SYSTEM AND METHOD FOR DETECTING THE PRESENCE OF A BODY IN A VEHICLE - Patent application

Abstract

A system and method for detecting a living body in a vehicle and generating an alert only if a child is detected. The radar detection system uses a vehicle vibration module, a time behavior analysis module, and a spatial feature module to analyze image data over time and space and detect false positives by distinguishing between a real child and other similar voxel clusters in the radar imagery.

Description

[Technical field]

[0001] This application claims the benefit of priority from No. 63 / 277,673, filed November 10, 2021, the contents of which are incorporated by reference in their entirety.

[0002] The present disclosure herein relates to a system and method for detecting the presence of a living body in a vehicle, and more particularly, to alerting a third party to the presence of an infant or other dependent that has been accidentally left in the passenger compartment. [Background technology]

[0003] Child Presence Detection is an assessment protocol aimed at reducing the risk to victims in situations where an infant or child is locked in a vehicle. Such protocols can also trigger an alert if a pet or other dependent is left behind in a closed passenger compartment.

[0004] The need to issue warnings and alerts is necessary in all cases where life is at risk from abandonment, but the accuracy of such warnings is essential to prevent excessive false alarms. Excessive false alarm rates pose a real risk to life, as they lead to warning fatigue. People who receive excessive warnings are more likely to ignore or possibly disable future warnings.

[0005] Furthermore, if there is a high degree of certainty in the accuracy of the detection, active interventions may be taken to mitigate the risk, e.g., vehicle windows may be opened, the air conditioning may be turned on, etc.

[0006] Therefore, there remains a need for a child presence detection system that has a high true positive detection rate and a low false positive detection rate.The invention described herein addresses the above-mentioned need. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] PCT / IB2022 / 055109 (International Publication No. WO2022 / 254347) and How to Write a Secret Prior Art Application Summary of the Invention

[0008] According to one aspect of the present invention, a system for detecting the presence of a body within a vehicle passenger compartment is introduced, the system including a radar unit comprising at least one transmitter antenna connected to an oscillator and configured to transmit electromagnetic waves into the vehicle passenger compartment, and at least one receiver antenna configured to receive electromagnetic waves reflected by an object within the vehicle passenger compartment and operable to generate raw data.

[0009] The system further includes a processor unit configured to receive data from the radar unit and operable to generate an alert instruction based on the data, and an alert generator configured to generate a child present detection (CPD) alert.

[0010] The processor typically includes modules such as a vehicle vibration detection module operable to detect vibrations in the vehicle cabin, a temporal behavior analysis module operable to analyze temporal characteristics of movement and identify oscillatory characteristics of breathing, and a spatial characteristic analysis module operable to analyze spatial characteristics of clusters of voxels detected within the vehicle cabin.

[0011] The system may further include a communication module, optionally in communication with the computer network, configured and operable to communicate the child presence detection alert to a third party.

[0012] Where appropriate, the system may further include a pre-processor configured to receive the raw data from the radar unit and operable to produce a filtered point cloud for model optimization. Additionally or alternatively, the system may include a frame buffer memory unit configured and operable to store the frame data.

[0013] Another aspect of the present invention is to teach a method for detecting the presence of a body in a vehicle cabin. The method may include the steps of providing a radar module, providing a vehicle vibration detection module, providing a time behavior analysis module, providing a spatial characteristic analysis module, and providing a presence detection unit. Thus, the method may further include at least one transmitter antenna transmitting electromagnetic waves into the vehicle cabin, at least one receiver antenna receiving the electromagnetic waves reflected by an object in the vehicle cabin, transferring data from the radar to a processor, the vehicle vibration detection module generating a vehicle vibration indicator, the time behavior analysis module generating a time movement indicator, the spatial characteristic analysis module generating a spatial characteristic indicator, transferring the feature vector to the presence detection unit, the presence detection unit processing the feature vector, and issuing an alert if a body is detected.

[0014] If appropriate, the step of the vehicle vibration detection module generating the vehicle vibration indicator may include obtaining a series of three-dimensional frames of image data, removing static objects from the image data, generating a two-dimensional moving target indication matrix, and summing the lowest intensity values ​​of pixels in the two-dimensional moving target indication matrix.

[0015] Optionally, removing static objects from the image data may include selecting a frame capture rate, collecting raw data from a first frame, waiting a time delay, collecting raw data from a second frame, and subtracting the first frame data from the second frame data.

[0016] Again optionally, the step of generating a two-dimensional moving target indication matrix comprises selecting a maximum intensity voxel

number

number

number

[0017] In various embodiments, the step of the temporal behavior analysis module generating the temporal motion index includes identifying clusters of hyperintense voxels in the three-dimensional image data, and for each cluster, the processor unit matching a series of complex values ​​for each voxel, and for each voxel, determining a center point of a complex plane, determining a phase value for each voxel in each frame, generating a smooth waveform representing phase changes over time for each voxel in each frame, selecting a subset of voxels indicative of a breathing pattern, and calculating the temporal motion index.

