Pedestrian recognition method, system and device based on radio frequency waves and storage medium

Abstract

The application provides a pedestrian recognition method, system and device based on radio frequency waves and a storage medium. The method comprises the following steps: acquiring a radio frequency thermal map group in a target area, acquiring pedestrian image data captured in the target area according to a preset capturing condition, and extracting pedestrian capturing information from the pedestrian image data; mapping the pedestrian capturing information to the radio frequency thermal map group to cut out a pedestrian radio frequency thermal image group from the radio frequency thermal map group; extracting pedestrian features from the pedestrian radio frequency thermal image group; and comparing the extracted pedestrian features with pedestrian features stored in a database to identify whether the captured pedestrian is a warehouse entry personnel. The method based on RF-PersonReID can effectively manage warehouse entry and exit personnel, complete the statistics of warehouse entry and exit personnel information and warehouse residence time, and effectively solve the visual interference caused by personal dress change or the same dress.

Description

Technical Field

[0001] This invention belongs to the field of image processing technology and relates to a recognition method, particularly a pedestrian recognition method, system, device, and storage medium based on radio frequency waves. Background Technology

[0002] Modern warehouse management involves various aspects, among which personnel access control is a complex and tedious task. Warehouses vary widely due to the diverse types of goods they store, such as cold storage, hazardous materials warehouses, refrigerated warehouses, and temperature-controlled warehouses. Warehouse operators have strict management requirements and regulations regarding the frequency and duration of personnel entry and exit. Each warehouse typically requires a dedicated administrator to monitor and record the entry and exit information of warehouse staff.

[0003] Traditional warehouse personnel access management primarily relies on RFID (Radio Frequency Identification) technology and manual supervision. RFID requires personnel to actively register or record information, which can lead to situations where personnel use identification cards or other identification documents. Manual supervision, on the other hand, requires significant manpower and is susceptible to negligence and errors by supervisors.

[0004] The existing technology has the following technical defects:

[0005] (1) The same person may change their clothes before and after entering and leaving the warehouse, and image-based technology cannot effectively solve the problem of changing clothes.

[0006] (2) Some warehouses require uniforms or protective clothing, but images cannot effectively identify individuals dressed in the same way.

[0007] (3) Drastic changes in ambient lighting will reduce the recognition effect.

[0008] Therefore, how to provide a pedestrian recognition method, system, device, and storage medium based on radio frequency waves to solve the shortcomings of existing technologies such as their inability to address visual interference caused by personal disguises or identical clothing has become an urgent technical problem for those in the field. Summary of the Invention

[0009] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide a pedestrian recognition method, system, device and storage medium based on radio frequency waves, so as to solve the problem that the prior art cannot deal with visual interference caused by personal disguise, change of clothes or similar attire.

[0010] To achieve the above and other related objectives, the present invention provides a pedestrian recognition method based on radio frequency waves, comprising: acquiring a radio frequency heatmap group within a target area, and simultaneously acquiring pedestrian image data captured within the target area according to preset capture conditions; extracting pedestrian capture information from the pedestrian image data; mapping the pedestrian capture information to the radio frequency heatmap group to extract a pedestrian radio frequency heatmap group from the radio frequency heatmap group; extracting pedestrian features from the pedestrian radio frequency heatmap group; and comparing the extracted pedestrian features with pedestrian features stored in a database to identify whether the captured pedestrian is a person registered in the database.

[0011] In one embodiment of the present invention, the step of acquiring a radio frequency thermal image group within a target area includes: acquiring first radio frequency thermal image data collected in the target area by a horizontal antenna array and second radio frequency thermal image data collected in the target area by a vertical antenna array; and stitching the first radio frequency thermal image data and the second radio frequency thermal image data together to form a radio frequency thermal image group.

[0012] In one embodiment of the present invention, the pedestrian capture information includes capture time and captured pedestrian detection boxes; the step of extracting pedestrian capture information from the pedestrian image data includes: performing pedestrian detection on the pedestrian image data through a preset pedestrian target detection network to select captured pedestrian detection boxes; and extracting capture time and captured pedestrian detection boxes that match preset capture conditions from the pedestrian image data.

[0013] In one embodiment of the present invention, the step of mapping the pedestrian capture information to the radio frequency thermal map group to extract the pedestrian radio frequency thermal image group from the radio frequency thermal map group includes: mapping the capture time matching the preset capture conditions to the radio frequency thermal map in the corresponding radio frequency thermal map group; and extracting the pedestrian radio frequency thermal image one by one from the matched radio frequency thermal map according to the coordinates of the captured pedestrian detection frame to form the pedestrian radio frequency thermal image group.

