Object detection method and system, electronic equipment and computer readable storage medium

By acquiring and processing channel data between wireless devices, generating characterization vectors and performing support vector model detection, the problem of low detection accuracy of wireless device objects is solved, the detection accuracy is improved and the calculation amount is reduced, and it is suitable for embedded devices. .

CN118965274AInactive Publication Date: 2024-11-15ZHEJIANG DAHUA TECH CO LTD
View PDF 11 Cites 0 Cited by

Patent Information

Application Number
CN202411419535.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-12
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

During object detection, existing wireless devices have low detection accuracy due to the small amount of perceptual data obtained, which cannot meet users' requirements for high bandwidth, low latency and anti-interference capabilities.

Method used

By acquiring the first channel data and the second channel data in the object detection system, combining data fusion and dimensionality reduction processing, a representation vector of channel data is generated and inputted into the target support vector model for detection, improving the accuracy of object detection.

Benefits of technology

It improves the accuracy of object detection and reduces the amount of data calculation, making it suitable for embedded devices with smaller calculation requirements, realizing efficient object detection on wireless devices.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118965274A_ABST
    Figure CN118965274A_ABST
Patent Text Reader

Abstract

The invention discloses an object detection method and system, electronic equipment and a computer readable storage medium, the method is applied to the object detection system, the object detection system comprises a first terminal and a second terminal, the first terminal and the second terminal are arranged in the same target signal area, and the method comprises the following steps: obtaining first channel data and second channel data, the first channel data is formed by the first terminal on the basis of receiving a wireless signal sent by a second terminal within a preset time period, and the second channel data is formed by the second terminal on the basis of the wireless signal sent by the first terminal within the preset time period; determining a representation vector of the channel data based on the first channel data and the second channel data; and inputting the representation vector of the channel data into the target support vector model to obtain a detection result of whether an object exists in the target signal region predicted by the target support vector model. Therefore, the accuracy of object detection is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to an object detection method, system, electronic device and computer-readable storage medium. Background Art

[0002] With the popularity of wireless devices, such as laptops, mobile phones, tablets, smart door locks, cameras and other smart home devices, users have higher and higher requirements for these wireless devices. In addition to supporting high bandwidth, low latency and strong anti-interference capabilities, wireless devices can also support perception functions.

[0003] Currently, directly performing object detection using a small amount of acquired perception data results in low detection accuracy.

[0004] Therefore, there is an urgent need for an object detection method that can improve the accuracy of object detection. Summary of the invention

[0005] The main technical problem solved by the present application is to provide an object detection method, system, electronic device and computer-readable storage medium, which can improve the accuracy of object detection.

[0006] To solve the above technical problems, the present application provides an object detection method, which is applied to an object detection system, wherein the object detection system includes a first terminal and a second terminal, wherein the first terminal and the second terminal are arranged in the same target signal area, and the method includes: acquiring first channel data and second channel data, wherein the first channel data is formed by the first terminal based on receiving a wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within the preset time period; determining a characterization vector of the channel data based on the first channel data and the second channel data; and inputting the characterization vector of the channel data into a target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model.

[0007] In one embodiment, the step of determining a characterization vector of channel data based on the first channel data and the second channel data includes: performing data fusion processing on the first channel data and the second channel data to obtain fused channel data; performing data dimensionality reduction processing on the fused channel data to obtain a characterization vector of the channel data.

[0008] In one embodiment, the step of performing data dimensionality reduction processing on the fused channel data to obtain a characterization vector of the channel data includes: performing principal component extraction processing on the fused channel data to obtain a principal component vector; and using the principal component vector as the characterization vector of the channel data.

[0009] In one embodiment, the method further includes: obtaining a channel data set to be trained, the channel data set to be trained including the channel data to be trained and labeling information of each channel data to be trained, the labeling information representing whether there is an object in the target signal area when the corresponding channel data to be trained is obtained; inputting the channel data set to be trained into a support vector model to be trained to obtain prediction information of each channel data to be trained output by the support vector model to be trained; calculating a loss function between the prediction information and the labeling information of each channel data to be trained; training the support vector model with the goal of minimizing the loss value of the loss function to obtain a target support vector model that meets the requirements.