[0018] Optionally, the step of calculating a time shift index includes calculating a spectral peak index, a respiration per minute (RPM) index, or a circle fit index.

[0019] Additionally or alternatively, the step of the spatial characteristic analysis module generating the spatial feature index includes acquiring a series of three-dimensional frames of image data of an active area including the vehicle passenger compartment and surroundings, identifying voxel clusters within the active area, counting clusters within a target area to thereby obtain a cluster count index, counting the number of voxels in each cluster, selecting a maximum number of voxels to thereby obtain a maximum cluster size index, acquiring a cluster depth index, acquiring a target maximum voxel index, acquiring an active area maximum voxel index, and acquiring a maximum voxel range index.

[0020] Variously, obtaining the cluster depth index may include calculating the difference between the maximum and minimum extent of the voxels in each cluster and selecting the value closest to the infant reference value. Obtaining the target maximum voxel index may include selecting the highest moving target indication value within the active region. Obtaining the active region maximum voxel index may include selecting the highest MTI value within the active region. Further, obtaining the maximum voxel extent index may include selecting the extent of the active region maximum voxel.

[0021] In other embodiments of the present invention, the method may further include distinguishing between a child and a pet. Illustratively, the method may include identifying a child-sized target, transmitting a pet stimulus signal insignificant to a human, and associating the child-sized target with a pet if increased activity is detected in the child-sized target. Thus, where appropriate, transmitting the pet stimulus signal insignificant to a human includes transmitting the pet stimulus signal at a frequency inaudible to the human ear.

[0022] For a better understanding of the embodiments and to show how they may be carried into effect, reference will now be made, purely by way of example, to the accompanying drawings in which:

[0023] Now, with reference to the drawings in detail, it is emphasized that the matters shown are presented only for the purpose of illustrative discussion of the selected embodiments, and to provide what is believed to be the most effective and easily understood explanation of the principles and conceptual aspects. In this regard, structural details are not shown in more detail, and the matters necessary for basic understanding are given priority, and the explanation taken together with the drawings makes it clear to those skilled in the art how to implement the selected various embodiments. [Brief description of the drawings]

[0024] [Figure 1A] FIG. 1 is a block diagram that generally illustrates selected components of a system for detecting the presence of a body within a vehicle. [Figure 1B] 1 is a flow chart that generally illustrates data flow between components of a system for detecting the presence of a body and components of a system for warning about a body in a vehicle. [Figure 1C] 1 is a flow chart that generally illustrates selected actions in a method for detecting the presence of a body within a vehicle. [Figure 2A] 4 is a flow chart that generally illustrates a possible method for generating a vehicle vibration indicator. [Figure 2B] 1 is a flow chart that generally illustrates possible steps for removing static objects from image data. [Figure 2C] 1 is a flow chart that generally illustrates possible steps for generating an MTI matrix. [Figure 3A] FIG. 2 is a diagram of a segment of a three-dimensional image showing selected voxels. [Figure 3B] FIG. 13 is a diagram of a set of voxels sharing the same angular coordinate but with different r coordinates. [Figure 3C] FIG. 1 is an example of a 2D MTI image matrix. [Figure 3D] 1A-1C are diagrams of example aligned MTI intensity profiles showing typical MTI profile characteristics of a moving object in a stationary environment. [Figure 3E] FIG. 1 is an example of aligned MTI intensity profiles showing typical MTI profile characteristics of a vibration environment. [Figure 4] 1 is a flow chart that generally illustrates a possible method for generating a time movement indicator; [Figure 5A] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 5B] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 5C] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 5D] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 5E] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 5F] 1 is an example plot of a series of complex values ​​representing reflected radiation at a single voxel within a target region across multiple frames. [Figure 6] 1 is a flow chart that generally illustrates a possible method for generating a spatial feature index. [Figure 7A] 1 is a flow chart that generally illustrates a possible method for pet mitigation. [Figure 7B] 1 is a flow chart that generally illustrates how the activity level (AL) or bounding volume (BV) covered by a cluster of pixels can be used to indicate the presence of a pet, such as a dog. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] Aspects of the present disclosure relate to systems and methods for detecting the presence of a living organism in a vehicle and generating an alert, and in particular, the present invention relates to a child presence detection system that has a high true positive detection rate and a low false positive detection rate.

[0026] It has been found that false alarms in detection systems are often caused by objects placed in the passenger cabin or cabin conditions that meet the triggering conditions of a sensor that causes an alarm when no actual infant or child is present in the cabin.

[0027] Several false alarm triggering conditions have been identified, for example, a water bottle in a car cabin shaken by wind, by hand, or by any other means may generate an oscillating signal superficially similar to a breathing child. Therefore, the vehicle vibration detection module may apply various methods as disclosed herein to enable a shaken vehicle to be detected. In this way, shaken vehicle type false alarm triggers may be avoided.

