Human body detection method and system based on multiple sensors

Abstract

The invention relates to a human body detection method and system based on multiple sensors, and the method comprises the steps: collecting ambient light data according to an ambient light sensor; adjusting detection parameters of a radar sensor and an infrared sensor based on the ambient light data; acquiring radar data and infrared heat radiation data through the adjusted radar sensor and infrared sensor; based on the ambient light data, the radar data and the infrared thermal radiation data, acquiring an ambient light change feature, a target dynamic feature and an infrared thermal radiation feature; and based on the ambient light change feature, the target dynamic feature and the infrared thermal radiation feature, judging the existence state of the human body through multi-feature fusion judgment logic. The problem of how to improve the human body detection accuracy in related technologies is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of human body detection, in particular to a human body detection method and system based on multiple sensors. BACKGROUND

[0002] In the field of smart home, security monitoring, human presence detection is the core basic technology to realize scene linkage and security protection.

[0003] At present, the mainstream human presence detection equipment on the market mostly adopts single sensor or dual sensor design, among which single PIR (pyroelectric infrared) sensor or single radar sensor is most widely used. It can determine the presence of human body by detecting the change of infrared radiation emitted by the human body or by the movement or micro-motion of objects in the environment. However, it has obvious defects: it is easy to be disturbed by the environment, and when there are items shaking, air conditioner vibration, pet activity, and floor heating in the detection range, it is easy to produce false positives, leading to false triggering of smart home linkage (such as turning on the light without reason, alarm), affecting user experience; although dual sensor (such as PIR+microwave radar) improves the anti-interference ability to some extent, it still has limitations: only two kinds of detection signal fusion judgment logic, it is still difficult to accurately filter the subtle interference in complex environment (such as the movement of light and thin objects driven by the wind accompanied by temperature change).

[0004] Therefore, at present, for the human body detection method based on multiple sensors, how to improve the accuracy of human body detection has not yet proposed an effective solution. SUMMARY

[0005] The embodiments of the present application provide a human body detection method and system based on multiple sensors to at least solve the problem of how to improve the accuracy of human body detection in related technologies.

[0006] In a first aspect, the embodiments of the present application provide a human body detection method based on multiple sensors, including an ambient light sensor, a radar sensor and an infrared sensor, the method comprising: collecting ambient light data according to the ambient light sensor; adjusting the detection parameters of the radar sensor and the infrared sensor based on the ambient light data; collecting radar data and infrared thermal radiation data through the adjusted radar sensor and infrared sensor; obtaining ambient light change characteristics, target dynamic characteristics and infrared thermal radiation characteristics based on the ambient light data, the radar data and the infrared thermal radiation data; determining the human presence state through a multi-feature fusion judgment logic based on the ambient light change characteristics, the target dynamic characteristics and the infrared thermal radiation characteristics.

[0007] In an embodiment, the adjusting the detection parameters of the radar sensor and the infrared sensor based on the ambient light data comprises: comparing the illumination intensity in the ambient light data with a preset illumination intensity threshold to obtain a comparison result; adjusting the micro-motion detection sensitivity of the radar sensor and adjusting the trigger threshold of the infrared sensor based on the comparison result.

[0008] In an embodiment, the adjusting the micro-motion detection sensitivity of the radar sensor and adjusting the trigger threshold of the infrared sensor based on the comparison result comprises: if the comparison result is that the illumination intensity is greater than the preset illumination intensity threshold, generating a first control instruction to increase the micro-motion detection sensitivity of the radar sensor and generating a second control instruction to decrease the trigger threshold of the infrared sensor; if the comparison result is that the illumination intensity data is less than or equal to the preset illumination intensity threshold, generating a third control instruction to decrease the micro-motion detection sensitivity of the radar sensor and generating a fourth control instruction to increase the trigger threshold of the infrared sensor.

[0009] In an embodiment, based on the ambient light data, the radar data and the infrared thermal radiation data, an ambient light change feature, a target dynamic feature and an infrared thermal radiation feature are obtained, comprising: performing mean filtering processing on the ambient light data to obtain mean filtered ambient light data; and based on the mean filtered ambient light data, the ambient light change feature is obtained; performing clutter suppression processing on the radar data to obtain clutter suppression processed radar data; and based on the clutter suppression processed radar data, the target dynamic feature is obtained; performing edge detection and noise filtering processing on the infrared thermal radiation data to obtain edge detection and noise filtering processed infrared thermal radiation data; and based on the processed infrared thermal radiation data, the infrared thermal radiation feature is obtained.

[0010] In an embodiment, based on the mean filtered ambient light data, the ambient light change feature is obtained, comprising: based on the filtered ambient light data, calculating a difference value between a current sampling value and a previous sampling value to obtain an illumination instantaneous change amount; comparing an absolute value of the illumination instantaneous change amount with a preset abrupt change threshold to determine whether an illumination abrupt change event occurs; based on the illumination abrupt change events occurring within a preset time window, calculating the number of abrupt changes per unit time to obtain an illumination fluctuation frequency; Based on the filtered ambient light data within the preset time window, its maximum and minimum values ​​are determined, and the illuminance change amplitude is calculated based on the difference between the maximum and minimum values. The instantaneous change in illuminance, the determination result of the illuminance abrupt change event, the illuminance fluctuation frequency, and the illuminance change amplitude are used as the characteristics of ambient light change.

[0011] In one embodiment, obtaining the target dynamic characteristics based on radar data after clutter suppression processing includes: Differential processing is performed on the dual-channel echo signal of the processed radar data to obtain a differential signal; Based on the differential signal, the Doppler frequency shift features of the target are extracted; Calculate the target's motion parameters based on the Doppler frequency shift characteristics; The motion parameters of the target are used as the target's dynamic features.

[0012] In one embodiment, obtaining the infrared thermal radiation characteristics based on the processed infrared thermal radiation data includes: Based on the processed infrared thermal radiation data, a temperature distribution matrix is ​​constructed, which contains the temperature value of each pixel in the array. Temperature clustering analysis is performed on the temperature distribution matrix to group spatially adjacent pixels with temperature differences within a preset range into a temperature cluster, resulting in at least one temperature cluster. Each temperature cluster includes the temperature value of the cluster center and the spatial distribution range of the cluster region. The temperature clustering is used as an infrared thermal radiation feature.

[0013] In one embodiment, the step of determining the human body's presence state through multi-feature fusion judgment logic based on the ambient light change characteristics, the target dynamic characteristics, and the infrared thermal radiation characteristics includes: Determine whether the ambient light change characteristics match the human activity light change pattern. The target's motion state is determined based on the target's dynamic characteristics; Based on the infrared thermal radiation characteristics, determine whether there are target characteristics that match human thermal radiation; Based on the judgment results of the human activity light variation pattern, the judgment results of the target motion state, and the judgment results of the target characteristics of human thermal radiation, a fusion logic judgment is performed to determine the human body's existence state.