[0014] In one embodiment of the present invention, the step of extracting pedestrian features from the pedestrian radio frequency thermal image includes: performing image processing on the pedestrian radio frequency thermal image group to obtain a pedestrian radio frequency thermal image group with uniform width and height; and extracting pedestrian features from the image-processed pedestrian radio frequency thermal image group.

[0015] In one embodiment of the present invention, the step of comparing the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person entering the database includes: performing a similarity comparison between the extracted pedestrian features and the pedestrian features stored in the database one by one; if the comparison is successful, the captured pedestrian is identified as a person entering the database, and their entry and exit records and personnel number are marked; if the comparison fails, the captured pedestrian is identified as a non-person entering the database, and they are added to the database and assigned a personnel number.

[0016] In one embodiment of the present invention, the step of comparing the extracted pedestrian features with the pedestrian features stored in the database one by one includes: calculating the similarity between the pedestrian features and all the pedestrian features pre-stored in the database one by one; when the similarity between the pedestrian feature and a pre-stored pedestrian feature is greater than or equal to a preset similarity threshold, the comparison is considered successful; when the similarity between the pedestrian feature and a pre-stored pedestrian feature is less than the preset similarity threshold, the comparison is considered unsuccessful.

[0017] Another aspect of the present invention provides a pedestrian recognition system based on radio frequency waves, comprising: an acquisition module for acquiring a radio frequency heat map group within a target area, and simultaneously acquiring pedestrian image data captured within the target area according to preset capture conditions; an information extraction module for extracting pedestrian capture information from the pedestrian image data; a matting module for mapping the pedestrian capture information onto the radio frequency heat map group to extract a pedestrian radio frequency heat map group from the radio frequency heat map group; a feature extraction module for extracting pedestrian features from the pedestrian radio frequency heat map group; and a feature comparison module for comparing the extracted pedestrian features with pedestrian features stored in a database to identify whether the captured pedestrian is an entrant to the database.

[0018] In another aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the program, when executed by a processor, implements the pedestrian recognition method based on radio frequency waves.

[0019] The final aspect of the present invention provides a pedestrian recognition device based on radio frequency waves, comprising: a processor and a memory; the memory for storing a computer program; the processor being connected to the memory and for executing the computer program stored in the memory, so that the terminal executes the pedestrian recognition method based on radio frequency waves.

[0020] As described above, the pedestrian recognition method, system, device, and storage medium based on radio frequency waves of the present invention have the following beneficial effects:

[0021] This invention utilizes RF-Person ReID (Radio Frequency Person Re-Identification) to effectively manage warehouse personnel entering and exiting. It primarily involves statistically analyzing information such as staff entry and exit times and their dwell time within the warehouse. Simultaneously, RF signals can effectively penetrate clothing and reflect off the human body. The reflected wireless signal is received by a receiver and converted into a thermal image, revealing information such as the body shape of the personnel entering and exiting the warehouse. Using this information as identification features effectively solves the visual interference caused by individuals changing clothes or wearing identical attire. Attached Figure Description

[0022] Figure 1A The diagram shown is a flowchart of an embodiment of the pedestrian recognition method based on radio frequency waves of the present invention.

[0023] Figure 1B The diagram shows an implementation scenario of the pedestrian recognition method based on radio frequency waves according to one embodiment of the present invention.

[0024] Figure 2A The diagram shown is a schematic representation of the image acquisition process in one embodiment of the pedestrian recognition method based on radio frequency waves of the present invention.

[0025] Figure 2B The diagram shows a process for extracting pedestrian information in one embodiment of the pedestrian recognition method based on radio frequency waves of the present invention.

[0026] Figure 2C The diagram shows a flowchart of extracting a pedestrian radio frequency thermal image group in one embodiment of the pedestrian recognition method based on radio frequency waves of the present invention.

[0027] Figure 2D The diagram shows a pedestrian feature extraction process in one embodiment of the pedestrian recognition method based on radio frequency waves of the present invention.

[0028] Figure 2E The diagram shown is a feature comparison flowchart of a pedestrian recognition method based on radio frequency waves according to an embodiment of the present invention.

