Personnel presence detection system and personnel movement detection system
By applying algorithms such as low-pass filters, Kalman filters, differentiators, and median filters to single-pixel thermal sensors, the problem of low accuracy and sharpness in personnel detection of existing two-dimensional multi-pixel thermal sensors is solved, achieving efficient and low-cost personnel presence and movement detection.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-09
- Publication Date
- 2026-03-13
AI Technical Summary
Existing two-dimensional multi-pixel thermal sensors are expensive and lack accuracy and sharpness when used for personnel presence and motion detection.
By employing a single-pixel thermal sensor combined with a low-pass filter, Kalman filter, differentiator, median filter, and convolution procedure, and by setting first and second thresholds, the temperature signal and signal change rate are processed to achieve the detection of human presence and movement.
It improves the accuracy and speed of personnel presence and movement detection while reducing hardware costs.
Smart Images
Figure CN121657153A_ABST
Abstract
Description
Technical Field
[0001] The specification discloses an object detection technology that utilizes thermal radiation, and in particular relates to a system that detects the presence or movement of people by processing temperature fluctuations. Background Technology
[0002] Current technologies for detecting the presence or movement of people employ a two-dimensional multi-pixel thermal sensor, which functions as a thermal imager or a sensor array with multiple thermal sensing elements. However, compared to traditional one-dimensional single-pixel thermal sensors, these thermal imagers are expensive for achieving the same level of accuracy and sharpness in detecting the presence or movement of people. Summary of the Invention
[0003] In view of the above-mentioned shortcomings, and with the aim of providing immediate and accurate personnel presence detection or personnel movement detection, and for applications in security and monitoring, healthcare, smart homes, retail, customer analytics, and search and rescue operations, the publication proposes a personnel presence detection system and a personnel movement detection system.
[0004] In one embodiment of the personnel presence detection system, the system includes a thermal sensor, a lens mounted on the thermal sensor, and a controller. The lens converges thermal radiation incident from a spatial region onto the thermal sensor, which senses the thermal radiation generated from the spatial region and outputs a temperature signal. The controller is coupled to the thermal sensor and sets a first threshold and a second threshold, where the first threshold is a value greater than the second threshold. The controller performs personnel presence detection based on the temperature signal, the first threshold, and the second threshold. The controller includes a low-pass filter and a Kalman filter. The controller inputs the temperature signal to the low-pass filter, which outputs a filtered temperature signal. The controller can also calculate a first difference signal by calculating the difference between the temperature signal and the filtered temperature signal. Then, the controller inputs the first difference signal to the Kalman filter, causing the Kalman filter to output a second difference signal.
[0005] Furthermore, according to one embodiment, when the second difference signal is less than the first threshold and greater than the second threshold, the controller determines that no personnel are present; when the second difference signal is greater than both the first threshold and the second threshold, the controller determines that personnel are present; and when the second difference signal is less than both the first threshold and the second threshold, the controller determines that personnel are present.
[0006] In one embodiment of the personnel movement detection system, the system includes a thermal sensor, a lens mounted on the thermal sensor, and a controller. The lens is used to converge thermal radiation emitted from a spatial region toward the thermal sensor. The thermal sensor senses the thermal radiation generated from the spatial region and outputs a temperature signal. The controller is coupled to the thermal sensor and sets a first threshold and a second threshold, wherein the first threshold is greater than the second threshold. Personnel movement detection is performed based on the temperature signal, the first threshold, and the second threshold. The controller includes a differentiator and a median filter. The controller inputs the temperature signal to the differentiator, causing the differentiator to output the rate of change of the temperature signal. The controller inputs the rate of change of the temperature signal to the median filter, causing the median filter to output the filtered rate of change of the temperature signal.
[0007] In one embodiment, when the rate of change of the filtered temperature signal is less than a first threshold and greater than a second threshold, the controller determines that no personnel are moving; when the rate of change of the filtered temperature signal is greater than both the first threshold and the second threshold, the controller determines that personnel are moving; and when the rate of change of the filtered temperature signal is less than both the first threshold and the second threshold, the controller determines that personnel are moving.
