Methods for identifying people left inside vehicles, electronic devices, vehicles, media and products

Abstract

This application discloses a method, electronic device, vehicle, medium, and product for identifying occupants left in a vehicle. This application relates to the field of vehicle technology. The method for identifying occupants left in a vehicle is applied to an electronic device, which includes a wireless signal transmitting module. Specifically, the method includes: controlling the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, and determining multiple target channel state information based on the multiple wireless detection signals; generating a target liveness image based on the multiple target channel state information, and extracting image features from the target liveness image; and determining the presence of a target occupant in the cabin when the image features match preset human body features. This application achieves the technical effect of improving the accuracy of identifying occupants left in the cabin.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and more particularly to a method for identifying occupants left in a vehicle, electronic equipment, vehicle, storage medium, and computer program product. Background Technology

[0002] With the continuous development of the automotive industry, new energy vehicles have become the preferred mode of transportation for more and more users' daily travel. In order to improve the safety of new energy vehicles, technicians usually install multiple camera devices in the vehicle's cabin to better inspect the interior.

[0003] In related technologies, vehicles typically control cameras to capture images of the cabin when the vehicle is locked, thereby obtaining image data containing the interior of the cabin, and determining whether there are any people left in the cabin based on the image data.

[0004] However, because the camera's view of the cabin can be obstructed by seats, luggage, or other debris while filming the interior, the accuracy of identifying whether there are people left in the cabin is low. This greatly increases the risk of accidents involving people left in the vehicle and reduces the vehicle's safety. Summary of the Invention

[0005] The main objective of this application is to provide a method, electronic device, vehicle, storage medium, and computer program product for identifying persons left in a vehicle, aiming to solve the technical problem of low accuracy in identifying persons left in the cabin in related technologies.

[0006] To achieve the above objectives, this application proposes a method for identifying occupants left inside a vehicle. This method is applied to an electronic device, which includes a wireless signal transmitting module. The method for identifying occupants left inside a vehicle includes:

[0007] The wireless signal transmitting module is controlled to transmit multiple wireless detection signals to the vehicle's cabin, and multiple target channel status information is determined based on the multiple wireless detection signals;

[0008] A target liveness image is generated based on multiple target channel state information, and image features of the target liveness image are extracted;

[0009] If the image features are detected to match the preset human features, it is determined that there are target personnel left behind in the cockpit.

[0010] In one embodiment, the step of determining multiple target channel state information based on multiple wireless detection signals includes:

[0011] The initial channel state information matched by each of the multiple wireless detection signals is detected, and each initial channel state information is processed to obtain complete channel state information.

[0012] The complete channel state information is filtered to determine the target channel state information, wherein the target channel state information is the channel state information corresponding to the wireless detection signal of a detected live target.

[0013] In one embodiment, the step of detecting initial channel state information matching each of the plurality of wireless detection signals includes:

[0014] Detect the wireless received signal that matches each of the multiple wireless detection signals;

[0015] Determine the signal change ratio between each of the plurality of wireless detection signals and the matched wireless signal;

[0016] Based on the signal change ratio matched by each of the multiple wireless detection signals, the initial channel state information matched by each of the multiple wireless detection signals is determined.

[0017] In one embodiment, the step of processing each of the initial channel state information to obtain each complete channel state information includes:

[0018] Determine the initial amplitude parameter and initial phase parameter included in each of the initial channel state information;

[0019] Phase expansion is performed on each of the initial amplitude parameters and each of the initial phase parameters, and the phase-expanded initial amplitude parameters and each of the initial phase parameters are filtered and eliminated to obtain each target amplitude parameter and each target phase parameter.

[0020] The target amplitude parameters and their respective matching target phase parameters are fused to obtain complete channel state information.

[0021] In one embodiment, the step of filtering each of the complete channel state information to determine the target channel state information includes:

[0022] Obtain preset standard channel state information, wherein the standard channel state information is the channel state information corresponding to the detection of a wireless detection signal of a non-living detection target;

[0023] Each complete channel state information is compared with the standard channel state information, and the complete channel state information that is inconsistent with the standard channel state information is determined as the target channel state information.

[0024] In one embodiment, the step of generating a target liveness image based on multiple target channel state information includes:

[0025] Extract the one-dimensional vector features contained in each of the multiple target channel state information;

[0026] The one-dimensional vector features are fused to obtain two-dimensional vector features, and a target liveness image is generated based on the two-dimensional vector features.

[0027] In one embodiment, the step of extracting image features from the target live image includes:

[0028] The target liveness image is subjected to phase-splitting processing to obtain multiple sub-liveness images;

[0029] Encode multiple sub-liveness images to obtain convolutional features, and mix the convolutional features to obtain pose fusion data;

[0030] Extract the target features contained in the pose fusion data, and determine the target features as the image features of the target liveness image.

[0031] In one embodiment, after determining that a target person remains in the cockpit, the method further includes:

[0032] Based on the image features, the identity information of the target person left behind is determined, and a message indicating that the person has been left behind is generated based on the identity information.

[0033] In one embodiment, the step of determining the identity information of the target abandoned person based on the image features includes:

[0034] Determine the preset personnel features that match the image features;

[0035] The preset identification information that matches the preset personnel characteristics is determined as the identity identification information of the target remaining personnel.

[0036] In addition, to achieve the above objectives, this application also proposes an electronic device comprising: a wireless signal transmitting module, a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for identifying persons left in a vehicle as described above.

[0037] In addition, to achieve the above objectives, this application also proposes a vehicle that includes the electronic equipment described above.

[0038] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the method for identifying persons left in the vehicle as described above.

[0039] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the method for identifying persons left in a vehicle as described above.

[0040] The method for identifying persons left behind in a vehicle provided in this application embodiment is applied to an electronic device. The electronic device includes a wireless signal transmitting module. By controlling the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, multiple target channel state information is determined based on the multiple wireless detection signals. A target liveness image is generated based on the multiple target channel state information, and image features of the target liveness image are extracted. If the image features are detected to match preset human body features, it is determined that a target person left behind exists in the cabin.

