Data processing method and data processing system

Abstract

The invention provides a data processing method and a data processing system, which can improve the detection precision of a living body in a vehicle. The data processing method is used for processing data acquired by a millimeter wave radar, and is characterized by comprising the following steps of: acquiring measured data in a vehicle; and a step of eliminating noise in the in-vehicle measured data on the basis of a noise map generated on the basis of in-vehicle environment data collected when it is confirmed that no living body exists in the vehicle.

Description

Technical Field

[0001] This invention relates to data processing methods and data processing systems. Background Technology

[0002] In the field of automotive safety, the detection of living beings such as children and pets inside vehicles is an important research direction. To achieve in-vehicle living being detection, occupant detection technology based on millimeter-wave radar is commonly used. Millimeter-wave radar is a radar device that emits discontinuous radar signals and analyzes the returned radar signals to sense the presence, distance, speed, etc., of targets. Specifically, in millimeter-wave radar-based occupant detection technology, after the driver leaves the vehicle, the millimeter-wave radar emits signals for a short period (e.g., 10 seconds), and analyzes the vital characteristics in the collected radar signals to identify living beings inside the vehicle (e.g., children, sleeping adults, or pets). When a living being is detected, an alarm is triggered to ensure the safety of the passengers. Furthermore, in existing millimeter-wave radar-based occupant detection technologies, the data from the collected radar signals needs to be processed to reduce the impact of noise in the vehicle's interior environment before analysis and identification.

[0003] Common data processing methods for noise reduction include real-time methods such as signal strength enhancement, resolution improvement, circuit filtering, and software filtering. However, since the acquired radar signals contain signals from both the target (e.g., a living organism) and noise, existing data processing methods for noise reduction suffer from various problems.

[0004] First, in existing data processing methods, when reducing noise based on noise separation parameters, the acquired radar signals simultaneously contain signals from both the target and noise sources, leading to low reliability of these noise separation parameters. Furthermore, obtaining highly reliable noise separation parameters requires highly complex calculation methods, increasing maintenance costs and reducing efficiency. Second, in existing data processing methods, achieving real-time noise reduction is difficult because the algorithm's accuracy is directly proportional to its processing time; therefore, it's impossible to simultaneously balance algorithm accuracy and processing time. To achieve sufficiently high algorithm accuracy, performance must be reduced and processing time increased, resulting in delayed alarms and hindering rescue efforts. Finally, existing data processing methods discard historical data from the previous detection and process new data each time a living being is detected inside the vehicle. This prevents sensors and other components from fully utilizing idle resources, further limiting the performance and efficiency of the data processing method.

[0005] Therefore, when using existing data processing methods to detect living beings inside vehicles, it is particularly susceptible to the influence of noise in the vehicle environment, which increases false detections, reduces detection accuracy, and requires a huge amount of data processing.

[0006] Furthermore, in another existing data processing method, pre-set data representing the vehicle's interior environment at the time of manufacture is used to eliminate noise in the collected radar signal each time a living being is detected inside the vehicle. However, the interior environment changes with vehicle use, for example, due to the addition of items inside the vehicle. Therefore, using pre-set data representing the interior environment at the time of manufacture cannot accurately eliminate noise, leading to reduced detection accuracy. Summary of the Invention

[0007] The present invention was made in view of the above-mentioned technical problems, and its purpose is to provide a data processing method and a data processing system that can improve the detection accuracy of living beings inside vehicles.

[0008] According to the data processing method of the present invention, data collected using millimeter-wave radar is processed, characterized in that it includes: a step of collecting in-vehicle measured data; and a step of eliminating noise in the in-vehicle measured data based on a noise map, wherein the noise map is generated based on in-vehicle environmental data collected when it is confirmed that there are no living beings in the vehicle.

[0009] Therefore, the data processing method of this invention collects in-vehicle environmental data to generate a noise map when it is confirmed that there are no living beings inside the vehicle, and then eliminates noise from the measured data inside the vehicle based on the noise map. This improves the accuracy of detecting living beings inside the vehicle when analyzing the noise-eliminated measured data.

