CN-121997147-A - Unmanned aerial vehicle identification method, unmanned aerial vehicle identification device, electronic equipment and storage medium

CN121997147ACN 121997147 ACN121997147 ACN 121997147ACN-121997147-A

Abstract

The invention relates to the field of artificial intelligence, and provides an unmanned aerial vehicle identification method, an unmanned aerial vehicle identification device, electronic equipment and a storage medium, wherein the method comprises the steps of obtaining electromagnetic signals of an unmanned aerial vehicle to be identified; the unmanned aerial vehicle recognition method based on the electromagnetic signal comprises the steps of extracting unmanned aerial vehicle features corresponding to the electromagnetic signal based on a feature extraction model, wherein the feature extraction model comprises a feature extraction module, the feature extraction module is used for carrying out state space modeling by applying time sequence features and space features of the electromagnetic signal, long-range features obtained by applying the state space modeling are used for constructing unmanned aerial vehicle features, matching the unmanned aerial vehicle features with registration features of each unmanned aerial vehicle in an unmanned aerial vehicle library, determining recognition results of unmanned aerial vehicles to be recognized based on the matching results, and the registration features are obtained based on the feature extraction model. The method, the device, the electronic equipment and the storage medium provided by the invention ensure the full fusion of the information of the electromagnetic signals in the time domain and the frequency domain, thereby realizing the efficient and accurate unmanned aerial vehicle identification in the urban low-altitude complex frequency spectrum environment.

Inventors

JIANG JUN
WANG JIANSHE
NI JIANGFAN
XU MINQIANG
FANG SIAN
LIU LIN
Xie Dacan
SI HUAJIAN

Assignees

合肥讯飞数码科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260407

Claims (11)

1. A method of unmanned aerial vehicle identification, comprising: acquiring an electromagnetic signal of the unmanned aerial vehicle to be identified; The unmanned aerial vehicle characteristic detection system comprises a unmanned aerial vehicle characteristic detection module, a characteristic extraction module, a state space modeling module and a control module, wherein the unmanned aerial vehicle characteristic detection module is used for detecting the electromagnetic signal according to the state space modeling module; matching the unmanned aerial vehicle characteristics with registration characteristics of each unmanned aerial vehicle in an unmanned aerial vehicle library, determining an identification result of the unmanned aerial vehicle to be identified based on the matching result, and obtaining the registration characteristics based on the characteristic extraction model.
2. The unmanned aerial vehicle recognition method according to claim 1, wherein the extracting unmanned aerial vehicle features corresponding to the electromagnetic signals based on the feature extraction model comprises: Extracting time sequence characteristics of the electromagnetic signals based on a time sequence extraction unit in the characteristic extraction module; Extracting spatial features of a frequency domain representation of the electromagnetic signal based on a frequency domain extraction unit in the feature extraction module; Based on a state space unit in the feature extraction module, carrying out state space modeling on the time sequence feature and the space feature to obtain the long-range feature; Based on a feature output unit in the feature extraction module, combining the frequency domain representation and the long-range features to construct high-level features; and constructing the unmanned aerial vehicle features by applying the high-level features based on an output module in the feature extraction model.
3. The unmanned aerial vehicle recognition method of claim 2, wherein the constructing a high-level feature based on the feature output unit in the feature extraction module, in combination with the frequency domain representation and the long-range feature, comprises: And extracting residual features of the frequency domain representation based on a feature output unit in the feature extraction module, and constructing the high-level features by combining the long-range features, the residual features and the frequency domain representation.
4. The unmanned aerial vehicle recognition method of claim 2, wherein the feature extraction model comprises a plurality of cascaded feature extraction modules, and the high-level features output by the former feature extraction module are electromagnetic signals input by the latter feature extraction module.
5. The unmanned aerial vehicle identification method of claim 4, wherein the constructing the unmanned aerial vehicle feature using the high-level feature based on the output module in the feature extraction model comprises: And based on the output modules in the feature extraction model, fusing the high-level features output by each of the plurality of cascaded feature extraction modules to obtain the unmanned aerial vehicle features.
6. The unmanned aerial vehicle identification method of any of claims 1 to 5, wherein the training step of the feature extraction model comprises: Acquiring a sample electromagnetic signal of a sample unmanned aerial vehicle; Based on an initial model, extracting sample unmanned aerial vehicle characteristics of the sample electromagnetic signals; and carrying out parameter iteration on the initial model based on the differences among the characteristics of the sample unmanned aerial vehicles belonging to the same sample unmanned aerial vehicle and the differences among the characteristics of the sample unmanned aerial vehicles belonging to different sample unmanned aerial vehicles to obtain the characteristic extraction model.
7. The unmanned aerial vehicle identification method of claim 6, wherein the sample electromagnetic signal comprises an original electromagnetic signal and an electromagnetic signal that data-augments the original electromagnetic signal.
8. The unmanned aerial vehicle identification method of any of claims 1 to 5, further comprising: acquiring an electromagnetic signal of the unmanned aerial vehicle to be registered; Extracting unmanned aerial vehicle features corresponding to electromagnetic signals of the unmanned aerial vehicle to be registered based on the feature extraction model; determining registration characteristics of the unmanned aerial vehicle to be registered based on unmanned aerial vehicle characteristics corresponding to electromagnetic signals of the unmanned aerial vehicle to be registered; and storing the registration characteristics of the unmanned aerial vehicle to be registered and the unmanned aerial vehicle identification into the unmanned aerial vehicle library.
9. An unmanned aerial vehicle recognition device, characterized by comprising: the acquisition unit is used for acquiring electromagnetic signals of the unmanned aerial vehicle to be identified; The system comprises an inference unit, a characteristic extraction module and a control unit, wherein the inference unit is used for extracting unmanned aerial vehicle characteristics corresponding to the electromagnetic signals based on a characteristic extraction model, the characteristic extraction model comprises a characteristic extraction module, the characteristic extraction module is used for carrying out state space modeling by applying time sequence characteristics and space characteristics of the electromagnetic signals, and constructing the unmanned aerial vehicle characteristics by applying long-range characteristics obtained by the state space modeling; The unmanned aerial vehicle identification device comprises an identification unit, a feature extraction model and a feature extraction model, wherein the identification unit is used for matching the unmanned aerial vehicle features with the registration features of each unmanned aerial vehicle in the unmanned aerial vehicle library, determining the identification result of the unmanned aerial vehicle to be identified based on the matching result, and obtaining the registration features based on the feature extraction model.
10. An electronic device comprising a memory, a processor and a computer program stored on the memory and running on the processor, characterized in that the processor implements the unmanned aerial vehicle identification method of any of claims 1 to 8 when the computer program is executed.
11. A non-transitory computer readable storage medium, having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the drone identification method of any of claims 1 to 8.