[0041] In this embodiment, during operation, the electronic device first controls its configured wireless signal transmission module to transmit multiple wireless detection signals to the vehicle's cabin. The electronic device then detects these multiple wireless detection signals to determine the target channel state information that matches each wireless detection signal that detects a live target. Subsequently, the electronic device processes the multiple target channel state information to generate a target live image containing the detected live target. The electronic device then extracts the image features contained in the target live image. Finally, the electronic device identifies the image features, and if the identified image features match a preset human body feature, it determines that the detected live target is a person left behind in the cabin.

[0042] Thus, this application solves the technical problem of low accuracy in identifying people left in the cabin in related technologies. Specifically, by utilizing the characteristic that wireless detection signals generate different channel state information when they come into contact with a living person, this application can accurately identify whether a living person in the cabin is a person left behind. This avoids interference from obstacles during the cabin detection process, thereby improving the accuracy of identifying people left behind in the cabin and greatly increasing vehicle safety. Attached Figure Description

[0043] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 This is a schematic diagram of the electronic device involved in an embodiment of the method for identifying occupants left in a vehicle according to this application;

[0046] Figure 2 A flowchart illustrating the method for identifying persons left inside a vehicle according to this application;

[0047] Figure 3 This is a schematic diagram of a target liveness image related to an embodiment of the method for identifying persons left inside a vehicle in this application;

[0048] Figure 4 This is a simplified flowchart illustrating the method for identifying persons left inside the vehicle in this application.

[0049] Figure 5 This is a schematic diagram of the module structure of the identification device for people left in a vehicle according to an embodiment of this application;

[0050] Figure 6 This is a schematic diagram of the hardware operating environment involved in the method for identifying people left in the vehicle in this application embodiment.

[0051] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0053] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0054] In this embodiment, for ease of description, please refer to Figure 1 , Figure 1 This is a schematic diagram of the electronic device involved in an embodiment of the method for identifying occupants left in a vehicle, as described below. Figure 1As shown, the execution entity is an electronic device with an internal wireless signal transmission module, or a mobile terminal, data storage control terminal, PC, or other terminal connected to an electronic control unit associated with the electronic device. The wireless signal transmission module comprises two parts: a TCAM (Telematics & Connectivity Antenna Module) and a DHU (Digital Cockpit Head Unit). The TCAM includes a first Wi-Fi Service, a Core Service, and an MCU (Microcontroller Unit). Similarly, the DHU includes a second Wi-Fi Service, a QNX Service (QNX Virtual Machine), an Android Service (Android Subsystem), an Algorithm Service (Data Processing Unit), and an MCU.

[0055] Based on the aforementioned electronic equipment, this application presents an overall concept for a method to identify persons left inside a vehicle.

[0056] With the continuous development of the automotive industry, new energy vehicles have become the preferred mode of transportation for an increasing number of users' daily travel. To improve the safety of new energy vehicles, technicians typically install multiple cameras inside the vehicle's cabin for better monitoring. In these technologies, the cameras usually capture images of the cabin when the vehicle is locked, obtaining data to determine if any occupants are left inside. However, because the camera's view can be obstructed by seats, luggage, or other objects, the accuracy of identifying unaccompanied persons inside the cabin can be low, significantly increasing the risk of accidents and reducing vehicle safety.

[0057] To address the above issues, this application provides a method for identifying persons left behind in a vehicle. This method is applied to an electronic device, which includes a wireless signal transmitting module. The method comprises: controlling the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, and determining multiple target channel state information based on the multiple wireless detection signals; generating a target liveness image based on the multiple target channel state information, and extracting image features from the target liveness image; and determining the presence of a target person left behind in the cabin when the image features match preset human body features.

[0058] Thus, this application solves the technical problem of low accuracy in identifying people left in the cabin in related technologies. Specifically, by utilizing the characteristic that wireless detection signals generate different channel state information when they come into contact with a living person, this application can accurately identify whether a living person in the cabin is a person left behind. This avoids interference from obstacles during the cabin detection process, thereby improving the accuracy of identifying people left behind in the cabin and greatly increasing vehicle safety.

[0059] Based on the overall concept of the method for identifying occupants left inside a vehicle, embodiments of this application provide a method for identifying occupants left inside a vehicle, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the method for identifying persons left inside a vehicle according to this application.

[0060] In this embodiment, the method for identifying persons left inside the vehicle is applied to an electronic device, which includes a wireless signal transmitting module. The method for identifying persons left inside the vehicle includes steps S10 to S30:

[0061] Step S10: Control the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, and determine multiple target channel status information based on the multiple wireless detection signals;

[0062] It should be noted that the wireless detection signal can specifically be a Wi-Fi signal. Furthermore, the target state channel information refers to the CSI (Channel State Information) data corresponding to the Wi-Fi signal from the detected liveness detection target inside the vehicle cabin. This CSI data is understood to be a complex decimal sequence representing the signal-to-wave ratio between the wireless detection signal and the corresponding wireless received signal.

[0063] In this embodiment, when the electronic device detects that the vehicle has entered the locked state, it first calls the wireless signal transmitting module to continuously transmit multiple wireless detection signals into the vehicle's cabin. The electronic device then controls the wireless signal transmitting module to receive the wireless reflection signals generated by each of the multiple wireless detection signals. Based on the received wireless detection signals and wireless reflection signals, the wireless signal transmitting module obtains multiple initial channel state information. The electronic device then controls the wireless signal transmitting module to input the obtained multiple initial channel state information into the data processing unit configured within the electronic device. The data processing unit filters the multiple initial channel state information to determine the target channel state information that matches each of the multiple wireless detection signals that detect a liveness detection target.

[0064] For example, when the electronic device detects that the vehicle has entered a locked state, it first controls its own wireless transmission module to activate the TCAM within the wireless transmission module. This causes the first Wi-Fi Service within the TCAM to transmit three first Wi-Fi signals into the cabin. The first Wi-Fi Service then receives the first reflected signals generated by each of the three first Wi-Fi signals and determines the initial CSI data matching each of the three first Wi-Fi signals. The TCAM then inputs the acquired initial CSI data into the DHU within the wireless transmission module, thereby causing the initial CSI data to enter the data processing unit Algorithm Service configured within the DHU. Simultaneously, the electronic device controls the DHU within the wireless transmission module to activate, causing the second Wi-Fi Service within the DHU to transmit three second Wi-Fi signals into the cabin. The second Wi-Fi Service then receives the second reflected signals generated by each of the three second Wi-Fi signals and determines the initial CSI data matching each of the three second Wi-Fi signals. The DHU then inputs the acquired initial CSI data into its own data processing unit Algorithm Service. After acquiring each initial CSI data, the Service filters each initial CSI data to determine the target CSI data corresponding to each target's first Wi-Fi signal and each target's second Wi-Fi signal that detected a living person in the cockpit.

