An occupancy detection method, device and medium in an automotive cabin

Abstract

Embodiments of the present application disclose a seat detection method, device and medium in a vehicle cabin. The method comprises: obtaining original echo data of a target object in the cabin collected by a millimeter wave radar, determining current output data, performing target detection on the target object based on the current output data to obtain a target detection result, determining target parameter information according to the target detection result, determining coordinate information of the target object according to a cabin environment region space model and the target parameter information, and determining a seat detection result in the cabin according to the coordinate information and a preset clustering algorithm. In the embodiments of the present application, the target object is detected by the current output data to obtain parameter information corresponding to the detection result, the coordinate information of the target object is determined based on the parameter information, and the seat detection result in the cabin is determined in combination with the preset clustering algorithm, so that the problems of missed detection and false detection in target object detection are solved, and the accuracy and robustness of seat detection are improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle cabin occupancy detection technology, and more particularly to a method, device and medium for detecting occupancy in an automobile cabin. Background Technology

[0002] The development of intelligent vehicles is booming, and intelligent cockpits can greatly enhance the user's driving experience. Passenger perception is a crucial component of this, aiming to detect the location of rear-seat occupants and provide feedback to the owner, enabling them to accurately understand the vehicle's interior. Current technologies typically employ several methods for occupant detection within the vehicle cabin. For example, using height and position sensors to determine a passenger's position based on their conductivity. However, this device can only identify passengers in their seats and requires initial calibration after each seat adjustment or maintenance. Another method uses video image-based techniques to detect occupant positions, employing background subtraction and skin color-based facial recognition algorithms to separate the occupant from the image, then using ellipse fitting to fit the upper body contour and tracking it. However, this method uses camera sensors, posing privacy risks.

[0003] Furthermore, traditional positioning algorithms for detecting occupancy issues often suffer from missed or false detections in scenarios where the target is stationary, there are moving and stationary targets simultaneously, or the target is in a difficult-to-detect location such as footspace. Additionally, the unstable instantaneous state of a person during movement can easily lead to a single target being falsely reported as multiple targets. Therefore, there is an urgent need for an in-vehicle cabin occupancy detection method to detect occupancy within the car's passenger compartment. Summary of the Invention

[0004] In view of this, the present invention provides a method, device and medium for detecting occupancy in a car cabin, which can solve the problems of missed detection, false detection and false alarm when detecting target objects, and can effectively detect moving targets and stationary weak targets in the scene, thereby improving the accuracy and robustness of occupancy detection in the cabin.

[0005] According to one aspect of the present invention, an embodiment of the present invention provides a method for detecting occupancy in a car cabin, the method comprising:

[0006] Acquire raw echo data of at least one target object inside the cockpit collected by millimeter-wave radar, and determine the current output data of the raw echo data;

[0007] Based on the current output data, target detection is performed on the target object to obtain a target detection result, and target parameter information is determined based on the target detection result; wherein, the target detection includes the detection of motion data and micro-motion data of the target object;

[0008] The coordinate information of the target object is determined based on the pre-constructed cabin environment area spatial model and the target parameter information;

[0009] The occupancy detection result of the target object in the cockpit is determined based on the coordinate information and the preset clustering algorithm.

[0010] According to another aspect of the present invention, embodiments of the present invention also provide an occupancy detection device in a car cabin, the device comprising:

[0011] The output data determination module is used to acquire the raw echo data of at least one target object in the cockpit collected by the millimeter-wave radar, and determine the current output data of the raw echo data.

[0012] The parameter information determination module is used to perform target detection on the target object based on the current output data to obtain a target detection result, and determine target parameter information based on the target detection result; wherein, the target detection includes the detection of motion data and micro-motion data of the target object;

[0013] The coordinate determination module is used to determine the coordinate information of the target object based on the pre-built cabin environment area spatial model and the target parameter information;

[0014] The result determination module is used to determine the occupancy detection result of the target object in the cockpit based on the coordinate information and a preset clustering algorithm.

[0015] According to another aspect of the present invention, embodiments of the present invention also provide an electronic device, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the occupancy detection method in the car cabin according to any embodiment of the present invention.

[0019] According to another aspect of the present invention, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions for causing a processor to execute and implement the occupancy detection method in a car cabin as described in any embodiment of the present invention.

[0020] The technical solution of this invention uses the current output data of the original echo data corresponding to the target object in the cockpit collected by millimeter-wave radar. By installing the radar in the vehicle cabin to measure the displacement of the human chest and the signals of various parts, the location of the target point can be obtained without privacy risks. Based on the current output data, motion data detection and micro-motion data detection of the target object are performed to obtain the target detection result. The target parameter information is determined based on the target detection result. The coordinate information of the target object is determined based on the spatial model of the cockpit environment area and the target parameter information. The occupancy detection result of the target object in the cockpit is determined based on the coordinate information and the preset clustering algorithm. This can solve the problems of missed detection, false detection and false alarm when detecting target objects. It can effectively detect moving targets and stationary weak targets in the scene, and improve the accuracy and robustness of occupancy detection in the cockpit.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a method for detecting occupancy in a car cabin according to an embodiment of the present invention;

[0024] Figure 2 A flowchart illustrating another method for detecting occupancy in a car cabin according to an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of a one-dimensional fast time dimension image distribution of raw echo data provided in an embodiment of the present invention;

[0026] Figure 4 This is a schematic diagram of a one-dimensional fast time dimension image distribution for motion data through mean cancellation, provided in an embodiment of the present invention.

[0027] Figure 5 This is a schematic diagram of a one-dimensional fast temporal image distribution obtained by inter-frame cancellation of micro-motion data according to an embodiment of the present invention;

[0028] Figure 6 This is a schematic diagram showing the detection result of a constant false alarm rate (CFAR) detection method provided in an embodiment of the present invention.

[0029] Figure 7 This is a schematic diagram of the spatial spectrum of a moving target provided in an embodiment of the present invention;

[0030] Figure 8 This is a schematic diagram of a target angle estimation result provided in an embodiment of the present invention;

[0031] Figure 9 This is a schematic diagram of a modeling of the rear passenger compartment of a car, provided in an embodiment of the present invention.

[0032] Figure 10 This is a schematic diagram of a coordinate information clustering result provided in an embodiment of the present invention;

[0033] Figure 11 This is a structural block diagram of a space occupancy detection device in an automobile cabin, provided in an embodiment of the present invention.