[0014] In one embodiment, determining whether the ambient light change characteristics match the human activity light change pattern includes: Based on the illuminance abrupt change events and illuminance fluctuation frequency in the ambient light change characteristics, determine whether the current ambient light change pattern conforms to the preset human activity light change pattern. If the light pattern matches the described human activity pattern, it is determined that someone is present. If the light variation pattern does not conform to the human activity pattern and the illuminance change amplitude is below the threshold, the system is determined to be unmanned. The step of determining the target motion state based on the target dynamic characteristics includes: Based on the target speed value in the motion parameters, determine whether it is within the preset range of human activity speed. If so, determine that there is a moving human body signal. Based on the micro-motion characteristic values ​​in the motion parameters, it is determined whether they conform to the preset micro-motion characteristics of living organisms. If so, it is determined that there is a micro-motion human body signal. If neither the moving human body signal nor the micro-movement human body signal is detected, it is determined to be a state of no movement; The step of determining whether there are target characteristics matching human body thermal radiation based on the infrared thermal radiation characteristics includes: Based on whether the temperature value of the cluster center of each temperature cluster is within the preset human body temperature threshold range, and whether the spatial distribution range of the cluster area conforms to the preset human body projection characteristics, it is determined whether there is a static target that conforms to the human body thermal radiation characteristics. If the temperature value of the cluster center of a temperature cluster is within the range of the human body temperature threshold, and the pixel area covered by the cluster region is within the range of the human body projection area, then it is determined that there is a static target that conforms to the characteristics of human body thermal radiation; if the temperature value of the cluster center of a temperature cluster is not within the range of the human body temperature threshold, then it is determined that there is no static target that conforms to the characteristics of human body thermal radiation. The determination of the human body's state of existence is based on a fusion logic judgment using the judgment results of the human body's activity light variation pattern, the target's motion state, and the target characteristics of the human body's thermal radiation. This includes: Based on the judgment results of the target motion state and the judgment results of the target characteristics of human body thermal radiation, a fusion judgment is performed to obtain a preliminary human body existence state. Based on the judgment results of the human activity light change pattern, the preliminary human existence state is verified and corrected; By using a weighting mechanism, the initial human presence status is combined with the verification and correction results to determine whether there is a person or no one.

[0015] Secondly, a multi-sensor-based human body detection system is provided, the system being used to execute the method described above. The system includes a module for acquiring ambient light data, a module for adjusting detection parameters, a module for acquiring radar data and infrared thermal radiation data, a feature acquisition module, and a module for determining the presence of a human body; wherein: The ambient light data acquisition module is used to acquire ambient light data based on the ambient light sensor. The detection parameter adjustment module is used to adjust the detection parameters of the radar sensor and the infrared sensor based on the ambient light data. The radar data and infrared thermal radiation data acquisition module is used to acquire radar data and infrared thermal radiation data through the adjusted radar sensor and infrared sensor. The feature acquisition module is used to acquire ambient light change features, target dynamic features and infrared thermal radiation features based on the ambient light data, the radar data and the infrared thermal radiation data. The module for determining the presence of a human body is used to determine the presence of a human body based on the ambient light change characteristics, the target dynamic characteristics, and the infrared thermal radiation characteristics, through multi-feature fusion judgment logic.

[0016] Thirdly, embodiments of this application provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements a multi-sensor-based human body detection method as described in the first aspect above.

[0017] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a multi-sensor-based human body detection method as described in the first aspect above.

[0018] The human body detection method and system based on multiple sensors provided in this application have at least the following technical effects.

[0019] Ambient light data collected by an ambient light sensor provides fundamental information on ambient light intensity, laying the data foundation for adaptive environmental detection. Based on this data, the detection parameters of the radar and infrared sensors are adjusted, achieving adaptive optimization of detection according to the environment. This overcomes the shortcomings of traditional sensors with fixed parameters and unstable performance in varying lighting conditions, significantly improving environmental adaptability across all time periods and scenarios, and reducing detection blind spots or false alarms caused by changes in ambient light. Through the adjusted radar and infrared sensors, radar data and infrared thermal radiation data are collected, resulting in higher-quality and more targeted data. With optimized parameters, the radar can more accurately capture human movement and subtle vital signs, while the infrared sensor can effectively focus on thermal radiation signals within the human body's temperature range. Based on ambient light data, radar data, and infrared thermal radiation data, characteristics of ambient light changes, target dynamics, and infrared thermal radiation are acquired; complex sensor signals are transformed into features characterizing key attributes of human presence, preparing for the final intelligent fusion judgment. Based on ambient light change characteristics, target dynamics, and infrared thermal radiation characteristics, a multi-feature fusion judgment logic determines the state of human presence. It effectively solves the problems of single sensors being susceptible to specific interference (such as PIR being afraid of heat sources and radar being afraid of metal shaking) and the problems of missed detection and false detection still existing in simple dual-detection logic. Ultimately, it achieves full coverage and accurate detection of dynamic, micro-movement and static human bodies, and solves the problem of how to improve the accuracy of human body detection in related technologies.

[0020] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0021] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This is a flowchart of a human body detection method based on multiple sensors; Figure 2 This is a flowchart illustrating step S102 according to an exemplary embodiment; Figure 3 This is a flowchart illustrating step S104 according to an exemplary embodiment; Figure 4 This is a flowchart illustrating step S105 according to an exemplary embodiment; Figure 5 This is a system structure block diagram of a multi-sensor-based human body detection system according to an exemplary embodiment; Figure 6 This is a structural block diagram of an electronic device according to an exemplary embodiment. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application.

[0023] Obviously, the accompanying drawings described below are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar scenarios based on these drawings without any inventive effort. Furthermore, it is understood that although the efforts made in this development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, any changes to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0024] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0025] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," and "third" used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0026] In this document, it should be understood that the terms used may be technical means used to implement part of the present invention or other summary technical terms. For example, the terms may include: Illuminance abrupt change event: refers to a rapid and significant change in ambient light intensity that occurs within an extremely short time (e.g., milliseconds) and exceeds a preset threshold. This event is usually caused by a human body moving within the sensor's field of view, blocking or reflecting ambient light, and is a key primary characteristic for determining human movement.

[0027] Illuminance fluctuation frequency: refers to the average rate at which illuminance abrupt events occur within a preset time window. This frequency characterizes the speed of light changes and can be used to distinguish between regular fluctuations caused by continuous human activities (such as walking) and irregular changes caused by other accidental disturbances. It is an important parameter for determining whether an activity is "human-like".