[0065] It should be noted that, in this embodiment and another embodiment, when the electronic device detects that the vehicle has entered the locked state, it can first control the initialization of the QNX virtual machine (QNX Service) configured in the DHU, so as to start the second Wifi Service and Android subsystem (Android Service) in the DHU through the QNX virtual machine. At the same time, the electronic device also controls the TCAM and DHU to conduct Socket communication to determine that the TCAM can send the acquired initial CSI data to the Algorithm Service in the DHU.

[0066] In addition, in this embodiment and another embodiment, after the electronic device controls the TCAM to start, it also needs to control the Core Service configured in the TCAM and the QNX virtual machine configured in the DHU to detect the communication status, so as to promptly initiate a re-communication request when the heartbeat communication between the TCAM and the DHU is detected to be interrupted.

[0067] In this way, by calling the wireless transmission module to transmit multiple wireless detection signals to the cockpit, the vehicle can obtain the initial channel state information generated by each of the multiple wireless detection signals, and based on the initial channel state information, determine the target wireless detection signal that detects a live target among the multiple wireless detection signals, and determine the target channel state information that matches each target wireless detection signal.

[0068] In one feasible implementation, the step of "determining multiple target channel state information based on multiple wireless detection signals" in step S10 above may specifically include steps S101 to S102:

[0069] Step S101: Detect the initial channel state information matched by each of the multiple wireless detection signals, and process each initial channel state information to obtain each complete channel state information;

[0070] Step S102: Filter each of the complete channel state information to determine the target channel state information, wherein the target channel state information is the channel state information corresponding to the wireless detection signal of a detected live target.

[0071] It should be noted that the initial channel state information is unfiltered and uncleaned CSI data. Furthermore, the complete channel state information is cleaned and filtered CSI data that does not contain outliers.

[0072] In this embodiment, after the electronic device controls the wireless signal transmitting module to continuously transmit multiple wireless detection signals into the cockpit, it further controls the wireless signal transmitting module to receive the wireless reflection signals generated by each of the multiple wireless detection signals. The electronic device then controls the wireless signal transmitting module to obtain the initial channel state information (without cleaning processing) matching each of the multiple wireless detection signals and the wireless reflection signals, based on the received signals. The electronic device then controls the wireless signal transmitting module to input each initial channel state information into a data processing unit configured within the electronic device. The data processing unit then performs cleaning and noise reduction processing on each initial channel state information to obtain complete channel state information that does not contain outliers. Afterward, the data processing unit filters each complete channel state information to determine the target channel state information contained in each complete channel state information that corresponds to the wireless detection signal that detected the liveness detection target.

[0073] For example, after the electronic device controls the first Wi-Fi Service configured in the TCAM to continuously transmit three first Wi-Fi signals into the cockpit, it controls the first Wi-Fi Service to continuously receive the first reflected signals generated by each of the three first Wi-Fi signals. Based on each first reflected signal and each first Wi-Fi signal, it determines the initial CSI data matched by each of the three first Wi-Fi signals. Simultaneously, after the electronic device controls the second Wi-Fi Service configured in the DHU to continuously transmit three second Wi-Fi signals into the cockpit, it controls the second Wi-Fi Service to continuously receive the second reflected signals generated by each of the three second Wi-Fi signals. Based on each second reflected signal and each second Wi-Fi signal, it determines the initial CSI data matched by each of the three second signals. The electronic device then controls the TCAM and DHU to input their respective initial CSI data into the data processing unit Algorithm Service configured in the DHU. The Algorithm Service performs data cleaning operations on each initial CSI data to eliminate abnormal data contained in each initial CSI data, thereby obtaining complete CSI data without abnormal data. Then, the Algorithm... The Service filters each complete CSI data for each first Wi-Fi signal and each second Wi-Fi signal, thereby identifying the target first Wi-Fi signal and each target second Wi-Fi signal that have detected a liveness detection target. The Algorithm Service then uses the complete CSI data corresponding to each target first Wi-Fi signal and each target second Wi-Fi signal as target CSI data to determine whether the detected live target is a human body.

[0074] In this way, by calling the wireless transmission module to transmit multiple wireless detection signals to the cockpit, the electronic device can obtain the initial channel state information generated by each of the multiple wireless detection signals, and based on the initial channel state information, determine the target wireless detection signal that detects a live target among the multiple wireless detection signals, and determine the target channel state information that matches each target wireless detection signal.

[0075] In one feasible implementation, the step of "detecting the initial channel state information that matches each of the plurality of wireless detection signals" in step S101 above may specifically include steps S1011 to S1013:

[0076] Step S1011: Detect the wireless received signal that matches each of the plurality of wireless detection signals;

[0077] Step S1012: Determine the signal change ratio between each of the plurality of wireless detection signals and the matched wireless signal;

[0078] Step S1013: Based on the signal change ratio matched by each of the multiple wireless detection signals, determine the initial channel state information matched by each of the multiple wireless detection signals.

[0079] In this embodiment, after the electronic device controls the wireless signal transmitting module to continuously transmit multiple wireless detection signals into the cockpit, it further controls the wireless signal transmitting module to receive the wireless reflection signals generated by each of the multiple wireless detection signals. Then, the wireless signal transmitting module compares the multiple wireless detection signals with their respective matched wireless received signals to determine the signal change ratio between each of the multiple wireless detection signals and the matched wireless received signals. Finally, the wireless signal transmitting module determines the signal change ratio corresponding to each of the multiple wireless detection signals as the initial channel state information matched by each of the multiple wireless detection signals, and inputs each initial channel state information to the aforementioned data processing unit.

