Target radial length estimation method, device, equipment and medium based on HRRP sequence

By performing specific processing on high-resolution distance image sequences and combining with the bidirectional CFAR threshold method, the problem of low target radial length estimation accuracy under low signal-to-noise ratio is solved, achieving higher recognition accuracy and robustness.

CN118731922BActive Publication Date: 2024-11-22NAT UNIV OF DEFENSE TECH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411235169.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-11-22
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

In the prior art, under low signal-to-noise ratio conditions, the target radial length estimation accuracy is low, and it is difficult to accurately locate the edge scattering point of the support area, resulting in insufficient target recognition accuracy.

Method used

By normalizing the one-dimensional high-resolution distance image sequence, Gaussian filtering smoothing, Sobel operator convolution, non-maximum suppression and dual-threshold detection, combined with the bidirectional CFAR threshold method, the start and end positions of the support area are determined by combining with the bidirectional CFAR threshold method. , estimate the radial length of the target.

Benefits of technology

The accuracy of target radial length estimation is improved, especially under low signal-to-noise ratio conditions, which enhances the detection robustness of edge signals in the support area and significantly improves the accuracy of target recognition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118731922B_ABST
    Figure CN118731922B_ABST
Patent Text Reader

Abstract

The present application relates to a method, device, equipment and medium for estimating the radial length of a target based on an HRRP sequence, wherein a one-dimensional high-resolution range image sequence is normalized, and then the normalized sequence is smoothed and convolved by using Gaussian filtering and a one-dimensional Sobel operator in sequence to obtain a one-dimensional high-resolution range image gradient sequence, and the gradient sequence is subjected to non-maximum suppression, and then processed by a double threshold detection method to obtain a double threshold detection sequence, and the double threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence, and the radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position and the range resolution of the broadband radar system that obtains the one-dimensional high-resolution range image sequence. The present method can effectively solve the problem of poor accuracy in estimating the radial length of a target in the prior art.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of target detection and recognition, and in particular to a method, device, equipment and medium for estimating the radial length of a target based on a HRRP sequence. Background Art

[0002] With the continuous development of radar technology, the upper limit of the working bandwidth of radar equipment has been continuously expanded, and broadband radars capable of capturing high-resolution range profiles (HRRP) have emerged. With the support of high-resolution imaging technology, radar targets are no longer simplified to single scattering points, but are presented as a collection of multiple scattering points, which are closely arranged in the distance dimension to form a series of adjacent peaks and produce the outline of HRRP. Therefore, HRRP is highly correlated with the physical structural characteristics of the target surface, and the target length information, as one of the target separability information, can be used to roughly classify the target or as a feature for target identification. Therefore, it is of great significance to study how to use HRRP to extract the length characteristics of the target.

[0003] Extracting the radial length of the target from the HRRP sequence is essentially a support area edge detection problem for a one-dimensional noisy sequence. In practice, a two-way constant false alarm rate (CFAR) threshold method is often used to distinguish the target signal support area and the noise area. However, the radial length estimation accuracy of the existing methods is low under low signal-to-noise ratio conditions, and it is easy to miss detection when the edge signal of the target support area is weak, making it difficult to accurately locate the scattering points at the edge of the support area. Summary of the invention

[0004] Based on this, it is necessary to provide a target radial length estimation method, device, equipment and medium based on HRRP sequence that can improve the estimation accuracy in order to solve the above technical problems.

[0005] A method for estimating radial length of a target based on HRRP sequence, the method comprising:

[0006] Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0007] The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0008] Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0009] The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0010] The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0011] In one embodiment, the one-dimensional high-resolution range image sequence is normalized using the following formula:

[0012] ;

[0013] In the above formula, and represent the minimum and maximum values ​​in the one-dimensional high-resolution range image sequence, respectively, represents the one-dimensional high-resolution range image sequence, denotes the normalized sequence, the superscript Indicates transpose.

[0014] In one embodiment, the Gaussian filter is used to smooth the normalized sequence using the following formula:

[0015] ;

[0016] in, ;

[0017] In the above formula, represents the standard deviation of the Gaussian distribution, represents a discrete index, Represents a positive integer, which is the Gaussian kernel The size of represents the smoothed sequence.

[0018] In one embodiment, the one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence, using the following formula:

[0019] ;

[0020] In the above formula, express The absolute value of the one-dimensional high-resolution distance image gradient sequence obtained after convolution with the one-dimensional Sobel operator, represents the one-dimensional Sobel operator.