[0034] Figure 12 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0035] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0037] In one embodiment, Figure 1This is a flowchart of a method for detecting occupancy in a car cabin according to an embodiment of the present invention. This embodiment is applicable to situations where occupancy in a car cabin is detected. The method can be executed by an occupancy detection device in the car cabin, which can be implemented in hardware and / or software and can be configured in an electronic device.

[0038] like Figure 1 As shown, the method for detecting occupancy in the car cabin in this embodiment includes the following specific steps:

[0039] S110. Acquire raw echo data of at least one target object inside the cockpit collected by the millimeter-wave radar, and determine the current output data of the raw echo data.

[0040] The raw echo data refers to the ADC raw data obtained after the receiver acquires multi-channel received signals and performs a series of processes such as mixing, filtering, and sampling. The raw echo data contains various information corresponding to the target object collected by the millimeter-wave radar, including, but not limited to, motion data and micro-motion data of one or more target objects. The current output data can be understood as the current output data obtained by performing a series of data processing steps on the raw echo data.

[0041] In this embodiment, the millimeter-wave radar system used in the vehicle cabin has multiple transmitting and receiving antennas. It transmits Linear Frequency Modulation Continuous Wave (LFMCW) signals and acquires multi-channel received signals at the receiving end. After a series of processing steps including mixing, filtering, and sampling, the raw ADC data is obtained. In this embodiment, the millimeter-wave radar collects raw echo data of at least one target object within the cabin. The target object can include, but is not limited to, animals and users; users can include children, adults, etc.

[0042] In this embodiment, the raw data frames of the target object inside the cockpit collected by the millimeter-wave radar can be acquired, processed to obtain raw echo data, and then subjected to a one-dimensional Fourier transform to obtain the current output data corresponding to the raw echo data. In some embodiments, the returned raw echo data can also be acquired, processed by Doppler filtering to obtain Doppler data, and then used as the corresponding current output data. This embodiment does not limit the method of determining the current output data of the raw echo data. It should be noted that the millimeter-wave radar can detect the breathing, heartbeat, and other movements of living beings inside the vehicle cabin by transmitting signals, and the time delay of its received signal will change with the movement of the living beings. The received signal of the millimeter-wave radar is processed by mixing, filtering, etc. to obtain an intermediate frequency signal. The intermediate frequency signal contains the motion information of the living beings. By processing the intermediate frequency signal through cancellation, accumulation, detection, positioning, clustering, etc., the position information of the living beings inside the vehicle cabin can be finally realized.

[0043] S120. Based on the current output data, target detection is performed on the target object to obtain the target detection result, and the target parameter information is determined according to the target detection result; wherein, target detection includes the detection of motion data and micro-motion data of the target object.

[0044] The target detection result refers to the detection results corresponding to the detection of motion data and micro-motion data of the target object, respectively. It should be noted that motion data can be the motion data composed of the traces of a living target with body movement; micro-motion data refers to the data composed of the traces of a living target in a static state; of course, motion data can include, but is not limited to, the speed of the living object's movement, the signal amplitude generated by the movement, the distance of movement, etc.; micro-motion data can include, but is not limited to, the relevant data generated by the breathing, heartbeat and other movements of living beings in the cabin.

[0045] In this implementation, since target detection includes the detection of motion data and micro-motion data of the target object, the target detection result generated by the target detection includes the detection results of motion data and micro-motion data of the target object. Target parameter information can be determined based on the target detection result. This target parameter information may include, but is not limited to, the target angle information and distance information corresponding to the target object when performing micro-motion data detection and motion data detection.

[0046] In this embodiment, when target detection involves detecting the motion data of a target object, the mean of each distance unit can be calculated sequentially along the slow time dimension for each channel dimension based on the current output data. The mean of each distance unit is then subtracted from the current output data corresponding to each distance unit to obtain a first data matrix. A two-dimensional Fourier transform is performed on the first data matrix to obtain first output data. The first output data is then subjected to modulo operation and accumulated along the channel dimension to obtain a range-Doppler spectrum data matrix. A constant false alarm rate (CFAR) detection method is then used to perform target detection on the range-Doppler spectrum data matrix to obtain a list of target points corresponding to the motion data. A list of target points is used as the first target detection result of the motion data corresponding to the target object. In some embodiments, when the target detection is the detection of the micro-motion data of the target object, the difference between the current output data and the previous output data is calculated to obtain a second output result. The second output result is accumulated within the frame from two dimensions: the channel dimension and the slow time dimension. The accumulated result is subjected to a modulo operation to obtain the one-dimensional imaging data of the target object in the distance dimension. The constant false alarm rate detection method is used to perform micro-motion target detection on the one-dimensional imaging data to obtain a second target point list. The second target point list is used as the target detection result of the micro-motion data corresponding to the target object.

[0047] In some embodiments, the relevant angle measurement method of angle estimation can be used to obtain the target angle value corresponding to the target object from the relevant data information in the target detection results of motion data and the relevant data information in the target detection results of micro-motion data. The first distance information of motion data is obtained from the target detection results of motion data, and the second distance information of micro-motion data is obtained from the target detection results of micro-motion data. The target angle value, the first distance information and the second distance information are used as target parameter information.

[0048] S130. Determine the coordinate information of the target object based on the pre-constructed cockpit environment area spatial model and target parameter information.

[0049] The cockpit environment spatial model is constructed with the radar's location as the origin, based on the millimeter-wave radar's installation position within the vehicle cabin and its line-of-sight direction. Coordinate information refers to the coordinates of the target object within the corresponding area of ​​the cockpit environment spatial model.

[0050] In this embodiment, the coordinate transformation method corresponding to the target parameter information can be determined based on the installation position of the millimeter-wave radar in the vehicle cabin. The transformed coordinates corresponding to the target parameter information are then determined based on the coordinate transformation method, and the transformed target coordinates are used as the coordinates of the target object in the radar coordinate system. It can be understood that the cockpit domain space model constructed will differ depending on the installation position and line-of-sight direction of the millimeter-wave radar within the vehicle cabin, thus requiring different coordinate transformation methods for the target parameter information.

[0051] S140. Determine the occupancy detection results of target objects in the cockpit based on coordinate information and preset clustering algorithm.

[0052] The preset clustering algorithm can include, but is not limited to, density-based clustering algorithms such as DBSCAN, partition-based clustering, and hierarchical clustering. The preset clustering algorithm can be used to perform relevant clustering processing on the coordinate information of the target object. The occupancy detection result can be understood as the detection result of one or more target objects within the car cabin, which can include, but is not limited to, the detection result of liveness of the target objects in the current cabin, the detection result of the current location area of ​​each person, that is, the number of occupied seats in the detection scene, and which seats are occupied.