[0080] For example, after the electronic device controls the TCAM to continuously transmit three first Wi-Fi signals into the cockpit via the first Wi-Fi Service, it further controls the TCAM to receive the first Wi-Fi reflection signals formed by the reflection of each first Wi-Fi signal after contacting an obstacle. Then, the TCAM compares each first Wi-Fi signal with its corresponding matching first Wi-Fi reflection signal to determine the first signal-to-wave ratio between each first Wi-Fi signal and its matching first Wi-Fi reflection signal. Finally, the TCAM determines the first signal-to-wave ratio corresponding to each first Wi-Fi signal as the initial CSI data corresponding to each first Wi-Fi signal. Similarly, the electronic device controls the DHU to configure the second Wi-Fi... After the Service continuously transmits three second Wi-Fi signals into the cockpit, it further controls the DHU to receive the second Wi-Fi reflection signals formed by the reflection of each second Wi-Fi signal after contacting an obstacle. Then, the DHU compares each second Wi-Fi signal with its corresponding second Wi-Fi reflection signal to determine the second signal wave ratio formed between each second Wi-Fi signal and its corresponding second Wi-Fi reflection signal. Finally, the TCAM determines the second signal wave ratio corresponding to each second Wi-Fi signal as the second initial CSI data corresponding to each second Wi-Fi signal. Finally, the electronic equipment controls the TCAM and DHU to input the collected initial CSI data into the data processing unit Algorithm Service configured in the DHU.

[0081] In this way, by calling the wireless transmission module to transmit multiple wireless detection signals to the cockpit, the electronic device can obtain the initial channel state information generated by each of the multiple wireless detection signals.

[0082] In one feasible implementation, the step of "processing each of the initial channel state information to obtain each complete channel state information" in step S101 above may specifically include steps S1014 to S1016:

[0083] Step S1014: Determine the initial amplitude parameter and initial phase parameter contained in each of the initial channel state information;

[0084] Step S1015: Perform phase expansion on each of the initial amplitude parameters and each of the initial phase parameters, and filter and eliminate the phase-expanded initial amplitude parameters and each of the initial phase parameters to obtain each target amplitude parameter and each target phase parameter;

[0085] Step S1016: Fuse each of the target amplitude parameters and their respective matching target phase parameters to obtain complete channel state information.

[0086] In this embodiment, after acquiring each initial channel state information, the data processing unit first reads the initial amplitude parameters and initial phase parameters contained in each initial channel state information. Then, the data processing unit performs phase expansion operation on each initial amplitude parameter and initial phase parameter. The data processing unit then performs filtering and elimination operation on each initial amplitude parameter and initial phase parameter after phase expansion to eliminate the outliers contained in each initial amplitude parameter and initial phase parameter, thereby obtaining each target amplitude parameter and each target phase parameter. Finally, the data processing unit fuses each target amplitude parameter and its matching target phase parameter to obtain each complete channel state information that does not contain outliers.

[0087] For example, after obtaining the initial CSI data input from TCAM and DHU respectively, the Algorithm Service first extracts the initial amplitude parameters and initial phase parameters contained in each initial CSI data. Then, the Algorithm Service performs phase expansion processing on each initial amplitude parameter and initial phase parameter, so that the expanded amplitude curve and phase curve are restored to continuous curves. At the same time, the Algorithm Service performs median filtering and uniform filtering processing on each expanded initial amplitude parameter and initial phase parameter to eliminate outliers in the time and frequency domains, thereby obtaining target amplitude parameters and target phase parameters with target outliers eliminated. Finally, the Algorithm Service fuses each target amplitude parameter and its matching target phase parameter to obtain complete CSI data without outliers.

[0088] It should be noted that the initial CSI data suffers from random amplitude and phase drift and inversion. Therefore, without denoising the initial CSI data, these outlier values ​​will affect the final detection results. Furthermore, while the amplitude and phase in the initial CSI data can be calculated, phases and amplitudes exceeding the preset function range will fold, resulting in discontinuities. Thus, a phase unrolling operation is required to restore the phases and amplitudes to a continuous state. The specific calculation process for phase unrolling is existing technology and will not be elaborated here. Similarly, the Algorithm Service's median and uniform filtering processes for the unrolled phase and amplitude parameters are also existing technology and will not be elaborated here.

[0089] In this way, the electronic equipment can remove the outliers contained in each initial channel state information, thereby obtaining complete channel state information without outliers, thus ensuring the accuracy of the detection results for personnel left in the cockpit.

[0090] In one feasible implementation, step S102 may specifically include steps S1021 to S1022:

[0091] Step S1021: Obtain preset standard channel state information, wherein the standard channel state information is the channel state information corresponding to the wireless detection signal that detects a non-live target;

[0092] Step S1022: Compare each of the complete channel state information with the standard channel state information, so as to determine the complete channel state information that is inconsistent with the standard channel state information as the target channel state information.

[0093] It should be noted that the standard channel state information is the CSI data corresponding to the Wi-Fi signal that detects a non-live target. It can be understood that when multiple Wi-Fi signals detect a non-live target, the signal change ratio between each Wi-Fi signal and its corresponding Wi-Fi reflection signal should be consistent. That is, when the CSI data corresponding to multiple Wi-Fi signals changes, it can be determined that the Wi-Fi signal has detected a moving live target.

[0094] In this embodiment, after obtaining each complete channel state information, the data processing unit first reads the storage module configured in the electronic device to obtain the preset standard channel state information. The data processing unit compares each obtained complete channel state information with the standard channel state information to obtain multiple comparison results. Finally, the data processing unit reads the multiple comparison results and determines the complete channel state information that is inconsistent with the standard channel state information as the target channel state information corresponding to the wireless detection signal that detects a live target based on the multiple first comparison results.

[0095] For example, after obtaining each complete CSI data, the Algorithm Service first reads the storage module configured in the electronic device to obtain the preset standard CSI data corresponding to when the Wi-Fi signal detects a non-living target. Then, the Algorithm Service compares each complete CSI data with the standard CSI data to obtain multiple comparison results. Finally, the Algorithm Service determines the target comparison results among the multiple comparison results, where the comparison results are inconsistent between the complete CSI data and the standard CSI data. Based on the multiple target comparison results, the Algorithm Service filters the multiple Wi-Fi signals, thereby determining the Wi-Fi signal corresponding to the target comparison result as the target Wi-Fi signal that has detected a live target.

[0096] In this way, the electronic device can identify the target wireless detection signal that detects a live target among multiple transmitted wireless detection signals, and then determine the target channel state information that matches each target wireless detection signal.

