Classification method and device based on radar target detection, equipment and medium

Abstract

The invention relates to the technical field of target detection, in particular to a classification method and device based on radar target detection, equipment and a medium. The method comprises the steps of obtaining returned radar signals, obtaining N signal segments, sequentially detecting each signal segment, determining time corresponding to a signal of target displacement starting as first time, determining time corresponding to a signal of target displacement ending as second time, and determining a target signal segment set according to the first time and the second time, and extracting a fusion feature corresponding to the target signal fragment set, and performing classification prediction on the fusion feature through a preset support vector machine classifier to obtain a category result corresponding to a classification prediction value. According to the invention, the target signal segment corresponding to the target is determined, the signal features of the target signal segment are extracted, and the signal features are classified by using the preset classifier, and only the target signal segment needs to be processed and classified, so that the processing of all signal segments is avoided, and the use power consumption of the radar is reduced.

Description

Technical Field

[0001] The present invention relates to the field of target detection technology, and in particular to a classification method, device, equipment and medium based on radar target detection. Background Art

[0002] With the development and progress of science and technology, the production and life of modern society are becoming more and more intelligent. There is an increasing demand for personnel detection, tracking and identification technologies in various fields such as smart homes, smart buildings, smart medical care, and smart cities. By detecting the status of people in the scene (location, movement, identity, etc.), it can effectively link access control, lighting, air conditioning and other electrical and electronic equipment, providing reliable technical support for the realization of intelligent production and life.

[0003] In the prior art, the use of low-power Doppler radar for intelligent detection has been widely used, such as: smart hotels that check whether there are people, smart elderly care that helps the elderly, smart lighting that detects whether there are people, and smart security that detects whether there are alarms. However, the output signal of the radar is a complex time series data. When a target moves within the monitoring range, the power consumption of the radar will change, and the phase of the radar signal will also change according to the direction of movement. Many features can be extracted from these changes, which increases the burden of target detection classification and the functional consumption of the radar. Therefore, in the classification process of low-power radar target detection, how to reduce power consumption has become an urgent problem to be solved. Summary of the invention

[0004] Based on this, it is necessary to provide a classification method, device, equipment and storage medium based on radar target detection to address the above-mentioned technical problems, so as to solve the problem of high functional consumption of radar.

[0005] A first aspect of an embodiment of the present application provides a classification method based on radar target detection, the classification method comprising: Acquire the returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1; Detecting each signal segment in turn to obtain a signal representing a target start displacement and a signal representing a target end displacement, and determining that a time corresponding to the signal representing the target start displacement is a first time and a time corresponding to the signal representing the target end displacement is a second time; According to the first time and the second time, determining M signal segments between the first time and the second time as a target signal segment set; Extracting signal features of each signal segment in the target signal segment set, and performing feature fusion on the signal features of each signal segment in the target signal segment set to obtain fusion features corresponding to the target signal segment set; The fusion feature is classified and predicted by a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and a category result corresponding to the classification prediction value is obtained based on the classification prediction value.

[0006] A second aspect of an embodiment of the present application provides a classification device based on radar target detection, the classification device comprising: An acquisition module is used to acquire the returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1; A target detection module is used to detect each signal segment in turn, obtain a signal representing a target start displacement and a signal representing a target end displacement, and determine that a time corresponding to the signal representing the target start displacement is a first time and a time corresponding to the signal representing the target end displacement is a second time; a target signal segment set determining module, configured to determine, according to the first time and the second time, M signal segments between the first time and the second time as a target signal segment set; A feature extraction module, used to extract signal features of each signal segment in the target signal segment set, perform feature fusion on the signal features of each signal segment in the target signal segment set, and obtain fusion features corresponding to the target signal segment set; The classification module is used to perform classification prediction on the fusion feature through a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and obtain a category result corresponding to the classification prediction value based on the classification prediction value.

[0007] In a third aspect, an embodiment of the present invention provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the classification method as described in the first aspect when executing the computer program.