[0028] Another false alarm trigger may be an oscillating object, such as a pendulum, spring, clock, or the like, that is typically characterized by a highly periodic motion. A temporal behavior analysis module may therefore be provided to analyze the temporal characteristics of the motion and identify those oscillations that are characteristic of real breathing.

[0029] Further characteristics may be used to distinguish between true and false alarms, such as spatial characteristics relating to size and shape of a suspected child, for example. Thus, a spatial characteristics analysis module may be provided to analyze characteristics of clusters of voxels detected within the vehicle cabin to identify clusters that may indicate a real child or the like.

[0030] It is further noted that even when a real living being is detected inside the vehicle, this living being may be a pet such as a dog, cat, or the like, and therefore a pet mitigation module may be provided to distinguish pets from humans when necessary.

[0031] Where necessary, detailed embodiments of the present invention are disclosed herein, but it should be understood that the disclosed embodiments are merely illustrative of the present invention, which may be embodied in various and alternative forms. The figures are not necessarily to scale, and some features may be exaggerated or minimized to show details of specific components. Therefore, specific structural and functional details disclosed herein should not be construed as limiting, but merely as a representative basis for teaching those skilled in the art how to employ the present invention in various ways.

[0032] Where appropriate, in various embodiments of the present disclosure, one or more tasks as described herein may be performed by a data processor, such as a computing platform or distributed computing system for executing a plurality of instructions. Optionally, the data processor includes or has access to volatile memory for storing instructions, data, or the like. Additionally or alternatively, the data processor may have access to non-volatile storage, illustratively a magnetic hard disk, flash drive, removable media, or the like, for storing instructions and / or data.

[0033] It is expressly pointed out that the systems and methods disclosed herein may not be limited in their application to the details and arrangements of construction of the components or methods set forth in the description or illustrated in the drawings and examples. The systems and methods disclosed herein may be implemented in other forms or by various means and techniques.

[0034] Alternative methods and materials similar or equivalent to those described herein may be used in the practice or testing of embodiments of the present disclosure. Nevertheless, certain methods and materials are described herein for illustrative purposes only. The materials, methods, and examples are not necessarily intended to be limiting. Thus, various embodiments may omit, substitute, or add various procedures or components, as appropriate. Illustratively, methods may be performed in different orders than described, and various steps may be added, omitted, or combined. Additionally, aspects and components described with respect to certain embodiments may be combined in various other embodiments.

[0035] Reference is now made to the block diagram of Figure 1A, which illustrates generally selected components of a system 100 for detecting the presence of a body within a vehicle. System 100 includes a radar unit 120 and a processor 140.

[0036] The radar 120 typically includes at least one array of radio frequency transmitter antennas 122 and at least one array of radio frequency receiver antennas 124. The radio frequency transmitter antennas are connected to an oscillator 126 (a radio frequency signal source) and are configured and operable to transmit electromagnetic waves toward a target area 200. The radio frequency receiver antennas 124 are configured to receive the electromagnetic waves reflected back from objects 210 within the target area 200.

[0037] Thus, the transmitter may be configured to produce a beam of electromagnetic radiation, such as microwave radiation or the like, directed toward a monitored area 200, such as a vehicle cabin or the like. The receiver may include at least one receiving antenna, or an array of receiver antennas, configured and operable to receive electromagnetic waves reflected by objects within the monitored area.

[0038] The raw data generated by the receiver is typically a set of complex values ​​representing magnitude and phase measurements corresponding to waves scattered back from objects in front of the array. A spatial reconstruction process is applied to the measurements to reconstruct the amplitude (scattered intensity) in the three-dimensional coordinates of the object within the target region. Thus, each three-dimensional cross-section of a volume within the target region may be represented by a voxel defined by four values ​​corresponding to an x-coordinate, a y-coordinate, a z-coordinate, and an amplitude value.

[0039] Typically, the receiver is connected to a pre-processing unit 130 configured and operable to process the amplitude matrix of the raw data generated by the receiver and capable of producing a filtered point cloud suitable for model optimization.

[0040] Thus, where appropriate, the pre-processing unit may include an amplitude filter operable to select voxels having amplitudes above a required threshold, and a voxel selector operable to reduce the number of voxels in the filtered data, illustratively by sampling the data or by clustering neighboring voxels. In this manner, the filtered point cloud may be output to a processor. It is further noted that the filtered point cloud may be further simplified by setting the amplitude value of each voxel to one when the amplitude is above a threshold and to zero when the amplitude is below the threshold.

[0041] The processor 140 in communication with the pre-processor unit may include modules such as a vehicle vibration detection module 142, a temporal behavior analysis module 144, a spatial characteristic analysis module 146, optionally a pet mitigation module 147, and an alert generator 148 configured to receive a feature vector including a combination of the feature indicators generated by the analysis modules and may be operable to generate a child presence detection (CPD) alert based on the received data.