[0029] Figure 3 The diagram shown is a schematic representation of the principle structure of a pedestrian recognition system based on radio frequency waves according to an embodiment of the present invention.

[0030] Component designation explanation

[0031] 31 Acquisition Module

[0032] 32 Information Extraction Module

[0033] 33. Background Removal Module

[0034] 34 Feature Extraction Module

[0035] 35 Feature Comparison Module

[0036] Steps S11 to S15 Detailed Implementation

[0037] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0038] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0039] Example 1

[0040] This embodiment provides a pedestrian recognition method based on radio frequency waves, including:

[0041] Acquire a radio frequency thermal map set within the target area, and simultaneously acquire pedestrian image data captured within the target area according to preset capture conditions, and extract pedestrian capture information from the pedestrian image data;

[0042] The pedestrian capture information is mapped to the radio frequency thermal image group to extract the pedestrian radio frequency thermal image group from the radio frequency thermal image group;

[0043] Extract pedestrian features from the pedestrian radio frequency thermal image set;

[0044] The extracted pedestrian features are compared with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has been added to the database.

[0045] The following will describe in detail the radio frequency wave-based pedestrian recognition method provided in this embodiment with reference to the illustrations. Please refer to... Figure 1A and Figure 1B The figures show a flowchart of an embodiment of a pedestrian recognition method based on radio frequency waves according to the present invention and an implementation diagram of the pedestrian recognition method based on radio frequency waves in an application scenario. Figure 1A and Figure 1BAs shown, the pedestrian recognition method based on radio frequency waves specifically includes the following steps:

[0046] S11, acquire the radio frequency thermal map group within the target area, and simultaneously acquire pedestrian image data captured within the target area according to preset capture conditions.

[0047] Please see Figure 2A The diagram shows the image acquisition process in the pedestrian recognition method based on radio frequency waves of the present invention.

[0048] S111, acquire a set of radio frequency thermal maps within the target area. S111 includes the following steps:

[0049] S111A, acquire first radio frequency thermal image data collected in the target area by a horizontal antenna array and second radio frequency thermal image data collected in the target area by a vertical antenna array.

[0050] In this embodiment, a horizontal antenna array and a vertical antenna array are respectively set in the target area at the warehouse entrance. When the horizontal antenna array captures images of people entering and exiting the warehouse entrance, it obtains first radio frequency thermal image data in the horizontal direction; when the vertical antenna array captures images of people entering and exiting the warehouse entrance, it obtains second radio frequency thermal image data in the vertical direction.

[0051] S111B, the first radio frequency thermal image data and the second radio frequency thermal image data are stitched together to form a radio frequency thermal image group.

[0052] S112, according to the preset capture conditions, captures images of pedestrians within the target area according to the preset capture conditions to obtain pedestrian image data.

[0053] In this embodiment, the camera captures images of pedestrians within the target area according to the capture conditions, and the captured pedestrian image data is an RGB image.

[0054] The capture conditions can be preset according to user needs. Specifically, capture can be achieved either when a pedestrian passes through a certain location within the target area, or when a pedestrian passes through a specific location.

[0055] S12, extract pedestrian capture information from the pedestrian image data. The pedestrian capture information includes: capture time, captured pedestrian detection box, image ID, image location information, image description information, etc.

[0056] Please see Figure 2B The diagram shows a flowchart of step S12 in the radio frequency wave-based pedestrian recognition method of the present invention. Figure 2B As shown, step S12 includes the following steps:

[0057] S121, pedestrian detection is performed on the pedestrian image data using a preset pedestrian target detection network to select a pedestrian detection box.

[0058] After the image capture is completed, pedestrian capture information is extracted from the pedestrian image data using a pedestrian target detection network. The pedestrian target detection network includes, but is not limited to, one or more methods combined, such as SSD, YOLO, FRCNN, and CenterNet.

[0059] Specifically, based on a pedestrian target detection network, the network model is used to detect pedestrian targets in the captured RGB image to determine pixel positions and sizes, ultimately obtaining information about the pedestrian detection box, such as the coordinates (x, y, w, h) of the top-left corner vertex of the detection box. Here, x, y, w, and h represent the x-coordinate and y-coordinate of the top-left corner vertex, the width of the detection box, and the height of the detection box, respectively.

[0060] Similarly, pedestrian detection boxes can also be obtained through multi-target tracking. Specifically, based on the captured RGB images, pedestrians in the images are tracked and identified. By combining features such as the size change, position and displacement of the target box, and target image information, the same pedestrian target in different images is marked, and finally, pedestrian detection boxes are obtained.