[0021] In one embodiment, non-maximum suppression is performed on the one-dimensional high-resolution range image gradient sequence, which is expressed as:

[0022] .

[0023] In one embodiment, the double threshold detection method is used to process the sequence after non-maximum suppression, using the following formula:

[0024] ;

[0025] In the above formula, and is the preset threshold, represents a dual-threshold detection sequence obtained by dual-threshold detection.

[0026] In one embodiment, the radial length is estimated using the following formula according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence:

[0027] ;

[0028] In the above formula, and Respectively represent the starting position and the ending position of the support area, represents the distance resolution.

[0029] The present application also provides a target radial length estimation device based on HRRP sequence, the device comprising:

[0030] A normalization processing module, used for acquiring a one-dimensional high-resolution range image sequence, and performing normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0031] A smoothing and convolution processing module, used for smoothing the normalized sequence by using Gaussian filtering, and then convolving the smoothed sequence by using a one-dimensional Sobel operator to obtain a one-dimensional high-resolution range image gradient sequence;

[0032] A non-maximum suppression and dual-threshold detection method processing module is used to perform non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and to process the sequence after non-maximum suppression using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0033] A bidirectional CFAR threshold method processing module, used to process the dual threshold detection sequence by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0034] The target radial length estimation module is used to estimate the radial length of the target in the one-dimensional high-resolution range image sequence according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for obtaining the one-dimensional high-resolution range image sequence.

[0035] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0037] The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0038] Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0039] The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0040] The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0041] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0042] Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0043] The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0044] Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0045] The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0046] The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0047] The above-mentioned target radial length estimation method, device, equipment and medium based on HRRP sequence obtains a normalized sequence by normalizing the one-dimensional high-resolution range image sequence, smoothes the normalized sequence by Gaussian filtering, and then obtains a one-dimensional high-resolution range image gradient sequence by convolution operation by a one-dimensional Sobel operator. After non-maximum suppression is performed on the one-dimensional high-resolution range image gradient sequence, the sequence after non-maximum suppression is processed by a double threshold detection method to obtain a double threshold detection sequence. The double threshold detection sequence is processed by a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence. According to the starting position and the ending position of the support area and the range resolution of the broadband radar system that obtains the one-dimensional high-resolution range image sequence, the radial length of the target in the one-dimensional high-resolution range image sequence is estimated. The use of this method can effectively solve the problem of poor accuracy in the estimation of the radial length of the target in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 is a flow chart of a method for estimating radial length of a target based on a HRRP sequence in one embodiment;

[0049] Figure 2 It is a schematic diagram of processed data at each stage in a process of sequentially performing Gaussian filtering, first-order Sobel operator gradient calculation, and non-maximum suppression double threshold detection on a normalized one-dimensional high-resolution range image sequence in one embodiment;

[0050] Figure 3 A schematic diagram of bidirectional CFAR sliding window detection in one embodiment;

[0051] Figure 4 is a schematic diagram of implementation steps of a target radial length estimation method based on HRRP sequence in another embodiment;

[0052] Figure 5 The left and right boundaries obtained by this method and the traditional threshold method under different signal-to-noise ratios in an experiment are shown in Figure 2. Figure 5 (a) shows the left and right boundaries when SNR=26. Figure 5 (b) The left and right boundaries obtained when SNR=1;

[0053] Figure 6 This is a schematic diagram of the comparison between the measured value and the true value obtained by this method and the traditional threshold method in an experiment, where: Figure 6 (a) is a schematic diagram of the relative error between the measured values ​​and the true values ​​obtained by using this method and the traditional threshold method for various different targets. Figure 6 (b) is a schematic diagram showing the comparison of the measured mean and true value obtained by using this method and the traditional threshold method for various targets;

[0054] Figure 7 is a structural block diagram of a target radial length estimation device based on HRRP sequence in one embodiment;

[0055] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0057] In the prior art, when the bidirectional constant false alarm rate (CFAR) threshold method is directly used to estimate the target length of a one-dimensional high-resolution range profile (HRRP), the estimation accuracy is low under low signal-to-noise ratio conditions, and it is easy to miss detection when the edge signal of the target support area is weak, and it is difficult to accurately locate the scattering points at the edge of the support area. Figure 1 As shown, a target radial length estimation method based on HRRP sequence is provided, comprising the following steps:

[0058] Step S100 : acquiring a one-dimensional high-resolution range image sequence, and performing normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence.