[0097] Step S20: Generate a target liveness image based on multiple target channel state information, and extract image features from the target liveness image;

[0098] In this embodiment, after obtaining multiple target channel state information, the data processing unit generates a target liveness image containing the detected liveness detection target based on the potential spatial features contained in each of the multiple target channel state information. At the same time, the data processing unit extracts the image features of the target liveness image.

[0099] For example, after obtaining the target CSI data, the Algorithm Service further inputs the target CSI data into its own RCNN, so that the RCNN performs convolution processing on the target CSI data to extract the latent spatial features contained in the target CSI data, and constructs a target liveness image containing only the liveness detection target based on the latent spatial features. At the same time, the RCNN processes the target liveness image to extract the image features of the target liveness image.

[0100] In this way, the electronic device can construct a target liveness image containing the liveness detection target based on the determined target channel state information, and extract the image features contained in the target liveness image to determine whether the liveness detection target is a human body based on the image features.

[0101] In one feasible implementation, the step of "generating a target liveness image based on multiple target channel state information" in step S20 above may specifically include steps S201 to S202:

[0102] Step S201: Extract the one-dimensional vector features contained in each of the multiple target channel state information;

[0103] Step S202: Fuse the one-dimensional vector features to obtain two-dimensional vector features, and generate a target liveness image based on the two-dimensional vector features.

[0104] It should be noted that the one-dimensional vector feature is the latent spatial feature contained within the target channel state. Specifically, the one-dimensional feature vector contains an amplitude tensor and a phase tensor, where each tensor is 150×3×3 in size (5 consecutive samples, 30 frequencies, 3 transmitters and 3 receivers). Thus, RCNN (Regions with Convolutional Neural Networks) can extract the spatial features contained in the one-dimensional vector feature through the last two dimensions of each tensor.

[0105] In this embodiment, after determining multiple target channel state information, the data processing unit first inputs the multiple target channel state information into its own configured convolutional neural network model, so that the convolutional neural network model calls multiple encoders to extract the one-dimensional feature vector contained in each of the multiple target channel state information. Then, the data processing unit fuses the one-dimensional feature vectors to obtain two-dimensional fused features, and converts the two-dimensional fused features into a feature map in the spatial domain. Then, the converted two-dimensional feature map is upsampled to obtain a target liveness image containing the liveness detection target.

[0106] For example, please refer to Figure 3 , Figure 3This is a schematic diagram of a target liveness image related to an embodiment of the method for identifying occupants left inside a vehicle, as described in this application. After identifying multiple target CSI data points, the Service first inputs these data points into its configured RCNN. The RCNN then calls two different encoders to process each target CSI data point, extracting amplitude and phase tensors of size 150×3×3 for each target CSI data point. The RCNN then defines these amplitude and phase tensors as 1D features. Next, the RCNN extracts the spatial information contained in each 1D feature based on the last two dimensions (3 transmitters and 3 receivers) of each amplitude and phase tensor. It then fuses these 1D features to obtain a fused 1D feature. The RCNN then reshapes the fused 1D feature into an initial 24×24 2D feature map and extracts the spatial information contained in the initial 2D feature map using its configured two convolutional blocks to obtain a 6×6 2D feature vector. Finally, the RCNN upsamples the 6×6 2D feature vector to adjust it to a size similar to... Figure 3 The image shown is 3×720×1280 in size and contains the target liveness image in the image domain of the liveness detection target.

[0107] In this way, the electronic device can construct a target liveness image containing the liveness detection target based on the determined target channel state information, and extract the image features contained in the target liveness image to determine whether the liveness detection target is a human body based on the image features.

[0108] In one feasible implementation, the step of "extracting image features of the target live image" in step S20 above may specifically include steps S203 to S205:

[0109] Step S203: Perform phase-splitting processing on the target liveness image to obtain multiple sub-liveness images;

[0110] Step S204: Encode the multiple sub-live images to obtain each convolutional feature, and mix the convolutional features to obtain pose fusion data;

[0111] Step S205: Extract the target features contained in the pose fusion data, and determine the target features as the image features of the target liveness image.

[0112] In this embodiment, after constructing the target liveness image through a convolutional neural network (CNN) model, the data processing unit further calls the CNN model to perform phase-splitting processing on the target liveness image to obtain multiple sub-liveness images. Then, the CNN model calls the encoder to encode the multiple sub-liveness images to identify the convolutional features corresponding to each of the multiple sub-liveness images. The CNN model then fuses the convolutional features to obtain pose fusion data. Finally, the CNN model performs inference merging operations on the pose fusion data and samples the inference-merged pose fusion data to extract the target features. The CNN model then determines the target features as the image features corresponding to the target liveness image.

[0113] For example, after constructing the target liveness image, RCNN needs to perform phase segmentation on the target liveness image based on color channels, thereby obtaining multiple sub-liveness images with differences in color channels, such as the red channel image corresponding to the R channel, the green channel image corresponding to the G channel, and the blue channel image corresponding to the B channel. Then, RCNN calls the CNN encoder to encode the multiple sub-liveness images separately to identify the convolutional features corresponding to each of the multiple sub-liveness images. RCNN then fuses the obtained convolutional features to obtain preliminary pose fusion data. Finally, RCNN performs inference operations through 8-bit 3-channel RGB reshape, thereby merging the preliminary pose fusion data into 24-bit data. RCNN then performs upsampling encoding and downsampling decoding operations on the 24-bit data in sequence to output a 3-channel image. RCNN then determines the 3-channel image as the image features corresponding to the target liveness image.

[0114] In this way, the electronic device can construct a target liveness image containing the liveness detection target based on the determined target channel state information, and extract the image features contained in the target liveness image to determine whether the liveness detection target is a human body based on the image features.

[0115] Step S30: If the image features are detected to match the preset human features, it is determined that there are target personnel left behind in the cockpit;

[0116] In this embodiment, after the data processing unit extracts the image features, it further compares the image features with preset human body features. When the data processing unit finds that the image features and human body features match, it determines that the above-mentioned liveness detection target is a human body, and thus determines that there are target personnel left behind in the cabin.