[0010] In one embodiment, the step of obtaining the first channel data includes: obtaining first original data through the first terminal; performing outlier removal processing on the first original data to obtain processed first original data; and performing low-pass filtering on the processed first original data to obtain the first channel data.

[0011] In one embodiment, the method further includes: if the detection result indicates that there is an object in the target signal area, matching the object with objects in a preset object library to obtain a matching result; in response to the matching result indicating that the object is other objects, reporting is performed, and the other objects are objects outside the preset object library.

[0012] In one embodiment, the method further includes: in response to the matching result characterizing that the object is a target object in the preset object library, obtaining a captured image, wherein the captured image includes the target object; performing matching processing based on the attribute features of the target object in the captured image and the preset attribute features in the preset attribute feature library; if there is a match, reporting the attribute information of the target object.

[0013] To solve the above technical problems, the present application provides an object detection system, which includes: an acquisition module, used to acquire the first channel data and the second channel data, the first channel data is formed by the first terminal based on receiving the wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within the preset time period; a vector determination module, used to determine the characterization vector of the channel data based on the first channel data and the second channel data; a detection module, used to input the characterization vector of the channel data into a target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model.

[0014] In order to solve the above technical problems, the present application provides an electronic device, including a memory and a processor, wherein the memory stores program instructions, and the processor retrieves the program instructions from the memory to execute the above object detection method.

[0015] In order to solve the above technical problems, the present application provides a computer-readable storage medium including program data stored therein, and the program data is used to implement the above object detection method when executed by a processor.

[0016] The above scheme obtains the first channel data and the second channel data, the first channel data is formed by the first terminal based on receiving the wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within a preset time period; determines the characterization vector of the channel data based on the first channel data and the second channel data; inputs the characterization vector of the channel data into the target support vector model to obtain the detection result of whether there is an object in the target signal area predicted by the target support vector model. On the one hand, based on the wireless signals sent by the first terminal and the second terminal, the channel data in the target signal area can be obtained more comprehensively, and then the accuracy of object detection can be improved when performing object detection based on the more comprehensive channel data. On the other hand, performing object detection on the characterization vector of the channel data through the target support vector model can reduce the amount of data calculation, thereby realizing object detection on embedded devices with low computational requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work, among which: Figure 1is a flowchart of an exemplary embodiment of an object detection method shown in the present application; Figure 2 It is a schematic diagram of the effect of the actual application scenario of the object detection method shown in this application; Figure 3 yes Figure 1 FIG. 1 is a flow chart of an exemplary embodiment of step S120 in the object detection method; Figure 4 is a block diagram of an object detection system shown in an exemplary embodiment of the present application; Figure 5 It is a structural schematic diagram of an embodiment of an electronic device provided by the present application; Figure 6 It is a structural schematic diagram of an embodiment of a computer-readable storage medium provided by the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. It will be appreciated that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only some but not all structures related to the present application are shown in the drawings. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the art without making creative work are within the scope of protection of the present application.

[0019] First of all, it should be noted that with the popularity of wireless devices, such as laptops, mobile phones, tablets, smart door locks, cameras and other smart home devices, users have higher and higher requirements for these wireless devices. Wireless devices can support perception functions in addition to supporting high bandwidth, low latency and strong anti-interference capabilities. At present, direct object detection through the acquisition of a small amount of perception data leads to the problem of low detection accuracy.

[0020] Based on this, the present application proposes an object detection method, an electronic device and a computer-readable storage medium. Figure 1 , Figure 1 It is a flowchart of an exemplary embodiment of the object detection method shown in the present application.

[0021] The execution subject of the object detection method may be a terminal device or a server or other processing device, wherein the terminal device may be a user equipment (UE), a computer, a mobile device, a user terminal, a terminal, a cellular phone, a cordless phone, a personal digital assistant (PDA), a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc. The execution subject of the object detection method may also be an object detection system. In some possible implementations, the object detection method may be implemented by a processor calling a computer-readable instruction stored in a memory.