[0010] Furthermore, the data processing method of this invention uses in-vehicle environmental data collected when it is confirmed that no living beings are present in the vehicle to generate a noise map to process the measured data inside the vehicle. Therefore, compared with methods that use pre-set in-vehicle environmental data at the factory to eliminate noise, noise can be eliminated more accurately, and the detection accuracy of living beings inside the vehicle can be improved.

[0011] According to the data processing method of the present invention, the in-vehicle environment data includes at least one of time domain data, frequency domain data, and spatial data.

[0012] Therefore, the data processing method of this invention converts in-vehicle environmental data collected after confirming the absence of living beings inside the vehicle into at least one of three dimensions: time-domain data, frequency-domain data, and spatial data. Thus, information such as the time, type, source, and location of noise are determined in each of these three dimensions. This allows for more accurate noise elimination and improves the detection accuracy of living beings inside the vehicle. Furthermore, it reduces the computational load when detecting living beings inside the vehicle, thereby improving work efficiency.

[0013] According to the data processing method of the present invention, in the step of eliminating noise in the in-vehicle measured data, based on the noise map, noise information in the in-vehicle measured data is identified according to the time domain data, the frequency domain data and the spatial data, and noise reduction method corresponding to the noise information is used to eliminate the noise.

[0014] Therefore, through the data processing method of this invention, noise information such as time, type, source, and location is determined in three dimensions: time domain data, frequency domain data, and spatial data. This allows for the appropriate application of corresponding noise reduction methods based on the different noise conditions represented by the noise information, leading to more accurate noise elimination and improved detection accuracy of living beings inside vehicles. Furthermore, it also improves work efficiency.

[0015] According to the data processing method of the present invention, in the step of eliminating noise in the in-vehicle measured data, the data of the time range in the in-vehicle measured data is eliminated based on the information representing the time range of the noise occurrence in the time domain data.

[0016] According to the data processing method of the present invention, in the step of eliminating noise in the in-vehicle measured data, the data of the frequency range in the in-vehicle measured data is eliminated based on the information representing the frequency range of the noise in the frequency domain data.

[0017] According to the data processing method of the present invention, in the step of eliminating noise in the in-vehicle measured data, the data of the location range in the in-vehicle measured data is eliminated based on the information in the spatial data representing the location range where the noise occurs.

[0018] Therefore, the data processing method of this invention eliminates noise in three dimensions: time domain data, frequency domain data, and spatial data. Furthermore, data processing is performed to eliminate noise in the in-vehicle measurement data used for detecting living beings inside the vehicle. This allows for more accurate noise elimination and improves the detection accuracy of living beings inside the vehicle. Additionally, by eliminating unwanted data, the computational load during in-vehicle detection is reduced, improving work efficiency.

[0019] According to the data processing method of the present invention, in the step of confirming that there is no living being in the vehicle, the confirmation is based on the user's operation.

[0020] According to the data processing method of the present invention, the feature portion of the measured data inside the vehicle is amplified by a filtering algorithm.

[0021] According to the data processing system of the present invention, data acquired using millimeter-wave radar is processed, characterized in that the data processing method according to any one of claims 1 to 7 is used.

[0022] Invention Effects

[0023] The data processing method and data processing system of the present invention can improve the detection accuracy of living beings inside vehicles.

[0024] Furthermore, the data processing method and data processing system according to the present invention can accurately eliminate noise and reduce the computational load when detecting living beings inside a vehicle, thereby improving work efficiency. Attached Figure Description

[0025] Figure 1 This is a block diagram representing an example of the structure of a data processing system.

[0026] Figure 2A This is a block diagram illustrating an example of the configuration of a millimeter-wave radar.

[0027] Figure 2B This is a schematic diagram showing the spatial distribution of millimeter-wave radar inside the vehicle's cabin.

[0028] Figure 3 This is a flowchart illustrating the process of detecting living beings inside a vehicle.