[0008] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the classification method as described in the first aspect is implemented.

[0009] Compared with the prior art, the present invention has the following beneficial effects: Acquire the returned radar signal, perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size, obtain N signal segments, N is an integer greater than 1, detect each signal segment in turn, obtain a signal characterized by the start displacement of the target and a signal characterized by the end displacement of the target, determine the time corresponding to the signal of the start displacement of the target as the first time, determine the time corresponding to the signal of the end displacement of the target as the second time, determine the M signal segments between the first time and the second time as the target signal segment set according to the first time and the second time, extract the signal feature of each signal segment in the target signal segment set, perform feature fusion on the signal feature of each signal segment in the target signal segment set, obtain the fusion feature corresponding to the target signal segment set, classify and predict the fusion feature by a preset support vector machine classifier, obtain the classification prediction value corresponding to the fusion feature, and obtain the category result corresponding to the classification prediction value according to the classification prediction value. In the present invention, according to the phase change value in the motion displacement process, target detection is performed, the target signal segment corresponding to the target is determined, the signal feature of the target signal segment is extracted, and the signal feature is classified using a preset classifier, and only the target signal segment needs to be processed and classified, avoiding the processing of all signal segments, thereby reducing the power consumption of the radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1 This is a schematic diagram of an application environment of a classification method based on radar target detection provided by an embodiment of the present invention; Figure 2 It is a flow chart of a classification method based on radar target detection provided by one embodiment of the present invention; Figure 3 A schematic diagram of a classification method based on radar target detection provided by an embodiment of the present invention Figure 4 It is a structural schematic diagram of a classification device based on radar target detection provided by an embodiment of the present invention; Figure 5 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0014] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0015] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]", depending on the context.

[0016] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0017] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0018] The embodiments of the present invention can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0019] AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0020] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0021] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0022] A classification method based on radar target detection provided by an embodiment of the present invention can be applied in the following aspects: Figure 1 In an application environment, a client communicates with a server. The client includes, but is not limited to, a PDA, a desktop computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), and other computer devices. The server can be implemented with an independent server or a server cluster generated by multiple servers.

[0023] See also Figure 2 , is a flow chart of a classification method based on radar target detection provided by an embodiment of the present invention. The classification method based on radar target detection can be applied to Figure 1 The server in the above connects to the corresponding client, such as Figure 2 As shown, the classification method based on radar target detection may include the following steps.

[0024] S201: Acquire a returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments.

[0025] In step S201, the returned radar signal is subjected to windowing and framing processing, which can effectively improve the focusing of the signal. The radar signal is subjected to windowing and framing processing to obtain N signal segments, where N is an integer greater than 1.

[0026] In this embodiment, the returned radar signal is processed, and the sidelobe amplitude of the point spread function is generally suppressed by "windowing" processing, so that the obtained signal looks smoother and more focused at each scattering point. Conventional window functions generally have "symmetrical" waveforms and "bell-shaped" smoothing characteristics, such as Hanning windows, Hamming windows, triangular windows, etc. Due to the sharply reduced sidelobe levels, they can effectively improve the focusing of the image. According to the preset time threshold and the preset step size, the radar signal is windowed and framed to obtain N signal segments.

[0027] It should be noted that, in this embodiment, the radar may be a synthetic aperture radar (SAR) or an inverse synthetic aperture radar (ISAR), which is realized by transmitting multiple groups of large time-width-bandwidth signals to achieve high-resolution distance, and by the relative motion between the radar and the target to achieve high-resolution azimuth. In this embodiment, radar echo signals are collected according to the signal transceiver mechanism of the radar imaging system. Assuming that the radar transmits multiple observation sequences within the observation accumulation time, and the number of sampling points of a single observation sequence is multiple, the acquired radar echo signals can be constructed into a two-dimensional discrete echo data matrix with a data scale according to the distance-azimuth dimension.