[0022] Specifically, the object detection method of this embodiment includes the following steps: Step S110, acquiring first channel data and second channel data, wherein the first channel data is formed by the first terminal based on receiving a wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within a preset time period.

[0023] The target signal area is provided with a first terminal and a second terminal that communicate with each other. The first terminal may be a transmitter capable of sending and receiving wireless signals, and the second terminal may be a receiver capable of sending and receiving wireless signals. For example, the first terminal may be a gateway terminal, and the second terminal may be a node terminal. The target signal area may be a park, zoo, traffic road, and other scene locations covered by wireless signals. The wireless signal may be an electromagnetic wave signal or a wifi signal, etc. For details, please refer to Figure 2 , Figure 2 A schematic diagram of the actual application scenario effect of the object detection method is shown in FIG. A represents the first terminal, B represents the second terminal, and C represents the object.

[0024] As an example, the first terminal continuously sends a wireless signal within a preset time period. The second terminal receives the wireless signal sent by the first terminal, determines the first channel data based on the received wireless signal, and then sends the first channel data to the object detection system. The object detection system obtains the first channel data. Among them, considering the presence of high-frequency noise and other information in the channel data, the method for the second mobile terminal to determine the first channel data based on the received wireless signal can be that the second mobile terminal forms the first original data based on the received wireless signal, and obtains the processed first original data by performing outlier removal processing on the first original data; then, the processed first original data is subjected to low-pass filtering processing to obtain the first channel data.

[0025] The second terminal continuously sends a wireless signal within a preset time period. The first terminal receives the wireless signal sent by the second terminal, determines the second channel data based on the received wireless signal, and then sends the second channel data to the object detection system. It should be noted that if there is an object in the wireless signal area, the wireless signal sent by the first terminal to the second terminal will be blocked, and the wireless signal will change the propagation path to be transmitted to the second terminal, resulting in changes in the wireless signal received by the second terminal, and then changes in the communication data formed based on the changed wireless signal. Among them, considering the presence of high-frequency noise and other information in the channel data, the method for the first mobile terminal to determine the second channel data based on the received wireless signal can be that the first mobile terminal forms the second original data based on the received wireless signal, and obtains the processed second original data by removing outliers from the second original data; then the processed second original data is low-pass filtered to obtain the second channel data.

[0026] As another exemplary embodiment, the first terminal continuously sends a wireless signal within a preset time period. The second terminal receives the wireless signal sent by the first terminal, and sends the received wireless signal to the object detection system. The object detection system receives the wireless signal sent by the second terminal, and processes the received wireless signal to obtain the first channel data. For the object detection system to receive the wireless signal sent by the second terminal, and process the received wireless signal to obtain the first channel data, it can be that the object detection system forms the first original data based on the received wireless signal, and obtains the processed first original data by performing outlier removal processing on the first original data; then, the processed first original data is subjected to low-pass filtering processing to obtain the first channel data.

[0027] The second terminal continuously sends a wireless signal within a preset time period. The first terminal receives the wireless signal sent by the second terminal, and sends the received wireless signal to the object detection system. The object detection system receives the wireless signal sent by the first terminal, and processes the received wireless signal to obtain the second channel data. For the object detection system to receive the wireless signal sent by the first terminal, and process the received wireless signal to obtain the second channel data, it can be that the object detection system forms the second original data based on the received wireless signal, and obtains the processed second original data by removing the outliers from the second original data; then the processed second original data is low-pass filtered to obtain the second channel data.

[0028] Among them, the object detection system can perform outlier removal processing through the hampel filter.

[0029] Step S120: determining a characterization vector of the channel data based on the first channel data and the second channel data.

[0030] The channel data includes first channel data and second channel data.

[0031] The characterization vector of the channel data includes the characterization vector of the first channel data and the second channel data, which can be understood as all data features of the first channel data and the second channel data.

[0032] The object detection system determines a characterization vector of the channel data based on the first channel data and the second channel data. Exemplarily, the object detection system may perform feature vector extraction processing on the first channel data and use the extracted feature vector as the characterization vector of the first channel data; perform feature vector extraction processing on the second channel data and use the extracted feature vector as the characterization vector of the second channel data; and then combine the characterization vector of the first channel data and the characterization vector of the second channel data to obtain the characterization vector of the channel data.