[0008] In an implementation of the personnel movement detection system, the system includes a thermal sensor, a lens mounted on the thermal sensor, and a controller. The lens is used to converge thermal radiation incident from a spatial region onto the thermal sensor, which senses the thermal radiation from the spatial region and outputs a temperature signal. The controller is coupled to the thermal sensor and sets a first threshold and a second threshold, wherein the first threshold is greater than the second threshold. The controller performs personnel movement detection based on the temperature signal, the first threshold, and the second threshold. The controller also performs a convolution procedure on the temperature signal to obtain a convolution signal.
[0009] In one embodiment, when the convolution signal is less than a first threshold and greater than a second threshold, the controller determines that no person has moved. When the convolution signal is greater than both the first and second thresholds, the controller determines that someone has moved. When the convolution signal is less than both the first and second thresholds, the controller determines that someone has moved.
[0010] Furthermore, in the convolution procedure, the controller obtains a number of samples to be convolved and performs the following steps for each sample to be convolved: the controller obtains the values of the current state and the previous state of the temperature signal, calculates the time difference between the current state and the previous state, and obtains multiple convolution values by multiplying the values of the current state, the previous state, and the time difference. The convolution signal is obtained by calculating the sum of the multiple convolution values.
[0011] Furthermore, the controller includes a low-pass filter. The controller inputs the convolutional signal to the low-pass filter, and the low-pass filter outputs a filtered convolutional signal. When the filtered convolutional signal is less than a first threshold and greater than a second threshold, the controller determines that no personnel have moved. When the filtered convolutional signal is greater than both the first and second thresholds, the controller determines that personnel have moved. When the filtered convolutional signal is less than both the first and second thresholds, the controller determines that personnel have moved.
[0012] Therefore, in personnel presence detection systems or personnel movement detection systems, the algorithm executed in the controller significantly improves the accuracy and speed of personnel presence or movement detection. The system only needs to use a single thermal sensor, such as a single-pixel thermal sensor, instead of using multiple thermal sensors. This reduces hardware costs.
[0013] To further understand the features and technical content of the present invention, please refer to the following detailed description and drawings of the present invention. However, the drawings provided are for reference and illustration only and are not intended to limit the present invention. Attached Figure Description
[0014] Figure 1A The diagram shows a block representation of a personnel presence detection system.
[0015] Figure 1B A schematic diagram of an embodiment of a thermal detection device shown in three-dimensional form;
[0016] Figure 1C A schematic diagram showing an embodiment of thermal radiation directed towards a thermal sensor from the lens aggregation space region;
[0017] Figure 2 This diagram illustrates an example of an implementation where personnel are present during the detection process.
[0018] Figure 3 A flowchart illustrating an embodiment of the personnel presence detection system's operational procedures;
[0019] Figure 4 The diagram shows an example of a testing process involving personnel.
[0020] Figure 5 Flowcharts illustrating several scenarios where personnel are detected;
[0021] Figure 6 Flowchart of other embodiments for detecting the presence of personnel;
[0022] Figure 7 Flowchart of another embodiment of the detection of personnel;
[0023] Figure 8 A block diagram showing an embodiment of the personnel movement detection system;
[0024] Figure 9 A flowchart illustrating an embodiment of the operation procedures of a personnel movement detection system;
[0025] Figure 10 The diagram shows an example of a personnel movement inspection process;
[0026] Figure 11 Flowcharts showing several implementation scenarios for personnel movement detection;
[0027] Figure 12 A flowchart illustrating an embodiment of the operation procedures of a personnel movement detection system;
[0028] Figure 13 This diagram shows another embodiment of the personnel moving inspection process;
[0029] Figure 14 The flowchart shown is an example of an implementation of the convolution procedure;
[0030] Figure 15 Flowcharts showing implementation examples of several scenarios for personnel movement detection; and
[0031] Figure 16 A flowchart showing another embodiment of several scenarios for people movement detection. Detailed Implementation
[0032] The document discloses a human presence detection system or human motion detection system implemented through a thermal detection device that combines hardware and software. One of the objectives is to improve the accuracy of human presence and motion detection with a low-cost and high-efficiency solution. This technical solution enables immediate and accurate human presence or motion detection technology applicable to applications such as security and surveillance, healthcare, smart homes, retail, consumer analytics, and search and rescue operations.