[0028] Since the transmitted signal has a certain degree of broadening during radar imaging, and the first-order or high-order relative motion between the radar and the target introduces a Doppler frequency in the echo spectrum, the radar echo signal often appears as a high-order polynomial phase signal (PPS). In this embodiment, after constructing the echo data matrix, a translation compensation factor is constructed according to the echo data matrix, and the echo data matrix is ​​subjected to motion compensation processing according to the translation compensation factor to obtain the compensated echo data matrix. The windowing function is determined according to the preset time threshold and the preset step size requirement, and the compensated echo data matrix is ​​subjected to time domain windowing processing according to the windowing function to obtain N signal segments.

[0029] S202: Detect each signal segment in turn to obtain a signal representing a target start displacement and a signal representing a target end displacement, and determine that the time corresponding to the signal of the target start displacement is a first time and the time corresponding to the signal of the target end displacement is a second time.

[0030] In step S202, by detecting the acquired radar return signal, the start displacement signal and the end displacement signal when the target passes through the radar detection area are determined, and according to the corresponding target start displacement signal and target end displacement signal, the time corresponding to the signal is determined, thereby determining the first time and the second time.

[0031] In this embodiment, the determination of the existence of the target needs to analyze the target data information contained in the echo, and there are usually other interference information such as noise and clutter in the echo, and compete with the target echo. Once it is determined that the echo contains a target signal, the speed, azimuth and radial distance of its current movement can be further determined by analysis. Due to the complexity of the radar echo composition, it is often difficult to analyze the signal composition through a single time-frequency domain method. In this embodiment, the constant false alarm detection technology is used to detect the corresponding target. Usually, the constant false alarm detector in the constant false alarm detection technology estimates the statistical characteristics of the interference through the received echo data itself, and compares the amplitude of the complex signal, the square of the amplitude, etc. (expressed as E) with the threshold T to obtain the final result, thereby obtaining a signal characterized as the start displacement of the target and a signal characterized as the end displacement of the target. After detecting the corresponding target, the time corresponding to the signal of the start displacement of the target is determined according to the detected signal as the first time and the time corresponding to the signal of the end displacement of the target is the second time.

[0032] It should be noted that in the process of radar target detection, the response echo obtained by the output end of the receiver is usually a composite signal after the superposition of components such as target, clutter and noise. The noise includes external noise such as the sun and the Milky Way and the receiver thermal noise which usually occupies an absolute advantage, while clutter refers to echo components that are useless for radar target detection, such as echoes scattered by high-rise buildings, trees, land, clouds, and sea. The existence of noise and clutter inevitably affects the accuracy of target detection. In the ideal case where the signal-to-noise ratio of the unit to be detected is known, the existence of the target can be judged by threshold detection based on a fixed threshold, in which the interference level is usually assumed to be a fixed constant. However, in the actual working environment of the radar, parameters such as the interference level and the signal-to-noise ratio are unpredictable, which greatly limits the application of fixed threshold detection and the control of false alarm probability. The constant false alarm detection technology is an improvement on the fixed threshold. It follows the Newman-Pearson criterion and dynamically estimates the threshold value through the clutter samples around the unit to be detected, thereby maximizing the detection probability of the target while maintaining a certain false alarm probability.

[0033] Optionally, detecting each signal segment in sequence to obtain a signal representing a target start displacement and a signal representing a target end displacement includes: Detect the phase change value in each signal segment in turn, and when the phase change value in the continuous signal segments is greater than a preset threshold, extract the continuous signal segments as the signal segments corresponding to the target; The first signal segment in the continuous signal segments is characterized as a signal of the start displacement of the target, and the last signal segment in the continuous signal segments is characterized as a signal of the end displacement of the target.

[0034] In this embodiment, the echo of the fixed target is usually stronger, so that the echo of the dynamic target is "swamped" in it, so it is necessary to try to distinguish the two. Because the Doppler frequency in the fixed target echo is zero, and the Doppler frequency contained in the clutter of the moving target is concentrated above zero frequency, after their echoes are phase-detected, the phase of the output signal will change slowly over time, which is reflected in the amplitude that changes slowly over time. On the contrary, after the dynamic target echo is output through phase detection, its phase changes greatly over time, which is reflected in the amplitude that changes over time. Therefore, if the phase detection output of the same distance unit in adjacent repetition cycles is subtracted, the fixed target echo will be completely cancelled, and the dynamic target echo will be retained, that is, the dynamic target signal is obtained.