[0053] In this embodiment, a preset clustering algorithm can be used to sequentially cluster the coordinate points in the coordinate information to obtain the corresponding clustering results. The average coordinates of the core points of the same class are then calculated based on the clustering results. The distribution of these coordinates in the vehicle cabin spatial model determines which area has occupied parking spaces. Simultaneously, the number of target users in the current vehicle cabin is determined based on the category of the clustered core points. It should be noted that this preset clustering algorithm defines two key parameters: the neighborhood search radius and the minimum number of objects contained within the search neighborhood. In some embodiments, the real-time detection of whether a parking space is correctly occupied can also be determined based on the coordinate information of the target object, using a top-down depth map fused with a convolutional neural network. This embodiment does not impose any limitations on this method.

[0054] The technical solution of this invention uses the current output data of the original echo data corresponding to the target object in the cockpit collected by millimeter-wave radar. By installing the radar in the vehicle cabin to measure the displacement of the human chest and the signals of various parts, the location of the target point can be obtained without privacy risks. Based on the current output data, motion data detection and micro-motion data detection of the target object are performed to obtain the target detection result. The target parameter information is determined based on the target detection result. The coordinate information of the target object is determined based on the spatial model of the cockpit environment area and the target parameter information. The occupancy detection result of the target object in the cockpit is determined based on the coordinate information and the preset clustering algorithm. This can solve the problems of missed detection, false detection and false alarm when detecting target objects. It can effectively detect moving targets and stationary weak targets in the scene, and improve the accuracy and robustness of occupancy detection in the cockpit.

[0055] In one embodiment, Figure 2 This is a flowchart of another method for detecting occupancy in a car cabin according to an embodiment of the present invention. Based on the above embodiments, this embodiment acquires the original echo data of at least one target object in the cabin collected by millimeter-wave radar and determines the current output data of the original echo data; performs target detection on the target object based on the current output data to obtain the target detection result, and determines the target parameter information based on the target detection result; determines the coordinate information of the target object based on the pre-constructed cabin environment area spatial model and the target parameter information; and further refines the determination of the occupancy detection result of the target object in the cabin based on the coordinate information and the preset clustering algorithm.

[0056] like Figure 2 As shown, the occupancy detection method in the car cabin of this embodiment may specifically include the following steps:

[0057] S210. Acquire the raw data frames of the target objects inside the cockpit collected by the millimeter-wave radar. The raw data frames contain three dimensions: fast time dimension, slow time dimension, and channel dimension.

[0058] The original data frame refers to the data frame corresponding to the current frame of millimeter-wave radar signal received. A complete frame of ADC data can be represented as a three-dimensional matrix.

[0059] In this embodiment, millimeter-wave radar can be installed inside the vehicle according to the vehicle model and the requirements for detecting the target object. The number and installation location of the millimeter-wave radar can also be installed as desired according to the requirements. This embodiment does not impose any restrictions here.

[0060] In this embodiment, the raw data frames of the target object inside the cockpit collected by the millimeter-wave radar are acquired. These raw data frames contain three dimensions: fast time dimension, slow time dimension, and channel dimension. Since a complete frame of ADC data is usually in the form of a three-dimensional matrix, for example, the i-th frame of data can be represented as X. i (N s N c (,M), where N s The number of sampling points in a single chirp is called X. i The fast time dimension of a matrix, N c The number of chirps contained in a single receive channel of each frame of ADC data is called X. i The slow-time dimension of the matrix, M, represents the number of virtual receive channels, and is called X. i The channel dimension of the matrix.

[0061] S220. Process the original data frame to obtain the original echo data, multiply the original echo data along the fast time dimension with a preset window function and perform a one-dimensional Fourier transform to obtain the current output data corresponding to the original echo data.

[0062] The preset window functions can be selected according to application requirements, such as the Hamming window and the Blackman window.

[0063] In this embodiment, when acquiring the original data frames of the target object inside the cockpit collected by the millimeter-wave radar, a linear frequency modulated continuous wave (LFMCW) signal can be transmitted, and multi-channel received signals can be acquired at the receiving end. After a series of processing steps such as mixing, filtering, and sampling, the original ADC data is obtained. The original echo data is multiplied along the fast time dimension by a preset window function and then subjected to a one-dimensional Fourier transform to obtain the current output data corresponding to the original echo data. For example, for the i-th frame data X in the original echo data... i (N s N c M) is multiplied along the fast time dimension by the Hamming window function, and then subjected to N point multiplication. r (N r ≥N s The one-dimensional Fourier transform processing of the original echo data of the i-th frame yields the 1DFFT (one-dimensional Fourier transform) output data matrix, denoted as X1. i (N r N c ,M), where N r The number of sampling points in a single chirp is called X. i The fast time dimension X of the matrix i The fast time dimension of a matrix, N cThis represents the number of chirps contained in a single receive channel of each frame of ADC data, where M represents the number of virtual receive channels, referred to as X. i The channel dimension of the matrix.

[0064] In this embodiment, to better understand how the original echo data is multiplied along the fast time dimension by a preset window function and then subjected to a one-dimensional Fourier transform to obtain the corresponding one-dimensional fast time dimension distribution of the current output data, Figure 3 This is a schematic diagram of a one-dimensional fast time dimension image distribution of raw echo data provided in an embodiment of the present invention, as shown below. Figure 3 As shown, the horizontal axis represents the number of chirps, and the vertical axis represents the fast time dimension, which is also the distance dimension.

[0065] S230. Target detection is the detection of motion data of target objects. Based on the current output data, the mean value of each distance unit in the m-th channel is calculated along the slow time dimension.

[0066] Where m takes the value 1≤m≤M, and the mean is expressed by the formula: Where, N r N represents the fast time dimension. c M represents the slow time dimension, and M represents the channel dimension.

[0067] In this embodiment, target detection, which involves detecting the motion data of a target object, typically includes mean cancellation, slow-time dimension FFT, and constant false alarm rate (CFAR). In this embodiment, mean cancellation is used to filter out echoes reflected by objects that are stationary relative to the radar. It can be performed by sequentially calculating the mean of each range cell along the slow-time dimension for the m-th channel based on the current output data. Specifically, based on the 1DFFT output data X1... i (N r N c For the m-th (1≤m≤M) channel data, the mean value of each distance unit along the slow time dimension (velocity dimension) is calculated and denoted as m. i (N r The mean is expressed by the formula: (M, m), where m takes the value 1 ≤ m ≤ M. Where, N r N represents the fast time dimension. c M represents the slow time dimension, and M represents the channel dimension.