[0028] Illuminance variation amplitude: This refers to the difference between the maximum and minimum ambient light intensity within a preset time window. This amplitude quantifies the drasticness of the light change, helping to filter out subtle light variations caused by minor disturbances (such as small flying insects), ensuring that the system only focuses on significant light changes caused by the movement of substantial targets (such as human bodies).

[0029] Doppler frequency shift characteristic: refers to the change in the frequency of the echo signal received by the radar sensor relative to the frequency of the transmitted signal. This change is caused by the radial velocity of the moving target (such as a human body) relative to the radar, and is a core physical quantity for radar to detect the target's speed and direction of movement and determine its motion state (moving / stationary).

[0030] Human Activity Light Variation Pattern: This refers to a standardized combination of light variation features defined based on two characteristics: abrupt illuminance events and illuminance fluctuation frequency, combined with preset judgment conditions (such as abruptness threshold and frequency range). This pattern aims to characterize the reflection or occlusion effects caused by typical human activity in ambient lighting. When the real-time extracted illuminance change features match this pattern, the system can preliminarily infer the presence of human movement in the environment.

[0031] Infrared thermal radiation characteristics: These refer to the structured descriptions extracted from the raw data collected by thermopile infrared array sensors after temperature cluster analysis, used to characterize potential heat sources. They typically include parameters for one or more temperature clusters (such as the center temperature of each cluster, the area / spatial distribution of covered pixels, shape features, etc.), and serve as a direct basis for determining the existence of static or micro-dynamic heat sources that match human body temperature and size characteristics.

[0032] In a first aspect, embodiments of this application provide a human body detection method based on multiple sensors, including an ambient light sensor, a radar sensor, and an infrared sensor. Figure 1 This is a flowchart of a multi-sensor-based human body detection method, such as... Figure 1 As shown, the method includes: Step S101: Collect ambient light data based on the ambient light sensor.

[0033] Step S102: Adjust the detection parameters of the radar sensor and infrared sensor based on ambient light data.

[0034] Step S103: Collect radar data and infrared thermal radiation data using the adjusted radar sensor and infrared sensor.

[0035] Step S104: Based on ambient light data, radar data, and infrared thermal radiation data, acquire ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics.

[0036] Step S105: Based on the characteristics of ambient light change, target dynamics and infrared thermal radiation, determine the state of human presence through multi-feature fusion judgment logic.

[0037] In summary, this application provides a multi-sensor-based human body detection method. It collects ambient light data from an ambient light sensor, providing fundamental information on ambient light intensity and laying the data foundation for adaptive detection. Based on the ambient light data, the detection parameters of the radar and infrared sensors are adjusted, achieving adaptive optimization of detection according to the environment. This overcomes the shortcomings of traditional sensors with fixed parameters and unstable performance in varying lighting conditions, significantly improving environmental adaptability across all time periods and scenarios, and reducing detection blind spots or false alarms caused by changes in ambient light. Through the adjusted radar and infrared sensors, radar data and infrared thermal radiation data are collected, resulting in higher quality and more targeted data. With optimized parameters, the radar can more accurately capture human movement and subtle vital signs, while the infrared sensor can effectively focus on thermal radiation signals within the human body's temperature range. Based on ambient light data, radar data, and infrared thermal radiation data, ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics are obtained; complex sensor signals are transformed into features characterizing key attributes of human presence, preparing for final intelligent fusion judgment. Based on ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics, the presence status of the human body is determined through multi-feature fusion judgment logic. It effectively solves the problems of single sensors being susceptible to specific interference (such as PIR being afraid of heat sources and radar being afraid of metal shaking) and the problems of missed detection and false detection still existing in simple dual-detection logic. Ultimately, it achieves full coverage and accurate detection of dynamic, micro-movement and static human bodies, and solves the problem of how to improve the accuracy of human body detection in related technologies.

[0038] In one embodiment, step S101 involves acquiring ambient light data using an ambient light sensor. Specifically, this includes the following steps: The ambient light sensor uses a high-speed, high-precision digital ambient light sensor. It is selected to have near-infrared suppression capability and a spectral response that matches the human eye's visual curve. It is used to accurately collect ambient light intensity. When a person or object moves within the range of the illuminance sensor, it will affect the reflection of ambient light and affect the illuminance sampling, thus causing a sudden change in the illuminance value. It outputs an I2C signal to the main control module to provide a basis for the dynamic adjustment of the subsequent detection logic.

[0039] Specifically, the ambient light sensor can be the Texas Instruments OPT4001, which has a 28-bit effective dynamic range, a detection range of 312.5 μlux-117 klux, a spectral response that matches the human eye's visual curve, strong near-infrared suppression capability, and communicates with the main control module via an I2C interface.

[0040] By employing an ambient light sensor, stable, reliable, and accurate data reflecting the true visual lighting environment can be acquired. This not only provides high-quality raw input for subsequent analysis of ambient light variation characteristics, but more importantly, it ensures the synchronization of ambient light data with radar and infrared data. In one embodiment,Figure 2 This is a flowchart illustrating step S102 according to an exemplary embodiment, as follows: Figure 2 As shown, step S102 involves adjusting the detection parameters of the radar sensor and infrared sensor based on ambient light data. Specifically, this includes the following steps: Step S1021: Compare the light intensity in the ambient light data with the preset light intensity threshold to obtain the comparison result; Step S1022: Based on the comparison results, adjust the micro-motion detection sensitivity of the radar sensor and the trigger threshold of the infrared sensor. Specifically, this includes: If the comparison result is that the light intensity is greater than the preset light intensity threshold, a first control command is generated to improve the micro-motion detection sensitivity of the radar sensor, and a second control command is generated to reduce the trigger threshold of the infrared sensor. If the comparison result shows that the light intensity data is less than or equal to the preset light intensity threshold, a third control command is generated to reduce the micro-motion detection sensitivity of the radar sensor, and a fourth control command is generated to increase the trigger threshold of the infrared sensor.

[0041] Optionally, if the light intensity is higher than the set threshold (e.g., in daytime scenarios), the micro-motion detection sensitivity of the 24G radar module is appropriately increased and the trigger threshold of the thermopile infrared array sensor is reduced to avoid interference from strong light on the detection of the thermopile infrared array sensor; if the light intensity is lower than the set threshold (e.g., in nighttime scenarios), the detection sensitivity of the radar module is reduced (to avoid interference from small object swaying at night) and the trigger threshold of the thermopile infrared array sensor is increased to improve detection accuracy. Step S102 achieves a qualitative leap in the sensor system, transforming it from "fixed sensing" to "environmentally adaptive sensing." Under strong light, the focus is on enhancing the radar's ability to detect subtle movements of life and compensating for potential interference with infrared detection. Under weak light, the radar's sensitivity to non-living disturbances is suppressed, and the confidence level of infrared detection is improved. This fundamentally optimizes the sensor's signal-to-noise ratio in different environments, enabling it to maintain optimal detection performance under complex conditions such as day-night cycles and drastic changes in lighting. This significantly improves the system's all-weather robustness and overall anti-interference capability, providing more reliable and consistent data input for subsequent fusion and judgment.