[0029] Figure 4 This is a flowchart representing the generation of the noise map.

[0030] Figure 5 This is a schematic diagram illustrating the process of eliminating noise from measured data inside a vehicle based on a noise map.

[0031] Explanation of reference numerals in the attached figures

[0032] 1. Data processing system; 11. Communication IF; 12. Data processing unit; 13. Memory; 2. Millimeter-wave radar; 21. Radar antenna; 22. Radar chip; 23. Power supply; 24. Communication IF; 3. Operation terminal Detailed Implementation

[0033] Hereinafter, the data processing method and data processing system of the present invention will be described with reference to the accompanying drawings. In the various embodiments described below, the same reference numerals are used for the same parts as in the prior drawings, and detailed descriptions of those parts are omitted; descriptions will focus primarily on the different parts.

[0034] The data processing system of this invention consists of multiple functional modules. It can be installed as software in a standalone computer or other device with a CPU (central processing unit) and memory, or it can be distributed across multiple devices. The processor executes the various functional modules of the data processing system stored in memory. The circuitry implementing the data processing system is capable of sending and receiving data or acquiring data via networks such as the Internet.

[0035] Furthermore, in the following description, the data processing method and system for detecting living beings inside a car are used as an example. However, the application scope of the data processing method and system of the present invention is not limited to this. For example, it can also be used to detect living beings inside other vehicles besides cars. In addition, in the description of the present invention, "living being" does not mean a person in the narrow sense, but also includes various living beings such as pets.

[0036] (Data Processing System 1)

[0037] The following is for reference Figure 1 The data processing system 1 will be described. Figure 1 This is a block diagram representing an example of the structure of data processing system 1.

[0038] like Figure 1 As shown, the data processing system 1 includes a communication IF 11, a data processing unit 12, and a memory 13.

[0039] Communication IF11 is an interface used for communication of various data or information. Data processing system 1 is connected, for example, to a millimeter-wave radar 2 and an operation terminal 3 external to data processing system 1 in a communicative manner via communication IF11. Communication IF11 also facilitates communication of various data or information between the components included in data processing system 1.

[0040] The data processing unit 12 is a functional unit for processing received data. The data processing unit 12 processes the data acquired using the millimeter-wave radar 2 using the data processing method described later.

[0041] The memory 13 is a device for storing various types of data or information. The memory 13 can be a storage medium that can be read by a processor (e.g., magnetic storage medium, electromagnetic storage medium, optical storage medium, semiconductor memory), or a drive device for reading and writing data or information between itself and the storage medium. For example, the memory 13 stores computer programs necessary for the data processing unit 12 to implement the data processing methods described later.

[0042] Furthermore, the above description assumes that the data processing system 1, the millimeter-wave radar 2, and the operation terminal 3 are configured independently. However, this is not a limitation; for example, the data processing system 1 can also be installed in the millimeter-wave radar 2 or the operation terminal 3. This is as long as the data processing system 1 can process the data acquired by the millimeter-wave radar 2 using the data processing methods described later.

[0043] Reference Figure 2A as well as Figure 2B The millimeter-wave radar 2 will be explained. Figure 2A This is a block diagram illustrating an example of the configuration of millimeter-wave radar 2. Figure 2B This is a schematic diagram showing the spatial distribution of millimeter-wave radar inside the vehicle's cabin.

[0044] Millimeter-wave radar 2 is a radar device that detects the presence, location, and time of living beings inside a vehicle by emitting millimeter-wave radar signals and analyzing the collected data (radar signals). For example... Figure 2A As shown, the millimeter-wave radar 2 includes a radar antenna 21, a radar chip (ASIC) 22, a power supply 23, and a communication IF 24.

[0045] like Figure 2B As shown, the spatial range of the millimeter-wave radar 2 is, for example, set around the cockpit (e.g., around the perimeter of the cockpit). Figure 2B (The dashed circle in the image). This is because the effective detection range of millimeter-wave radar 2 is typically within 2 meters, and there are usually many obstacles inside the cockpit. Furthermore, the spatial location and number of millimeter-wave radar 2 units are not limited to this; other locations and numbers are also possible.