[0033] In implementations of personnel presence detection systems or personnel movement detection systems, the thermal detection device uses a thermal sensor to sense thermal radiation generated from the surrounding environment. When the sensed thermal radiation is converted into a digital signal, an algorithm is executed by a controller to process fluctuations in the digital temperature value to determine whether any personnel are present or moving in front of the thermal detection device. Therefore, once the digital temperature value reaches a detection threshold, the presence or movement of personnel is detected. Specifically, the thermal sensor of the thermal detection device can be an infrared (IR) sensor, which can be used to sense infrared radiation emitted by objects (such as the human body).
[0034] [Example 1: Personnel Presence Detection System]
[0035] refer to Figure 1A The diagram shown is a block illustration of a personnel presence detection system, with reference to... Figure 1B A schematic diagram of an embodiment of the three-dimensional thermal detection device 10 is shown, and further references are provided. Figure 1C The diagram shows an embodiment of thermal radiation directed from the lens aggregation space region toward the thermal sensor.
[0036] The figure shows an embodiment of the thermal detection device 10 proposed by the personnel presence detection system. The main components of the thermal detection device 10 include a lens 101, a thermal sensor 103 and a circuit board 105. The controller 107 is electrically connected to the circuit board 105 of the thermal detection device 10. The solution shown in the figure is that the controller 107 can be an external component or a component built into the thermal detection device 10.
[0037] Circuit board 105 implements a control circuit for controlling the operation of thermal sensor 103. Circuit board 105 is electrically connected to the controller 107 of the personnel presence detection system and can drive thermal sensor 103 to operate according to the control signal generated by controller 107. For example, it drives thermal sensor 103 to sense thermal radiation 110 radiated by an object and converts the thermal radiation 110 into a temperature signal. The lens 101 shown is mounted on thermal sensor 103. Lens 101 is implemented as a radiation converging device to converge the thermal radiation 110 incident on thermal sensor 103 from the spatial region. In this way, thermal sensor 103 can sense the converged thermal radiation 110 generated from the spatial region and output a temperature signal.
[0038] The controller 107 is used to operate the functions of the thermal detection device 10. For example, the controller 107 drives the thermal sensor 103 to sense thermal radiation 110 and generate a temperature signal. The controller 107 is used to execute algorithms to perform personnel presence detection. In an embodiment, the controller 107 sets a first threshold and a second threshold, where the first threshold is greater than the second threshold. When a temperature signal is received from the thermal detection device 10, which has a circuit board 105, via a specific connection (such as a signal line or a flexible circuit board), the controller 107 can perform personnel presence detection based on the temperature signal, the first threshold, and the second threshold.
[0039] The personnel presence detection system is equipped with a host 109, which can be used to monitor the operation of the personnel presence detection algorithm executed by the controller 107. The results of personnel presence detection are provided to the host 109. For example, the host 109 can use a display to visualize the progress and results of personnel presence detection.
[0040] Furthermore, according to Figure 1AIn the illustrated embodiment, controller 107 includes a low-pass filter 171 and a Kalman filter 173. Controller 107 inputs a temperature signal to low-pass filter 171. In one embodiment, based on the processing time of low-pass filter 171, a phase delay can be introduced into the temperature signal, and then a filtered temperature signal is output. Subsequently, controller 107 calculates the difference between the temperature signal and the filtered temperature signal to derive a first difference signal, and then inputs the first difference signal to Kalman filter 173, causing Kalman filter 173 to output a second difference signal.
