A target new parameter identification method based on identification confidence level evaluation, medium and device
By using a method based on recognition confidence level assessment, the model recognition confidence, number of electromagnetic parameters and confidence ratio of the target waveform are evaluated, which solves the problem of low recognition accuracy of new electromagnetic parameters and improves the model recognition accuracy.
Patent Information
- Application Number
- CN202311339017.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-10-16
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2043-10-16
AI Technical Summary
With the development of radar technology, the accuracy of identifying new electromagnetic parameters has decreased, affecting the radar identification accuracy of electronic reconnaissance systems.
The method based on recognition confidence level assessment first evaluates the model recognition confidence of the target waveform, then evaluates the number of electromagnetic parameters and the confidence ratio of the recognition as the target. When the conditions are met, electromagnetic parameters with recognition confidence within a certain range are used as new electromagnetic parameters to update the target model recognition library or for online learning of the recognition model.
It improved the accuracy of model identification for new electromagnetic parameters of the target.
Smart Images

Figure CN117370928B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of target recognition technology, and more specifically, to a method, medium, and apparatus for identifying new target parameters based on a confidence level assessment. Background Technology
[0002] In the field of signal reconnaissance, radiation source identification is a crucial capability. However, with the development of radar technology, radars often adopt new electromagnetic parameters, leading to a decrease in the accuracy of identification based on these new parameters. This severely impacts the radar identification accuracy of electronic reconnaissance systems. Summary of the Invention
[0003] The present invention aims to provide a method, medium, and device for identifying new target parameters based on confidence level assessment, so as to improve the recognition accuracy.
[0004] This invention provides a method for identifying novel target parameters based on confidence level assessment, comprising:
[0005] First, the confidence level of the target waveform for model identification is evaluated. Second, the number of electromagnetic parameters in the target waveform is evaluated. Third, the proportion of electromagnetic parameters that are identified as the target is evaluated. When all three evaluations meet the conditions, the electromagnetic parameters with a confidence level within a certain range are used as the new electromagnetic parameters of the target. This updates the target model identification library or is used for online learning of the identification model, thereby improving the model identification accuracy of the new electromagnetic parameters of the target.
[0006] Furthermore, assume that there are a total of M electromagnetic parameters in the target waveform of the radiation source, denoted as Para. m There are N target models Tgt (m = 1, 2, ..., M). n (n = 1, 2, ..., N). The model identification result corresponding to each electromagnetic parameter is represented as A. m,n (m = 1, 2, ..., M; n = 1, 2, ..., N), its specific form is as follows:
[0007]
[0008] Assuming that after fusing M electromagnetic parameters, the identification confidence level of each model in the target waveform identification result is B. n (n = 1, 2, ..., N);
[0009] The target new parameter identification method based on confidence level assessment includes the following steps:
[0010] Step 1: Normalize the confidence scores of the N target models for each electromagnetic parameter, and obtain the normalized result.
[0011] Step 2: Set the recognition confidence threshold of the target waveform to ThresB, and search for the recognition confidence B n (n=1,2,…,N) to find the maximum recognition confidence and its corresponding target. Assume the index of the maximum value of the recognition confidence B n (n=1,2,…,N) is Idx, and the corresponding target is Tgt Idx , the recognition confidence of this target is B Idx ; if B Idx <ThresB, it indicates that the recognition confidence of the target waveform is insufficient to support the evaluation processing of new electromagnetic parameters; if B Idx ≥ThresB, proceed to further evaluation according to Step 3;
[0012] Step 3: Set the electromagnetic parameter quantity threshold in the target waveform to ThresNum. If the electromagnetic parameter quantity M in the target waveform is less than ThresNum, it is determined that the electromagnetic parameter quantity is insufficient to support the evaluation processing of new electromagnetic parameters; if M≥ThresNum, proceed to further evaluation according to Step 4;
[0013] Step 4: Set the high confidence threshold of electromagnetic parameters to ThresA High , and the low confidence threshold to ThresA Low ; set the high confidence proportion threshold of electromagnetic parameters to ThresRatio, count the proportion Ratio of A m,Idx ,(m=1,2,…,M) in the electromagnetic parameter recognition results that is greater than ThresA High , if Ratio<ThresRatio, it indicates that the proportion of high-confidence parameters is insufficient to support the recognition processing of new electromagnetic parameters; if Ratio≥ThresRatio, proceed to evaluate and extract new electromagnetic parameters of the target according to Step 6;
[0014] Step 5: Among the electromagnetic parameters, the electromagnetic parameters satisfying the condition A m,Idx >ThresA Low &A m,Idx <ThresA High are taken as the new electromagnetic parameters of the target Tgt Idx ;
[0015] Step 6: The new electromagnetic parameters of the target Tgt Idx are used to update the target recognition library or for online learning of the target recognition model, so as to improve the model recognition accuracy of the new target parameters.