[0046] The operating terminal 3 is a terminal operated by a user (e.g., a driver). The operating terminal 3 can also function as a client of the data processing system 1. The operating terminal 3 can be a smartphone, tablet computer, vehicle-mounted terminal, wearable device, mobile terminal, or handheld terminal, etc. Furthermore, the operating terminal 3 has an input interface for receiving various operations from the user. The operating terminal 3 may also have a display device for displaying various data or information. The display device can also be a touch panel display device that also functions as an input interface.

[0047] Specifically, the operating terminal 3 receives the user's confirmation that no living beings are present in the vehicle and sends this confirmation to the data processing system 1. When it is necessary to collect in-vehicle environmental data (described later), the operating terminal 3 can prompt the user via a display device to confirm the absence of living beings in the vehicle. For example, the operating terminal 3 can periodically prompt the user to collect in-vehicle environmental data based on preset time intervals. Additionally, when the millimeter-wave radar 2 detects the presence of a living being in the vehicle, the operating terminal 3 can also issue an alarm to the user or other terminals via a display device when a living being is detected in the vehicle.

[0048] (The process for detecting living beings inside a vehicle)

[0049] The following is for reference Figure 3 The procedure for detecting living beings inside a vehicle is explained. Figure 3 This is a flowchart illustrating the process of detecting living beings inside a vehicle.

[0050] In step S1, the data processing system 1 determines whether the car door is closed. If the door is closed ("Yes" in step S1), proceed to step S2. If the door is not closed ("No" in step S1), continue the determination in step S1. Alternatively, in step S1, the data processing system may also determine whether the door is closed and locked instead of whether it is closed, proceeding to step S2 only if the door is closed and locked, and continuing the determination in step S1 otherwise.

[0051] In step S2, the millimeter-wave radar 2 collects in-vehicle measured data. Here, "in-vehicle measured data" refers to data collected by the millimeter-wave radar 2, for example, each time the driver leaves the vehicle and closes the door, in order to detect the presence of any living beings inside the vehicle. Next, the process proceeds to step S3.

[0052] In step S3, the data processing system 1 processes the in-vehicle measured data acquired by the millimeter-wave radar 2 in step S2 based on the noise map to eliminate noise in the in-vehicle measured data. Here, the noise map used by the data processing system 1 is generated through the noise map generation process described later. The noise map generation process and the method of eliminating noise in the in-vehicle measured data based on the noise map will be explained in detail later. Next, proceed to step S4.

[0053] In step S4, the millimeter-wave radar 2 performs data filtering on the in-vehicle measured data after noise removal by the data processing system 1 in step S3. Then, the process proceeds to step S5.

[0054] In step S5, the millimeter-wave radar 2 filters the in-vehicle measured data from living organisms after data filtering in step S4. Then, it proceeds to step S6.

[0055] In step S6, the millimeter-wave radar 2 scores the data from living beings selected in step S5, and detects the presence of living beings inside the vehicle based on the scores. Then, the process proceeds to step S7.

[0056] In step S7, if the millimeter-wave radar 2 detects the presence of a living being inside the vehicle ("Yes" in step S7), it proceeds to step S8. If it detects no living being inside the vehicle ("No" in step S7), the process ends.

[0057] In step S8, the millimeter-wave radar 2 sends an alarm to the user or other terminals, for example, via the display device of the operating terminal 3, indicating the presence of a living being inside the vehicle. Then, the process ends.

[0058] Here, in this invention, the methods such as "data filtering", "screening data from living organisms", and "scoring processing" mentioned in the above description are not limited in any way, and can be accomplished by any known method.

[0059] For example, in step S4, the feature parts in the measured data inside the vehicle can be amplified using a filtering algorithm to complete the data filtering.

