the implementations of the first aspect or the first aspect.

In a fifth aspect, embodiments of the present disclosure also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the DETR-based method of detecting the quasi-periodic radiation phenomenon of the tensor satellite No. one in any of the implementations of the first aspect or the first aspect.

According to the method for detecting the quasi-periodic radiation phenomenon of the tensor first satellite based on the DETR, which is disclosed by the embodiment of the invention, when the problems of large density difference, complex background, targets with different dimensions and the like of the quasi-periodic radiation phenomenon of the DETR are caused, the detection precision and the detection speed are improved, and the method can be deployed in satellite equipment.

Drawings

The foregoing is merely an overview of the present invention, and the present invention is further described in detail below with reference to the accompanying drawings and detailed description.

Fig. 1 is a schematic flow chart of a DETR-based method for detecting quasi-periodic radiation phenomenon of a first satellite in a tensor scale;

Fig. 2 is a flow chart of a DETR-based method for detecting quasi-periodic radiation phenomenon of a first satellite in a tensor scale according to an embodiment of the present disclosure;

fig. 3 is a schematic flow chart of global feature acquisition by an encoder in a transducer of the DETR according to an embodiment of the present disclosure;

fig. 4 is a schematic flow chart of feature extraction by an encoder according to an embodiment of the present disclosure through a multi-layer attention fusion mechanism;

FIG. 5 is a schematic diagram of proton cyclotron frequency provided in an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of results of a test on a test set provided in an embodiment of the present disclosure;

FIG. 7 is a schematic diagram of a result of detection on a completely new data set according to an embodiment of the present disclosure;

fig. 8 is a schematic structural diagram of a DETR-based quasi-periodic radiation phenomenon detection system of a first satellite in a tensor scale according to an embodiment of the present disclosure; and

Fig. 9 is a schematic diagram of an electronic device according to an embodiment of the disclosure.

Detailed Description

Embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.

Other advantages and effects of the present disclosure will become readily apparent to those skilled in the art from the following disclosure, which describes embodiments of the present disclosure by way of specific examples. It will be apparent that the described embodiments are merely some, but not all embodiments of the present disclosure. The disclosure may be embodied or practiced in other different specific embodiments, and details within the subject specification may be modified or changed from various points of view and applications without departing from the spirit of the disclosure. It should be noted that the following embodiments and features in the embodiments may be combined with each other without conflict. All other embodiments, which can be made by one of ordinary skill in the art without inventive effort, based on the embodiments in this disclosure are intended to be within the scope of this disclosure.

It is noted that various aspects of the embodiments are described below within the scope of the following claims. It should be apparent that the aspects described herein may be embodied in a wide variety of forms and that any specific structure and/or function described herein is merely illustrative. Based on the present disclosure, one skilled in the art will appreciate that one aspect described herein may be implemented independently of any other aspect, and that two or more of these aspects may be combined in various ways. For example, an apparatus may be implemented and/or a method practiced using any number of the aspects set forth herein. In addition, such apparatus may be implemented and/or such methods practiced using other structure and/or functionality in addition to one or more of the aspects set forth herein.

In addition, in the following description, specific details are provided in order to provide a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects may be practiced without these specific details.

The embodiment of the invention provides a method for detecting quasi-periodic radiation phenomenon of a Zhangheng first satellite based on DETR, which comprises the following steps: collecting electric field waveform data of ELF wave bands of the EFD data of the tensor satellite, and converting the electric field waveform data to obtain a time-frequency diagram; selecting a quasi-periodic radiation phenomenon with remarkable significance from the time-frequency diagram, and marking the quasi-periodic radiation phenomenon with marked data to form a data set; dividing the data set into a training set and a testing set; building a neural network, inputting the labeling data of the training set into the network for training until the testing requirements of the testing set are met, and obtaining a quasi-periodic radiation intelligent detection model; and detecting the time-frequency diagram to be detected by using the quasi-periodic radiation intelligent detection model to obtain a quasi-periodic radiation detection result.

Detecting the time-frequency diagram to be detected by using the quasi-periodic radiation intelligent detection model to obtain a quasi-periodic radiation detection result, comprising the following steps of:

the image is first input into a convolutional neural network, where EFFICIECTNETV network is selected, whereby the input image can be converted into high-level semantic features for use by the decoder module. A high-Efficiency Channel Attention (ECA) module is then added, which can help us better extract the desired features in a complex context.