[0042] The communication module 160 is configured and operable to communicate the child presence detection alert to a third party. Optionally, the communication module 160 may be in communication with a computer network 162, such as the Internet, via which the communication module 160 may communicate the alert to a third party, illustratively via a phone, computer, wearable device, or the like.

[0043] In yet other embodiments, the CPD alert may initiate active intervention that may be taken to mitigate the risk, by way of example, the vehicle windows may be opened, the air conditioning may be turned on, or the like.

[0044] Referring now to the flow chart of FIG. 1B, FIG. 1B indicates how data may flow between components of the system to generate a CPD alert. The radar module 120 may generate raw data that is passed to a processor 140 that generates a feature vector. The feature vector is used by a presence detection unit 150. The presence detection unit 150 is operable to determine whether a child is indeed present and, if appropriate, communicate an alert instruction to an alert generator 156. The presence detection unit 150 may include a dimensionality reduction unit 152 operable to transform the multi-dimensional feature vector for principal component analysis, and a classifier 154, such as a support vector machine operable to classify the feature vector into presence detected or non-presence detected.

[0045] Reference is now made to the flow chart of Figure 1C, which diagrammatically illustrates selected actions in a method for detecting the presence of a body in a vehicle, the method including providing a radar module 1001, providing a vehicle vibration detection module 1002, providing a time behavior analysis module 1003, providing a spatial characteristic analysis module 1004, and providing a presence detection unit (PDU) 1005.

[0046] Thus, an alert may be generated by the radar scanning the target area 1006 and transferring the raw data to a processor 1007, the vehicle vibration detection module generating a vibration indicator 1008, the time behavior analysis module generating a time movement indicator such as a spectral peak indicator, an RPM indicator, and a circle fit indicator or the like 1009, and the spatial feature analysis module generating a spatial feature indicator 1010.

[0047] The method may continue with the processor transferring 1011 a feature vector containing these indicators to the presence detection unit, the presence detection unit processing 1012 the feature vector, and determining 1013 whether a living body has been detected.

[0048] If no body is detected 1014, the radar continues scanning the target area 1006. If a body is detected, an alert is generated 1015 and the radar continues scanning the target area 1006 further as before.

[0049] If necessary, an additional step of applying pet mitigation 1016 may be included to distinguish between human and animal bodies.

[0050] 2A, a flow chart is presented outlining a possible method 2008 for generating a vehicle vibration indicator as described above. Optionally, the vehicle vibration sensing module may obtain 2028 a series of three-dimensional frames representing captured radar images of the target area, remove static objects from the image data 2048, thereby optionally generating a two-dimensional Moving Target Indication (MTI) matrix 2068, in which the lowest intensity values ​​of pixels, e.g., the lowest 5 percentiles, in the MTI matrix are summed 2088, thereby providing an indication of background movement in the target area.

[0051] It is expected that the lowest intensity pixel 302 in a stationary vehicle should be very low, as indicated in the MTI intensity profile of Figure 3D, yet it is expected that the lowest intensity pixel in a shaking vehicle could be much higher 304, as indicated in the MTI intensity profile of Figure 3E.

[0052] The sum of the lowest intensity pixels of the MTI matrix can serve as an effective vehicle vibration index.

[0053] A possible manner for removing 2048 static objects from image data is depicted in the flow chart of Figure 2B. A temporal filter may be applied to select 2481 a frame capture rate, collect 2482 raw data from a first frame, wait 2483 a time delay possibly determined by the frame capture rate, collect 2484 raw data from a second frame, and subtract 2485 the first frame data from the second frame data. In this manner, a filtered image may be produced in which the static background has been removed and only the moving target data remains.

[0054] By storing multiple frames in a frame buffer memory unit, the temporal filter may be further improved by applying a Moving Target Indication (MTI) filter as described in Applicant's co-pending International Patent Application No. PCT / IB2022 / 055109, which is incorporated herein in its entirety.

[0055] The MTI may be applied to the data signal before it is transferred to the image reconstruction block or directly to the image data. The MTI may estimate background data, illustratively using an infinite impulse response (IIR) low-pass filter (LPF). This background data is subtracted from the image data to separate reflections from moving objects. It is noted that such a process may be achieved by subtracting the average of several previous frames from the current frame. Optionally, the average may be calculated by an IIR or FIR low-pass filter, such as the LPF implementation described above.

[0056] The MTI IIR filter time constant, or the duration over which the IIR response is averaged, is typically fixed to best suit the requirements, i.e., short to better suit dynamic targets, or long to suit stationary or slow targets.

[0057] Thus, the MTI method may include steps such as selecting a filter time constant, applying an IIR filter for the duration of the selected time constant, applying a low-pass filter, and removing background from the raw data.