[0060] In step S6, the data from living beings selected in step S5 are scored with an exist_score (0-1.0), and a pre-set threshold thresh_normal (e.g., 0.6) is used. If the exist_score is greater than the threshold thresh_normal, it is determined that a living being exists in the vehicle. If the exist_score is less than or equal to the threshold thresh_normal, it is determined that no living being exists in the vehicle. Ideally, when a driver is present in the vehicle, the exist_score will reach 0.8. Conversely, in certain noisy scenarios or at certain times, the exist_score may drop to 0.7 or even lower.

[0061] (The process of generating noise maps)

[0062] The following is for reference Figure 4 The process of generating noise maps is explained. Figure 4 This is a flowchart illustrating the process of generating a noise map. Figure 3 In step S3 shown, data processing system 1 is based on... Figure 4The noise map generation process shown (steps S301 to S306) generates a noise map to eliminate noise in the measured data inside the vehicle.

[0063] In this invention, "in-vehicle environment data" refers to the data representing the in-vehicle environment collected by millimeter-wave radar 2 when it is confirmed that there are no living beings inside the vehicle based on the user's operation of the operating terminal 3. It can be considered that "in-vehicle environment data" reflects a collection of noise data inside the vehicle.

[0064] Data processing system 1 periodically prompts the user via operating terminal 3 whether a noise map needs to be generated, and collects in-vehicle environmental data when the user selects "yes". The periodicity of prompting the user via operating terminal 3 to generate a noise map could be, for example, once a month. In addition, data processing system 1 can also generate a new noise map if the time elapsed since the last noise map generation exceeds a preset time interval.

[0065] In step S301, the data processing system 1 determines whether the user has confirmed that there are no living beings inside the vehicle. Here, the data processing system 1 may, for example, prompt the user via the operating terminal 3 to confirm that there are no living beings inside the vehicle. If the user confirms that there are no living beings inside the vehicle based on the operation of the operating terminal 3, the process proceeds to step S302.

[0066] In step S302, the millimeter-wave radar 2 collects in-vehicle environmental data. Since the user confirmed the absence of living beings in the vehicle in step S301, the in-vehicle environmental data collected by the millimeter-wave radar 2 does not include information from living beings, but includes information from noise within the vehicle environment. Furthermore, the in-vehicle environmental data collected by the millimeter-wave radar 2 includes at least one of time-domain data, frequency-domain data, and spatial data. Next, steps S303A, S304A, and S305A are performed.

[0067] In step S303A, the data processing system 1 generates time-domain data from the in-vehicle environment data acquired by the millimeter-wave radar 2. Here, the generated time-domain data includes data representing the time-varying changes in the in-vehicle environment data. Furthermore, the data processing system 1 generates the time-domain data, for example, through analog-to-digital conversion or signal queuing. Next, the process proceeds to step S303B.

[0068] In step S303B, the data processing system 1 identifies noise information in the time-domain data generated in step S303A. Here, the method for "identifying noise" is not limited and can be performed using any known method. Furthermore, the "noise information" in this invention includes, for example, information indicating the time, type, source, and location of the noise. Next, the process proceeds to step S303C.

[0069] In step S303C, the data processing system 1 determines the time range of noise occurrence based on the noise information identified in step S303B.

[0070] Specifically, when identifying noise information in time-domain data, the noise information is, for example, information representing the time-domain changes of the noise.

[0071] For example, when a car door is closed, the vehicle will generate significant shaking noise for a short period of time. The time-domain data of the in-vehicle environment collected at this time will exhibit a large waveform amplitude, and the larger the amplitude, the higher the probability of the noise's presence. By identifying the waveform with large amplitude within a short period, the time-domain information of the noise caused by the door closing can be identified in the time-domain data. At this point, based on the changes in the time-domain data, the time range of noise occurrence can be determined. For example, the interval with large waveform amplitude can be defined as the time range of noise occurrence. Furthermore, the time-domain data of the in-vehicle environment can also distinguish between permanent and temporary noise. This is because temporary noise will cause a significant change in the waveform of the time-domain data within a short period of time. Next, proceed to step S306.