After adding position coding, the improved transducer encoder and decoder structure is input, the decoder aligns the position in the feature map with the object query vector through a self-attention mechanism to predict the category and position of the object, and the full connection layer maps the feature vector to the output space of the object detection, and generates the final object detection result through a series of full connection operations, activation functions and parameterized conversion.

A physical constraint on the proton cyclotron frequency is then added at the time of prediction to eliminate some erroneous prediction results.

Fig. 1 is a schematic diagram of a flow of a DETR-based method for detecting quasi-periodic radiation phenomenon of a first satellite in tensor.

Fig. 2 is a flow chart of a DETR-based method for detecting quasi-periodic radiation phenomenon of a first satellite in tensor balance corresponding to fig. 1.

As shown in fig. 1, at step S110, an input image is converted into high-level semantic features by a convolutional neural network, resulting in a feature map.

In the embodiment of the invention, the convolutional neural network is EFFICIECTNETV <2 > network.

In an embodiment of the present invention, the converting, by a convolutional neural network, an input image into a high-level semantic feature includes: acquiring global features through deformable convolution in an encoder; wherein the encoder is further provided with a multi-layer deformable sparse attention fusion mechanism; the obvious individual features are parsed by a decoder.

In an embodiment of the present invention, the obtaining the global feature by deformable convolution in the encoder includes: and calculating K offsets, and adding the K offsets and the reference points to obtain coordinates of K attention points.

More specifically, as shown in fig. 3, the DETR acquires global features through an encoder (Encoder) in the transducer, and the Decoder (Decoder) parses the obvious individual features, so that the detection accuracy of the overlapping target is higher in the target detection process. Deformable convolution (Deformable) is an effective way to focus on sparse space positioning, and the optimal sparse space sampling method of Deformable convolution and the relational modeling capability of a transform are combined, so that convergence of the DETR is accelerated, the complexity of a problem is reduced, and therefore a Deformable attention mechanism is introduced into the model of the invention. A key issue with the original transducer attention is that it will focus on all spatial locations. However, the default attention module only focuses on a small number of key sampling points near the reference point, and calculates K offsets by query, and then adds the K offsets to the reference point to calculate K "coordinates of the attention point" instead of h×w points of the whole feature map (where H is the height of the feature map, W is the width of the feature map, and h×w is all points of the feature map).

As shown in fig. 4, in the model proposed by the present invention, the global sampling adopted by Multi-Head self-Attention in the original Transformer Encoder is replaced by the variability Attention mechanism (Def-Attention) proposed in the performable: a strategy for sparse sampling extracts finer features with less computation.

In addition, the multilayer attention fusion mechanism designed at Encoder can improve the feature learning capability and enhance the feature extraction effect.

More specifically, step S120 is next followed.

Extracting a target detection result from the high-level semantic features based on the efficient channel attention module extraction at step S120; wherein the target detection result includes a position code.

Next, the process goes to step S130.

At step S130, the final output feature map is mapped into a sequence form, then input to an encoder, and position encoding is added; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; and interacting the position in the finally output feature map with the object query vector based on a self-attention mechanism in the decoder to obtain the representation information of the feature, mapping the feature to a prediction space through a plurality of fully connected layers, and predicting the category and the position of the target.

In an embodiment of the invention, the position codes are fixed sine functions.

In an embodiment of the present invention, the object query vectors are a set of learnable vectors, each representing a potential target object or target class; these vectors are typically initialized to random values and then optimized by back propagation during training.

The choice of prediction space is closely related to the characteristics of the dataset. During training, the model will learn to adjust parameters in the prediction space to generate predictions that match the true target class and attributes.

In an embodiment of the present invention, the method further includes: physical constraints of proton cyclotron frequency are added at the time of prediction.

In an embodiment of the present invention, the adding physical constraint of proton cyclotron frequency in prediction includes: calculating proton cyclotron frequency of predicted orbit by observed magnetic field strength based on the following formula _：

Table 1:

more specifically, the proton cyclotron frequency of the predicted orbit was calculated by tensing the magnetic field strength observed by the CDSM of HPM number one.

The lowest end to the uppermost end of the image in the vertical axis direction are respectively 0-2.5kHZ, the latitudinal value is approximately-69 degrees to 69 degrees, the latitudinal value and the frequency value are interpolated to the number of the pixel points, and therefore the position of the image where the pixel points are located can be obtained through the latitudinal value and the frequency value (the proton convolution frequency value of the current latitudinal value can be calculated, and the frequency is reflected on the image to the pixel size of the vertical axis, namely the coordinate y, and the frequency value is the coordinate x). When (x 1, y 1) at the center of the prediction frame is obtained, a verification is made as to whether the value of y1 at the x1 position is greater than the pixel value where the proton cyclotron frequency is located, and if so, the prediction frame is discarded.