[0058] It has been found that MTI can produce artifacts such as ghosts when an object is suddenly removed from the background. For instance, when a chair is moved, a person goes to sleep, a wall is briefly closed, or similar events occur, and subsequent background subtraction can cause such events to leave a shadow in the image in the previously occupied location. Since the signal is complex, it is not possible to distinguish between a real object and its dark shadow.

[0059] Similarly, an obscured stationary object in the background can appear dynamic when the object suddenly appears when not obscured by a moving object in the foreground.

[0060] Furthermore, slow changes of interest may be suppressed; by way of illustration, reflections from motionless people sitting or lying down may change very little over time and therefore their effect may be attenuated by background subtraction.

[0061] Thus, a three-dimensional MTI array may be generated, from which a two-dimensional MTI matrix may be generated, as illustratively illustrated in FIG.

number

number

number

[0062] As shown in FIG. 3A, a 3D MTI array can include a 3D array of voxels, each voxel having spherical coordinates

number

number

number

[0063] The 3D MTI array of the target region is reduced to a 2D MTI matrix by constructing a matrix containing the 2D array of pixels. Hence, a unique MTI value

number

number

[0064] It is noted that although a spherical coordinate system is described herein, equivalent methods may use other three-dimensional coordinate systems, such as cylindrical, Cartesian, or the like.

[0065] FIG. 3B shows a set of voxels that share the same angular coordinate but have different r coordinates.

number

[0066] An example of such a two-dimensional MTI matrix is ​​provided in Figure 3C. By constructing a two-dimensional MTI matrix and ordering the pixels starting with the pixel having the highest MTI intensity and proceeding to pixels with lower and lower MTI intensities, an MTI intensity distribution profile may be generated.

[0067] Figure 3D shows an example of an aligned MTI intensity profile that shows typical MTI profile characteristics of a moving object in a stationary environment. It is noted that most of the pixels show no movement at all, and only pixels that indicate moving objects have high MTI values.

[0068] In contrast, Figure 3E shows an example of an aligned MTI intensity profile that exhibits a typical MTI profile characteristic of a vibrating environment, where vibrations are detected uniformly in all directions. It is noted that every pixel now indicates movement, as indicated by the relative uniformity of the MTI profile. This profile is to be expected in a rocking vehicle.

[0069] Reference is now made to the flow chart of FIG. 4, which illustrates a possible method 400 for the temporal behavior analysis module to generate a time movement indicator for use by the presence detection unit.

[0070] The method includes identifying 401 clusters of hyperintense voxels in the three-dimensional image data of the target region, and for each cluster, a processor unit matches 402 a series of complex values ​​per voxel representing reflected radiation of associated voxels in a plurality of frames, determining 403 a center point of a complex plane for each voxel, determining 404 a phase value for each voxel in each frame, generating 405 a smooth waveform representing phase change over time for each voxel, selecting 406 a subset of voxels that may be indicative of a breathing pattern, and calculating time migration indices such as a spectral peak index 407, a breaths per minute (RPM) index 408, and a circle fit index 409.

[0071] Various values ​​of the time shift index may be selected. For illustrative purposes, the ratio of the maximum fast Fourier transform of the unwrapped phase to the intermediate fast Fourier transform

number

[0072] An example of an RPM index may be given by calculating the following values:

number

[0073] Again by way of illustration, a circle fit index, which may indicate how closely a complex value fits a circle in the complex plane, is the standard deviation of the ratio of the magnitude of a complex vector to its maximum magnitude:

number

[0074] Optionally, the processor may generate a series of frames, each frame including an array of complex values ​​representing radiation reflected from each voxel of the target region during a given time segment.

[0075] In one embodiment, the method of the present invention monitors multiple voxels in parallel over a period of time. The signal received by the receiver may be given by:

number

[0076] For each monitored voxel, a reference center point

number

[0077] The phase value for each voxel in each frame may be determined by the processor matching a series of complex values ​​for each voxel representing reflected radiation for associated voxels in multiple frames, determining a center point of a complex plane for each voxel, and calculating the arc tangent of the ratio of the imaginary and real components of the difference between the frame value and the center point.

[0078] Thus, taking into account the reference center point, the phase of a voxel v at a given instant n may be calculated as follows:

number

[0079] The phase values ​​may be rounded to create a smooth waveform representing the phase changes over time. The phase may be unwrapped to produce a smooth waveform with no discontinuities greater than π according to the following formula:

number

[0080] In another embodiment, phase unwrapping may be based on a prediction of the next phase based on several previous phases. Such a prediction may be used to lower frame rates and / or improve resilience to noise while avoiding cycle slips. By way of illustration, the following predictors describe the phase momentum well:

number

[0081] Figures 5A and 5D show illustrative plots of a series of complex values ​​representing reflected radiation at a single voxel in a target region across multiple frames. It is noted that the complex values ​​form an approximate circle in the complex plane centered at a single point. The periodicity may appear to give rise to a characteristic oscillatory function, such as that shown in Figures 5B and 5E, which represent the phase variation over time (using frame number as a proxy for time) of the plots of Figures 5A and 5B, respectively, across multiple frames. Figures 5C and 5F show the corresponding variation in frequency space. Such oscillatory functions may illustratively represent respiration or pulse rate.