[0042] In one embodiment, step S103 involves acquiring radar data and infrared thermal radiation data using the adjusted radar sensor and infrared sensor. Specifically, this includes the following steps: Optionally, the radar sensor uses the 24GHz band, FMCW (Frequency Modulated Continuous Wave), and a one-transmit-two-receive technology, possessing high range resolution and velocity detection capabilities. It is used to detect the movement status, distance, and micro-motion information (such as human breathing and slight movements) of targets within its range, and outputs digital or analog signals to the main control module (MCU). Specifically, the HiSilicon Hi38610 series integrated radar module is selected, with an operating frequency band of 24.0-24.25GHz, a detection range of 0.1-8m, a detection angle of 120°, and supports UART interface output of digital signals. It has a built-in micro-motion detection algorithm that can distinguish between living and non-living objects.

[0043] The infrared sensor can be a thermopile infrared array sensor, using an 8x8 pixel thermopile array, which can quickly capture the infrared signal of a stationary human body. This sensor also identifies stationary human bodies through temperature contrast and outputs an I2C signal to the main control module. Specifically, the Heiman HTPA8x8L2.1 / 0.8HiC thermopile infrared array sensor is selected, which is an 8x8 pyroelectric array sensor using n-poly / p-poly Si technology, with an I²C digital interface, a focal length of 0.8mm, a field of view of 51x51°, a maximum frame rate of 88Hz, and a power supply voltage of 3.3-3.6V.

[0044] In one embodiment, Figure 3 This is a flowchart illustrating step S104 according to an exemplary embodiment, as follows: Figure 3 As shown, step S104 involves acquiring ambient light variation characteristics, target dynamic characteristics, and infrared thermal radiation characteristics based on ambient light data, radar data, and infrared thermal radiation data. This specifically includes the following steps: Step S1041: Perform mean filtering on the ambient light data to obtain mean-filtered ambient light data; based on the mean-filtered ambient light data, obtain the ambient light change characteristics; specifically including: Based on the filtered ambient light data, the difference between the current sample value and the previous sample value is calculated to obtain the instantaneous change in illuminance; The absolute value of the instantaneous change in illuminance is compared with a preset abrupt change threshold to determine whether an illuminance abrupt change event has occurred. Based on the illuminance abrupt change events that occur within a preset time window, the number of abrupt changes per unit time is calculated to obtain the illuminance fluctuation frequency. Based on the filtered ambient light data within a preset time window, the maximum and minimum values ​​are determined, and the amplitude of illuminance change is calculated based on the difference between the maximum and minimum values. The instantaneous change in illuminance, the determination results of illuminance abrupt events, the frequency of illuminance fluctuations, and the amplitude of illuminance changes are used as characteristics of ambient light changes.

[0045] Optionally, the raw ambient light data stream collected by the ambient light sensor is first subjected to mean filtering, for example, by taking the arithmetic mean of five consecutive sampling periods (each with a period of 100ms) to filter out instantaneous electrical noise or light flicker. Then, the algorithm performs calculations based on this filtered, smoothed illuminance data sequence: first, the numerical difference between two consecutive sampling points is calculated as the millisecond-level instantaneous change in illuminance; the absolute value of this change is compared with a preset abrupt change threshold set based on the amplitude of reflected light changes caused by typical human movement; if the threshold is exceeded, an illuminance abrupt change event is marked; the system counts the number of such events within a configurable preset time window (e.g., 2 seconds) and divides by the window duration to calculate the illuminance fluctuation frequency (unit: Hz); simultaneously, the algorithm records the maximum and minimum illuminance values ​​within this time window, and the difference is the illuminance change amplitude. Finally, the system packages this set of quantitative features—including instantaneous change, abrupt change events (Boolean values), fluctuation frequency (numerical values), and change amplitude (numerical values)—into a structured ambient light change feature output.

[0046] The beneficial effects are as follows: mean filtering effectively eliminates random interference and ensures the stability of features; mutation detection can quickly respond to changes in light occlusion or reflection caused by human movement; fluctuation frequency analysis helps distinguish between regular human activities (such as walking) and irregular interference (such as wind blowing curtains); amplitude calculation can filter out minute ambient light fluctuations. The feature set extracted in this way provides a multi-dimensional and quantitative basis for determining whether "light changes are caused by human movement," significantly improving the anti-interference capability of the ambient light dimension and the accuracy of representing real human activities. Step S1042: Perform clutter suppression processing on the radar data to obtain clutter-suppressed radar data; based on the clutter-suppressed radar data, obtain the target dynamic characteristics; specifically including: Differential processing is performed on the dual-channel echo signals of the processed radar data to obtain differential signals; Based on the differential signal, the Doppler frequency shift features of the target are extracted; Calculate the target's motion parameters based on the Doppler frequency shift characteristics; The motion parameters of the target are used as the target's dynamic characteristics.

[0047] Optionally, the raw digital signal reported by the 24G dual-transmitter radar sensor (such as the HiSilicon Hi38610) via the UART interface is first subjected to clutter suppression processing, for example, using a median filtering algorithm to filter out static or quasi-static echoes generated by environmental background (such as stationary walls) and fixed electromagnetic interference. Subsequently, considering the unique dual-receiver channel architecture of this radar sensor, differential processing is performed on the suppressed echo signals received by the two channels, i.e., calculating the difference between the two channel signals. This step effectively cancels common-mode interference and enhances the signal changes caused by the asymmetric motion of the target (especially the human body). Based on the obtained differential signal, the Doppler frequency shift characteristics of the target are extracted using frequency domain analysis methods such as Fast Fourier Transform (FFT). This frequency shift value directly corresponds to the radial velocity of the target. Further, the target's motion parameters are calculated based on the Doppler frequency shift characteristics, typically including the target's instantaneous velocity, direction of motion, and micro-motion characteristics of living organisms (such as specific low-frequency modulation caused by respiration) identified through time-frequency analysis of the Doppler signal (such as observing its periodicity or amplitude characteristics). Ultimately, this set of motion parameters, encompassing velocity, direction, and micro-motion characteristics, is encapsulated into a structured target dynamic characteristic output. Specifically, the "one transmit, two receive" configuration refers to: One transmit antenna (TX) is used to continuously or intermittently radiate electromagnetic wave signals of a specific frequency (e.g., 24 GHz frequency-modulated continuous wave) into space. Two receive antennas (RX1 and RX2) are spatially separated at a specific distance (typically a fraction of the wavelength) to simultaneously receive the echo signals reflected from the target.