The satellite device can obtain a magnetic field strength B, the proton charge q is 1.60217733 ×10 ^-19 C, the proton mass m is 1.672621637 ×10 ^-27 KG, and the red line is the proton cyclotron frequency as shown in fig. 5.

Table 2:

table 2 shows that the invention is obviously superior to other existing algorithms in terms of speed and accuracy of reasoning.

The results of the test on the test set are shown in fig. 6;

Compared with the original DETR, the method provided by the application has the advantage that the accuracy is improved by 1%. In addition, the number of parameters is reduced by 17%, so that the model size is only 34.27M. The model of the application is superior to other algorithms in terms of detection speed, exhibiting similar performance to YOLOv. In summary, these findings support the conclusion that the proposed method provides excellent performance in detecting quasi-periodic radiation events, not only exceeding the prior art in terms of accuracy, but also providing significant detection speeds. The combination of increased accuracy, reduced parameter count, and enhanced detection speed localizes this new model as a robust and efficient solution to the QP event detection task.

In this study, the present application starts with the original DETR model and a comparison experiment was performed on the same dataset at each improvement step. Through these experiments, the performance of the enhanced functions proposed by the present application was evaluated and analyzed. The results obtained from these experiments were measured and compared by using the mAP metric. The mAP-based evaluation verifies that the improvements introduced by the present application effectively enable the model to more accurately extract features of quasi-periodic radiation. By conducting these comparative experiments on consistent datasets, the effectiveness of the proposed modifications in enhancing the model to accurately capture the unique features of quasi-periodic radiation can be demonstrated.

The result of the detection on the new data set is shown in FIG. 7 _.

In order to evaluate the practical applicability of the model of the present application in future scenarios, quasi-periodic radiation event detection was performed on data collected at month 2023, 6 using the automatic detection model proposed by the present application. In 1918 pictures, 164 quasi-periodic radiation events were detected, 87 of which were correctly identified, and 32 quasi-periodic radiation events were not identified. The result of 6 months in 2023 shows that the accuracy is 53.0% and the recall is 81.5%. These findings indicate that the model of the present application accurately identifies most QP events.

Fig. 8 illustrates a DETR-based tensor first satellite quasi-periodic radiation phenomenon detection system 800 provided by the present invention, including a feature extraction module 810, a prediction module 820, and a conversion module 830.

The feature extraction module 810 is configured to convert an input image into high-level semantic features through a convolutional neural network, so as to obtain a feature map; and

The prediction module 820 is used for mapping the final output feature map into a sequence form, inputting the sequence form into an encoder, and adding position codes; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; and interacting the position in the finally output feature map with the object query vector based on a self-attention mechanism in the decoder to obtain the representation information of the feature, mapping the feature to a prediction space through a plurality of fully connected layers, and predicting the category and the position of the target.

In an embodiment of the present invention, the system further includes:

Referring to fig. 9, the disclosed embodiment also provides an electronic device 90, which includes:

At least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the DETR-based method of detecting quasi-periodic radiation phenomenon of a tensor first satellite in the foregoing method embodiments.

The disclosed embodiments also provide a non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the DETR-based tensor first satellite quasi-periodic radiation phenomenon detection method in the foregoing method embodiments.

The disclosed embodiments also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the DETR-based tensor first satellite quasi-periodic radiation phenomenon detection method of the foregoing method embodiments.

Referring now to fig. 9, a schematic diagram of an electronic device 90 suitable for use in implementing embodiments of the present disclosure is shown. The electronic devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and the like, and stationary terminals such as digital TVs, desktop computers, and the like. The electronic device shown in fig. 9 is merely an example, and should not impose any limitations on the functionality and scope of use of embodiments of the present disclosure.

As shown in fig. 9, the electronic device 90 may include a processing means (e.g., a central processor, a graphics processor, etc.) 901, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 902 or a program loaded from a storage means 908 into a Random Access Memory (RAM) 903. In the RAM 903, various programs and data necessary for the operation of the electronic device 90 are also stored. The processing device 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input/output (I/O) interface 905 is also connected to the bus 904.

In general, the following devices may be connected to the I/O interface 905: input devices 906 including, for example, a touch screen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, and the like; an output device 907 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 908 including, for example, magnetic tape, hard disk, etc.; and a communication device 909. The communication means 909 may allow the electronic device 90 to communicate with other devices wirelessly or by wire to exchange data. While an electronic device 90 having various means is shown, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may be implemented or provided instead.