[0068] S240. Subtract the mean value of each distance cell from the current output data corresponding to each distance cell to obtain the first data matrix.

[0069] In this embodiment, for the current output data corresponding to each distance unit, the mean of each distance unit is subtracted sequentially to obtain the first data matrix after mean cancellation. This first data matrix can be denoted as X2. i (Nr N c M), where the first data matrix is ​​expressed by the formula: X2 i (N r ,k,M)=

[0070] X1 i (N r ,k,M)-m i (N r M), k = 1…N c .

[0071] In this embodiment, to facilitate a better understanding of the mean cancellation process, Figure 4 This is a schematic diagram of a one-dimensional fast time dimension image distribution for motion data through mean cancellation, as provided in an embodiment of the present invention. Figure 4 As shown, the horizontal axis represents the number of chirps, and the vertical axis represents the fast time dimension, which is also the distance dimension. It can be seen that mean cancellation can eliminate most of the stationary clutter in the scene.

[0072] S250. Window the first data matrix along the slow time dimension and perform a two-dimensional Fourier transform to obtain the first output data.

[0073] In this embodiment, the first data matrix is ​​windowed along the slow time dimension and subjected to a two-dimensional Fourier transform to obtain the first output data. Specifically, the first data matrix X2 is based on mean cancellation. i (N r N c For the m-th (1≤m≤M) channel data, windowing is applied along the slow time dimension, and then N is performed. v (N v ≥N c The FFT process at a given point yields the first output data of the 2DFFT, denoted as X21. i (N r N v ,M), where the window function can be selected according to application requirements, such as Hamming window, Blackman window, etc.

[0074] S260. The first output data is processed by taking the modulus and accumulated along the channel dimension to obtain the range Doppler spectrum data matrix. The constant false alarm rate (CFAR) detection method is used to detect targets in the range Doppler spectrum data matrix to obtain the first target point list corresponding to the motion data.

[0075] The first target point list refers to the list of target points corresponding to the motion data of the target object. The first target point list may include the first target point number, the first distance unit where the target point is located, the first velocity unit where the target point is located, and the first signal amplitude.

[0076] In this embodiment, when the constant false alarm rate (CFAR) detection method detects the target object, the detection method often needs to be selected according to the application scenario. Commonly used CFAR detection methods are not limited to CFAR detection methods based on sorting statistics or CFAR detection methods based on unit averages.

[0077] In this embodiment, the first output data is modulo-processed and accumulated along the channel dimension to obtain a range-Doppler spectrum data matrix, also known as the RD spectrum. A constant false alarm rate (CFAR) detection method is then used to perform target detection on the range-Doppler spectrum data matrix to obtain a list of first target points corresponding to the motion data. Specifically, for the first output data of the 2DFFT, denoted as X21... i (N r N v The data matrix X211 is obtained by taking the modulus of the matrix M and accumulating it along the channel dimension. i (N r N v Then, target detection is performed based on the RD spectrum, extracting important information such as the distance and velocity units where all signal peaks are located and the amplitude, and forming a target point list. The target point list includes the first target point number, the first distance unit where the target point is located, the first velocity unit where the target point is located, and the first signal amplitude.

[0078] S270. Use the first target point list as the first target detection result of the motion data corresponding to the target object.

[0079] In this embodiment, the first target point trace list, composed of the first target point number, the first distance unit where the target point is located, the first velocity unit where the target point is located, and the first signal amplitude, is used as the first target detection result of the motion data corresponding to the target object.

[0080] S280, Target detection includes the detection of micro-motion data of the target object; the difference between the current output data and the previous output data is used to obtain the second output result.

[0081] In this embodiment, target detection, which involves detecting the micro-motion data of the target object, typically includes inter-frame cancellation, intra-frame accumulation, micro-motion target detection, and filtering. In this embodiment, the difference between the current output data and the previous output data is used to obtain the second output result, which can be understood as the process of inter-frame cancellation, i.e., the process of calculating the difference between the current data frame and the previous data frame. Its purpose is to eliminate stationary clutter in the scene while preserving as much micro-motion target information as possible. This is performed after the fast time dimension FFT, typically by subtracting the corresponding 1DFFT output data of the previous frame from the 1DFFT output data of the current data frame. For example, the 1DFFT data of the i-th frame X1... i (Nr N c The previous frame, i.e., the (i-1)th data frame, is X1. i-1 (N r N c If the i-th frame and the previous frame (i-1) are considered as M, then the inter-frame cancellation output is denoted as Xzhen. i (N r N c ,M), then Xzhen i It can be obtained from the following formula: Xzhen i (N r N c M) = X1 i (N r N c ,M)-X1 i-1 (N r N c M) where N r N represents the fast time dimension. c X1 represents the slow time dimension, M represents the channel dimension; i (N r N c M) represents the current output data, X1 i-1 (N r N c M) represents the previous output data.

[0082] In this embodiment, to facilitate a better understanding of the fast temporal dimension image distribution corresponding to inter-frame cancellation, Figure 5 This is a schematic diagram of a one-dimensional fast temporal image distribution obtained by inter-frame cancellation of micro-motion data according to an embodiment of the present invention, as shown below. Figure 5 As shown, the horizontal axis represents the number of chirps, and the vertical axis represents the fast time dimension, which is also the distance dimension. It can be seen that inter-frame cancellation can preserve the information of micro-moving targets in the scene.

[0083] S290. The second output result is accumulated intra-frame from both the channel dimension and the slow time dimension to obtain the accumulated result.

[0084] In this embodiment, the second output result is accumulated intra-frame from both the channel dimension and the slow time dimension. This process can be called intra-frame accumulation. The signal after inter-frame cancellation is correlated and accumulated before CFAR detection. The purpose of this intra-frame accumulation is to highlight the signals of slightly moving targets within the observation range, facilitating accurate target detection in the next step. This is because after inter-frame cancellation, the echo signal energy of absolutely stationary targets such as seats is significantly reduced, while the echo signal of slightly moving targets is preserved. For example, let the accumulated result be denoted as Xleijia. i(N r ,1), then Xleijia i It can be obtained from the following formula: Where, N r N represents the fast time dimension c , where represents the slow time dimension, and M represents the channel dimension.

[0085] S2100. Perform a modulus operation on the accumulated result to obtain one-dimensional imaging data of the target object in the distance dimension, and use the constant false alarm rate detection method to perform micro-movement target detection on the one-dimensional imaging data to obtain a second target point list.