[0082] Voxels indicative of respiratory and pulsation characteristics may be found by selecting a subset of voxels that fit selection rules, such as by using a metric that evaluates the suitability of these voxels. Such metrics can include fitting a model of an arc of a circle, fitting a predefined pattern of pulsation waveform with strong periodicity, and the like, and the spatial location of the voxels.

[0083] In many cases, the voxels that are best suited for respiration tracking are those near the rib cage and stomach of a person who is breathing, in other cases the most appropriate choices are other voxels such as reflections from walls or ceilings, or the movement of other objects due to respiration.

[0084] An inverse fit metric may be calculated for the phase value associated with each voxel, and the selected voxels will be those with an inverse fit metric above a predefined threshold. Illustratively, the metric may assess the accuracy of fitting the data to a model described herein. Relative stability may be measured from the distance between the received signal and an estimated reference center point.

number

[0085] The metric may be calculated as follows, for example:

number

[0086] Additionally or alternatively, a time-dependent function may be calculated for the phase value associated with each voxel, and voxels may be selected that have periodic characteristics indicative of a pulse, such as duration of systole, duration of diastole, pulse rate, and the like, and combinations thereof.

[0087] Such a metric is the unwrapped phase

number

[0088] Illustratively, cyclic characteristics indicative of respiration may include, for example, an inspiration to expiration ratio of 1:1 to 1:6, a respiration rate of, for example, 1 second to 10 seconds, and illustratively, a cyclic characteristic indicative of a subject's pulse at rest may include, for example, a pulse or heart rate of 45 to 150 beats per minute, and a diastolic to systolic ratio of approximately 2:1.

[0089] The above two metrics may be smoothed and then combined into a single metric that represents the suitability of each voxel for pulse extraction.

[0090] In one possible embodiment of the present invention, a single voxel is selected for pulse determination based on the above metric with hysteresis to avoid frequent jumps between voxels.

[0091] In another embodiment, multiple voxels with high metric values ​​are selected and their waveforms are averaged by using SVD (PCA) after weighting by the suitability metric.

[0092] The lower frequency oscillations of the phase signal indicative of respiration may be filtered from the phase profile signal to leave the higher phase oscillations indicative of heart rate.

[0093] In yet other embodiments, voxel selection may use additional metrics such as signal quality (SNR) to ensure that the signal extracted from this voxel should have a good enough signal to be useful, and breath detection to ensure that the observed signal matches a real-world breathing signal. These two metrics may be combined to determine whether a voxel is suitable for selection.

[0094] Referring now to the flow chart of Fig. 6, a possible method 600 for generating spatial feature indices is presented. The method includes acquiring 3D image data of an active region including a target region and its surroundings 601, identifying voxel clusters within the target region 602, counting the clusters within the target region to obtain a cluster number index 603, counting the number of voxels within each cluster and selecting the maximum number to obtain a maximum cluster size index 604, calculating the difference between the maximum and minimum extents of voxels within each cluster 606, and selecting the value closest to the infant reference value 607 to obtain a cluster depth index 605, selecting the highest moving target indication (MTI) value within the target region to obtain a target maximum voxel index 608, selecting the highest MTI value within the active region to obtain an active region maximum voxel index 609, and selecting the extent of the active region maximum voxel to obtain a maximum voxel extent index 610.

[0095] As mentioned above, if desired, a pet mitigation module may be provided to distinguish pets, such as dogs, from children. Thus, when a body is detected within the vehicle, pet mitigation may be applied, the results of which may determine whether an alert is generated or the nature of such alert.

[0096] Because radar images of some pets can be similar in size and shape to children, pet mitigation may require high-resolution imaging to distinguish between them. Moreover, even with high-resolution imaging devices such as cameras, processing times may be too long and automatic recognition may be further complicated by the wide variety of different pets and their behaviors.

[0097] Referring back to FIG. 1A according to one system, the pet mitigation module 147 may be in communication with a pet stimulation signal generator 149 configured to generate a signal that a pet will typically react to differently than a human. By way of illustration, an ultrasonic signal that is audible to a dog's ear but not to a human's may elicit a physical reaction from the dog that can be detected by the radar 120, while the human body would remain unaffected. Alternative solutions may include pitches or tones that a pet may react to differently than a human in a predictable manner. Such behavioral solutions should be robust and be able to encompass many options for pet size, location, posture, etc.