[0033] Step S130 , inputting the characterization vector of the channel data into the target support vector model to obtain a detection result of whether there is an object in the target signal region predicted by the target support vector model.

[0034] The target support vector model, also known as the support vector machine SVM (Support Vector Machine), is used to predict whether there is an object in the target signal area based on the characterization vector of the channel data.

[0035] For the determination method of the target support vector model, the object detection system can train the support vector model to be trained by the acquired channel data set to be trained, and obtain the target support vector model that meets the requirements. Among them, the channel data set to be trained includes the channel data to be trained and the annotation information of each channel data to be trained, and the annotation information represents whether there is an object in the signal area when the corresponding channel data to be trained is obtained. It should be noted that the signal area can be a target signal area or other signal area. Specifically, the object detection system inputs the channel data set to be trained into the support vector model to be trained, and the support vector model to be trained performs prediction based on the input channel data set to be trained to obtain the prediction information of each channel data to be trained output by the support vector model to be trained; calculates the loss function between the prediction information and the annotation information of each channel data to be trained; and trains the support vector model with the loss value of the loss function becoming smaller as the target to obtain the target support vector model that meets the requirements. For the method of training the support vector model with the loss value of the loss function becoming smaller as the target to obtain the target support vector model that meets the requirements, for example, the loss value of the loss function between the prediction information and the annotation information is less than the loss threshold, and then the corresponding support vector model is determined as the target support vector model that meets the requirements. Another exemplary method is to count the number of model training times. When the number of model training times is greater than a threshold and the loss value of the loss function between the prediction information and the annotation information is less than a loss threshold, the corresponding support vector model is determined as a target support vector model that meets the requirements.

[0036] It can be seen that the object detection method of the embodiment of the present application obtains the first channel data and the second channel data, the first channel data is formed by the first terminal based on receiving the wireless signal sent by the second terminal within the preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within the preset time period; determines the characterization vector of the channel data based on the first channel data and the second channel data; inputs the characterization vector of the channel data into the target support vector model, and obtains the detection result of whether there is an object in the target signal area predicted by the target support vector model. On the one hand, based on the wireless signals sent by the first terminal and the second terminal, the channel data in the target signal area can be obtained more comprehensively, and then the accuracy of object detection can be improved when performing object detection based on the more comprehensive channel data. On the other hand, performing object detection on the characterization vector of the channel data through the target support vector model can reduce the amount of data calculation, thereby realizing object detection on embedded devices with low computational requirements.

[0037] On the basis of the above-mentioned embodiments, when determining that there is an object in the target signal area, the object detection system of the embodiment of the present application can judge whether the object is an object in the preset object library, and then issue an early warning of the object in the target signal area. As an example, if the detection result indicates that there is an object in the target signal area, the object detection system matches the object with the object in the preset object library to obtain a matching result; in response to the matching result indicating that the object is other objects, a reporting process is performed. Among them. Other objects are objects outside the preset object library. In actual application scenarios, such as a residential parking lot, if it is detected that a vehicle in the target signal area does not exist in the preset object library, that is, the detected vehicle does not belong to the vehicles registered in the object detection system, the corresponding information will be reported to the staff so that the staff can manage it in time.

[0038] As another example, if the detection result indicates that there is an object in the target signal area, the object detection system matches the object with the objects in the preset object library to obtain a matching result; in response to the matching result indicating that the object is a target object in the preset object library, the object detection system does not perform a reporting process. In actual application scenarios, such as detecting vehicles in a residential parking lot, if the vehicle detected in the target signal area exists in the preset object library, that is, the detected vehicle belongs to a vehicle registered in the object detection system, no reporting process is performed.