[0035] In this embodiment, when the phase change value in a continuous signal segment is greater than a preset threshold, the continuous signal segment is extracted as a signal segment corresponding to the target, the first signal segment in the continuous signal segment is characterized as a signal for the start of target displacement, and the last signal segment in the continuous signal segment is characterized as a signal for the end of target displacement.

[0036] S203: According to the first time and the second time, determine M signal segments between the first time and the second time as a target signal segment set.

[0037] In step S203, the target signal segment set is the echo signal obtained when the moving target enters the radar monitoring range.

[0038] In this embodiment, the acquired echo signal is windowed to obtain N signal segments. Different signal segments correspond to time periods. In order to reduce radar power consumption, when processing different signal segments, only the signal segments containing the target need to be processed. Therefore, the M signal segments between the first time and the second time are determined as a target signal segment set, and the signal segments in the target signal segment set are processed.

[0039] According to the first time and the second time, determining M signal segments between the first time and the second time as a target signal segment set includes: Calculate the time difference between the first time and the second time according to the first time and the second time; When the time difference is greater than a preset time limit threshold, a time period equal to the preset time limit threshold is extracted from the first time and the second time, and the signal segments in the time period are determined as a target signal segment set.

[0040] In this embodiment, in order to avoid errors in target detection caused by non-target echo signals, the time difference corresponding to the first time and the second time is calculated based on the first time and the second time. When the time difference is greater than the preset time limit threshold, the number of signal segments that need to be processed will increase. In order to improve the response speed of the radar, when the time difference is greater than the preset time limit threshold, a time period equal to the preset time limit threshold is extracted from the first time and the second time, and the signal segments in the time period are determined as the target signal segment set.

[0041] S204: extracting signal features of each signal segment in the target signal segment set, and performing feature fusion on the signal features of each signal segment in the target signal segment set to obtain fusion features corresponding to the target signal segment set.

[0042] In step S204, the signal features of each signal segment are extracted, and the signal features include the corresponding signal amplitude, signal phase and spectrum features. When the signal features of each signal segment in the target signal segment set are fused, the signal features of each signal segment are spliced ​​to obtain the corresponding fused features.

[0043] In this embodiment, the signal features of each signal segment are extracted, and the signal features include the corresponding signal amplitude, signal phase and spectrum features. The signal features of the following formula include: in, represents the signal amplitude during period t, represents the signal phase during time period t, Represents the spectrum characteristics of the time period t. When extracting the corresponding spectrum characteristics, the signal spectrum of each signal segment can be normalized to obtain the sequence length of the signal spectrum. In specific implementation, it can be defined as a rectangular signal sequence, a circular signal sequence, a triangular signal sequence, etc. The signal sequences of the two groups of reference signals are calculated using the symmetric Holder coefficient technology to obtain the high-order spectrum characteristics of each signal segment. Among them, the Holder coefficient technology is an existing technology. The symmetric Holder coefficient is evolved from the resemblance coefficient, which simplifies the resemblance coefficient and reduces the algorithm complexity. Compared with the resemblance coefficient, it can better describe the difference between the two signals. Similarly, the symmetric Holder coefficient technology is used to extract the high-order spectrum characteristics multiple times to obtain several groups of spectrum characteristics of the signal spectrum of each signal segment. Several groups of spectrum characteristics are combined to obtain the spectrum feature set spectrum characteristics of the radar signal.