[0061] The multi-target tracking algorithm can employ methods such as DeepSort and IouTracking.

[0062] S122, extract the capture time and captured pedestrian detection box that match the preset capture conditions from the pedestrian image data.

[0063] S13, the pedestrian capture information is mapped to the radio frequency thermal map group to extract the pedestrian radio frequency thermal image group from the radio frequency thermal map group. Please refer to... Figure 2C The diagram shows a flowchart of the process for extracting pedestrian radio frequency thermal image groups in the pedestrian recognition method based on radio frequency waves of the present invention. Figure 2C As shown, step S13 includes the following steps:

[0064] S131, the capture time that matches the preset capture conditions and the pedestrian detection frame of the captured pedestrian are mapped to the radio frequency heat map in the corresponding radio frequency heat map group.

[0065] The pedestrian detection box at a certain moment is matched with the radio frequency heatmap group at the same moment, and the pedestrian detection box is used to select the image in the heatmap group at the same moment.

[0066] S132, based on the coordinates of the pedestrian detection box, extract the pedestrian radio frequency thermal images one by one from the matched radio frequency thermal map to form a pedestrian radio frequency thermal image group.

[0067] S14, extract pedestrian features from the pedestrian radio frequency thermal image set. (See also...) Figure 2D The diagram shows a pedestrian feature extraction process in one embodiment of the radio frequency wave-based pedestrian recognition method of the present invention. Figure 2D As shown, step S14 includes the following steps:

[0068] S141, perform image processing on the pedestrian radio frequency thermal image group to obtain a pedestrian radio frequency thermal image group with uniform width and height.

[0069] The images with different width and height dimensions in the pedestrian radio frequency thermal image group obtained in step S132 are scaled to obtain a pedestrian radio frequency thermal image group with the same width and height dimensions.

[0070] S142, extract pedestrian features from the processed pedestrian radio frequency thermal image set.

[0071] The processed pedestrian radio frequency thermal images are used to extract pedestrian features through a recognition network. This embodiment uses an MGN network as an example. Specifically, the pedestrian radio frequency thermal images are used as input parameters to train the network recognition model. The recognition network model calculates a feature for each image, ultimately resulting in several pedestrian features. The dimensionality of the pedestrian features is determined according to user needs; in this embodiment, different dimensions such as 1024, 2048, and 512 can be used.

[0072] In this embodiment, a method of merging the recognition network and the pedestrian feature extraction network is adopted to efficiently complete the extraction of pedestrian map features.

[0073] S143, pedestrian features are transmitted to the database for feature comparison in the following steps. The database contains pre-stored pedestrian features.

[0074] S15, compare the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person registered in the database. Please refer to [link / reference]. Figure 2E The diagram shows a feature comparison process in one embodiment of the radio frequency wave-based pedestrian recognition method of the present invention. Figure 2E As shown, step S15 includes the following steps:

[0075] S151, the extracted pedestrian features are compared one by one with the pedestrian features stored in the database for similarity. S151 includes the following steps:

[0076] S151A, the similarity of the pedestrian features with all pedestrian features pre-stored in the database is calculated one by one.

[0077] The similarity calculation formula is as follows:

[0078] Similarity = 1 - dist(X, Y)

[0079] Wherein, Similarity represents the similarity; Dist represents the distance calculation function, and common Euclidean or cosine distance calculation formulas are given on the next page; X represents pedestrian features with a size of Nx1; Y represents any pedestrian feature stored in the database with a size of Nx1.

[0080] The Euclidean distance and cosine distance calculation formulas used in this embodiment are as follows:

[0081] The Euclidean distance formula is:

[0082]

[0083] The formula for cosine distance is:

[0084]

[0085] Where, Dist – distance calculation function; X – pedestrian feature vector, with size Nx1; Y – any pedestrian feature vector stored in the database, with size Nx1; x i - represents the i-th element in the pedestrian feature vector X; y i - represents the i-th element in any pedestrian feature vector Y stored in the database;

[0086] Based on the above calculation formula, the similarity between the pedestrian features and all pedestrian features pre-stored in the database is calculated.

[0087] S151B, when the similarity between the pedestrian feature and a pre-stored pedestrian feature is greater than or equal to a preset similarity threshold, the comparison is successful; when the similarity between the pedestrian feature and a pre-stored pedestrian feature is less than the preset similarity threshold, the comparison fails.