[0039] As another example, if the detection result indicates that there is an object in the target signal area, the object detection system matches the object with the objects in the preset object library to obtain a matching result; in response to the matching result indicating that the object is a target object in the preset object library, the object detection system further performs object warning processing according to the attribute information of the target object. Specifically, the object detection system determines that the detected object is a target object in the preset object library, then obtains a captured image, which includes the target object; matches the attribute features of the target object in the captured image with the preset attribute features in the preset attribute feature library; if they match, the attribute information of the target object is reported. It should be noted that the attribute information can be the behavior information of the object, etc. In practical applications, for example, when detecting the activities of stray animals in a community, if the animal detected by the object detection system is an animal in the preset object library, the behavior information of the stray animal is determined according to the feature information of the stray animal in the captured image, and if the behavior of the stray animal is fighting, it is reported so that the staff can intervene.

[0040] See also Figure 3 , Figure 3 1 is a flow chart of an exemplary embodiment of step S120 in the object detection method shown in the present application. Specifically, the process of determining the characterization vector of the channel data based on the first channel data and the second channel data in step S120 includes the following steps: Step S310: performing data fusion processing on the first channel data and the second channel data to obtain fused channel data.

[0041] The object detection system performs data fusion processing on the first channel data and the second channel data to obtain fused channel data. Exemplarily, the object detection system may perform dimension fusion on the first channel data and the second channel data to obtain fused channel data. For example, the first channel data is represented as , the second channel data is expressed as , then the fused channel data is expressed as For specific expressions, please refer to the following:

[0042]

[0043]

[0044] Step S320, performing data dimension reduction processing on the fused channel data to obtain a representation vector of the channel data.

[0045] The object detection system can use the principal component analysis algorithm PCA (Principal Component Analysis) to perform data dimensionality reduction processing. Specifically, the object detection system performs principal component extraction processing on the fused channel data to obtain the principal component vector; the principal component vector is used as the representation vector of the channel data. The object detection system can also use linear discriminant analysis, multi-dimensional scaling transformation or autoencoder to perform data dimensionality reduction processing.

[0046] It can be seen that the object detection method of the embodiment of the present application performs data fusion processing on the first channel data and the second channel data to obtain fused channel data; and performs data dimension reduction processing on the fused channel data to obtain a representation vector of the channel data. In this way, even when the amount of collected channel data is large, the data processing volume can be maintained to improve the object detection efficiency and avoid the problem of low detection efficiency caused by excessive data processing volume.

[0047] Figure 4 FIG. 1 is a block diagram of an object detection system shown in an exemplary embodiment of the present application. Figure 4 As shown, the exemplary object detection system 400 includes: an acquisition module 410, a vector determination module 420 and a detection module 430. Specifically: The acquisition module 410 is used for acquiring the first channel data and the second channel data, wherein the first channel data is formed by the first terminal based on receiving the wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within a preset time period.

[0048] The vector determination module 420 is configured to determine a characterization vector of the channel data based on the first channel data and the second channel data.

[0049] The detection module 430 is used to input the characterization vector of the channel data into the target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model.

[0050] In this exemplary object detection system, first channel data and second channel data are obtained, the first channel data is formed by the first terminal based on receiving the wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within a preset time period; a characterization vector of the channel data is determined based on the first channel data and the second channel data; the characterization vector of the channel data is input into the target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model. On the one hand, based on the wireless signals sent by the first terminal and the second terminal, the channel data in the target signal area can be obtained more comprehensively, and then the accuracy of object detection can be improved when performing object detection based on the more comprehensive channel data. On the other hand, performing object detection on the characterization vector of the channel data through the target support vector model can reduce the amount of data calculation, thereby realizing object detection on embedded devices with low computational requirements.

[0051] The functions of each module can be found in the object detection method embodiment, which will not be repeated here.

[0052] To implement the object detection method of the above embodiment, the present application proposes another electronic device, for details, please refer to Figure 5 , Figure 5 It is a structural schematic diagram of an embodiment of an electronic device provided by the present application.

[0053] The electronic device 500 includes a memory 501 and a processor 502 , wherein the memory 501 and the processor 502 are coupled.

[0054] The memory 501 is used to store program data, and the processor 502 is used to execute the program data to implement the object detection method of the above embodiment.

[0055] In this embodiment, the processor 502 may also be referred to as a CPU (Central Processing Unit). The processor 502 may be an integrated circuit chip having signal processing capabilities. The processor 502 may also be a general-purpose processor, 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. A general-purpose processor may be a microprocessor or the processor 502 may also be any conventional processor, etc.