[0044] Optionally, extracting the signal feature of each signal segment in the target signal segment set, and performing feature fusion on the signal feature of each signal segment in the target signal segment set to obtain a fusion feature corresponding to the target signal segment set, including: Extracting the signal amplitude feature, signal phase feature and signal spectrum feature in each signal segment in the target signal segment set, wherein the signal amplitude feature, signal phase feature and signal spectrum feature constitute the signal feature of each signal segment; The signal amplitude feature, the signal phase feature and the signal spectrum feature in each signal segment are spliced ​​to obtain the splicing feature in each signal segment; The concatenated features in each signal segment in the target signal segment set are fused to obtain the fused features corresponding to the target signal segment set.

[0045] In this embodiment, when obtaining the fusion feature, the splicing features in each signal segment in the target signal segment set are fused, for example, , , are the signal features corresponding to the signal segments in the target signal segment set, respectively , , in, Indicates the signal amplitude of 1 period, Indicates the signal phase for one period of time. represents the spectrum characteristics of 1 period, Indicates the signal amplitude of 2 periods, Indicates the signal phase of 2 periods, represents the spectral characteristics of 2 time periods, Indicates the signal amplitude of 3 periods, Indicates the signal phase of 3 periods, represents the spectral features of the three time periods. When performing feature fusion, , , Perform splicing processing to obtain the corresponding fusion features, as shown below: Among them, f is the fusion feature.

[0046] S205: Perform classification prediction on the fused features through a preset support vector machine classifier to obtain a classification prediction value corresponding to the fused features, and obtain a category result corresponding to the classification prediction value based on the classification prediction value.

[0047] In step S205, a preset support vector machine classifier is used to perform classification prediction on the fused features to obtain a classification prediction value corresponding to the fused features, and a classification result of the corresponding target is determined according to the classification prediction value.

[0048] In this embodiment, the above-mentioned fusion features are input into a preset support vector machine classifier, so that the fusion features are processed by the preset support vector machine classifier for behavioral category recognition, and the category results corresponding to the fusion features are obtained. First, according to the preset support vector machine classifier and the one-to-many classification rule, the above-mentioned fusion features are input into the preset support vector machine classifier, wherein the number of classifiers in the preset support vector machine classifier is the same as the number of categories, and the corresponding output results are obtained in each classifier, and then the classification scores corresponding to each of the above-mentioned output results are calculated respectively according to the above-mentioned output results. After all the above-mentioned classification scores are obtained, the specified results with the largest classification scores are screened out from all the above-mentioned output results, and the categories corresponding to the above-mentioned specified results are used as classification results. Finally, the above-mentioned classification results are determined as corresponding category results. For example, it is assumed that there are 3 categories, which are respectively used as 3 categories. We make three preset support vector machine classifiers, namely (class 1), (class 2), and (class 3). Now we have a fused feature X, which is sent to the three preset support vector machine classifiers respectively. If the classification value of (class 1) is the largest among all the preset support vector machine classifiers, then the fused feature X belongs to class 1.

[0049] A returned radar signal is obtained, and according to a preset time threshold and a preset step size, the radar signal is subjected to windowing and framing processing to obtain N signal segments, where N is an integer greater than 1. Each signal segment is detected in turn to obtain a signal characterized by a start displacement of the target and a signal characterized by an end displacement of the target, and the time corresponding to the signal of the start displacement of the target is determined as a first time, and the time corresponding to the signal of the end displacement of the target is determined as a second time. According to the first time and the second time, M signal segments between the first time and the second time are determined as a target signal segment set, where M is an integer greater than 1 and less than N. A signal feature of each signal segment in the target signal segment set is extracted, and the signal feature of each signal segment in the target signal segment set is subjected to feature fusion to obtain a fusion feature corresponding to the target signal segment set. The fusion feature is classified and predicted by a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and a category result corresponding to the classification prediction value is obtained according to the classification prediction value. In the present invention, target detection is performed according to the phase change value in the motion displacement process, the target signal segment corresponding to the target is determined, the signal characteristics of the target signal segment are extracted, and the signal characteristics are classified using a preset classifier. Only the target signal segment needs to be processed and classified, avoiding the processing of all signal segments, thereby reducing the power consumption of the radar.