[0098] Referring now to FIG. 7, a possible method 700 for applying pet mitigation is presented. A target is identified 701 and identified as a living being 702. A pet stimulation signal may be generated and transmitted 703 to the vehicle cabin. The pet stimulation signal is selected such that only pets will respond to the signal. Thus, if increased activity of the target is detected 704, such as significant movement between the seat and footwell or the like, this activity illustratively indicates an unrestrained pet in the vehicle cabin, rather than a restrained child. It will be appreciated that the movement of the target may be tracked illustratively by constructing a bounding box around the pixel cluster and tracking the position of the centroid of the bounding box over time. Other possible activity tracking methods may be used as would occur to one skilled in the art, such as the one illustrated in FIG. 8.

[0099] Technical Notes The technical and scientific terms used herein shall have the same meaning as commonly understood by those skilled in the art to which this disclosure pertains. Nevertheless, it is expected that many related systems and methods will be developed during the life of the patent that expires from this application. Thus, the scope of terms such as computing unit, network, display, memory, server, etc. is intended to include all such new technologies a priori.

[0100] As used herein, the term "about" refers to at least ±10%.

[0101] The terms "comprise," "comprising," "including," "including," "having," and cognates thereof mean "including but not limited to" and indicate the inclusion of the listed elements but not the exclusion of other elements as a whole. Such terms include the terms "consisting of" and "consisting essentially of."

[0102] The phrase "consisting essentially of" means that the composition or method may include additional elements and / or steps, but only if the additional elements and / or steps do not materially alter the basic and novel characteristics of the claimed composition or method.

[0103] As used herein, the singular forms "a," "an," and "the" can include plural references unless the context clearly dictates otherwise. Illustratively, the term "a compound" or "at least one compound" can include a plurality of compounds, including mixtures of multiple compounds.

[0104] The word "exemplary" is used herein to mean "serving as an example, instance, or illustration." Any embodiment described as "exemplary" is not necessarily to be construed as preferred or advantageous over other embodiments or to exclude the incorporation of features from other embodiments.

[0105] The word "optionally" is used herein to mean "is provided in some embodiments and is not provided in other embodiments." Any particular embodiment of the present disclosure may include multiple "optional" features, so long as the multiple "optional" features do not conflict.

[0106] Whenever a numerical range is indicated herein, it is meant to include any cited numbers (fractional or integer) within the indicated range. The phrases "ranging between" a first indicated number and a second indicated number, and "ranging from" a first indicated number to a second indicated number, are used interchangeably herein and are meant to include the first and second indicated numbers, and all fractional and integer numbers therebetween. It should therefore be understood that the description of the range format is merely for convenience and brevity, and should not be taken as an inflexible limitation on the scope of the disclosure. Thus, the description of a range should be considered to have specifically disclosed all possible subranges and individual numerical values ​​within the range. By way of example, description of a range such as 1 to 6 should be considered to have specifically disclosed subranges such as 1 to 3, 1 to 4, 1 to 5, 2 to 4, 2 to 6, 3 to 6, etc., and individual numbers within that range, by way of example only, 1, 2, 3, 4, 5, and 6, as well as non-integer intermediate values. This applies regardless of the broadness of the range.

[0107] It is understood that certain features of the disclosure that are described for clarity in the context of separate embodiments may also be provided in combination in a single embodiment. Conversely, various features of the disclosure that are described for brevity in the context of a single embodiment may also be provided separately or in any suitable subcombination or as appropriate in any other described embodiment of the disclosure. Certain features described in the context of various embodiments should not be considered essential features of those embodiments unless the embodiment is inoperable without those elements.

[0108] While this disclosure has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art. Accordingly, it is intended to embrace all such alternatives, modifications, and variations that fall within the spirit and broad scope of the appended claims.

[0109] All publications, patents, and patent applications mentioned herein are hereby incorporated by reference in their entirety to the same extent as if each individual publication, patent, or patent application was specifically and individually indicated to be incorporated by reference herein. Furthermore, citation or identification of any reference in this application should not be taken as an admission that such reference is available as prior art to the present disclosure. To the extent section headings are used, they should not be taken in a necessarily limiting sense.

[0110] The scope of the disclosed subject matter is defined by the appended claims and includes both combinations and subcombinations of the various features described above, as well as variations and modifications thereof, that would occur to one skilled in the art upon reading the foregoing description.

Claims

1. 1. A system for detecting the presence of a body within a vehicle passenger compartment, comprising: at least one transmitter antenna coupled to the oscillator and configured to transmit electromagnetic waves into the vehicle passenger compartment; and at least one receiver antenna configured to receive electromagnetic waves reflected by objects within the vehicle passenger compartment and operable to generate raw data; A radar unit comprising: a processor unit configured to receive data from the radar unit and operable to generate a warning command based on the data; an alert generator configured to generate a child presence detection (CPD) alert; Equipped with The processor, a vehicle vibration sensing module operable to sense vibrations in the vehicle cabin; a temporal behavior analysis module operable to analyze temporal characteristics of movement and to identify respiratory oscillation characteristics; a spatial characteristic analysis module operable to analyze spatial characteristics of clusters of voxels detected within the vehicle cabin; A system comprising:

2. The system of claim 1 , further comprising a communication module configured and operable to communicate the child presence detection alert to a third party.