[0048] The beneficial effects are that it can not only reliably detect human movement, but also sensitively capture micro-motion information related to vital signs such as breathing and slight movements, thus achieving full coverage detection of both "dynamic human bodies" and "static but alive human bodies." This greatly compensates for the deficiency of traditional PIR sensors in detecting stationary human bodies, and through precise speed and feature analysis, it effectively filters out interference caused by non-living irregular movements such as fan blade rotation and air conditioner vibration. Step S1043: Perform edge detection and noise filtering on the infrared thermal radiation data to obtain edge-detected and noise-filtered infrared thermal radiation data; based on the processed infrared thermal radiation data, obtain the infrared thermal radiation features. Specifically, this includes: Based on the processed infrared thermal radiation data, a temperature distribution matrix is ​​constructed, which contains the temperature value of each pixel in the array. Temperature clustering analysis is performed on the temperature distribution matrix. Spatially adjacent pixels with temperature differences within a preset range are grouped into a temperature cluster to obtain at least one temperature cluster. Each temperature cluster includes the temperature value of the cluster center and the spatial distribution range of the cluster region. Temperature clustering is used as a feature of infrared thermal radiation.

[0049] Optionally, the raw infrared thermal radiation data from the thermopile infrared array sensor is first processed with edge detection and noise filtering. For example, Gaussian filtering is used to suppress random thermal noise, and operators such as Sobel or Canny are used to detect the edge contours of the temperature field to initially distinguish the boundaries of different heat sources. After processing, the system constructs a temperature distribution matrix containing 64 temperature values ​​based on the readings of the 8x8 pixel array, where each element corresponds to the calibrated temperature value of a pixel. Subsequently, the algorithm performs temperature clustering analysis on the matrix: it traverses all pixels, and if two pixels are spatially adjacent (using 4-connectivity or 8-connectivity criteria) and their temperature difference is less than a preset temperature similarity threshold, they are grouped into the same temperature cluster. This process is iterated until all pixels are grouped, ultimately generating one or more independent thermal regions. For each cluster, the system calculates the average temperature of all its member pixels as the cluster center temperature value and statistically analyzes the pixel coordinate range covered by the cluster to determine its spatial distribution range (such as minimum bounding rectangle, area, aspect ratio, and other geometric features). Ultimately, these clustered objects (containing information about their central temperature and spatial extent) are structured and packaged as infrared thermal radiation features output.

[0050] The beneficial effects are as follows: by subsequently analyzing whether the center temperature of each cluster falls within the range of human body temperature, and whether its spatial distribution range (area, shape) conforms to the projection characteristics of the human body in the sensor's field of view, the system can achieve high-precision and high-specificity identification of stationary or slightly moving human bodies. This fundamentally solves the problem of traditional single infrared sensors being susceptible to interference from non-human heat sources, significantly improving detection accuracy and anti-interference capabilities in complex thermal environments.

[0051] In one embodiment, Figure 4 This is a flowchart illustrating step S105 according to an exemplary embodiment, as follows: Figure 4 As shown, step S105 involves determining the presence of a human body based on ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics using multi-feature fusion logic. This specifically includes the following steps: Step S1051: Determine whether the ambient light change characteristics match the light change pattern of human activity; specifically including: Based on the illuminance abrupt events and illuminance fluctuation frequency in the characteristics of ambient light changes, determine whether the current ambient light change pattern conforms to the preset human activity light change pattern. If the light pattern matches the human activity pattern, it is determined that there are people in the scene. If the light variation pattern does not conform to human activity and the amplitude of the illuminance change is lower than the threshold, it is determined to be an unmanned state.

[0052] Optionally, within a unit of time (e.g., 1-2 seconds), the number of illuminance abrupt events must exceed a minimum threshold (e.g., ≥2 times), and the calculated illuminance fluctuation frequency must fall within the frequency range corresponding to typical human activities (e.g., walking, waving) (e.g., 0.5 Hz to 3 Hz). The system compares real-time characteristics with this pattern: if both the number of illuminance abrupt events and the fluctuation frequency meet preset conditions, the current light change is determined to conform to the human activity pattern, generating a preliminary judgment of "personnel presence"; if either condition is not met, and the illuminance change amplitude is simultaneously detected to be continuously below a set threshold (indicating only weak, stable ambient light fluctuations), the current light change is determined to be without human activity, generating a preliminary judgment of "no human presence". This judgment process is performed periodically, forming a dynamic state estimation based on light changes. It can effectively identify light intensity changes with specific rhythms and amplitudes caused by human movement, while filtering out irregular or low-frequency / small-amplitude interferences caused by cloud drift, light flickering, or changes in the brightness of electrical appliance screens.

[0053] Step S1052: Determine the target's motion state based on its dynamic characteristics; specifically including: Based on the target speed value in the motion parameters, determine whether it is within the preset range of human activity speed. If so, determine that there is a moving human signal. Based on the micro-motion characteristic values ​​in the motion parameters, it is determined whether they conform to the preset micro-motion characteristics of living organisms. If so, it is determined that there is a micro-motion human body signal. If no moving or micro-moving human body signals are detected, the state is determined to be without movement. Optionally, firstly, the target velocity value in the parameters is read and compared with a preset range of human activity velocity based on human kinematics statistics (e.g., the macroscopic movement velocity range is set to 0.1 m / s to 3 m / s). If the velocity value falls within this range, a moving human signal is determined to exist. Simultaneously, the system analyzes the micro-motion characteristic values ​​in the parameters and matches them with a preset database of vital micro-motion characteristics (e.g., typical frequencies and modulation patterns representing breathing and heartbeat). If the match is successful, a micro-motion human signal is determined to exist. If no signal is triggered in either of the above two determinations, the final determination is a state of no movement.

[0054] Step S1053: Determine whether there are target characteristics that match human body thermal radiation based on infrared thermal radiation characteristics; specifically including: Based on whether the temperature value of the cluster center of each temperature cluster is within the preset human body temperature threshold range, and whether the spatial distribution range of the cluster area conforms to the preset human body projection characteristics, it is determined whether there is a static target that conforms to the human body thermal radiation characteristics. If the temperature value of the cluster center of a temperature cluster is within the human body temperature threshold range, and the pixel area covered by the cluster region is within the human body projection area, then it is determined that there is a static target that conforms to the characteristics of human body thermal radiation; if the temperature value of the cluster center of a temperature cluster is not within the human body temperature threshold range, then it is determined that there is no static target that conforms to the characteristics of human body thermal radiation. Optionally, one or more temperature clusters obtained from temperature clustering analysis are traversed. For each cluster, the cluster center temperature value (i.e., the average temperature of all pixels within the cluster) is first read and compared with a pre-stored human body temperature threshold range (e.g., set to 30°C to 38°C to cover surface temperature changes while clothed). The system then calculates the pixel area covered by the cluster (i.e., the total number of pixels belonging to the cluster) and determines whether this area falls within a preset human projection area range (this range is calculated based on the sensor installation height, viewing angle, and typical adult / child torso projection size; for example, it may correspond to 5 to 50 pixels). Only when a temperature cluster simultaneously satisfies both the center temperature within the human body temperature threshold range and the pixel area within the human projection area range is the system determined that the cluster represents a static target conforming to human thermal radiation characteristics. Conversely, if the center temperature of a cluster exceeds the human range (e.g., close to room temperature or too high), it is directly determined to be a non-human heat source regardless of its area size. The system ultimately outputs a comprehensive judgment based on all cluster analyses, indicating whether at least one such static human target exists. It can reliably and accurately identify truly stationary or minimally moving human bodies in complex thermal environments.