[0056] The present application also provides a computer-readable storage medium, such as Figure 6 As shown, the computer-readable storage medium 600 is used to store program data 601. When the program data 601 is executed by the processor, it is used to implement the object detection method in the method embodiment of the present application.

[0057] The method involved in the embodiment of the object detection method of the present application, when implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a device, such as a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or part of the contribution to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0058] The above description is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An object detection method, characterized in that: The method is applied to an object detection system, the object detection system includes a first terminal and a second terminal, the first terminal and the second terminal are arranged in the same target signal area, and the method includes: Acquire first channel data and second channel data, where the first channel data is formed by the first terminal based on receiving a wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within the preset time period; Determine a characterization vector of channel data based on the first channel data and the second channel data; The characterization vector of the channel data is input into a target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model.

2. The object detection method according to claim 1, characterized in that: The step of determining a characterization vector of channel data based on the first channel data and the second channel data comprises: Performing data fusion processing on the first channel data and the second channel data to obtain fused channel data; Performing data dimension reduction processing on the fused channel data to obtain a representation vector of the channel data.

3. The object detection method according to claim 2, characterized in that: The step of performing data dimensionality reduction processing on the fused channel data to obtain a representation vector of the channel data includes: Performing principal component extraction processing on the fused channel data to obtain a principal component vector; The principal component vector is used as a characterization vector of the channel data.

4. The object detection method according to claim 1, characterized in that: The method further comprises: Acquire a channel data set to be trained, wherein the channel data set to be trained includes channel data to be trained and annotation information of each channel data to be trained, wherein the annotation information indicates whether there is an object in a signal area when acquiring the corresponding channel data to be trained; Inputting the channel data set to be trained into the support vector model to be trained, and obtaining prediction information of each channel data to be trained output by the support vector model to be trained; Calculate the loss function between the predicted information and the labeled information of each channel data to be trained; The support vector model is trained with the goal of reducing the loss value of the loss function to obtain a target support vector model that meets the requirements.

5. The object detection method according to claim 1, characterized in that: The step of acquiring the first channel data comprises: Acquiring first original data through the first terminal; Performing outlier removal processing on the first original data to obtain processed first original data; The processed first original data is subjected to low-pass filtering to obtain the first channel data.

6. The object detection method according to claim 1, characterized in that: The method further comprises: If the detection result indicates that there is an object in the target signal area, matching the object with objects in a preset object library to obtain a matching result; In response to the matching result indicating that the object is other objects, a reporting process is performed, wherein the other objects are objects outside the preset object library.

7. The object detection method according to claim 6, characterized in that: The method further comprises: In response to the matching result indicating that the object is a target object in the preset object library, acquiring a captured image, wherein the captured image includes the target object; Performing matching processing according to the attribute features of the target object in the collected image and the preset attribute features in the preset attribute feature library; If there is a match, the attribute information of the target object is reported.

8. An object detection system, characterized in that: The object detection system comprises: an acquisition module, configured to acquire the first channel data and the second channel data, wherein the first channel data is formed by the first terminal based on receiving a wireless signal sent by the second terminal within a preset time period, and the second channel data is formed by the second terminal based on the wireless signal sent by the first terminal within the preset time period; A vector determination module, configured to determine a characterization vector of channel data based on the first channel data and the second channel data; The detection module is used to input the characterization vector of the channel data into a target support vector model to obtain a detection result of whether there is an object in the target signal area predicted by the target support vector model.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores program instructions, and the processor retrieves the program instructions from the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: include: Program data is stored, and when the program data is executed by a processor, it is used to implement the method according to any one of claims 1 to 7.

Citation Information

Patent Citations

  • Tumbling detecting method, device, and system

    CN109394229A

  • Entrepreneurship park personnel activity identification method and device and terminal equipment

    CN111488828A

  • Intrusion detection method and device based on CSI, and computer storage medium

    CN111741471A

  • Dynamic channel state information reporting method, equipment, device and storage medium

    CN114071557A

  • Method for acquiring channel information

    CN114642019A