[0016] Further, the calculation formula for the normalization processing is:
[0017] Furthermore, the high confidence threshold ThresA for electromagnetic parameters High ThresA is a low-confidence threshold set by humans based on experience. Low Determined based on the quantity N of the target model.
[0018] The present invention also provides a computer terminal storage medium storing computer terminal executable instructions, which are used to execute the above-described target new parameter identification method based on recognition confidence level assessment.
[0019] The present invention also provides a computing device, comprising:
[0020] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described target novel parameter identification method based on identification confidence level assessment.
[0021] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0022] This invention first evaluates the model identification confidence level of the target waveform, then evaluates it using the number of electromagnetic parameters in the target waveform, and finally evaluates it using the proportion of electromagnetic parameters that are identified as the target with confidence. When all three evaluations meet the conditions, the electromagnetic parameters with identification confidence levels within a certain range are used as the new electromagnetic parameters of the target, thereby updating the target model identification library or using it for online learning of the identification model, thus improving the model identification accuracy of the new electromagnetic parameters of the target. Attached Figure Description
[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.
[0024] Figure 1 This is a flowchart of a target new parameter identification method based on confidence level assessment in an embodiment of the present invention. Detailed Implementation
[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0026] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0027] Example
[0028] like Figure 1 As shown, this embodiment proposes a target new parameter identification method based on a graded evaluation of identification confidence. First, the method evaluates the target waveform's model identification confidence. Second, it evaluates the number of electromagnetic parameters in the target waveform. Third, it evaluates the proportion of electromagnetic parameters that identify the target with confidence. When all three evaluations meet the conditions, electromagnetic parameters with identification confidence within a certain range are used as the new electromagnetic parameters of the target, thereby updating the target model identification library or target model identification model, thus improving the model identification accuracy of the new electromagnetic parameters. Specifically, the method includes the following steps:
[0029] Assume that there are a total of M electromagnetic parameters in the target waveform of the radiation source, denoted as Para. m There are N target models Tgt (m = 1, 2, ..., M). n (n = 1, 2, ..., N). The model identification result corresponding to each electromagnetic parameter is represented as A. m,n (m = 1, 2, ..., M; n = 1, 2, ..., N), its specific form is as follows:
[0030]
[0031] Assuming that after fusing M electromagnetic parameters, the identification confidence level of each model in the target waveform identification result is B. n (n = 1, 2, ..., N).
[0032] Step 1: Normalize the confidence scores of the N target models for each electromagnetic parameter, and obtain the normalized result. The formula for normalization is:
[0033]
[0034] Step 2: Set the recognition confidence threshold of the target waveform as ThresB, and search for the recognition confidence B n (n=1,2,…,N) for the maximum recognition confidence and its corresponding target. Assume that the serial number of the maximum value of the recognition confidence B n (n=1,2,…,N) is Idx, and the corresponding target is Tgt Idx , and the recognition confidence of this target is B Idx . If B Idx <ThresB, it indicates that the recognition confidence of the target waveform is insufficient to support the evaluation processing of new electromagnetic parameters; if B Idx ≥ThresB, further evaluation is performed according to Step 3.