[0088] S152, if the comparison is successful, the captured pedestrian is identified as an inbound person, and their entry / exit record and personnel number are marked; if the comparison fails, the captured pedestrian is identified as a non-inbound person, and they are put into the warehouse and assigned a personnel number. This allows for the acquisition of inbound and outbound information, ultimately enabling the statistical analysis of the warehouse's inbound / outbound personnel and their dwell time.

[0089] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following... Figure 1AThe pedestrian recognition method based on radio frequency waves.

[0090] At any possible level of technical detail, this application can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this application.

[0091] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, (but not limited to) electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0092] The computer-readable program described herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer-readable program instructions from the network and forwards these instructions to a computer-readable storage medium in the respective computing / processing device. The computer program instructions used to perform the operations of this application may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, integrated circuit configuration data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and procedural programming languages ​​such as "C" or similar programming languages. Computer-readable program instructions may execute entirely on a user's computer, partially on a user's computer, as a standalone software package, partially on a user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of this application.

[0093] Example 2

[0094] This embodiment provides a pedestrian recognition system based on radio frequency waves, including:

[0095] The acquisition module is used to acquire radio frequency thermal map groups within the target area, and simultaneously acquire pedestrian image data captured within the target area according to preset capture conditions.

[0096] The information extraction module is used to extract pedestrian capture information from the pedestrian image data;

[0097] The image matting module is used to map the pedestrian capture information to the radio frequency thermal image group, so as to extract the pedestrian radio frequency thermal image group from the radio frequency thermal image group;

[0098] The feature extraction module is used to extract pedestrian features from the pedestrian radio frequency thermal image set;

[0099] The feature comparison module is used to compare the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has been added to the database.

[0100] The following will describe in detail the radio frequency wave-based pedestrian recognition system provided in this embodiment, with reference to the accompanying drawings. Please refer to... Figure 3 The diagram shown is a schematic representation of the principle structure of a pedestrian recognition system based on radio frequency waves according to an embodiment of the present invention. Figure 3 As shown, the pedestrian recognition system based on radio frequency waves includes: an acquisition module 31, an information extraction module 32, an image matting module 33, a feature extraction module 34, and a feature comparison module 35.

[0101] The acquisition module 31 is used to acquire radio frequency thermal map groups within the target area, and simultaneously acquire pedestrian image data captured within the target area according to preset capture conditions.

[0102] Specifically, the acquisition module 31 is used to acquire a set of radio frequency (RF) thermal images within the target area. Specifically, it acquires first RF thermal image data collected within the target area via a horizontal antenna array and second RF thermal image data collected within the target area via a vertical antenna array. In this embodiment, a horizontal antenna array and a vertical antenna array are respectively set within the target area at the warehouse entrance. When the horizontal antenna array acquires images of people entering and exiting the warehouse entrance, it obtains first RF thermal image data in the horizontal direction; when the vertical antenna array acquires images of people entering and exiting the warehouse entrance, it obtains second RF thermal image data in the vertical direction. The first and second RF thermal image data are then stitched together to form a set of RF thermal images.

[0103] Meanwhile, the acquisition module 31 is used to acquire pedestrian image data captured within the target area according to preset capture conditions. In this embodiment, the camera captures pedestrians within the target area, and the captured pedestrian image data is an RGB image.

[0104] The capture conditions can be preset according to user needs. Specifically, capture can be achieved either when a pedestrian passes through a certain location within the target area, or when a pedestrian passes through a specific location.

[0105] The information extraction module 32 is used to extract pedestrian capture information from the pedestrian image data.

[0106] After capturing the image, the information extraction module 32 uses a preset pedestrian target detection network to perform pedestrian detection on the pedestrian image data to select the captured pedestrian detection box. Specifically, a pedestrian target detection algorithm can be used to perform pedestrian target detection on the captured RGB image to finally obtain the information of the pedestrian detection box.

[0107] The pedestrian target detection algorithms used in this embodiment include YOLO, SSD, FRCNN, CenterNet, etc., and the above algorithms can run on processing servers or terminals such as GPU, NPU, DSP, FPGA, and AI chips.

[0108] The image matting module 33 is used to map the pedestrian capture information to the radio frequency thermal image group, so as to extract the pedestrian radio frequency thermal image group from the radio frequency thermal image group.