[0041] According to one embodiment, when a first difference signal is received, the Kalman filter 173 performs smoothing processing on the first difference signal and outputs a smoothed version of the first difference signal, which is the second difference signal. It is worth noting that the average value of the first difference signal and the second difference signal is zero, therefore it is not necessary to obtain the ambient temperature and calculate its average value. Thus, the first difference signal and the second difference signal are values centered around zero. Furthermore, the first threshold is an upper limit value and can be a value greater than zero; the second threshold is a lower limit value and can be any other value less than zero. Based on the above description, the present invention provides the settings for the detection system. Figure 2 Next, a schematic diagram of an example scenario where the presence of an executor is detected is displayed.
[0042] The image shows as follows Figure 1A and Figure 1B The diagram shows an embodiment of the thermal detection device 10, which includes a lens 101, a thermal sensor 103, and a circuit board 105. In this embodiment, the controller 107 may be a microcontroller (MCU) located in the host 200.
[0043] According to an embodiment, the thermal sensor 103 can be an infrared sensor used to sense the thermal radiation emitted by a person 20 located in front of the thermal detection device and directed toward the sensing window of the thermal sensor 103. The thermal radiation passes through a lens 101 mounted on the thermal sensor 103 and is focused by the lens 101 before being directed toward the thermal sensor 103, so that the thermal sensor 103 can effectively sense the thermal radiation.
[0044] After circuit board 105 performs analog-to-digital conversion to convert the analog signal sensed by thermal sensor 103 into a digital temperature signal, the controller 107 receives the digital temperature signal. The controller 107 then performs personnel presence detection based on the obtained temperature signal and the first and second thresholds set therein, according to hardware settings and algorithms.
[0045] It is worth mentioning that, unlike existing two-dimensional multi-pixel thermal sensors, the thermal sensor 103 used in the thermal detection device proposed in the publication can be a one-dimensional single-pixel thermal sensor (such as an infrared sensor). This provides a low-cost solution, but based on the embodiments proposed in the publication, it still has high accuracy in personnel detection.
[0046] Figure 3 A flowchart illustrating an embodiment of the personnel presence detection system's operational procedures is shown. Figure 4 The description includes a schematic diagram of an example of a testing process.
[0047] After the thermal detection device is initialized, the thermal sensor is driven to sense the thermal radiation emitted by the object (step S301). For example, after the thermal sensor senses the thermal radiation, the circuit board of the thermal sensor receives the sensed analog signal and then performs analog-to-digital conversion to convert the analog signal into a digital signal, such as the temperature signal 401 shown in the figure (step S303). Afterwards, the circuit board of the thermal sensor outputs the temperature signal 401 to the controller (step S305).
[0048] like Figure 4 As shown, the controller includes a low-pass filter 403 and a Kalman filter 407. The controller inputs the temperature signal 401 to the low-pass filter 403 (step S307) and outputs the filtered temperature signal 404 (step S309). Next, the controller uses an operator, such as a subtractor 405, to calculate the difference between the temperature signal 401' (which is consistent with the input temperature signal 401) and the filtered temperature signal 404 to obtain a first difference signal 406 (step S311). The subtractor 405 subtracts the value of the filtered temperature signal 404 from the temperature signal 401'.
[0049] Next, the first difference signal 406 is input to the Kalman filter 407 (step S313). For example, the Kalman filter 407 is used to smooth the first difference signal 406 and output a second difference signal 409, which is the smoothed first difference signal 406 (step S315).
[0050] In some embodiments, the low-pass filter 403 may be a weighted low-pass filter. The low-pass filter 403 may introduce a phase delay into the temperature signal 401 and may perform weighted moving average (WMA) filtering. When no person is detected, an arithmetic mean of the temperature signal 401 can be calculated. This arithmetic mean of the temperature signal 401 acts as a baseline signal. From this baseline of the temperature signal 401, a variable detection threshold can be calculated to make person presence detection more stable by reducing fluctuations (e.g., filtering out system noises).
[0051] In one implementation example, the thermal sensor defines a field of view (FOV). When no person is detected within the thermal sensor's FOV, a first threshold (e.g., an upper limit) and a second threshold (e.g., a lower limit) are set parallel, and a delta offset is defined from the temperature signal 401. When the system is running or no person is detected, the DC level of the temperature signal 401 is changed based on system imperfections, and the variable detection threshold is adjusted proportionally. This variable detection threshold mechanism effectively prevents false detections of person presence and allows users to correctly set the threshold, thus improving the detection range.