[0050] See also Figure 3 , is a flow chart of a classification method based on radar target detection provided by an embodiment of the present invention, such as Figure 3 , the classification method based on radar target detection may include the following steps: S301: Acquire a returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1.

[0051] S302: Detect each signal segment in turn to obtain a signal representing a target start displacement and a signal representing a target end displacement, and determine that the time corresponding to the signal of the target start displacement is a first time and the time corresponding to the signal of the target end displacement is a second time.

[0052] S303: According to the first time and the second time, determine M signal segments between the first time and the second time as a target signal segment set.

[0053] S304: extracting signal features of each signal segment in the target signal segment set, and performing feature fusion on the signal features of each signal segment in the target signal segment set to obtain fusion features corresponding to the target signal segment set.

[0054] Among them, the contents of the above steps S301 to S304 are the same as those of the above steps S201 to S204, and the description of the above steps S201 to S204 may be referred to, and will not be repeated here.

[0055] S305: constructing an initial support vector machine classifier according to the number of existing categories, wherein the number of classifiers in the initial support vector machine classifier is equal to the number of existing categories; S306: Obtain category sample data corresponding to different categories in the existing categories, train the initial support vector machine classifier according to the category sample data, obtain a trained support vector machine classifier, and use the trained support vector machine classifier as a preset support vector machine classifier.

[0056] In this embodiment, a support vector machine classifier is trained to obtain a preset support vector machine classifier, the number of existing categories is obtained, a plurality of initial support vector machine classifiers having the same number as the above existing categories are constructed, category sample data corresponding to different categories in the existing categories are obtained, corresponding training sets are extracted from the category sample data corresponding to different categories, and when constructing the i-th training set, the sample data belonging to the i-th category is marked as a positive training set, and the sample data not belonging to the i-th category is marked as a negative training set, that is, for each category of sample data, the behavior of that category is used as a positive sample, and other sample data is used as a negative sample, and finally, the above positive training set and the negative training set are used to train the corresponding initial support vector machine classifiers to obtain a trained support vector machine classifier, and the trained support vector machine classifier is used as the preset support vector machine classifier. S307: Perform classification prediction on the fused features through a preset support vector machine classifier to obtain a classification prediction value corresponding to the fused features, and obtain a category result corresponding to the classification prediction value based on the classification prediction value.

[0057] Among them, the content of the above step S307 is the same as that of the above step S205, and reference may be made to the description of the above step S205, which will not be repeated here.

[0058] See also Figure 4 , Figure 4 Schematic diagram of a classification device based on radar target detection provided by an embodiment of the present invention. In this embodiment, the terminal includes various units for executing Figures 2 to 3 For details, please refer to the steps in the corresponding embodiment. Figures 2 to 3 as well as Figures 2 to 3 For the convenience of explanation, only the parts related to this embodiment are shown. Figure 4 As shown, the classification device 40 includes: an acquisition module 41, a target detection module 42, a target signal segment set determination module 43, a feature extraction module 44, and a classification module 45.

[0059] The acquisition module 41 is used to acquire the returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1; The target detection module 42 is used to detect each signal segment in turn, obtain a signal representing the start displacement of the target and a signal representing the end displacement of the target, and determine that the time corresponding to the signal of the start displacement of the target is the first time and the time corresponding to the signal of the end displacement of the target is the second time; a target signal segment set determination module 43, configured to determine, according to the first time and the second time, M signal segments between the first time and the second time as a target signal segment set; A feature extraction module 44 is used to extract the signal feature of each signal segment in the target signal segment set, and perform feature fusion on the signal feature of each signal segment in the target signal segment set to obtain a fusion feature corresponding to the target signal segment set; The classification module 45 is used to perform classification prediction on the fused features through a preset support vector machine classifier to obtain a classification prediction value corresponding to the fused features, and obtain a category result corresponding to the classification prediction value according to the classification prediction value.