[0086] The second target point list refers to the list of target points corresponding to the micro-motion data of the target object. This list may include: the second target point number, the second distance cell where the target point is located, and the second signal amplitude. It should be noted that, since this is micro-motion target detection, the speed during micro-motion target detection is set to 0 by default.

[0087] In this embodiment, the obtained intra-frame accumulation result Xleijia i (N r 1) The modulus operation can obtain the one-dimensional imaging data Xcx of the target in the range dimension. i (N r ,1), for Xcx i Micro-motion target detection is performed in the fast time dimension to locate the range cell where the target is located, forming a second target point list. The second target point list includes, but is not limited to, the second target point number, the second range cell where the target point is located, and the second signal amplitude. It should be noted that the constant false alarm rate (CFAR) detection method in this embodiment can select relevant target detection methods according to requirements. These detection methods include, but are not limited to, CFAR detection methods based on ranking statistics and CFAR detection methods based on cell averaging. Based on the range cell corresponding to the detected target and the radar's detection range in the actual scene, targets that do not meet the requirements can be filtered out.

[0088] S2110. Use the second target point list as the second target detection result of the micro-motion data corresponding to the target object.

[0089] In this embodiment, the second target point list, composed of the second target point number, the second distance unit where the target point is located, and the second signal amplitude, is used as the second target detection result of the motion data corresponding to the target object.

[0090] S2120. Using an angle measurement method, the target angle is estimated for the first velocity unit and the first distance unit in the first target detection result of the motion data, and the second distance unit in the second target detection result of the micro-motion data, to obtain the target angle value corresponding to the target object.

[0091] Among them, the angle measurement method can also be called the angle estimation method. In use, the method is often selected according to the actual test scenario, accuracy requirements and computing resources. The angle estimation methods in this embodiment include digital beamforming, deterministic maximum likelihood estimation, minimum variance distortionless response and multiple signal classification.

[0092] In this embodiment, the target detection result includes a first target detection result based on micro-motion data and a second target detection result based on motion data. An angle measurement method can be used to estimate the target angle of the first velocity unit and the first distance unit in the first target detection result based on motion data, and the second distance unit in the second target detection result based on micro-motion data, to obtain the target angle value corresponding to the target object. Specifically, the first target detection result based on micro-motion data and the second target detection result based on motion data are determined. The first velocity unit and the first distance unit in the first target detection result, and the second distance unit in the second target detection result, are sequentially input into the output data matrix of the two-dimensional Fourier transform to obtain the multi-channel data of the corresponding target distance and velocity units, denoted as X8. i (M), which is an M*1 dimensional complex vector, and then the angle value of the target is obtained by using the angle measurement method.

[0093] S2130. Obtain a first distance unit from the first target detection result, and determine the first distance information of the motion data based on the first distance unit; obtain a second distance unit from the second target detection result, and determine the second distance information of the micro-motion data based on the second distance unit.

[0094] In this embodiment, a first distance unit is obtained from the first target detection result, and a first distance information of motion data is determined based on the first distance unit. A second distance unit is obtained from the second target detection result, and a second distance information of micro-motion data is determined based on the second distance unit.

[0095] S2140, Use the target angle value, first distance information and second distance information as target parameter information.

[0096] In this embodiment, the target angle value corresponding to the target object, the first distance information of the motion data determined by the first distance unit in the first target detection result, and the second distance information of the micro-motion data determined by the second distance unit in the second target detection result are used as target parameter information.

[0097] In this embodiment, to facilitate a better understanding of the results obtained using the angle measurement method, Figure 6 This is a schematic diagram illustrating the detection result of a constant false alarm rate (CFAR) detection method according to an embodiment of the present invention. Figure 6 As shown, constant false alarm rate (CFAR) detection is performed in the fast time dimension. Figure 6The horizontal axis represents distance, and the vertical axis represents the signal strength of the corresponding distance cell. To better understand the spatial spectrum calculation of the target's motion data, a list of target points is obtained. Figure 7 This is a schematic diagram of the spatial spectrum of a moving target provided in an embodiment of the present invention. Figure 8 This is a schematic diagram of a target angle estimation result provided in an embodiment of the present invention.

[0098] S2150. Determine the coordinate transformation method corresponding to the target parameter information based on the installation position of the millimeter-wave radar in the vehicle cabin.

[0099] In this embodiment, the construction of the cockpit environment area spatial model includes: taking the location of the radar as the origin of the rectangular coordinate system, and constructing the spatial model of the cockpit based on the installation position and line of sight of the millimeter-wave radar in the vehicle cabin. This can be understood as follows: first, a coordinate system needs to be established based on the actual scene and the installation position of the radar. Assuming that the location of the radar is the origin of the rectangular coordinate system and the line of sight of the radar is the y-axis of the rectangular coordinate system, each area of ​​the cabin space is constructed in the coordinate system according to its actual size. Then, the distance and angle of the moving target and the micro-moving target are obtained, and the coordinate information of the target object in the radar coordinate system is calculated.

[0100] To facilitate a better understanding of the construction of the cabin environment area spatial model, Figure 9 This is a schematic diagram of a car rear seat area modeling according to an embodiment of the present invention, defining the range of the area inside the car cabin and different location areas, such as... Figure 9 As shown, the coordinate point (0,0) is the location of the radar. In the figure, area 1 represents the foot space area of ​​the left rear seat, area 2 represents the foot space area of ​​the middle seat, area 3 represents the foot space area of ​​the right rear seat, and areas 4, 5 and 6 represent the left, middle and right rear seat areas respectively.

[0101] S2160. Determine the transformed coordinates corresponding to the target parameter information according to the coordinate transformation method, and use the target coordinate information after coordinate transformation as the coordinate information of the target object in the radar coordinate system.

[0102] In this embodiment, since the construction of the spatial model of the measured area is related to the radar's installation location and viewing angle, it is known that the coordinate transformation method corresponding to the target parameter information will be different depending on the radar's installation location. For example, taking the origin of the radar coordinate system and the rectangular coordinate system as the origin, the straight line length from the origin to the cockpit domain is the distance, and the angle between this origin and the y-axis is the angle just measured (i.e., the angle between the line and the vertical axis). Based on the distance and the angle, the target parameter information can be transformed into coordinates.

[0103] S2170. Select a coordinate point from the coordinate information in a preset processing order, wherein the coordinate information is distributed in the distribution area corresponding to the cockpit environment area spatial model.