[0052] For example, once a person is detected within the field of view of the thermal sensor, it indicates that the temperature signal 401 has changed relative to a fixed detection threshold. Otherwise, if no person is detected, such as when a person moves out of the field of view of the thermal sensor, the temperature signal 401 reaches a stable state, and the detection threshold is recalculated and updated to continue the next detection of the presence of a person.
[0053] In another embodiment, the thermal sensor can be implemented using an infrared sensor, which senses infrared radiation and outputs a digital temperature value after performing an analog-to-digital conversion, thus obtaining a temperature signal 401. The digital temperature value is proportional to the amount of infrared radiation.
[0054] In another embodiment, given the poor sensor accuracy and the fluctuations in the digital temperature value caused by changes in the ambient temperature, the personnel presence detection system employs a Kalman filter 407 to smooth the resulting first difference signal 406, resulting in a smoothed first difference signal 406.
[0055] As described in the above embodiment, the low-pass filter 403 introduces a phase delay into the temperature signal 401 and outputs a filtered temperature signal 404. When the first difference signal 406 is calculated, the Kalman filter 407 can be used to reduce high-frequency fluctuations caused by system imperfections and system noise.
[0056] Thus, when a person is detected within the field of view of the thermal sensor, and there is no fluctuation on the baseline representing stable person presence detection, a sharp change will occur in the second difference signal 409. Specifically, if the temperature of an indoor object is higher than the ambient temperature, the second difference signal 409 will show a sharp increase; conversely, if the temperature of an outdoor object is lower than the ambient temperature, the second difference signal 409 will show a sharp decrease, thereby enabling the detection of person presence.
[0057] Figure 5 Next, flowcharts of several scenario examples involving personnel in the inspection process are displayed.
[0058] As described above, the controller of the personnel presence detection system sets a first threshold and a second threshold, with the second threshold being smaller than the first threshold. When a second difference signal 409 is obtained, the controller compares the second difference signal 409 with the first threshold and the second threshold to perform personnel presence detection (step S501).
[0059] When the second difference signal is less than the first threshold and greater than the second threshold, meeting the first scenario, the controller determines that no personnel are present (step S503). When the second difference signal is greater than both the first and second thresholds, meeting the second scenario, the controller determines that personnel are present (step S505). When the second difference signal is less than both the first and second thresholds, meeting the third scenario, the controller determines that personnel are present (step S507).
[0060] Figure 6 This is another embodiment of the flowchart showing the detection process for the number of consecutive times that meet the second scenario.
[0061] The personnel presence detection system is set to confirm the presence of personnel in a second scenario, where the second difference signal is simultaneously greater than both the first and second thresholds.
[0062] In an example process, the controller of the personnel presence detection system counts the first consecutive number of times that the second scenario is met (step S601), and then determines whether the first consecutive number of times has reached the first preset quantity (step S603).
[0063] If the first consecutive count in the second scenario does not reach the first preset number (No), the controller can confirm that no person is detected (step S605); conversely, if the first consecutive count in the second scenario reaches the first preset number (Yes), the controller of the personnel presence detection system can confirm that a person is present (step S607).
[0064] Figure 7 A flowchart showing another embodiment of the detection process that considers the number of consecutive times that meet the third scenario.
[0065] The personnel presence detection system is configured with a second preset number of conditions to confirm personnel presence under a third scenario. The third scenario refers to the situation where the second difference signal is simultaneously less than both the first and second thresholds. In the personnel presence detection process, the controller counts the second consecutive number of times that meet the third scenario (step S701) and determines whether the second consecutive number of times reaches the second preset number (step S703).
[0066] If the second consecutive count in the third scenario does not reach the second preset number (No), it is determined that no personnel exist (step S705); otherwise, if the second consecutive count in the third scenario reaches the second preset number (Yes), it is confirmed that personnel are detected (step S707).