[0060] Optionally, the classification device 40 further includes: A construction module is used to construct an initial support vector machine classifier according to the number of existing categories, and the number of classifiers in the initial support vector machine classifier is equal to the number of existing categories; The training module is used to obtain category sample data corresponding to different categories in the existing categories, train the initial support vector machine classifier according to the category sample data, obtain the trained support vector machine classifier, and use the trained support vector machine classifier as the preset support vector machine classifier.

[0061] Optionally, the target detection module 42 includes: A detection unit, used to detect the phase change value in each signal segment in turn, and when the phase change value in the continuous signal segments is greater than a preset threshold, extract the continuous signal segments as the signal segments corresponding to the target; The characterization unit is used to characterize the first signal segment in the continuous signal segments as a signal for the target to start displacement, and to characterize the last signal segment in the continuous signal segments as a signal for the target to end displacement.

[0062] Optionally, the target signal segment set determination module 43 includes: A calculation unit, configured to calculate a time difference between the first time and the second time according to the first time and the second time; The judgment unit is used to extract a time period equal to the preset time limit threshold from the first time and the second time when the time difference is greater than the preset time limit threshold, and determine the signal segment in the time period as the target signal segment set.

[0063] Optionally, the feature extraction module 44 includes: An extraction unit, used to extract a signal amplitude feature, a signal phase feature and a signal spectrum feature from each signal segment in the target signal segment set, wherein the signal amplitude feature, the signal phase feature and the signal spectrum feature constitute a signal feature of each signal segment; A splicing unit, used for splicing the signal amplitude feature, the signal phase feature and the signal spectrum feature in each signal segment to obtain a splicing feature in each signal segment; The fusion unit is used to fuse the splicing features in each signal segment in the target signal segment set to obtain the fusion features corresponding to the target signal segment set.

[0064] It should be noted that the information interaction, execution process, etc. between the above-mentioned modules, units, and sub-units are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0065] Figure 5 Schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 5 As shown, the computer device of this embodiment includes: at least one processor ( Figure 5Only one is shown in the figure), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, any of the above-mentioned steps of the classification method based on radar target detection are implemented.

[0066] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that Figure 5 This is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, it may also include a network interface, a display screen, and an input device.

[0067] The processor may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0068] The memory includes a readable storage medium, an internal memory, etc., wherein the internal memory may be the memory of a computer device, and the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The readable storage medium may be a hard disk of a computer device, and in other embodiments, it may also be an external storage device of the computer device, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Further, the memory may also include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of a computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.

[0069] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned device can refer to the corresponding process in the above-mentioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, which can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may at least include: any entity or device capable of carrying computer program code, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disk. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electric carrier signals and telecommunication signals.

[0070] The present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by a computer program product. When the computer program product runs on a computer device, the computer device can implement the steps in the above-mentioned method embodiment when executing.

[0071] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0072] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0073] In the embodiments provided by the present invention, it should be understood that the disclosed devices / computer equipment and methods can be implemented in other ways. For example, the device / computer equipment embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0074] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0075] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A classification method based on radar target detection, characterized in that: The classification method includes: Acquire the returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1; Detecting each signal segment in turn to obtain a signal representing a target start displacement and a signal representing a target end displacement, and determining that a time corresponding to the signal representing the target start displacement is a first time and a time corresponding to the signal representing the target end displacement is a second time; According to the first time and the second time, determining M signal segments between the first time and the second time as a target signal segment set; Extracting signal features of each signal segment in the target signal segment set, and performing feature fusion on the signal features of each signal segment in the target signal segment set to obtain fusion features corresponding to the target signal segment set; The fusion feature is classified and predicted by a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and a category result corresponding to the classification prediction value is obtained based on the classification prediction value.

2. The classification method according to claim 1, characterized in that: The step of sequentially detecting each signal segment to obtain a signal representing a target start displacement and a signal representing a target end displacement includes: Detecting the phase change value in each signal segment in turn, and when the phase change value in the continuous signal segments is greater than a preset threshold, extracting the continuous signal segments as signal segments corresponding to the target; The first signal segment in the continuous signal segments is characterized as a signal of the start of displacement of the target, and the last signal segment in the continuous signal segments is characterized as a signal of the end of displacement of the target.