In particular, according to embodiments of the present disclosure, the processes described above with reference to flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method shown in the flowcharts. In such an embodiment, the computer program may be downloaded and installed from a network via the communication device 909, or installed from the storage device 908, or installed from the ROM 902. When executed by the processing device 901, performs the above-described functions defined in the methods of the embodiments of the present disclosure.

It should be noted that the computer readable medium described in the present disclosure may be a computer readable signal medium or a computer readable storage medium, or any combination of the two. The computer readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this disclosure, a computer-readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present disclosure, however, the computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, with the computer-readable program code embodied therein. Such a propagated data signal may take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination of the foregoing. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, fiber optic cables, RF (radio frequency), and the like, or any suitable combination of the foregoing.

The computer readable medium may be contained in the electronic device; or may exist alone without being incorporated into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring at least two internet protocol addresses; sending a node evaluation request comprising the at least two internet protocol addresses to node evaluation equipment, wherein the node evaluation equipment selects an internet protocol address from the at least two internet protocol addresses and returns the internet protocol address; receiving an Internet protocol address returned by the node evaluation equipment; wherein the acquired internet protocol address indicates an edge node in the content distribution network.

Or the computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: receiving a node evaluation request comprising at least two internet protocol addresses; selecting an internet protocol address from the at least two internet protocol addresses; returning the selected internet protocol address; wherein the received internet protocol address indicates an edge node in the content distribution network.

Computer program code for carrying out operations of the present disclosure may be written in one or more programming languages, including an object oriented programming language such as Java, smalltalk, C ++ and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, through the Internet using an Internet service provider).

The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units involved in the embodiments of the present disclosure may be implemented by means of software, or may be implemented by means of hardware. The name of the unit does not in any way constitute a limitation of the unit itself, for example the first acquisition unit may also be described as "unit acquiring at least two internet protocol addresses".

It should be understood that portions of the present disclosure may be implemented in hardware, software, firmware, or a combination thereof.

The foregoing is merely specific embodiments of the disclosure, but the protection scope of the disclosure is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the disclosure are intended to be covered by the protection scope of the disclosure. Therefore, the protection scope of the present disclosure shall be subject to the protection scope of the claims.

Claims

1. The method for detecting the quasi-periodic radiation phenomenon of the first satellite of Zhangheng based on DETR is characterized by comprising the following steps:

2. The DETR-based method for detecting quasi-periodic radiation phenomenon of a tensegrity satellite No. 1, said method further comprising: physical constraints of proton cyclotron frequency are added at the time of prediction.

3. The DETR-based method for detecting quasi-periodic radiation phenomenon of a tensor first satellite according to claim 2, wherein said adding physical constraints of proton cyclotron frequency at the time of prediction comprises:

4. The DETR-based method for detecting quasi-periodic radiation phenomenon of a tensor first satellite according to claim 1, wherein the convolutional neural network is EFFICIECTNETV network.

5. The DETR-based method for detecting quasi-periodic radiation phenomenon of a tensor first satellite according to claim 1, wherein said converting the input image into high-level semantic features through the convolutional neural network comprises:

Acquiring global features through deformable convolution in an encoder; wherein the encoder is further provided with a multi-layer deformable sparse attention fusion mechanism.

6. The DETR-based tensegrity detection method of satellite No. one quasiperiodic radiation phenomenon according to claim 5, wherein said obtaining global features by deformable convolution in the encoder comprises:

7. A DETR-based tensegrity satellite number one quasi-periodic radiation phenomenon detection system, the system comprising:

a prediction module configured to map the final output feature map into a sequence form, and then input to an encoder, and add a position code; the encoder globally models the sequence mapped by the feature map through a multi-layer self-attention mechanism and a feedforward neural network; and interacting the position in the finally output feature map with the object query vector based on a self-attention mechanism in the decoder to obtain the representation information of the feature, mapping the feature to a prediction space through a plurality of fully connected layers, and predicting the category and the position of the target.

8. The DETR-based tensegrity satellite one quasiperiodic radiation phenomenon detection system of claim 7, further comprising:

9. An electronic device, comprising:

At least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, which when executed by the at least one processor, cause the at least one processor to perform the DETR-based tensor first satellite quasi-periodic radiation phenomenon detection method according to any one of claims 1 to 6.

10. A non-transitory computer-readable storage medium storing computer instructions that, when executed by at least one processor, cause the at least one processor to perform the DETR-based tensor first satellite quasi-periodic radiation phenomenon detection method according to any one of claims 1 to 6.