[0109] Specifically, the image matting module 33 maps the capture time and the pedestrian detection box of the captured pedestrian that match the preset capture conditions to the radio frequency heatmap in the corresponding radio frequency heatmap group, and uses the pedestrian detection box to select the images in the heatmap group at the same time. Based on the coordinates of the pedestrian detection box, the pedestrian radio frequency heatmap images are matted one by one on the matched radio frequency heatmap to form a pedestrian radio frequency heatmap image group.

[0110] The feature extraction module 34 is used to extract pedestrian features from the pedestrian radio frequency thermal image group.

[0111] The feature extraction module 34 performs image scaling on the pedestrian radio frequency thermal image group to obtain a pedestrian radio frequency thermal image group with uniform width and height. The processed pedestrian radio frequency thermal image group is then used to extract pedestrian features through a recognition network.

[0112] Specifically, the feature extraction module 34 uses the pedestrian radio frequency thermal image group as the input parameter of the network recognition training model, calculates a feature for each image through the recognition network model, and finally obtains several pedestrian features.

[0113] The extracted pedestrian features are transmitted to a database for feature comparison in the following steps. The database contains pre-stored pedestrian features.

[0114] The feature comparison module 35 is used to compare the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has entered the database.

[0115] Specifically, the feature comparison module 35 calculates the similarity between the pedestrian feature and all pre-stored pedestrian features in the database. The similarity is calculated using a similarity calculation formula. A successful comparison occurs when the similarity between the pedestrian feature and a pre-stored pedestrian feature is greater than or equal to a preset similarity threshold. After a successful comparison, the captured pedestrian is identified as a person entering the warehouse, and their entry / exit record and personnel number are marked. This allows for the collection of entry and exit information for pedestrians, ultimately enabling the statistical analysis of the entry / exit and dwell time of warehouse personnel.

[0116] If the similarity between the pedestrian's features and a pre-stored pedestrian's features is less than a preset similarity threshold, the comparison fails. In this case, the captured pedestrian is identified as a non-databaseed person, added to the database, and assigned a personnel number.

[0117] It should be noted that the division of the various modules in the above system is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these modules can be implemented entirely in software through processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. For example, module x can be a separate processing element, or it can be integrated into a chip within the system. Alternatively, it can be stored as program code in the system's memory, and its function can be called and executed by a processing element. The implementation of other modules is similar. Moreover, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element mentioned here can be an integrated circuit with signal processing capabilities. During implementation, each step of the above method or each of the above modules can be completed through integrated logic circuits in the hardware of the processor element or through software instructions.

[0118] These modules can be one or more integrated circuits configured to implement the above methods, such as one or more Application Specific Integrated Circuits (ASICs), one or more digital processors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs). Alternatively, when a module is implemented using processing element scheduler code, the processing element can be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor capable of calling program code. Furthermore, these modules can be integrated together as a system-on-a-chip (SoC).

[0119] Example 3

[0120] This embodiment provides a pedestrian recognition device based on radio frequency waves. The radio frequency wave-based pedestrian recognition device includes: a processor, a memory, a transceiver, a communication interface, and a system bus. The memory and the communication interface are connected to the processor and the transceiver through the system bus and complete mutual communication. The memory is used to store computer programs, the communication interface is used to communicate with other devices, and the processor and the transceiver are used to run the computer programs, so that the radio frequency wave-based pedestrian recognition device performs the various steps of the radio frequency wave-based pedestrian recognition method described above.

[0121] The system bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This system bus can be divided into address bus, data bus, control bus, etc. For ease of representation, only one thick line is used in the diagram, but this does not indicate that there is only one bus or one type of bus. The communication interface is used to enable communication between the database access device and other devices (such as clients, read-write libraries, and read-only libraries). Memory may include Random Access Memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0122] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0123] The protection scope of the pedestrian recognition method based on radio frequency waves described in this invention is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting, or replacing steps in the prior art based on the principles of this invention is included within the protection scope of this invention.

[0124] In summary, the pedestrian recognition method, system, device, and storage medium based on radio frequency waves provided by this invention have the following beneficial effects:

[0125] This invention effectively manages warehouse personnel entering and exiting, and accurately and efficiently tracks their entry and exit information, as well as their dwell time in the warehouse. Furthermore, compared to RFID-based management devices, this invention effectively avoids unauthorized registration and confusion caused by using devices other than the registered owner, such as ID cards, wristbands, and mobile phones. Moreover, compared to AI recognition technologies that rely on image information, this invention can identify personnel without relying on visible features such as clothing or appearance, solving the visual interference problem caused by changes in clothing or identical attire; it also achieves accurate pedestrian identification even in poor lighting conditions. Therefore, this invention effectively overcomes the various shortcomings of existing technologies and has high industrial application value.