Description

Unmanned aerial vehicle identification method, unmanned aerial vehicle identification device, electronic equipment and storage medium Technical Field The invention relates to the technical field of artificial intelligence, in particular to an unmanned aerial vehicle identification method, an unmanned aerial vehicle identification device, electronic equipment and a storage medium. Background With the rapid popularization of Unmanned aerial vehicles (un-manned AERIAL VEHICLES, UAV), the importance of maintaining airspace, infrastructure and public privacy security is increasing for the detection and identification of Unmanned aerial vehicles. Currently, detection and recognition methods for unmanned aerial vehicles mainly include radar-based methods, visual image-based methods, sound-based methods, and Remote identity (Remote Identification, remote ID) based methods. However, in the low-altitude complex spectrum environment of the city, the radar-based method is easily interfered by multipath effect and ground clutter caused by dense buildings, the visual image-based method is easily submerged by abundant background noise of the city due to illumination change, bad weather and visual shielding, and the remote identity-based method is used as a cooperative monitoring technology and cannot cope with a non-cooperative unmanned aerial vehicle refusing broadcasting or technical tampering signals. Under such circumstances, how to realize effective detection and identification for unmanned aerial vehicles in a complex urban environment is still a problem to be solved in the art. Disclosure of Invention The invention provides an unmanned aerial vehicle identification method, an unmanned aerial vehicle identification device, electronic equipment and a storage medium, which are used for solving the defects that in the related technology, the unmanned aerial vehicle identification technology is very easy to be interfered and the unmanned aerial vehicle cannot be dealt with in a complex urban environment. The invention provides an unmanned aerial vehicle identification method, which comprises the following steps: acquiring an electromagnetic signal of the unmanned aerial vehicle to be identified; The unmanned aerial vehicle characteristic detection system comprises a unmanned aerial vehicle characteristic detection module, a characteristic extraction module, a state space modeling module and a control module, wherein the unmanned aerial vehicle characteristic detection module is used for detecting the electromagnetic signal according to the state space modeling module; matching the unmanned aerial vehicle characteristics with registration characteristics of each unmanned aerial vehicle in an unmanned aerial vehicle library, determining an identification result of the unmanned aerial vehicle to be identified based on the matching result, and obtaining the registration characteristics based on the characteristic extraction model. According to the unmanned aerial vehicle identification method provided by the invention, the unmanned aerial vehicle features corresponding to the electromagnetic signals are extracted based on the feature extraction model, and the unmanned aerial vehicle identification method comprises the following steps: Extracting time sequence characteristics of the electromagnetic signals based on a time sequence extraction unit in the characteristic extraction module; Extracting spatial features of a frequency domain representation of the electromagnetic signal based on a frequency domain extraction unit in the feature extraction module; Based on a state space unit in the feature extraction module, carrying out state space modeling on the time sequence feature and the space feature to obtain the long-range feature; Based on a feature output unit in the feature extraction module, combining the frequency domain representation and the long-range features to construct high-level features; and constructing the unmanned aerial vehicle features by applying the high-level features based on an output module in the feature extraction model. According to the unmanned aerial vehicle identification method provided by the invention, the feature output unit based on the feature extraction module is used for constructing high-level features by combining the frequency domain representation and the long-range features, and the unmanned aerial vehicle identification method comprises the following steps: And extracting residual features of the frequency domain representation based on a feature output unit in the feature extraction module, and constructing the high-level features by combining the long-range features, the residual features and the frequency domain representation. According to the unmanned aerial vehicle identification method provided by the invention, the feature extraction model comprises a plurality of cascaded feature extraction modules, and the high-level features output by the former feature extraction mod