[0117] For example, after extracting image features through the RCNN described above, the Algorithm Service further reads the storage module to obtain preset human body features. The Algorithm Service then compares the image features with the human body features. If the Algorithm Service determines that the comparison result is a match between the image features and the human body features, it determines that the liveness detection target detected by the target Wi-Fi signals is a human body, thereby determining that there are target personnel left behind in the cabin.

[0118] Similarly, if the Algorithm Service determines that the image features and human features do not match, it determines that the live detection targets detected by the Wi-Fi signals of the above targets are other animals, thereby determining that there are no target personnel left in the cabin.

[0119] In addition, in this embodiment and another embodiment, if the Algorithm Service determines that there are target personnel left behind in the cabin, it can also send the detection results to the QNX virtual machine configured in the DHU. The QNX virtual machine converts the detection results into CAN signals through the MCU connected to the DHU, and sends the CAN signals to hardware devices such as buzzers and LED headlights connected to the DHU through the BUS bus in the vehicle, so as to control the buzzers, LED headlights and other devices to enter the operating state.

[0120] In this embodiment, when the electronic device detects that the vehicle has entered a locked state, it first calls the wireless signal transmission module to continuously transmit multiple wireless detection signals into the vehicle's cabin. The electronic device then controls the wireless signal transmission module to receive the wireless reflection signals generated by each of the multiple wireless detection signals. Based on the received wireless detection signals and wireless reflection signals, the wireless signal transmission module obtains multiple initial channel state information. The electronic device then controls the wireless signal transmission module to input the obtained multiple initial channel state information into the data processing unit configured within the electronic device. The data processing unit filters the multiple initial channel state information to determine the target channel state information that matches each of the multiple wireless detection signals that detect a live target. Then, based on the potential spatial features contained in each of the multiple target channel state information, the data processing unit generates a target live image containing the detected live target. At the same time, the data processing unit extracts the image features of the target live image. Finally, the data processing unit compares the image features with preset human body features. If the comparison shows that the image features and human body features match, the data processing unit determines that the above-mentioned live target is a human body, and thus determines that there is a target person left in the cabin.

[0121] Thus, this application solves the technical problem of low accuracy in identifying people left in the cabin in related technologies. Specifically, by utilizing the characteristic that wireless detection signals generate different channel state information when they come into contact with a living person, this application can accurately identify whether a living person in the cabin is a person left behind. This avoids interference from obstacles during the cabin detection process, thereby improving the accuracy of identifying people left behind in the cabin and greatly increasing vehicle safety.

[0122] Based on the first embodiment of this application, a second embodiment of this application is proposed herein. In this second embodiment, content that is the same as or similar to the above embodiments can be referred to the above description and will not be repeated hereafter. Furthermore, after step S30 above, the method for identifying persons left inside a vehicle may further include step A10:

[0123] Step A10: Determine the identity information of the target person left behind based on the image features, and generate a message indicating that the person has been left behind based on the identity information.

[0124] In this embodiment, when the electronic device determines that there is a target person left behind in the cockpit, it can also determine the identity information of the target person left behind based on the extracted image features. At the same time, the electronic device inputs the identity information into the virtual machine unit configured in the wireless signal transmission module. The virtual machine unit then generates a leave-behind notification message based on the identity information. Afterwards, the virtual machine unit sends the leave-behind notification message to the user's mobile terminal, which is connected to the electronic device, through the Android subsystem running inside. The user's mobile terminal then displays the leave-behind notification message to the user, thereby informing the user of the identity of the person left behind in the cockpit.

[0125] For example, when an electronic device detects that a target person has been left behind in the cabin, it first queries a preset identity database based on the extracted image features to determine the name information that matches the image features. The electronic device then inputs the name information into the QNX virtual machine configured in the DHU, and the QNX virtual machine generates a message indicating that a person has been left behind based on the name information. After that, the QNX virtual machine sends the message to the Android subsystem it is running, and the Android subsystem sends the message to a mobile device that is connected to the electronic device. The target app on the mobile device then displays the message to the user, indicating that a target person has been left behind in the cabin and informing the user of the name information of the target person left behind in the cabin.

[0126] In addition, in this embodiment and another embodiment, after the electronic device identifies the identity information of the target person left behind, the virtual machine unit can determine a preset function control command that matches the target person left behind based on the identity information. Finally, the electronic device controls the various functional modules configured in the vehicle according to the function control command to adjust the functions such as speakers, air conditioning, and seat heating in the cabin, so that each functional module enters the operating mode that matches the function control command.

[0127] In this way, when the electronic device detects the presence of a target person left behind in the cabin, it can further identify the identity information of the target person left behind, generate a message of abandonment based on the identity information, and then send the message of abandonment to the mobile terminal held by the car owner, so that the car owner can know the specific information of the person left behind in the cabin.

[0128] In one feasible implementation, the step of "determining the identity information of the target remaining person based on the image features" in step A10 above may specifically include steps A101 to A102:

[0129] Step A101: Determine the preset personnel features that match the image features;

[0130] Step A102: The preset identification information matched with the preset personnel characteristics is determined as the identity identification information of the target remaining personnel.

[0131] In this embodiment, when the electronic device determines that there is a target person left behind in the cabin, it first reads the aforementioned storage module to obtain multiple preset personnel features and preset identification information that matches each of the multiple preset personnel features. The electronic device filters the multiple preset personnel features based on image features to determine the preset personnel features that match the image features. Finally, the electronic device determines the preset identification information corresponding to the preset personnel features that match the image features as the identity identification information corresponding to the target person left behind in the cabin.

[0132] For example, when the electronic device determines that there is a target person left behind in the cabin, it first reads the aforementioned storage module to obtain an identity database containing multiple preset personnel features and preset name information that matches each of the multiple preset personnel features. The DHU queries the identity database based on the obtained image features to match the image features with the multiple preset personnel features contained in the identity database, thereby determining the target preset personnel feature that matches the image features among the multiple preset personnel features. Finally, the DHU determines the preset name information that matches the target preset personnel feature in the identity database as the name identification information corresponding to the target person left behind in the cabin.

[0133] In this way, when the electronic device detects the presence of a target person left behind in the cabin, it can further identify the identity information of the target person left behind, generate a message of abandonment based on the identity information, and then send the message of abandonment to the mobile terminal held by the car owner, so that the car owner can know the specific information of the person left behind in the cabin.