[0055] Step S1054: Based on the judgment results of the human body activity light variation pattern, the target motion state, and the target characteristics of human body thermal radiation, a fusion logic judgment is performed to determine the state of human existence. Specifically, this includes: Based on the judgment results of the target's motion state and the judgment results of the target's thermal radiation characteristics, a fusion judgment is made to obtain the preliminary state of the human body's existence. Based on the judgment results of the light change pattern of human activity, the preliminary state of human existence is verified and corrected. By using a weighting mechanism, the initial state of human presence is combined with the results of verification and correction to determine whether there is a human presence or no human presence.

[0056] Optionally, in the fusion logic judgment process, firstly, the main control module performs a preliminary fusion based on the target motion state judgment result output by the radar module (including whether the target's movement speed is within the range of human activity of 0.1-3m / s, and whether there are micro-movement characteristics of living beings such as breathing) and the human body thermal radiation characteristic judgment result output by the thermopile infrared array sensor (such as the temperature distribution extracted by the 8x8 pixel array, and whether the cluster center matches the human body temperature range): when the radar confirms the presence of human movement or micro-movement signals, and the infrared array synchronously detects a static target that matches the human body thermal radiation characteristics, a preliminary "occupied" state judgment is generated; otherwise, it is marked as "unoccupied" tendency. Secondly, the main control module incorporates the light variation pattern judgment results of human activity collected by a high-speed, high-precision digital ambient light sensor (such as the amplitude and frequency of abrupt changes in illuminance data within a millisecond time window) to verify and correct the preliminary judgment. For example, in a scenario initially judged as "occupied," if the light variation pattern shows no abrupt changes or the abrupt changes are unrelated to human movement, it may be corrected to an interference signal. Conversely, if the preliminary judgment is "unoccupied" but the light variation pattern continues to show abnormal fluctuations, a re-detection is triggered. Finally, through an embedded weight allocation mechanism (such as assigning higher weight to radar micro-motion detection in high-light daytime environments, increasing the weight of the infrared array in low-light nighttime environments, and setting dynamic verification weights for illuminance data), the preliminary judgment results and the verification and correction results are weighted and fused to calculate, ultimately outputting a high-confidence "occupied" or "unoccupied" status signal.

[0057] Its beneficial effects are as follows: It effectively overcomes the limitations of a single sensor in complex scenarios: the initial fusion of radar and infrared data ensures full coverage recognition of both dynamic and static human bodies, while the introduction of light-varying modes increases the filtering capability against environmental interference (such as shaking or changes in light caused by non-human bodies). The weight allocation mechanism further optimizes the detection sensitivity under different environmental conditions, thereby reducing the false alarm rate (such as avoiding false triggers caused by pet activity, air conditioner vibration, etc.) while maintaining high sensitivity to real human activities, ultimately enhancing the reliability and user experience of smart home linkage control.

[0058] In one embodiment, when it is determined that the state of the human body has changed, a state change information is sent to the smart home system; the smart home system executes scene linkage control corresponding to the state of the human body based on the received state change information.

[0059] Optionally, when the current human presence status ("occupied" or "unoccupied") changes relative to the previously reported status (e.g., from "unoccupied" to "occupied," or vice versa) determined by multi-feature fusion logic, the smart home system gateway, upon receiving this information, automatically triggers device control commands corresponding to the human presence status based on pre-configured scene linkage rules. For example, when the status changes to "occupied," the system automatically executes a sequence of commands such as "turn on the main living room light and adjust the air conditioner to 26°C"; when the status changes to "unoccupied" and this status persists for more than a set delay (e.g., 30 seconds), it automatically executes a sequence of commands such as "turn off the lights, turn off the air conditioner, and arm the security camera." This achieves a seamless closed loop from "precise perception" to "intelligent execution," truly transforming sensor technology into a valuable user experience.

[0060] In one embodiment, the operating mode of the sensors is dynamically adjusted according to the detection results of each sensor: when the 24G radar detects human movement / micro-movement, the thermopile infrared array sensor is controlled to enter the high-frequency sampling mode; when the radar does not detect a movement signal, the thermopile infrared array sensor is controlled to enter the low-frequency inspection mode, while maintaining the low-power sampling of the high-precision illuminance sensor, thereby achieving a balance between detection accuracy and power consumption.

[0061] In summary, the human body detection system based on multiple sensors provided in this application provides basic information on ambient light intensity by collecting ambient light data from an ambient light sensor, laying the data foundation for adaptive detection. Based on the ambient light data, the detection parameters of the radar and infrared sensors are adjusted, achieving adaptive optimization of detection according to the environment. This overcomes the shortcomings of traditional sensors with fixed parameters and unstable performance in varying lighting environments, significantly improving environmental adaptability across all time periods and scenarios, and reducing detection blind spots or false alarms caused by changes in ambient light. Through the adjusted radar and infrared sensors, radar data and infrared thermal radiation data are collected, resulting in higher quality and more targeted data. With optimized parameters, the radar can more accurately capture human movement and subtle vital signs, while the infrared sensor can effectively focus on thermal radiation signals within the human body's temperature range. Based on ambient light data, radar data, and infrared thermal radiation data, ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics are obtained; complex sensor signals are transformed into features characterizing key attributes of human presence, preparing for final intelligent fusion judgment. Based on ambient light change characteristics, target dynamic characteristics, and infrared thermal radiation characteristics, the presence status of the human body is determined through multi-feature fusion judgment logic. It effectively solves the problems of single sensors being susceptible to specific interference (such as PIR being afraid of heat sources and radar being afraid of metal shaking) and the problems of missed detection and false detection still existing in simple dual-detection logic. Ultimately, it achieves full coverage and accurate detection of dynamic, micro-movement and static human bodies, and solves the problem of how to improve the accuracy of human body detection in related technologies.