[0059] Step S110 , using Gaussian filtering to smooth the normalized sequence, and then using a one-dimensional Sobel operator to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence.

[0060] Step S120, performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression using a dual-threshold detection method to obtain a dual-threshold detection sequence.

[0061] Step S130 , using a bidirectional CFAR threshold method to process the dual threshold detection sequence, to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence.

[0062] Step S140 , estimating the radial length of the target in the one-dimensional high-resolution range image sequence according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0063] In step S100, the one-dimensional high-resolution range image sequence is normalized using the following formula:

[0064] (1)

[0065] In formula (1), and represent the minimum and maximum values ​​in the one-dimensional high-resolution range image sequence, respectively, represents the one-dimensional high-resolution range image sequence, denotes the normalized sequence, the superscript Represents transposition. After normalization, the original HRRP sequence is mapped to Within the range.

[0066] In step S110, the normalized HRRP sequence is first smoothed by using a Gaussian filter to remove noise. The normalized sequence is smoothed by using a Gaussian filter, using the following formula:

[0067] (2)

[0068] in, ;

[0069] In formula (2), represents the standard deviation of the Gaussian distribution, represents a discrete index, Represents a positive integer, which is the Gaussian kernel The size of represents the smoothed sequence.

[0070] The Gaussian filtering process can be expressed as the normalized sequence of the input HRRP: and Gaussian kernel Perform convolution operation, as shown in formula (2).

[0071] Furthermore, the one-dimensional Sobel operator uses discrete differential approximation to calculate the first-order derivative of the signal, thereby determining the slope change characteristics of the signal. It can be defined as follows:

[0072] (3)

[0073] Then the one-dimensional Sobel operator is used to perform convolution operation on the smoothed sequence, which can be expressed as:

[0074] (4)

[0075] In formula (4), express The absolute value of the one-dimensional high-resolution distance image gradient sequence obtained after convolution with the one-dimensional Sobel operator.

[0076] In step S120, non-maximum suppression is performed on the one-dimensional high-resolution range image gradient sequence extracted in step S110. For each signal point in the sequence, the gradient strength of the signal point is compared with that of its adjacent signal points. If the gradient strength of the signal point is the largest, it is retained. Otherwise, it is suppressed and set to 0. Mathematically, it can be expressed as:

[0077] (5)

[0078] Furthermore, the extracted The sequence, i.e. the sequence after non-maximum suppression, is subjected to double threshold detection to determine the final boundary, where the boundary refers to the boundary of the support area of ​​the HRRP sequence, using the following formula:

[0079] (6)

[0080] In formula (6), and The preset threshold can be determined by experimental data statistics. represents a dual-threshold detection sequence obtained by dual-threshold detection.

[0081] like Figure 2 As shown, it is a schematic diagram of the processed data at each stage in the process of Gaussian filtering, first-order Sobel operator gradient calculation and non-maximum suppression double threshold detection on the normalized one-dimensional high-resolution range image sequence.

[0082] In step S130, the bidirectional CFAR threshold method is used to determine the start position and the end position of the signal support area in the dual threshold detection sequence.

[0083] Specifically, yes The sliding boundary extraction is performed on the sequence. Drawing on the idea of ​​radar detection, it is assumed that the boundary unit is , the protection unit is , the reference unit is and , then the length of the entire sliding window can be obtained as or ,and .

[0084] Furthermore, referring to the CFAR idea in radar target detection, the thresholds of the left and right boundaries are set as and , then the left and right boundary determination conditions are set to:

[0085] (7)

[0086] In formula (7), Indicates the HRRP gradient sequence value at the corresponding position, and They represent the mean of the gradient sequence in the reference unit window length. Assume that the false alarm rate is , the number of reference units is , then the thresholds of the left and right boundaries can be determined by the following formula:

[0087] (8)

[0088] In formula (8), The superscript indicates that the above formula is applicable to the calculation of the thresholds on both sides. is the output of the square law detector, which can be obtained as follows:

[0089] (9)

[0090] like Figure 3 As shown in the figure, it is a schematic diagram of bidirectional CFAR sliding window detection, in which the above boundary judgment conditions are used to search the boundary of the gradient sequence of the high-resolution range image from the left and right sides at the same time. After the boundary is determined, the search is stopped immediately, and the left and right boundary positions are recorded respectively. and .