[0134] For example, to help understand the implementation process of the method for identifying people left in a vehicle after combining the above embodiments, please refer to... Figure 4 , Figure 4 This is a simplified flowchart illustrating the method for identifying occupants left inside the vehicle in this application. Specifically:

[0135] In this embodiment, when the electronic device detects that the vehicle has entered the locked state, it first activates its configured TCAM and DHU. The Core Service in the TCAM checks whether the TCAM and DHU can communicate normally. If the Core Service detects that the TCAM and DHU are communicating normally, the electronic device controls the first Wifi Service in the TCAM to transmit multiple first Wifi signals into the cabin and controls the first Wifi Service to receive the first Wifi reflection signals generated by each first Wifi signal. The first Wifi Service then generates multiple first initial CSI data based on each first Wifi signal and each first Wifi reflection signal. The TCAM then sends the generated first initial CSI data to the data processing unit in the DHU.

[0136] Simultaneously, the electronic equipment controls the second Wi-Fi service within the DHU to transmit multiple second Wi-Fi signals into the cockpit and controls the second Wi-Fi... The Service receives the second Wi-Fi reflection signals generated by each second Wi-Fi signal. The second Wi-Fi Service then generates multiple second initial CSI data based on each second Wi-Fi signal and each second Wi-Fi reflection signal. The DHU then sends the generated second CSI data to the data processing unit. Subsequently, the data processing unit performs noise reduction processing on the acquired first initial CSI data and each second initial CSI data to obtain multiple complete CSI data. At this time, the data processing unit reads the storage module in the electronic device to obtain the preset CSI data corresponding to the Wi-Fi signal when a non-live detection target is detected. The complete CSI data and the preset CSI data are compared to obtain multiple comparison results. Based on the multiple comparison results, the data processing unit determines the target first Wi-Fi signal that detects a live detection target in each first Wi-Fi signal and the target second Wi-Fi signal that detects a live detection target in each second Wi-Fi signal. The data processing unit then filters each complete CSI data based on each target first Wi-Fi signal and each target second Wi-Fi signal to determine the target CSI data for whether a human body is detected in the user's cabin.

[0137] Next, the data processing unit inputs the CSI data of each target into a preset convolutional neural network model. The convolutional neural network model extracts the one-dimensional feature vector contained in the CSI data of each target and fuses the one-dimensional feature vectors to generate a target detection image containing the liveness detection target. The convolutional neural network model then extracts the image features of the target detection image and matches the image features with preset human body features.

[0138] Finally, after determining that the image features and human features match, the convolutional neural network identifies the liveness detection target as a person left behind in the cockpit. The data processing unit then queries the identity information of the person left behind based on the image features and inputs the identity information into the QNX virtual machine configured in the DHU. The QNX virtual machine generates a message of abandonment based on the identity information and sends the message of abandonment to the user's mobile terminal device through its own Android subsystem, so that the message of abandonment can be displayed to the user through the mobile terminal device.

[0139] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method for identifying people left in the vehicle in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0140] This application provides a device for identifying occupants left inside a vehicle. The device is characterized in that it is applied to an electronic device, which includes a wireless signal transmitting module. (Please refer to...) Figure 5 , Figure 5 This is a schematic diagram of the module structure of the identification device for people left in the vehicle according to an embodiment of this application, as shown below. Figure 5 As shown, the device includes:

[0141] The signal detection module 10 is used to control the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, and to determine multiple target channel status information based on the multiple wireless detection signals.

[0142] The feature extraction module 20 is used to generate a target liveness image based on multiple target channel state information and extract image features of the target liveness image;

[0143] The feature recognition module 30 is used to determine the presence of a target person left behind in the cabin when the image features are detected to match the preset human features.

[0144] In one feasible implementation, the signal detection module 10 is further configured to:

[0145] The initial channel state information matched by each of the multiple wireless detection signals is detected, and each initial channel state information is processed to obtain complete channel state information.

[0146] The complete channel state information is filtered to determine the target channel state information, wherein the target channel state information is the channel state information corresponding to the wireless detection signal of a detected live target.

[0147] In one feasible implementation, the signal detection module 10 is further configured to:

[0148] Detect the wireless received signal that matches each of the multiple wireless detection signals;

[0149] Determine the signal change ratio between each of the plurality of wireless detection signals and the matched wireless signal;

[0150] Based on the signal change ratio matched by each of the multiple wireless detection signals, the initial channel state information matched by each of the multiple wireless detection signals is determined.

[0151] In one feasible implementation, the signal detection module 10 is further configured to:

[0152] Determine the initial amplitude parameter and initial phase parameter included in each of the initial channel state information;

[0153] Phase expansion is performed on each of the initial amplitude parameters and each of the initial phase parameters, and the phase-expanded initial amplitude parameters and each of the initial phase parameters are filtered and eliminated to obtain each target amplitude parameter and each target phase parameter.

[0154] The target amplitude parameters and their respective matching target phase parameters are fused to obtain complete channel state information.

[0155] In one feasible implementation, the signal detection module 10 is further configured to:

[0156] Obtain preset standard channel state information, wherein the standard channel state information is the channel state information corresponding to the detection of a wireless detection signal of a non-living detection target;

[0157] Each complete channel state information is compared with the standard channel state information, and the complete channel state information that is inconsistent with the standard channel state information is determined as the target channel state information.

[0158] In one feasible implementation, the feature extraction module 20 is further configured to:

[0159] Extract the one-dimensional vector features contained in each of the multiple target channel state information;

[0160] The one-dimensional vector features are fused to obtain two-dimensional vector features, and a target liveness image is generated based on the two-dimensional vector features.

[0161] In one feasible implementation, the feature extraction module 20 is further configured to:

[0162] The target liveness image is subjected to phase-splitting processing to obtain multiple sub-liveness images;

[0163] Encode multiple sub-liveness images to obtain convolutional features, and mix the convolutional features to obtain pose fusion data;

[0164] Extract the target features contained in the pose fusion data, and determine the target features as the image features of the target liveness image.

[0165] In one feasible implementation, the feature recognition module 30 is further configured to:

[0166] Based on the image features, the identity information of the target person left behind is determined, and a message indicating that the person has been left behind is generated based on the identity information.

[0167] In one feasible implementation, the feature recognition module 30 is further configured to:

[0168] Determine the preset personnel features that match the image features;

[0169] The preset identification information that matches the preset personnel characteristics is determined as the identity identification information of the target remaining personnel.