[0126] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A pedestrian recognition method based on radio frequency waves, characterized in that, include: Acquire first radio frequency thermal image data acquired within the target area via a horizontal antenna array and second radio frequency thermal image data acquired within the target area via a vertical antenna array; The first radio frequency thermal image data and the second radio frequency thermal image data are then stitched together to form a radio frequency thermal image group; Simultaneously, pedestrian image data captured within the target area according to preset capture conditions is acquired, and pedestrian capture information is extracted from the pedestrian image data; The pedestrian image data is processed by a preset pedestrian target detection network to select the pedestrian detection box for capture, and pedestrian capture information containing the capture time matching the preset capture conditions and the pedestrian detection box is extracted from the pedestrian image data. The pedestrian capture information is mapped to the corresponding radio frequency thermal map. Based on the coordinates of the captured pedestrian detection box, pedestrian radio frequency thermal images are extracted one by one from the matched radio frequency thermal map to form a pedestrian radio frequency thermal image group. Extract pedestrian features from the pedestrian radio frequency thermal image set; The extracted pedestrian features are compared with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has been added to the database.

2. The pedestrian recognition method based on radio frequency waves according to claim 1, characterized in that, The steps for extracting pedestrian feature vectors from the pedestrian radio frequency thermal image include: Image processing is performed on the pedestrian radio frequency thermal image group to obtain pedestrian radio frequency thermal image groups with uniform width and height; Pedestrian features were extracted from the processed radio frequency thermal images of pedestrians.

3. The pedestrian recognition method based on radio frequency waves according to claim 2, characterized in that, The steps to compare the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has entered the database include: The extracted pedestrian features are compared one by one with the pedestrian features stored in the database. If the comparison is successful, the captured pedestrian is identified as an inbound person and their entry / exit record and personnel number are marked. If the comparison fails, the captured pedestrian is identified as a non-inbound person, and they are added to the database and assigned a personnel number.

4. The pedestrian recognition method based on radio frequency waves according to claim 3, characterized in that, The step of comparing the extracted pedestrian features with the pedestrian features stored in the database one by one includes: The similarity between the pedestrian features and all pedestrian features pre-stored in the database is calculated one by one. When the similarity between the pedestrian feature and a pre-stored pedestrian feature is greater than or equal to a preset similarity threshold, the comparison is considered successful. If the similarity between the pedestrian feature and a pre-stored pedestrian feature is less than a preset similarity threshold, the comparison fails.

5. A pedestrian recognition system based on radio frequency waves, characterized in that, include: The acquisition module is used to acquire first radio frequency thermal image data collected in the target area by a horizontal antenna array and second radio frequency thermal image data collected in the target area by a vertical antenna array. The first radio frequency thermal image data and the second radio frequency thermal image data are stitched together to form a radio frequency thermal image group; at the same time, pedestrian image data captured in the target area according to preset capture conditions are acquired. The information extraction module is used to perform pedestrian detection in the pedestrian image data through a preset pedestrian target detection network to select the pedestrian detection box to be captured, and to extract pedestrian capture information in the pedestrian image data, including the capture time that matches the preset capture conditions and the pedestrian detection box. The image matting module is used to map the pedestrian capture information to the corresponding radio frequency thermal map. Based on the coordinates of the captured pedestrian detection box, the module extracts pedestrian radio frequency thermal images one by one from the matched radio frequency thermal map to form a pedestrian radio frequency thermal image group. The feature extraction module is used to extract pedestrian features from the pedestrian radio frequency thermal image set; The feature comparison module is used to compare the extracted pedestrian features with the pedestrian features stored in the database to identify whether the captured pedestrian is a person who has been added to the database.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the pedestrian recognition method based on radio frequency waves as described in any one of claims 1 to 4.

7. A pedestrian recognition device based on radio frequency waves, characterized in that, include: Processor and memory; The memory is used to store computer programs; The processor is connected to the memory and is used to execute the computer program stored in the memory so that the radio frequency wave-based pedestrian recognition device performs the radio frequency wave-based pedestrian recognition method according to any one of claims 1 to 4.

Images