3. The system of claim 2 , wherein the communication module is in communication with a computer network.

4. 10. The system of claim 1, further comprising a pre-processor configured to receive raw data from the radar unit and operable to produce a filtered point cloud for model optimization.

5. 10. The system of claim 1, further comprising a frame buffer memory unit configured and operable to store frame data.

6. 1. A method for detecting the presence of a body within a vehicle passenger compartment, comprising: Providing a radar module; providing a vehicle vibration sensing module; providing a time behavior analysis module; providing a spatial feature analysis module; Providing a presence sensing unit; at least one transmitter antenna transmitting electromagnetic waves into the vehicle passenger compartment; receiving, with at least one receiver antenna, electromagnetic waves reflected by objects within the vehicle passenger compartment; transferring data from the radar to a processor; the vehicle vibration sensing module generating a vehicle vibration indicator; the time behavior analysis module generating a time movement index; said spatial feature analysis module generating a spatial feature index; transferring the feature vector to the presence detection unit; said presence detection unit processing said feature vector; if a body is detected, issuing an alert; A method comprising:

7. 4. The method of claim 3, wherein the step of the vehicle vibration sensing module generating a vehicle vibration indicator further comprises: acquiring a series of three-dimensional frames of image data; removing static objects from the image data; generating a two-dimensional moving target indication matrix; summing the lowest intensity values ​​of pixels in said two-dimensional moving target indication matrix; The method of claim 6, comprising:

8. removing static objects from the image data, selecting a frame capture rate; collecting raw data from a first frame; waiting for a time delay; collecting raw data from a second frame; subtracting the first frame data from the second frame data; The method of claim 7, comprising:

9. said step of generating a two-dimensional moving target indication matrix further comprising: Maximum Intensity Voxel [0010] The angular coordinate [0025] and identifying each pair of A value I equal to the intensity of the identified maximum voxel ｍａｘ Each pixel assigned [0030] constructing a two-dimensional matrix having The method of claim 7, comprising:

10. The step of the time behavior analysis module generating the time movement index comprises: identifying clusters of hyperintense voxels within the three-dimensional image data; for each cluster, said processor unit matching a sequence of complex values ​​for each voxel; determining, for each voxel, a centre point of a complex plane; determining a phase value for each voxel in each frame; generating a smooth waveform representing phase change over time for each voxel in each frame; selecting a subset of voxels indicative of a breathing pattern; calculating a time moving index; The method of claim 6, comprising:

11. The method of claim 10 , wherein the step of calculating a time shift index comprises the step of calculating a spectral peak index.

12. The method of claim 10, wherein the step of calculating a time movement index comprises the step of calculating a breaths per minute (RPM) index.

13. The method of claim 10 , wherein the step of calculating a time movement metric comprises the step of calculating a circle fit metric.

14. The step of the spatial feature analysis module generating a spatial feature index further comprises: acquiring a series of three-dimensional frames of image data of an active area including the vehicle cabin and surroundings; identifying voxel clusters within the active region; counting the clusters within the target region, thereby obtaining a cluster count indicator; counting the number of voxels in each cluster; selecting a maximum number of voxels, thereby obtaining a maximum cluster size index; obtaining a cluster depth metric; obtaining a target maximum voxel index; obtaining an active region maximum voxel index; obtaining a maximum voxel range index; The method of claim 6, comprising:

15. The step of obtaining a cluster depth indicator further comprises: calculating the difference between the maximum and minimum extents of the voxels in each cluster; selecting the value closest to the infant reference value; 15. The method of claim 14, comprising:

16. The method of claim 14 , wherein obtaining a target maximum voxel index comprises selecting a highest moving target indication value within the active region.

17. The method of claim 14 , wherein obtaining the active region maximum voxel index comprises selecting the highest MTI value within the active region.

18. The method of claim 14 , wherein obtaining a maximum voxel range indicator comprises selecting a range of the active region maximum voxel.

19. The method of claim 6 further comprising the step of distinguishing between children and pets.

20. The step of distinguishing between children and pets further comprises: identifying a child-sized target; transmitting a pet stimulation signal that is meaningless to humans; if increased activity is detected in the child-sized target, associating the child-sized target with a pet; The method of claim 6, comprising:

21. 7. The method of claim 6, wherein said step of transmitting a pet stimulation signal that is inaudible to humans comprises transmitting the pet stimulation signal at a frequency that is inaudible to the human ear.

Images