[0091] Finally, in step S140 , the projection length of the target in the radar line of sight direction is obtained according to the radar range resolution.

[0092] Furthermore, according to the starting and ending positions of the support area and the range resolution of the broadband radar system that obtains the one-dimensional high-resolution range image sequence, the radial length is estimated using the following formula:

[0093] (10)

[0094] In formula (10), and Respectively represent the starting and ending positions of the support area, Indicates the distance resolution.

[0095] In another embodiment, the specific implementation steps of the target radial length estimation method based on the HRRP sequence can also be as follows: Figure 4As shown, the method includes the following specific steps: first, the obtained HRRP sequence is normalized; second, the Gaussian filter is applied to smooth the sequence to remove noise; third, the Sobel operator is used to calculate the output sequence gradient; then, non-maximum suppression is performed on the obtained HRRP gradient sequence; then, double threshold detection is performed to further confirm the boundary information; then, the bidirectional CFAR threshold method is used to determine the starting and ending positions of the signal support area; finally, the estimated value of the target radial length is obtained from the radar range resolution.

[0096] In order to verify the effectiveness of the method proposed in this paper (this method), two simulation experiments are carried out to illustrate it.

[0097] Simulation experiment 1 is based on FEKO simulation HRRP data, which is obtained by converting the step frequency echo data within 3MHz~3.3MHz. Figure 5 The detection results of the traditional threshold method and this method under different signal-to-noise ratios are shown in Figure 5 The blue curve in the figure represents the one-dimensional range image. Figure 5 As shown in (a), for an aerial target within 10° of the head or tail, when SNR = 26dB, both methods can extract the correct size; Figure 5 As shown in (b), when SNR=1dB, the traditional threshold method fails, but this method still has a good detection effect, which shows that the detection effect of this method is robust.

[0098] Simulation experiment 2 gives the true value and measured average value of the length of 12 different aerial targets. The relative errors (|measured value - true value| / |true value|) of the traditional threshold method and this method for different targets are shown as follows: Figure 6 As shown in (a), compared with the traditional threshold method, the test results of this method on different targets are robust, with an average relative error of 0.12695, which is lower than the 0.22028 of the comparison method. Figure 6 (b) is a schematic diagram showing the comparison of the predicted values ​​and true values ​​obtained for various targets using this method and the traditional threshold method. Figure 6 (b) shows the deviation between the predicted value and the true value of 12 different aerial target lengths. Figure 6 The dotted line in (b) represents the ideal prediction curve. The position of the triangle is determined by the predicted value and the true value. After calculation, the Pearson correlation coefficient between the predicted value and the true value is R=0.975, which is a good effect.

[0099] In the above-mentioned target radial length estimation method based on HRRP sequence, a normalized sequence is obtained by normalizing a one-dimensional high-resolution range image sequence, a Gaussian filter is used to smooth the normalized sequence, and a one-dimensional Sobel operator is used to perform a convolution operation to obtain a one-dimensional high-resolution range image gradient sequence, and after non-maximum suppression is performed on the one-dimensional high-resolution range image gradient sequence, a double threshold detection method is used to process the sequence after non-maximum suppression to obtain a double threshold detection sequence, and the double threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence, and the radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system that obtains the one-dimensional high-resolution range image sequence. The method can effectively solve the problem of poor accuracy in the estimation of the radial length of the target in the prior art, and experiments have proved that the detection effect of the method is more robust and more accurate than that of the traditional method.

[0100] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0101] In one embodiment, Figure 7 As shown, a target radial length estimation device based on HRRP sequence is provided, comprising: a normalization processing module 200, a smoothing and convolution processing module 210, a non-maximum suppression and dual threshold detection method processing module 220, a bidirectional CFAR threshold method processing module 230 and a target radial length estimation module 240, wherein:

[0102] A normalization processing module 200 is used to obtain a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0103] The smoothing and convolution processing module 210 is used to use Gaussian filtering to smooth the normalized sequence, and then use a one-dimensional Sobel operator to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0104] A non-maximum suppression and dual-threshold detection method processing module 220 is used to perform non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and use a dual-threshold detection method to process the sequence after non-maximum suppression to obtain a dual-threshold detection sequence;

[0105] A bidirectional CFAR threshold method processing module 230 is used to process the dual threshold detection sequence by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0106] The target radial length estimation module 240 is used to estimate the radial length of the target in the one-dimensional high-resolution range image sequence according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for obtaining the one-dimensional high-resolution range image sequence.