[0072] In step S304A, the data processing system 1 generates frequency domain data from the in-vehicle environment data acquired by the millimeter-wave radar 2. Here, the generated frequency domain data includes data representing frequency variations in the in-vehicle environment data. Furthermore, the data processing system 1 generates the frequency domain data, for example, through filtering and integration methods. Next, the process proceeds to step S304B.

[0073] In step S304B, the data processing system 1 identifies noise information in the frequency domain data generated in step S304A. Then, it proceeds to step S304C.

[0074] In step S304C, the data processing system 1 determines the frequency range of the noise occurrence based on the noise information identified in step S304B.

[0075] Specifically, when identifying noise information in frequency data, the noise information is, for example, information representing changes in the frequency domain of the noise.

[0076] For example, when a liquid (such as mineral water) is placed inside a vehicle, the liquid continuously generates noise at a specific frequency, similar to biological respiration. The frequency domain data of the in-vehicle environment collected at this time exhibits a high SNR (signal-to-noise ratio) at that specific frequency, and a higher SNR indicates a higher probability of the presence of such noise. By identifying the specific frequency that generates the high SNR, information about the frequency domain of the noise caused by the liquid placed inside the vehicle can be identified in the frequency domain data. Additionally, for example, when a decorative element is suspended inside the vehicle, the swaying of this element also generates noise at a specific frequency. In this case, based on changes in the frequency domain data, the frequency range of the stationary vibration noise can be determined. For example, the specific frequency or frequency range that generates the high SNR can be determined as the frequency range of the noise. Next, proceed to step S306.

[0077] In step S305A, the data processing system 1 generates spatial data from the in-vehicle environment data acquired by the millimeter-wave radar 2. Here, the generated spatial data includes data representing spatial changes in the in-vehicle environment. Additionally, the data processing system 1 calculates spatial distance as spatial data, for example, using a timer and signal strength matrix offset. Next, the process proceeds to step S305B.

[0078] In step S305B, the data processing system 1 identifies noise information in the spatial data generated in step S305A. Then, it proceeds to step S305C.

[0079] In step S305C, the data processing system 1 determines the location range of the noise occurrence based on the noise information identified in step S305B.

[0080] Specifically, in the case of identifying noise information in spatial data, the noise information is, for example, information representing changes in the signal strength of noise in space.

[0081] For example, vehicle structure and various objects placed there can cause multipath effects. At a specific spatial location, there is a possibility that the signal strength of radar signals reflected by vehicle structure or objects can reach the signal strength of radar signals reflected by living organisms. In addition to the localized high signal strength caused by reflections from vehicle structure and objects, diffraction or refraction can also cause localized high signal strength. These localized high signal strengths can all reach the signal strength of radar signals reflected by living organisms. The spatial data of the in-vehicle environment collected at this time has the characteristic of generating high signal strength (amplitude) locally, and the higher the signal strength, the higher the probability of the presence of such noise. By identifying the locally generated high signal strength, spatial information about noise caused by vehicle structure and various objects placed there can be identified in the spatial data. Next, proceed to step S306.

[0082] Reference Figure 5 This paper explains the process of eliminating noise from measured in-vehicle data based on noise maps. Figure 5 This is a schematic diagram illustrating the process of eliminating noise from measured data inside a vehicle based on a noise map.

[0083] In step S306, the data processing system 1 generates a noise map based on the in-vehicle environment data collected by the millimeter-wave radar 2 when the user confirms that there are no living beings inside the vehicle. Specifically, the noise map is generated based on the time-domain data, frequency-domain data, and spatial data in the in-vehicle environment data. Here, the "noise map" can be considered as a database containing noise information. In the noise map, the noise information is stored in correspondence with noise reduction methods used to eliminate the noise. Additionally, as... Figure 5 As shown, since the noise map is generated based on temporal, frequency, and spatial data from the in-vehicle environment, the noise information in the noise map can be divided into temporal, frequency, and spatial characteristics. Specifically, the temporal characteristics include the noise identified in the time-domain data and its time range; the frequency-domain characteristics include the noise identified in the frequency-domain data and its frequency range; and the spatial characteristics include the noise identified in the spatial data and its location range. The process then ends.