[0104] In this embodiment, a series of discrete target points are distributed in the coordinate system. We usually know that areas with dense distribution of target points are likely generated by the same living target. In order to determine the number of occupied seats in the detection scene, that is, to determine which seats are occupied, a clustering method can be used. First, a coordinate point is selected from the coordinate information in a preset processing order. The coordinate information is distributed in the distribution area corresponding to the cabin environment area spatial model.

[0105] S2180. Using the coordinate point as the center, determine whether there is a target coordinate point with a preset number threshold within the preset range of the coordinate point. If it exists, execute S2190; if it does not exist, execute S2200.

[0106] In this embodiment, taking the coordinate point as the center, it is determined whether there are target coordinate points within the preset range of the coordinate point, which is a preset number threshold. If there are target coordinate points within the preset range, which are a preset number threshold, then the target coordinate points that reach the preset number threshold are classified as first category coordinate points. The mean coordinates of each first category coordinate point are determined, and the occupancy detection result of the target object is determined based on the distribution area of ​​the mean coordinates in the cabin environment area spatial model and the number of mean coordinates. This can be understood as determining the current number of people and the position of the human target based on the clustering results. If there are no targets, then the target coordinate points that do not reach the preset number threshold are classified as boundary points.

[0107] S2190. The target coordinate points that reach the preset quantity threshold are classified as first category coordinate points. The mean coordinates of each first category coordinate point are determined. The occupancy detection result of the target object is determined based on the distribution area of ​​the mean coordinates in the cabin environment area spatial model and the quantity of the mean coordinates.

[0108] In this embodiment, if there are target coordinate points with a preset number threshold within the preset range of the coordinate points, the target coordinate points that reach the preset number threshold are classified as first category coordinate points. The mean coordinates of each first category coordinate point are determined, and the occupancy detection result of the target object is determined based on the distribution area of ​​the mean coordinates in the cabin environment area spatial model and the number of mean coordinates. This can be understood as determining the current number of people and the position of the human target based on the clustering results.

[0109] S2200: Classify target coordinate points that have not reached the preset quantity threshold as boundary points.

[0110] In this embodiment, if there are no target coordinate points within a preset number threshold within the preset range of the coordinate points, then the target coordinate points that have not reached the preset number threshold are classified as boundary points.

[0111] In one embodiment, determining the occupancy detection result of a target object based on the distribution area of ​​the mean coordinates in the spatial model of the cabin environment and the number of mean coordinates includes:

[0112] The location of the target object is determined based on the distribution of the mean coordinates in the spatial model of the cabin environment.

[0113] The number of target objects within the cabin environment area is determined based on the number of mean coordinates;

[0114] The location area of ​​the target object and the number of target objects are used as the occupancy detection results of the target objects.

[0115] In this embodiment, specifically, the location area of ​​the target object is determined based on the distribution of mean coordinates in the cabin environment spatial model; the number of target objects in the cabin environment spatial model is determined based on the number of mean coordinates; and the location area of ​​the target object and the number of target objects are used as the occupancy detection result of the target object. This can be understood as follows: the obtained clustering results are used to calculate the average coordinates of core points of the same class, and based on the distribution of these coordinates in the vehicle interior spatial model, it is determined which area contains occupancy. Simultaneously, the number of people in the cabin is determined based on the category of the clustered core points.

[0116] In this embodiment, to facilitate a better understanding of the coordinate information clustering method, Figure 10 This is a schematic diagram of a coordinate information clustering result provided in an embodiment of the present invention, as shown below. Figure 10 As shown, by clustering the located target points using a density-based fast clustering method, the final occupancy information can be obtained.

[0117] The above technical solution of this invention involves: calculating the mean of each distance unit along the slow time dimension for the m-th channel based on the current output data; subtracting the mean of each distance unit from the current output data corresponding to each distance unit to obtain a first data matrix; windowing the first data matrix along the slow time dimension and performing a two-dimensional Fourier transform to obtain first output data; performing modulo processing on the first output data and accumulating it along the channel dimension to obtain a range-Doppler spectrum data matrix; using a constant false alarm rate (CFAR) detection method to perform target detection on the range-Doppler spectrum data matrix to obtain a first target point list corresponding to the motion data; using the first target point list as the first target detection result corresponding to the motion data of the target object; calculating the difference between the current output data and the previous output data to obtain a second output result; and then processing the second output result from the channel... The system performs intra-frame accumulation in two dimensions: the first dimension and the slow time dimension. The accumulated result is then moduloed to obtain one-dimensional imaging data of the target object in the distance dimension. A constant false alarm rate (CFAR) detection method is used to perform micro-motion target detection on the one-dimensional imaging data to obtain a second target point list. This second target point list is used as the second target detection result corresponding to the micro-motion data of the target object. Based on the target detection results, target parameter information is determined. Based on the coordinate information and a preset clustering algorithm, the occupancy detection result of the target object within the cockpit is determined. This method can further solve the problems of missed detections, false detections, and false alarms in target object detection. It can effectively detect stationary weak targets in the scene, especially for living beings in a sleeping state who only exhibit breathing and heartbeat movements. It can effectively detect the occupancy of targets and has very high robustness and accuracy.

[0118] In one embodiment, Figure 11 This is a structural block diagram of a vehicle cabin occupancy detection device according to an embodiment of the present invention. The device is suitable for detecting occupancy within a vehicle cabin and can be implemented in hardware or software. It can be configured in an electronic device to implement a vehicle cabin occupancy detection processing method according to an embodiment of the present invention.

[0119] like Figure 11 As shown, the device includes: an output data determination module 1110, a parameter information determination module 1120, a coordinate determination module 1130, and a result determination module 1140.

[0120] The output data determination module 1110 is used to acquire the original echo data of at least one target object in the cockpit collected by the millimeter-wave radar, and to determine the current output data of the original echo data.

[0121] The parameter information determination module 1120 is used to perform target detection on the target object based on the current output data to obtain a target detection result, and determine target parameter information based on the target detection result; wherein, the target detection includes the detection of motion data and micro-motion data of the target object;

[0122] The coordinate determination module 1130 is used to determine the coordinate information of the target object based on the pre-built cockpit environment area spatial model and the target parameter information;

[0123] The result determination module 1140 is used to determine the occupancy detection result of the target object in the cockpit based on the coordinate information and a preset clustering algorithm.