[0062] The embodiments of this application also have the following effects: 1. Strong anti-interference capability and low false alarm rate: This application adopts a three-technology fusion detection scheme of "ambient light + 24G radar + display infrared". Through multi-dimensional data cross-verification, it can effectively filter false triggers caused by single factors such as object shaking, air conditioner vibration, pet activity, and strong light interference. Compared with traditional single or dual-technology sensors, the detection accuracy is greatly improved. 2. Wide environmental adaptability: By collecting light data through a high-speed, high-precision ambient light sensor, the detection threshold and sensitivity of the radar and thermopile infrared array sensors are dynamically adjusted, so that the sensor can maintain stable detection performance in different environments such as strong light during the day, weak light at night, high temperature, and low temperature, avoiding detection failure caused by environmental factors. 3. Comprehensive detection: The 24G radar module can detect subtle human movements (such as sitting still or breathing) and movement status, making up for the temperature sensitivity of thermopile infrared array sensors; the thermopile infrared array sensor is sensitive to human infrared radiation and can quickly respond to human movement. The combination of the two achieves full coverage detection of both moving and static human bodies. 4. Adaptable to smart home scenarios: Through standardized communication modules, it can link with smart home systems and automatically trigger lighting, air conditioning, security and other scenarios according to whether someone is present or not, thereby improving the level of home intelligence and user experience; at the same time, the low power consumption design can be adapted to battery power supply, reducing installation and wiring costs and making it more applicable (such as corridors, studies and other scenarios without power interfaces).

[0063] Secondly, embodiments of this application provide a human body detection system based on multiple sensors. Figure 5 This is a system structure block diagram of a multi-sensor-based human body detection system according to an exemplary embodiment, such as... Figure 5 As shown, the system includes an ambient light data acquisition module 510, a detection parameter adjustment module 520, a radar data and infrared thermal radiation data acquisition module 530, a feature acquisition module 540, and a human presence determination module 550; wherein: The ambient light data acquisition module 510 is used to acquire ambient light data based on the ambient light sensor. The detection parameter adjustment module 520 is used to adjust the detection parameters of the radar sensor and the infrared sensor based on ambient light data. The radar data and infrared thermal radiation data acquisition module 530 is used to acquire radar data and infrared thermal radiation data through the adjusted radar sensor and infrared sensor. The feature acquisition module 540 is used to acquire ambient light change features, target dynamic features and infrared thermal radiation features based on ambient light data, radar data and infrared thermal radiation data. The module 550 for determining the presence of a human body is used to determine the presence of a human body based on the characteristics of ambient light changes, target dynamics, and infrared thermal radiation through multi-feature fusion judgment logic.

[0064] In summary, the human body detection system based on multiple sensors provided in this application effectively solves the problems of single sensors being susceptible to specific interferences (such as PIR being susceptible to heat sources and radar being susceptible to metal vibrations) and the problems of missed detections and false detections still existing in simple dual-detection logic. It achieves full coverage and accurate detection of dynamic, micro-moving, and static human bodies, and solves the problem of how to improve the accuracy of human body detection in related technologies.

[0065] It should be noted that the human body detection system based on multiple sensors provided in this embodiment is used to implement the above-described embodiments, and details already described will not be repeated. As used above, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the above embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0066] Thirdly, embodiments of this application provide an electronic device, Figure 6 This is a block diagram illustrating an electronic device according to an exemplary embodiment. (e.g.) Figure 6 As shown, the electronic device may include a processor 81 and a memory 82 storing computer program instructions.

[0067] Specifically, the processor 81 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0068] The memory 82 may include a mass storage device for data or instructions. For example, and not limitingly, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 82 may include removable or non-removable (or fixed) media. Where appropriate, the memory 82 may be internal or external to a data processing device. In a particular embodiment, the memory 82 is non-volatile memory. In a particular embodiment, the memory 82 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), an electrically alterable read-only memory (EAROM), or flash memory, or a combination of two or more of these. Where appropriate, the RAM can be Static Random-Access Memory (SRAM) or Dynamic Random-Access Memory (DRAM). DRAM can be Fast Page Mode Dynamic Random-Access Memory (FPMDRAM), Extended Data Out Dynamic Random-Access Memory (EDODRAM), Synchronous Dynamic Random-Access Memory (SDRAM), etc.

[0069] The memory 82 can be used to store or cache various data files that need to be processed and / or communicated, as well as possible computer program instructions executed by the processor 81.

[0070] The processor 81 reads and executes computer program instructions stored in the memory 82 to implement any of the multi-sensor-based human detection methods in the above embodiments.

[0071] In one embodiment, a multi-sensor-based human body detection device may further include a communication interface 83 and a bus 80. Wherein, as... Figure 6 As shown, the processor 81, memory 82, and communication interface 83 are connected through bus 80 and complete communication with each other.

[0072] The communication interface 83 is used to enable communication between the various modules, devices, units, and / or equipment in the embodiments of this application. The communication interface 83 can also enable data communication with other components such as external devices, image / data acquisition devices, databases, external storage, and image / data processing workstations.

[0073] Bus 80 includes hardware, software, or both, that couples together components of a multi-sensor-based human detection device. Bus 80 includes, but is not limited to, at least one of the following: Data Bus, Address Bus, Control Bus, Expansion Bus, and Local Bus. For example, and not as a limitation, bus 80 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 80 may include one or more buses. Although specific buses are described and illustrated in the embodiments of this application, this application considers any suitable bus or interconnection.

[0074] Fourthly, embodiments of this application provide a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements a multi-sensor-based human body detection method provided in the first aspect.

[0075] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0076] In a possible implementation, the present invention can also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, causes the terminal device to perform steps implementing the multi-sensor-based human body detection method provided in the first aspect.

[0077] The program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0078] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0079] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A multi-sensor based human detection method, characterized by, The method comprises an ambient light sensor, a radar sensor and an infrared sensor, and comprises: According to the ambient light sensor, ambient light data is collected; Based on the ambient light data, the detection parameters of the radar sensor and the infrared sensor are adjusted; Through the adjusted radar sensor and infrared sensor, radar data and infrared thermal radiation data are collected; Based on the ambient light data, the radar data and the infrared thermal radiation data, the ambient light change feature, the target dynamic feature and the infrared thermal radiation feature are obtained; Based on the ambient light change feature, the target dynamic feature and the infrared thermal radiation feature, the human presence state is determined through multi-feature fusion judgment logic.

2. The human body detection method based on multiple sensors according to claim 1, wherein, Based on the ambient light data, the detection parameters of the radar sensor and the infrared sensor are adjusted, comprising: The light intensity in the ambient light data is compared with the preset light intensity threshold to obtain a comparison result; Based on the comparison result, the micro-motion detection sensitivity of the radar sensor is adjusted, and the trigger threshold of the infrared sensor is adjusted.