[0170] This application provides an electronic device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the vehicle occupant identification method of Embodiment 1 described above.

[0171] The following is for reference. Figure 6 The diagram illustrates a structural schematic of an electronic device suitable for implementing the embodiments of this application. The electronic device in the embodiments of this application may include, but is not limited to, an electronic device with an internally configured wireless signal transmitting module, or a mobile terminal, data storage control terminal, PC, or other terminal connected to an electronic control unit associated with the electronic device. Figure 6 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0172] like Figure 6 As shown, the electronic device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the electronic device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. While electronic devices with various systems are shown in the figures, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0173] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0174] The electronic device provided in this application, employing the vehicle occupant identification method described in the above embodiments, can solve the technical problem of low accuracy in identifying occupants left in the cabin in related technologies. Compared with the prior art, the beneficial effects of the electronic device provided in this application are the same as those of the vehicle occupant identification method provided in the above embodiments, and other technical features of the electronic device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0175] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0176] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0177] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the vehicle occupant identification method in the above embodiments.

[0178] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0179] The aforementioned computer-readable storage medium may be included in an electronic device or may exist independently without being assembled into an electronic device.

[0180] The aforementioned computer-readable storage medium carries one or more programs that, when executed by an electronic device, cause the electronic device to: control the wireless signal transmitting module to transmit multiple wireless detection signals to the vehicle's cabin, and determine multiple target channel state information based on the multiple wireless detection signals; generate a target liveness image based on the multiple target channel state information, and extract image features from the target liveness image; and, if the image features are detected to match preset human features, determine that a target person remains in the cabin.

[0181] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can 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 can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0182] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0183] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0184] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for identifying occupants left in the vehicle, thereby solving the technical problem of low accuracy in identifying occupants left in the cabin in related technologies. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the method for identifying occupants left in the vehicle provided in the above embodiments, and will not be repeated here.

[0185] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method for identifying persons left in a vehicle as described above.

[0186] The computer program product provided in this application can solve the technical problem of low accuracy in identifying occupants left in the cabin in related technologies. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the vehicle occupant identification method provided in the above embodiments, and will not be repeated here.

[0187] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for identifying persons left inside a vehicle, characterized in that, The method for identifying persons left inside a vehicle is applied to an electronic device, which includes a wireless signal transmitting module. The method for identifying persons left inside a vehicle includes: The wireless signal transmitting module is controlled to transmit multiple wireless detection signals to the vehicle's cabin, and multiple target channel status information is determined based on the multiple wireless detection signals; A target liveness image is generated based on multiple target channel state information, and image features of the target liveness image are extracted; If the image features are detected to match the preset human features, it is determined that there are target personnel left behind in the cockpit; The step of generating a target liveness image from the state information includes: One-dimensional vector features contained in each of the multiple target channel state information are extracted using a convolutional neural network; The one-dimensional vector features are fused to obtain two-dimensional vector features, and a target liveness image is generated based on the two-dimensional vector features.

2. The method for identifying persons left inside a vehicle as described in claim 1, characterized in that, The step of determining multiple target channel state information based on multiple wireless detection signals includes: The initial channel state information matched by each of the multiple wireless detection signals is detected, and each initial channel state information is processed to obtain complete channel state information. The complete channel state information is filtered to determine the target channel state information, wherein the target channel state information is the channel state information corresponding to the wireless detection signal of a detected live target.

3. The method for identifying persons left inside a vehicle as described in claim 2, characterized in that, The step of detecting the initial channel state information that matches each of the plurality of wireless detection signals includes: Detect the wireless received signal that matches each of the multiple wireless detection signals; Determine the signal change ratio between each of the plurality of wireless detection signals and the matched wireless signal; Based on the signal change ratio matched by each of the multiple wireless detection signals, the initial channel state information matched by each of the multiple wireless detection signals is determined.

4. The method for identifying persons left inside a vehicle as described in claim 2, characterized in that, The step of processing each of the initial channel state information to obtain each complete channel state information includes: Determine the initial amplitude parameter and initial phase parameter included in each of the initial channel state information; Phase expansion is performed on each of the initial amplitude parameters and each of the initial phase parameters, and the phase-expanded initial amplitude parameters and each of the initial phase parameters are filtered and eliminated to obtain each target amplitude parameter and each target phase parameter. The target amplitude parameters and their respective matching target phase parameters are fused to obtain complete channel state information.

5. The method for identifying persons left inside a vehicle as described in claim 2, characterized in that, The step of filtering each of the complete channel state information to determine the target channel state information includes: Obtain preset standard channel state information, wherein the standard channel state information is the channel state information corresponding to the detection of a wireless detection signal of a non-living detection target; Each complete channel state information is compared with the standard channel state information, and the complete channel state information that is inconsistent with the standard channel state information is determined as the target channel state information.

6. The method for identifying persons left inside a vehicle as described in claim 1, characterized in that, The step of extracting image features from the target live image includes: The target liveness image is subjected to phase-splitting processing to obtain multiple sub-liveness images; Encode multiple sub-liveness images to obtain convolutional features, and mix the convolutional features to obtain pose fusion data; Extract the target features contained in the pose fusion data, and determine the target features as the image features of the target liveness image.

7. The method for identifying persons left inside a vehicle as described in claim 1, characterized in that, After determining that a target person remains in the cockpit, the method further includes: Based on the image features, the identity information of the target person left behind is determined, and a message indicating that the person has been left behind is generated based on the identity information.

8. The method for identifying persons left inside a vehicle as described in claim 7, characterized in that, The step of determining the identity information of the target abandoned person based on the image features includes: Determine the preset personnel features that match the image features; The preset identification information that matches the preset personnel characteristics is determined as the identity identification information of the target remaining personnel.

9. An electronic device, characterized in that, The electronic device includes: a wireless signal transmitting module, a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the method for identifying persons left in a vehicle as described in any one of claims 1 to 8.

10. A vehicle, characterized in that, The vehicle includes the electronic equipment as described in claim 9.

11. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the method for identifying persons left in the vehicle as described in any one of claims 1 to 8.

12. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the method for identifying persons left in a vehicle as described in any one of claims 1 to 8.

Images