[0124] In this embodiment of the invention, the current output data of the original echo data corresponding to the target object in the cockpit collected by millimeter-wave radar can be used to measure the displacement of the human chest and the signal of various parts of the body by installing the radar in the vehicle cabin, thereby obtaining the target point position without privacy risks. Based on the current output data, motion data detection and micro-motion data detection of the target object are performed to obtain the target detection result, and the target parameter information is determined according to the target detection result. The coordinate information of the target object is determined according to the spatial model of the cockpit environment area and the target parameter information. The occupancy detection result of the target object in the cockpit is determined according to the coordinate information and the preset clustering algorithm. This can solve the problems of missed detection, false detection and false alarm when detecting target objects, and can effectively detect moving targets and stationary weak targets in the scene, improving the accuracy and robustness of occupancy detection in the cockpit.

[0125] In one embodiment, the output data determination module 1110 includes:

[0126] The acquisition unit is used to acquire the original data frames of the target objects inside the cockpit collected by the millimeter-wave radar. The original data frames contain three dimensions: fast time dimension, slow time dimension, and channel dimension.

[0127] The output data determination unit is used to process the original data frame to obtain the original echo data, and multiply the original echo data along the fast time dimension by a preset window function and perform a one-dimensional Fourier transform to obtain the current output data corresponding to the original echo data.

[0128] In one embodiment, the target detection includes detecting motion data of the target object; the parameter information determination module 1120 includes:

[0129] The mean determination unit is used to calculate the mean of each distance unit along the slow time dimension for the m-th channel based on the current output data, where m takes the value 1≤m≤M, and the mean is expressed by the formula: Where, Nr N represents the fast time dimension. c M represents the slow time dimension, and M represents the channel dimension;

[0130] The first data matrix determination unit is used to subtract the mean value of each distance unit from the current output data corresponding to each distance unit to obtain the first data matrix; wherein, the first data matrix is ​​expressed by the formula: X2 i (N r ,k,M)=X1 i (N r ,k,M)-m i (N r M), k = 1…N c ;

[0131] The first output data determination unit is used to window the first data matrix along the slow time dimension and perform a two-dimensional Fourier transform to obtain the first output data.

[0132] The first target trace list determination unit is used to perform modulo processing on the first output data and accumulation processing along the channel dimension to obtain a range Doppler spectrum data matrix, and to perform target detection on the range Doppler spectrum data matrix using a constant false alarm rate (CFAR) detection method to obtain a first target trace list corresponding to the motion data; wherein, the first target trace list includes a first target point number, a first distance unit where the target point is located, a first velocity unit where the target point is located, and a first signal amplitude;

[0133] The first result determination unit is used to take the first target point list as the first target detection result corresponding to the motion data of the target object.

[0134] In one embodiment, the target detection includes detecting micro-motion data of the target object; the parameter information determination module 1120 further includes:

[0135] The second output result determination unit is used to calculate the difference between the current output data and the previous output data to obtain a second output result; wherein, the second output result is expressed by the formula: X3 i (N r N c M) = X1 i (N r N c ,M)-X1 i-1 (N r N c ,M), wherein the N r N represents the fast time dimension. c X1 represents the slow time dimension, M represents the channel dimension; i (N r Nc M) represents the current output data, X1 i-1 (N r N c M) represents the previous output data;

[0136] The accumulation result determination unit is used to perform intra-frame accumulation of the second output result from two dimensions: channel dimension and slow time dimension, to obtain the accumulation result, wherein the accumulation result is expressed by the formula: Where, N r N represents the fast time dimension c , represents the slow time dimension, and M represents the channel dimension;

[0137] The second target trace list determination unit is used to perform a modulo operation on the accumulated result to obtain one-dimensional imaging data of the target object in the distance dimension, and to perform micro-movement target detection on the one-dimensional imaging data using a constant false alarm rate detection method to obtain a second target trace list. The second target trace list includes: a second target point number, a second distance unit where the target point is located, and a second signal amplitude.

[0138] The second result determination unit is used to take the second target point list as the second target detection result corresponding to the micro-motion data of the target object.

[0139] In one embodiment, the target detection result includes a first target detection result of the micro-motion data and a second target detection result of the motion data; the parameter information determination module 1120 further includes:

[0140] An angle value determination unit is used to estimate the target angle of the first velocity unit and the first distance unit in the first target detection result of the motion data and the second distance unit in the second target detection result of the micro-motion data by using an angle measurement method, so as to obtain the target angle value corresponding to the target object;

[0141] The first distance determination unit is used to obtain a first distance unit from the first target detection result and determine the first distance information of the motion data based on the first distance unit;

[0142] The second distance determination unit is used to obtain the second distance unit from the second target detection result and determine the second distance information of the micro-motion data based on the second distance unit;

[0143] The parameter determination unit is used to use the target angle value, the first distance information, and the second distance information as the target parameter information.

[0144] In one embodiment, the construction of the cockpit environment area spatial model includes: constructing a spatial model of the cockpit based on the location of the radar as the origin of the rectangular coordinate system, according to the installation position of the millimeter-wave radar inside the vehicle cabin and the line of sight; the coordinate determination module 1130 includes:

[0145] The conversion mode determination unit is used to determine the coordinate conversion mode corresponding to the target parameter information based on the installation position of the millimeter-wave radar in the vehicle cabin.

[0146] The coordinate determination unit is used to determine the transformed coordinates corresponding to the target parameter information according to the coordinate transformation method, and to use the target coordinate information after coordinate transformation as the coordinate information of the target object in the radar coordinate system.

[0147] In one embodiment, the result determination unit 1140 further includes:

[0148] The coordinate point selection unit is used to select a coordinate point from the coordinate information in a preset processing order, wherein the coordinate information is distributed in the distribution area corresponding to the cockpit environment area spatial model;

[0149] The result determination unit is used to determine whether there are target coordinate points within a preset range of the coordinate point, with the coordinate point as the center. If there are, the target coordinate points that reach the preset number threshold are classified as first category coordinate points, the mean coordinates of each first category coordinate point are determined, and the occupancy detection result of the target object is determined based on the distribution area of ​​the mean coordinates in the cabin environment area spatial model and the number of the mean coordinates. If there are no, the target coordinate points that do not reach the preset number threshold are classified as boundary points.

[0150] In one embodiment, the result determination unit includes:

[0151] The first determining subunit is used to determine the location area of ​​the target object based on the distribution of the mean coordinates in the spatial model of the cabin environment area;

[0152] The first determining subunit is used to determine the number of target objects in the cabin environment area space based on the number of mean coordinates;

[0153] The result determination subunit is used to determine the location area of ​​the target object and the number of the target objects as the occupancy detection result of the target objects.

[0154] The vehicle cabin occupancy detection device provided in this embodiment of the invention can execute the vehicle cabin occupancy detection processing method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0155] In one embodiment, Figure 12 This is a schematic diagram of an electronic device provided for an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0156] like Figure 12 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0157] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0158] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as occupancy detection methods in a car cabin.