[0084] Therefore, the data processing system 1 generates a noise map based on the in-vehicle environment data collected when the user confirms that there are no living beings in the vehicle. Furthermore, in step S3 of the in-vehicle living being detection process, the data processing system 1 uses the generated noise map to eliminate noise from the measured in-vehicle data.

[0085] Specifically, when eliminating noise in the in-vehicle measured data, information processing system 1 divides the in-vehicle measured data according to three dimensions: time domain, frequency domain, and space. For example, ... Figure 5 As shown, information processing system 1 divides the measured data inside the vehicle into Range bin 1, Range bin 2, Range bin 3, and Range bin 4 according to the Range bin (spatial range) dimension. At this time, Figure 5 The right figure shows the time-domain and frequency data corresponding to the divided Range bins 1 to 4 in the in-vehicle measured data. However, the method and result of dividing the in-vehicle measured data are not limited to this; for example, the in-vehicle measured data can also be divided into Range bins 1 to 8.

[0086] Next, the information processing system 1 eliminates the noise in the measured data inside the vehicle based on the noise information (noise time domain characteristics, noise frequency domain characteristics, and noise spatial characteristics) in the noise map and the corresponding noise reduction method.

[0087] For example, based on the time-domain characteristics of noise in the noise map, the area indicated by the dashed box in the time domain represents the time range in which the noise occurs. At this point, information processing system 1, based on the information representing the time range of noise occurrence from the time-domain data of the in-vehicle environment data included in the noise map, eliminates data within that time range from the measured in-vehicle data. Thus, the following is obtained: Figure 5 The processed in-vehicle measured data is shown in the time domain dimension.

[0088] Furthermore, based on the noise frequency domain characteristics in the noise map, it can be seen that the range indicated by the single-dotted-line box in the frequency domain dimension represents the frequency range where the noise occurs. At this point, information processing system 1, based on the information representing the frequency range of noise occurrence from the frequency domain data of the in-vehicle environment data included in the noise map, eliminates data within that frequency range from the measured in-vehicle data. Thus, the following is obtained: Figure 5 The data shown is the frequency domain dimension of the processed in-vehicle measured data.

[0089] Furthermore, based on the spatial characteristics of the noise in the noise map, Rangebin 3, indicated by the gray box, represents the location range where the noise occurs. At this point, information processing system 1, based on the information representing the location range of the noise occurrence from the spatial data of the in-vehicle environment data included in the noise map, eliminates data representing that location range from the measured in-vehicle data. Thus, the following is obtained: Figure 5 The data shown is the spatial dimension of the processed in-vehicle measured data.

[0090] Therefore, the information processing system 1 uses the noise map to eliminate noise in the in-vehicle measured data. Next, the process of the data processing system 1 processing the in-vehicle measured data to eliminate noise ends, and the process proceeds to step S4 as described above.

[0091] (Invention Effects)

[0092] The data processing method and system of the present invention can improve the detection accuracy of living beings inside vehicles. Furthermore, they can accurately eliminate noise and reduce the computational load when detecting living beings inside vehicles, thereby improving work efficiency.

[0093] Specifically, when detecting the presence of living beings inside a vehicle based on in-vehicle measurement data, the information processing system 1 eliminates at least one of the data related to the time range, frequency range, and location range of noise occurrence in the in-vehicle measurement data. Therefore, the detection accuracy of living beings inside the vehicle based on the in-vehicle measurement data can be improved. Furthermore, by eliminating noise data from the in-vehicle measurement data, the computational load for detecting living beings inside the vehicle based on the in-vehicle measurement data can be reduced, thus improving work efficiency.

[0094] In addition, the noise map generated based on the in-vehicle environment data contains corresponding noise information and noise reduction methods. Therefore, the appropriate noise reduction methods can be used appropriately according to different noise conditions, and the noise reduction methods are highly targeted.