[0035] Step 3: Set the electromagnetic parameter quantity threshold in the target waveform as ThresNum. If the electromagnetic parameter quantity M in the target waveform is less than ThresNum, it is determined that the electromagnetic parameter quantity is insufficient to support the evaluation processing of new electromagnetic parameters; if M≥ThresNum, further evaluation is performed according to Step 4.
[0036] Step 4: Set the high confidence threshold of electromagnetic parameters as ThresA High , and the low confidence threshold as ThresA Low . Wherein, ThresA High is set artificially based on experience, while ThresA Low is determined according to the quantity N of target models, that is, the larger the model quantity, the smaller the low confidence threshold. Set the high confidence proportion threshold of electromagnetic parameters as ThresRatio, count the proportion Ratio of A m,Idx ,(m=1,2,…,M) in the electromagnetic parameter recognition results that is greater than ThresA High . If Ratio<ThresRatio, it indicates that the proportion of high-confidence parameters is insufficient to support the recognition processing of new electromagnetic parameters; if Ratio≥ThresRatio, the evaluation and extraction of new target electromagnetic parameters are performed according to Step 6.
[0037] Step 5: Among the electromagnetic parameters, the electromagnetic parameters that satisfy the condition A m,Idx >ThresA Low &A m,Idx <ThresA High are taken as the new electromagnetic parameters of the target Tgt Idx .
[0038] Step 6: The new electromagnetic parameters of the target Tgt Idx are used to update the target recognition library or for online learning of the target recognition model, so as to improve the model recognition accuracy of the new target parameters.
[0039] Furthermore, in some embodiments, a computer terminal storage medium is proposed, storing computer terminal executable instructions for executing the target novel parameter identification method based on recognition confidence level assessment as described in the preceding embodiments. Examples of computer storage media include magnetic storage media (e.g., floppy disks, hard disks, etc.), optical recording media (e.g., CD-ROMs, DVDs, etc.) or memory such as memory cards, ROMs, or RAMs. The computer storage medium can also be distributed across a network-connected computer system, for example, as an application store.
[0040] Furthermore, in some embodiments, a computing device is proposed, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the target novel parameter identification method based on identification confidence level assessment as described in the foregoing embodiments. Examples of computing devices include PCs, tablet computers, smartphones, or PDAs, etc.
[0041] An example:
[0042] Assume there are a total of 5 electromagnetic parameters in the target waveform of the radiation source, denoted as Para. m There are a total of 3 target models Tgt (m = 1, 2, ..., M). n (n = 1, 2, 3). The model identification result corresponding to each electromagnetic parameter is represented as A. m,n (m = 1, 2, ..., M; n = 1, 2, ..., N), its specific form is as follows:
[0043]
[0044] Assuming that after fusing M electromagnetic parameters, the confidence level of each model in the target waveform identification result is B. n =[0.12,0.83,0.05].
[0045] Step 1: Normalize the confidence scores of the N target models for each electromagnetic parameter to obtain the normalized result. Its value is:
[0046]
[0047] Step 2: Set the recognition confidence threshold for the target waveform to ThresB = 0.8, and find the recognition confidence level B. n The maximum recognition confidence and its target in (n = 1, 2, ..., N), assuming recognition confidence B. nThe maximum value index of (n = 1, 2, ..., N) is Idx = 2, and the corresponding target is Tgt. Idx The confidence level for identifying the target is B. Idx =0.83. Because B Idx If the result is ≥ThresB, then proceed to step 3 for further evaluation.
[0048] Step 3: Set the threshold for the number of electromagnetic parameters in the target waveform to ThresNum = 4. Since M = 5, all M ≥ ThresNum. Then proceed to step 4 for further evaluation.
[0049] Step 4: Set the high confidence threshold for the electromagnetic parameters to ThresA. High =0.7, low confidence threshold is ThresA Low =0.333. Set the high confidence threshold for electromagnetic parameters to ThresRatio = 0.5, and statistically analyze the A values in the electromagnetic parameter identification results. m,Idx (m=1,2,…,M) is greater than ThresA High The ratio Ratio = 0.6. Since Ratio ≥ ThresRatio, the evaluation and extraction of the new electromagnetic parameters of the target are carried out according to step 6.