[0159] In some embodiments, the occupancy detection processing method within a vehicle cabin can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the occupancy detection method within a vehicle cabin described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the occupancy detection method within a vehicle cabin by any other suitable means (e.g., by means of firmware).

[0160] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] Computer programs used to implement the methods of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable occupancy detection device in a vehicle cockpit, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on 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.

[0163] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0164] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0165] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0166] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and no limitation is imposed herein.

[0167] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for detecting occupancy in a car cabin, characterized in that, include: Acquire raw echo data of at least one target object inside the cockpit collected by millimeter-wave radar, and determine the current output data of the raw echo data; Based on the current output data, target detection is performed on the target object to obtain a target detection result, and target parameter information is determined based on the target detection result; wherein, the target detection includes the detection of motion data and micro-motion data of the target object; The coordinate information of the target object is determined based on the pre-constructed cabin environment area spatial model and the target parameter information; The occupancy detection result of the target object in the cockpit is determined based on the coordinate information and the preset clustering algorithm; The target detection result includes a first target detection result of the micro-motion data and a second target detection result of the motion data; the step of determining target parameter information based on the target detection result includes: An angle measurement method is used to estimate the target angle value corresponding to the target object by estimating the first velocity unit and the first distance unit in the first target detection result of the motion data and the second distance unit in the second target detection result of the micro-motion data. Obtain a first distance unit from the first target detection result, and determine the first distance information of the motion data based on the first distance unit; Obtain a second distance unit from the second target detection result, and determine the second distance information of the micro-motion data based on the second distance unit; The target angle value, the first distance information, and the second distance information are used as the target parameter information.

2. The method according to claim 1, characterized in that, The acquisition of raw echo data of at least one target object inside the cockpit collected by millimeter-wave radar, and the determination of the current output data of the raw echo data, includes: Acquire raw data frames of target objects inside the cockpit collected by millimeter-wave radar, wherein the raw data frames contain three dimensions: fast time dimension, slow time dimension, and channel dimension; The original data frame is processed to obtain the original echo data. The original echo data is then multiplied along the fast time dimension by a preset window function and subjected to a one-dimensional Fourier transform to obtain the current output data corresponding to the original echo data.

3. The method according to claim 1, characterized in that, The target detection includes detecting the motion data of the target object; the step of performing target detection on the target object based on the current output data to obtain the target detection result includes: Based on the current output data, the mean value of each distance unit along the slow time dimension of the m-th channel is calculated sequentially, where m takes the value 1 ≤ m ≤ M. The mean value is expressed by the formula: ,in, Indicates fast time dimension, Indicates the slow time dimension. Indicates the channel dimension; This represents the current output data; The first data matrix is ​​obtained by subtracting the mean value of each distance unit from the current output data corresponding to each distance unit; wherein, the first data matrix is ​​expressed by the formula: , ; The first data matrix is ​​windowed along the slow time dimension and then subjected to a two-dimensional Fourier transform to obtain the first output data. The first output data is subjected to modulo processing and accumulated along the channel dimension to obtain a range Doppler spectrum data matrix. The constant false alarm rate (CFAR) detection method is used to detect targets in the range Doppler spectrum data matrix to obtain a first target point list corresponding to the motion data. The first target point list includes a first target point number, a first distance unit where the target point is located, a first velocity unit where the target point is located, and a first signal amplitude. The first list of target points is used as the first target detection result corresponding to the motion data of the target object.

4. The method according to claim 1, characterized in that, The target detection includes detecting the micro-motion data of the target object; the step of obtaining the target detection result by performing target detection on the target object based on the current output data includes: The second output result is obtained by calculating the difference between the current output data and the previous output data; wherein, the second output result is expressed by the formula: , wherein Indicates fast time dimension, Indicates the slow time dimension. Indicates the channel dimension; This represents the current output data. This represents the previous output data; The second output result is accumulated intra-frame from both the channel dimension and the slow time dimension to obtain the accumulated result, which is expressed by the formula: ; The accumulated result is subjected to a modulo operation to obtain one-dimensional imaging data of the target object in the distance dimension, and a constant false alarm rate detection method is used to perform micro-motion target detection on the one-dimensional imaging data to obtain a second target point list, wherein the second target point list includes: the second target point number, the second distance cell where the target point is located, and the second signal amplitude; The second target point list is used as the second target detection result corresponding to the micro-motion data of the target object.

5. The method according to claim 1, characterized in that, The construction of the cockpit environment area spatial model includes: constructing a spatial model of the cockpit based on the location of the radar as the origin of the rectangular coordinate system, according to the installation position of the millimeter-wave radar inside the vehicle cabin and the line of sight; determining the coordinate information of the target object based on the pre-constructed cockpit environment area spatial model and the target parameter information includes: The coordinate transformation method corresponding to the target parameter information is determined based on the installation position of the millimeter-wave radar inside the vehicle cabin; The transformed coordinates corresponding to the target parameter information are determined according to the coordinate transformation method, and the transformed target coordinate information is used as the coordinate information of the target object in the radar coordinate system.

6. The method according to claim 1, characterized in that, The step of determining the occupancy detection result of the target object inside the cockpit based on the coordinate information and a preset clustering algorithm includes: Select a coordinate point from the coordinate information in a preset processing order, wherein the coordinate information is distributed in the distribution area corresponding to the cabin environment area spatial model; Using the coordinate point as the center, determine whether there are target coordinate points within a preset range of the coordinate point that meet a preset number threshold. If they do, classify the target coordinate points that meet the preset number threshold as first-class coordinate points, determine the mean coordinates of each first-class coordinate point, and determine the occupancy detection result of the target object based on the distribution area of ​​the mean coordinates in the cabin environment area spatial model and the number of the mean coordinates. If they do not meet the preset number threshold, classify the target coordinate points that do not meet the preset number threshold as boundary points.

7. The method according to claim 6, characterized in that, The determination of the occupancy detection result of the target object based on the distribution area of ​​the mean coordinates in the spatial model of the cabin environment and the number of the mean coordinates includes: The location of the target object is determined based on the distribution of the mean coordinates in the spatial model of the cabin environment area; The number of target objects within the cabin environment area is determined based on the number of mean coordinates. The location area where the target object is located and the number of the target objects are used as the occupancy detection results of the target objects.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the occupancy detection method in the vehicle cabin as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the occupancy detection method for a vehicle cabin as described in any one of claims 1-7.

Images