3. The human body detection method based on multiple sensors according to claim 2, characterized in that, Based on the comparison result, the micro-motion detection sensitivity of the radar sensor is adjusted, and the trigger threshold of the infrared sensor is adjusted, comprising: If the comparison result is that the light intensity is greater than the preset light intensity threshold, a first control instruction is generated to increase the micro-motion detection sensitivity of the radar sensor, and a second control instruction is generated to reduce the trigger threshold of the infrared sensor; If the comparison result is that the light intensity data is less than or equal to the preset light intensity threshold, a third control instruction is generated to reduce the micro-motion detection sensitivity of the radar sensor, and a fourth control instruction is generated to increase the trigger threshold of the infrared sensor.

4. The human body detection method based on multiple sensors according to claim 1, characterized in that, Based on the ambient light data, the radar data and the infrared thermal radiation data, the ambient light change feature, the target dynamic feature and the infrared thermal radiation feature are obtained, comprising: The ambient light data is subjected to mean filtering processing to obtain mean filtered ambient light data; based on the mean filtered ambient light data, the ambient light change feature is obtained; The radar data is subjected to clutter suppression processing to obtain clutter suppressed radar data; based on the clutter suppressed radar data, the target dynamic feature is obtained; The infrared thermal radiation data is subjected to edge detection and noise filtering processing to obtain edge detected and noise filtered infrared thermal radiation data; based on the processed infrared thermal radiation data, the infrared thermal radiation feature is obtained.

5. The human body detection method based on multiple sensors according to claim 4, characterized in that, Based on the mean filtered ambient light data, the ambient light change feature is obtained, comprising: Based on the filtered ambient light data, the difference between the current sampling value and the previous sampling value is calculated to obtain the illumination instantaneous change amount; According to the absolute value of the illumination instantaneous change amount and the preset sudden change threshold, it is determined whether an illumination sudden change event occurs; Based on the illumination sudden change events occurring within the preset time window, the number of sudden changes per unit time is calculated to obtain the illumination fluctuation frequency; determine a maximum value and a minimum value based on the filtered ambient light data within the preset time window, and calculate an illumination change amplitude based on a difference between the maximum value and the minimum value; use the illumination instantaneous change amount, the determination result of the illumination sudden change event, the illumination fluctuation frequency, and the illumination change amplitude as the ambient light change feature.

6. The human body detection method based on multiple sensors according to claim 5, characterized in that, the target dynamic feature is acquired based on the clutter-suppressed radar data, including: differential processing is performed on the dual-channel echo signal of the processed radar data to obtain a differential signal; a Doppler shift feature of the target is extracted based on the differential signal; a motion parameter of the target is calculated according to the Doppler shift feature; the motion parameter of the target is used as the target dynamic feature.

7. The human body detection method based on multiple sensors according to claim 6, characterized in that, the infrared thermal radiation feature is acquired based on the processed infrared thermal radiation data, including: a temperature distribution matrix is constructed based on the processed infrared thermal radiation data, and the temperature distribution matrix contains temperature values of each pixel point in the array; temperature clustering analysis is performed on the temperature distribution matrix, and pixel points that are spatially adjacent and have a temperature difference within a preset range are classified into a temperature cluster to obtain at least one temperature cluster, and each temperature cluster includes a cluster center temperature value and a spatial distribution range of a cluster region; the temperature cluster is used as the infrared thermal radiation feature.

8. The human detection method based on multi-sensor according to claim 7, characterized in that, the human body presence state is determined through multi-feature fusion judgment logic based on the ambient light change feature, the target dynamic feature, and the infrared thermal radiation feature, including: whether the ambient light change feature conforms to a human activity light change mode is determined; a target motion state is determined according to the target dynamic feature; whether the infrared thermal radiation feature conforms to a human body thermal radiation target feature is determined; a human body presence state is determined through fusion logic based on the determination result of the human activity light change mode, the determination result of the target motion state, and the determination result of the human body thermal radiation target feature.

9. The human body detection method based on multiple sensors according to any one of claims 5-8, characterized in that: whether the ambient light change feature conforms to the human activity light change mode is determined, including: whether the current ambient light change mode conforms to a preset human activity light change mode is determined according to the illumination sudden change event and the illumination fluctuation frequency in the ambient light change feature; if the human activity light change mode is met, it is determined that there is a human body; if the human activity light change mode is not met, and the illumination change amplitude is lower than a threshold value, it is determined that there is no human body; the target motion state is determined according to the target dynamic feature, including: whether the target speed value in the motion parameter is within a preset human activity speed range is determined, and if so, it is determined that there is a moving human body signal; whether the micro-motion feature value in the motion parameter conforms to a preset life micro-motion feature is determined, and if so, it is determined that there is a micro-motion human body signal; if neither the moving human body signal nor the micro-motion human body signal is detected, it is determined that there is no movement. The judging whether the target feature conforming to the human body heat radiation exists according to the infrared heat radiation feature comprises: judging whether the static target feature conforming to the human body heat radiation exists based on whether the cluster center temperature value of each temperature cluster is within the preset human body temperature threshold range and whether the spatial distribution range of the cluster region conforms to the preset human body projection feature; if the cluster center temperature value of a temperature cluster is within the human body temperature threshold range and the pixel area covered by the cluster region is within the human body projection area range, it is determined that the static target feature conforming to the human body heat radiation exists; if the cluster center temperature value of a temperature cluster is not within the human body temperature threshold range, it is determined that the static target feature conforming to the human body heat radiation does not exist; the fusion logic judgment based on the judgment result of the human body activity light variation mode, the judgment result of the target motion state and the judgment result of the target feature of the human body heat radiation to determine the human body existence state comprises: the fusion judgment based on the judgment result of the target motion state and the judgment result of the target feature of the human body heat radiation to obtain a preliminary human body existence state; the verification and correction of the preliminary human body existence state based on the judgment result of the human body activity light variation mode; the human or no human state is determined through the weight distribution mechanism based on the preliminary human body existence state and the verification and correction result.

10. A multi-sensor based human detection system, characterized by, The system is used for executing the method in any one of the above claims 1 to 9, and the system comprises an ambient light data acquisition module, an adjustment detection parameter module, a radar data and infrared heat radiation data acquisition module, a feature acquisition module and a human body existence state judgment module; wherein: the ambient light data acquisition module is used for acquiring ambient light data according to the ambient light sensor; the adjustment detection parameter module is used for adjusting the detection parameters of the radar sensor and the infrared sensor based on the ambient light data; the radar data and infrared heat radiation data acquisition module is used for acquiring radar data and infrared heat radiation data through the adjusted radar sensor and infrared sensor; the feature acquisition module is used for acquiring ambient light variation features, target dynamic features and infrared heat radiation features based on the ambient light data, the radar data and the infrared heat radiation data; the human body existence state judgment module is used for judging the human body existence state through multi-feature fusion judgment logic based on the ambient light variation features, the target dynamic features and the infrared heat radiation features.

Images