[0107] For the specific definition of the target radial length estimation device based on the HRRP sequence, please refer to the definition of the target radial length estimation method based on the HRRP sequence above, which will not be repeated here. Each module in the above-mentioned target radial length estimation device based on the HRRP sequence can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0108] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8 As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a target radial length estimation method based on an HRRP sequence is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0109] Those skilled in the art will understand that Figure 8The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0110] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0111] Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0112] The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0113] Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0114] The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0115] The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0116] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0117] Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence;

[0118] The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence;

[0119] Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence;

[0120] The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence;

[0121] The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

[0122] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0123] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0124] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A target radial length estimation method based on HRRP sequence, characterized in that: The method comprises: Acquire a one-dimensional high-resolution range image sequence, and perform normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence; The normalized sequence is smoothed by using a Gaussian filter, and then a one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence to obtain a one-dimensional high-resolution range image gradient sequence; Performing non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and processing the sequence after non-maximum suppression by using a dual-threshold detection method to obtain a dual-threshold detection sequence; The dual threshold detection sequence is processed by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence; The radial length of the target in the one-dimensional high-resolution range image sequence is estimated according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence.

2. The target radial length estimation method according to claim 1, characterized in that: The one-dimensional high-resolution range image sequence is normalized using the following formula: ; In the above formula, and represent the minimum and maximum values ​​in the one-dimensional high-resolution range image sequence, respectively, represents the one-dimensional high-resolution range image sequence, denotes the normalized sequence, the superscript Indicates transpose.

3. The target radial length estimation method according to claim 2, characterized in that: The Gaussian filter is used to smooth the normalized sequence using the following formula: ; in, ; In the above formula, represents the standard deviation of the Gaussian distribution, represents a discrete index, Represents a positive integer, which is the Gaussian kernel The size of represents the smoothed sequence.

4. The target radial length estimation method according to claim 3, characterized in that: The one-dimensional Sobel operator is used to perform a convolution operation on the smoothed sequence, using the following formula: ; In the above formula, express The absolute value of the one-dimensional high-resolution distance image gradient sequence obtained after convolution with the one-dimensional Sobel operator, represents the one-dimensional Sobel operator.

5. The target radial length estimation method according to claim 4, characterized in that: The non-maximum suppression is performed on the one-dimensional high-resolution range image gradient sequence, which is expressed as: 。 6. The target radial length estimation method according to claim 5, characterized in that: The double threshold detection method is used to process the sequence after non-maximum suppression, using the following formula: ; In the above formula, and is the preset threshold, represents a dual-threshold detection sequence obtained by dual-threshold detection.

7. The target radial length estimation method according to claim 6, characterized in that: According to the starting position and the ending position of the support area and the range resolution of the broadband radar system for acquiring the one-dimensional high-resolution range image sequence, the radial length is estimated using the following formula: ; In the above formula, and Respectively represent the starting position and the ending position of the support area, represents the distance resolution.

8. A target radial length estimation device based on HRRP sequence, characterized in that: The device comprises: A normalization processing module, used for acquiring a one-dimensional high-resolution range image sequence, and performing normalization processing on the one-dimensional high-resolution range image sequence to obtain a normalized sequence; A smoothing and convolution processing module, used for smoothing the normalized sequence by using Gaussian filtering, and then convolving the smoothed sequence by using a one-dimensional Sobel operator to obtain a one-dimensional high-resolution range image gradient sequence; A non-maximum suppression and dual-threshold detection method processing module is used to perform non-maximum suppression on the one-dimensional high-resolution range image gradient sequence, and to process the sequence after non-maximum suppression using a dual-threshold detection method to obtain a dual-threshold detection sequence; A bidirectional CFAR threshold method processing module, used to process the dual threshold detection sequence by using a bidirectional CFAR threshold method to determine the starting position and the ending position of the support area in the one-dimensional high-resolution range image sequence; The target radial length estimation module is used to estimate the radial length of the target in the one-dimensional high-resolution range image sequence according to the starting position and the ending position of the support area and the range resolution of the broadband radar system for obtaining the one-dimensional high-resolution range image sequence.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.