[0067] [Example 2: Personnel Movement Detection System]
[0068] Figure 8 A block diagram showing an embodiment of the personnel movement detection system is displayed.
[0069] The personnel movement detection system includes a thermal detection device 80 and a controller 807. The thermal detection device 80 further includes a lens 801, a thermal sensor 803, and a circuit board 805. The lens 801 is mounted on the thermal sensor 803. The controller 807 is electrically connected to the circuit board 805 of the thermal detection device 80 and is located on the connection between the thermal detection device 80 and the host computer 809. When the controller 807 executes the personnel movement detection algorithm, the host computer 809 can be used to monitor the personnel movement detection process and display the results of the personnel movement detection. For example, the display can be used to visualize the personnel movement detection process and the results of the personnel movement detection.
[0070] The controller 807 includes a differentiator 871 and a median filter 873. In one embodiment, the differentiator 871 generates an output signal proportional to the rate of change of the temperature signal received from the thermal detection device 80. Specifically, the differentiator 871 performs mathematical differentiation to output a signal that reflects the rate of change of the temperature signal.
[0071] The median filter 873 is used to smooth the rate of change of the temperature signal. The median filter replaces input entries with the median of each entry and its neighboring entries. For example, if a numerical sequence {0, 1, -1, 5, 2, 1} represents the rate of change of the temperature signal, the median filter 873 performs smoothing to output a numerical sequence {0, 0, 1, 2, 2, 1}. The entries in this output numerical sequence are smoothed to be closer to the values of the input numerical sequence. The entry value "5", which is higher than its neighboring entries "-1" and "2", is filtered by the median filter 873 and output as entry value "2" because value "2" is the median of the numerical sequences "-1", "5", and "2". Therefore, the controller 807 performs personnel movement detection based on the temperature signal and some prediction thresholds using the output value of the median filter 873.
[0072] Figure 9 Display the flowchart of the personnel movement detection system, and refer to... Figure 10 The diagram shown illustrates an example of a personnel movement detection system in operation.
[0073] exist Figure 9 In the illustrated embodiment, the thermal sensor 803 initially senses thermal radiation incident from the spatial region, and the lens 801 shown is used to focus the thermal radiation incident on the thermal sensor 803 (step S901). The thermal detection device 80 then outputs a temperature signal (step S903).
[0074] according to Figure 10 The schematic diagram of the system embodiment shown illustrates that the controller 807 inputs a temperature signal 111 to the differentiator 113 (step S905) to output a temperature signal change rate 114 (step S907). Then, the controller 807 inputs the temperature signal change rate 114 to the median filter 115 (step S909). One purpose of the median filter 115 is to remove high-frequency noise and rapid fluctuations from the signal. Through the operation of the median filter 115, the controller 807 outputs a filtered temperature signal change rate 117 (step S911).
[0075] The average of the rate of change of the temperature signal 114 and the rate of change of the filtered temperature signal 117 is zero. Therefore, it is not necessary to obtain the ambient temperature and calculate the average value. Specifically, the rate of change of the temperature signal is a value centered around zero.
[0076] For the purpose of personnel movement detection, according to an embodiment, the controller sets a first threshold (such as an upper limit) and a second threshold (such as a lower limit), i.e., the first threshold is greater than the second threshold. Furthermore, the first threshold can be a value greater than zero, while the second threshold can be a value less than zero. The controller can then perform personnel movement detection based on the temperature signal, the first threshold, and the second threshold.
[0077] Figure 11 Flowcharts showing several implementation scenarios for personnel movement detection.
[0078] The controller compares the rate of change of the filtered temperature signal with the first threshold and the second threshold to detect personnel movement (step S111). When the rate of change of the filtered temperature signal is smaller than the first threshold but larger than the second threshold, which meets the first scenario, the controller determines that no personnel are moving (step S113). When the rate of change of the filtered temperature signal is larger than both the first and second thresholds, which meets the second scenario, the controller determines that personnel are moving (step S115). When the rate of change of the filtered temperature signal is smaller than both the first and second thresholds, which meets the third scenario, the controller determines that personnel are moving (step S117).