[0095] Furthermore, as mentioned above, in situations such as closed car doors, the presence of liquids inside the vehicle, or the presence of decorative items hanging inside, existing methods for detecting in-vehicle life forms often result in scores approaching a pre-set threshold. However, in this invention, noise is eliminated from the measured data inside the vehicle based on a noise map, thus improving the accuracy of in-vehicle life form detection.

[0096] Furthermore, the data processing method of this invention is performed when the vehicle door is determined to be closed, thus fully utilizing the computing resources of the millimeter-wave radar 2 during idle periods. Existing methods for detecting living beings inside vehicles require heavy-load algorithms such as FFT, almost saturating the computing resources of the millimeter-wave radar 2. However, the data processing method of this invention eliminates noise data from the measured data inside the vehicle based on a noise map, thereby optimizing computing resources, reducing processing time, and improving work efficiency.

[0097] Furthermore, in existing data processing methods, pre-set factory-set in-vehicle environment data is used to eliminate noise from the measured in-vehicle data each time a living being is detected inside the vehicle. However, as mentioned above, the in-vehicle environment changes due to factors such as closed doors, the presence of liquids inside the vehicle, or the presence of decorative items. In such cases, using the factory-set in-vehicle environment data cannot accurately eliminate noise, leading to reduced detection accuracy. In contrast, the data processing method of this invention collects in-vehicle environment data to generate a noise map when it is confirmed that no living being is present inside the vehicle, and then eliminates noise from the measured in-vehicle data based on this noise map. Therefore, noise can be accurately eliminated.

[0098] The above embodiments are merely illustrative examples and are not intended to limit the present invention. Those skilled in the art can appropriately add, delete, or modify the constituent elements and designs based on the above embodiments; any modifications that incorporate the technical concept of the present invention are included within the scope of the present invention.

[0099] For example, the generated noise map can be uploaded to a cloud server, where a learning model deployed on the cloud server can learn from it. The learning model then distributes the learned noise map to various vehicles or terminals.

Claims

1. A data processing method for processing data acquired using millimeter-wave radar, characterized in that, include: Steps for collecting measured data inside the vehicle; as well as The steps for eliminating noise from the measured in-vehicle data based on noise maps. The noise map is generated based on in-vehicle environmental data collected when it is confirmed that there are no living beings inside the vehicle.

2. The data processing method according to claim 1, characterized in that, The in-vehicle environment data includes at least one of time-domain data, frequency-domain data, and spatial data.

3. The data processing method according to claim 2, characterized in that, In the step of eliminating noise from the measured in-vehicle data Based on the noise map, noise information in the measured in-vehicle data is identified according to the time-domain data, the frequency-domain data, and the spatial data. The noise is eliminated using a noise reduction method corresponding to the noise information.

4. The data processing method according to claim 3, characterized in that, In the step of eliminating noise from the measured in-vehicle data Based on the information in the time domain data representing the time range in which the noise occurs, the data within that time range in the in-vehicle measured data is eliminated.

5. The data processing method according to claim 3, characterized in that, In the step of eliminating noise from the measured in-vehicle data Based on the information representing the frequency range of the noise in the frequency domain data, the data representing the frequency range in the in-vehicle measured data is eliminated.

6. The data processing method according to claim 3, characterized in that, In the step of eliminating noise from the measured in-vehicle data Based on the information in the spatial data indicating the location range where the noise occurs, the data representing that location range in the in-vehicle measured data is eliminated.

7. The data processing method according to claim 1, characterized in that, In the step of confirming that there are no living beings inside the vehicle... The absence of any living being inside the vehicle is confirmed based on the user's actions.

8. The data processing method according to any one of claims 1 to 7, characterized in that, The feature components in the measured data inside the vehicle are amplified using a filtering algorithm.

9. A data processing system for processing data acquired using millimeter-wave radar, characterized in that, The data processing method according to any one of claims 1 to 7.

Images