[0050] Step 5: Among the electromagnetic parameters, those that satisfy condition A m,Idx ThresA Low &A m,Idx <ThresA High The electromagnetic parameters, namely the second and third electromagnetic parameters, are used as the target Tgt. Idx The new electromagnetic parameters.
[0051] Step 6: Move the target Tgt Idx The new electromagnetic parameters are used to update the target recognition library or for online learning of the target recognition model, thereby improving the accuracy of target model recognition based on the new parameters.
[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for identifying novel target parameters based on confidence level assessment, characterized in that, include: First, the confidence level of the target waveform is evaluated. Second, the number of electromagnetic parameters in the target waveform is evaluated. Third, the proportion of electromagnetic parameters that are identified as the target is evaluated. When all three evaluations meet the conditions, the electromagnetic parameters with a confidence level within a certain range are used as the new electromagnetic parameters of the target. This updates the target model identification library or is used for online learning of the identification model, thereby improving the model identification accuracy of the new electromagnetic parameters. Assume that there are a total of M electromagnetic parameters in the target waveform of the radiation source, denoted as: There are a total of N target models. The model identification results corresponding to each electromagnetic parameter are represented as follows: Its specific form is as follows: Assuming that after fusing M electromagnetic parameters, the identification confidence level of each model in the target waveform identification result is: ; The target new parameter identification method based on confidence level assessment includes the following steps: Step 1: Normalize the confidence scores of the N target models for each electromagnetic parameter, and obtain the normalized result. ; Step 2: Set the recognition confidence threshold for the target waveform to [value]. Find the confidence level for identification. The maximum recognition confidence and its target, assuming recognition confidence. The maximum value index is The corresponding target is The confidence level for identifying the target is ;if This indicates that the confidence level of the target waveform identification is insufficient to support the evaluation and processing of new electromagnetic parameters; if Then proceed to step 3 for further evaluation; Step 3: Set the threshold for the number of electromagnetic parameters in the target waveform to [value]. If the number of electromagnetic parameters in the target waveform If the number of electromagnetic parameters is insufficient to support the evaluation and processing of new electromagnetic parameters, then it is considered that the number of electromagnetic parameters is insufficient; if Then proceed to step 4 for further evaluation; Step 4: Set the high confidence threshold for the electromagnetic parameters to... The low confidence threshold is Set the high confidence threshold for electromagnetic parameters to [value]. In the statistical electromagnetic parameter identification results Greater than proportion ,if This indicates that the proportion of high-confidence parameters is insufficient to support the identification and processing of new electromagnetic parameters; if Then proceed to step 6 to evaluate and extract the new electromagnetic parameters of the target. Step 5: Among the electromagnetic parameters, those that meet the conditions Electromagnetic parameters as the target The new electromagnetic parameters; Step 6: Set the target The new electromagnetic parameters are used to update the target recognition library or for online learning of the target recognition model, thereby improving the accuracy of target model recognition based on the new parameters.
2. The target new parameter identification method based on confidence level assessment according to claim 1, characterized in that, The formula for normalization is: .
3. The target new parameter identification method based on confidence level assessment according to claim 1, characterized in that, High confidence threshold for electromagnetic parameters Human-based, experience-based, low-confidence threshold Determined based on the quantity N of the target model. .
4. A computer terminal storage medium storing computer terminal executable instructions, characterized in that, The computer terminal can execute instructions for performing the target new parameter identification method based on recognition confidence level assessment as described in any one of claims 1-3.
5. A computing device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the target new parameter identification method based on identification confidence level assessment as described in any one of claims 1-3.
Citation Information
Patent Citations
Target threat assessment method based on an electronic reconnaissance signal, model and model construction method based on electronic reconnaissance signal
CN109190667A
Radiation source individual information identification method and device, electronic equipment and medium
CN115390016A