3. The classification method according to claim 1, characterized in that: The determining, according to the first time and the second time, M signal segments between the first time and the second time as a target signal segment set includes: Calculate the time difference between the first time and the second time according to the first time and the second time; When the time difference is greater than a preset time limit threshold, a time period equal to the preset time limit threshold is extracted from the first time and the second time, and the signal segments in the time period are determined as a target signal segment set.

4. The classification method according to claim 1, characterized in that: The extracting the signal features of each signal segment in the target signal segment set, and fusing the signal features of each signal segment in the target signal segment set to obtain fused features corresponding to the target signal segment set, includes: Extracting a signal amplitude feature, a signal phase feature, and a signal spectrum feature from each signal segment in the target signal segment set, wherein the signal amplitude feature, the signal phase feature, and the signal spectrum feature constitute a signal feature of each signal segment; Splicing the signal amplitude feature, signal phase feature and signal spectrum feature in each signal segment to obtain a splicing feature in each signal segment; The splicing features in each signal segment in the target signal segment set are fused to obtain fused features corresponding to the target signal segment set.

5. The classification method according to claim 1, characterized in that: Before performing classification prediction on the fusion feature by using a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and obtaining a category result corresponding to the classification prediction value according to the classification prediction value, the method further includes: According to the number of the existing categories, construct an initial support vector machine classifier, wherein the number of classifiers in the initial support vector machine classifier is equal to the number of the existing categories; Obtain category sample data corresponding to different categories in the existing categories, train the initial support vector machine classifier according to the category sample data to obtain a trained support vector machine classifier, and use the trained support vector machine classifier as a preset support vector machine classifier.

6. A classification device based on radar target detection, characterized in that: The classification device comprises: An acquisition module is used to acquire the returned radar signal, and perform windowing and framing processing on the radar signal according to a preset time threshold and a preset step size to obtain N signal segments, where N is an integer greater than 1; A target detection module is used to detect each signal segment in turn, obtain a signal representing a target start displacement and a signal representing a target end displacement, and determine that a time corresponding to the signal representing the target start displacement is a first time and a time corresponding to the signal representing the target end displacement is a second time; a target signal segment set determining module, configured to determine, according to the first time and the second time, M signal segments between the first time and the second time as a target signal segment set; A feature extraction module, used to extract signal features of each signal segment in the target signal segment set, perform feature fusion on the signal features of each signal segment in the target signal segment set, and obtain fusion features corresponding to the target signal segment set; The classification module is used to perform classification prediction on the fusion feature through a preset support vector machine classifier to obtain a classification prediction value corresponding to the fusion feature, and obtain a category result corresponding to the classification prediction value based on the classification prediction value.

7. The classification device according to claim 6, characterized in that The classification device also includes: A construction module, used to construct an initial support vector machine classifier according to the number of the existing categories, wherein the number of classifiers in the initial support vector machine classifier is equal to the number of the existing categories; The training module is used to obtain category sample data corresponding to different categories in the existing categories, train the initial support vector machine classifier according to the category sample data, obtain a trained support vector machine classifier, and use the trained support vector machine classifier as a preset support vector machine classifier.

8. The classification device according to claim 6, characterized in that: The target detection module includes A phase detection unit, configured to detect a phase change value in each signal segment in turn, and when the phase change value in a continuous signal segment is greater than a preset threshold, extract the continuous signal segment as a signal segment corresponding to the target; The signal determination unit is used to characterize the first signal segment in the continuous signal segments as a signal for the target to start displacement, and characterize the last signal segment in the continuous signal segments as a signal for the target to end displacement.

9. A computer device, characterized in that: The computer device comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, and the processor implements the classification method according to any one of claims 1 to 5 when executing the computer program.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the classification method according to any one of claims 1 to 5 is implemented.

Images