[0079] The system calculates the change between the current temperature signal and the previous temperature signal in real time, and also calculates the time difference between the current and previous states in real time. Then, it calculates the derivative of the temperature signal using Equation 1 to measure the rate of change of the temperature signal. The function "Obj Derivative" refers to the rate of change of the temperature signal, the variable "ΔObj ADC" refers to the change of the temperature signal between the current and previous states, and the variable "Δtimestamp" refers to the time change.
[0080] Equation 1:
[0081]
[0082] The rate of change of the temperature signal is further filtered by a median filter (such as...). Figure 10 The displayed median filter (115) removes high-frequency noise and rapid signal fluctuations. When no person is detected within the field of view of the thermal sensor, the amplitude of the filtered "Obj Derivative" count signal approaches zero, and a user-defined threshold is set to achieve more accurate detection of distant person movement.
[0083] When someone passes through the thermal sensor's field of view, if the person approaches the sensor's field of view, the rate of change of the temperature signal, "Obj Derivative," increases and exceeds the upper limit, thus detecting movement. Conversely, if the person leaves the thermal sensor's field of view, the rate of change of the temperature signal, "Obj Derivative," decreases and exceeds the lower limit, also detecting movement.
[0084] [Example 3: Personnel Movement Detection System]
[0085] The personnel movement detection system performs personnel movement detection by processing the temperature signal change rate 114. The controller 807 calculates convolution, derivative and performs filtering on the temperature signal to generate a filtered and convoluted object temperature signal. Similarly, the derivative of the filtered temperature signal is used to compare with an adjustable detection threshold to activate personnel movement detection and display the detection results.
[0086] Figure 12 A flowchart illustrating an embodiment of the operation process of the convolution program in a personnel movement detection system is shown, along with reference to... Figure 13 The diagram shows an example of a personnel movement detection process.
[0087] The thermal detection device of the personnel movement detection system is used to sense thermal radiation (step S121) and output a temperature signal 131 (step S123). The temperature signal 131 can be a digital signal obtained by converting the sensed thermal radiation value.
[0088] To ensure a more stable personnel movement detection method, the controller coupled to the thermal sensor performs a convolution procedure 133 (step S125) on the temperature signal 131, and then outputs convolution signals 134 (step S127). The convolution procedure 133 can be found in [reference needed]. Figure 14 The process implementation diagram is shown below.
[0089] Subsequently, the controller inputs the convolution signal 134 to the low-pass filter 135 (step S129). In another embodiment, the low-pass filter 135 may be a weighted moving average filter (WMA). The weighted moving average filter assigns higher weights to the most recent signal to eliminate fluctuations in the input signal. The input signal is the convolution signal, and the controller then outputs the filtered convolution signal 137 (step S131).
[0090] It is worth mentioning that the average value of the convolution signal 134 and the filtered convolution signal 137 is zero, so the personnel movement detection system does not need to obtain the surrounding temperature and calculate the average value.
[0091] To perform personnel movement detection, the controller is coupled to a thermal sensor. The controller sets a first threshold and a second threshold, where the first threshold can be an upper limit value greater than zero, and the second threshold can be a lower limit value less than zero. Therefore, the first threshold is greater than the second threshold. In this way, the controller performs personnel movement detection based on the temperature signal 131 and the first and second thresholds.
[0092] The temperature signal is convolved with the previous state to generate a convolution signal. Then, the convolution signal is processed by a weighted moving average to reduce signal fluctuations. When no personnel are detected, the convolution signal can be a near-zero amplitude signal and becomes the baseline. The convolution signal can be calculated using Equation 2.
[0093] Equation 2:
[0094]
[0095] In Equation 2, the variable "Obj ADC(t)" represents the current state of the temperature signal at time "t", and the variable "previous Obj ADC(i)" represents the previous state of the temperature signal at time "i". The symbol "Δt" represents the time difference (delta timestamp), and "N" represents the number of samples after convolution.
[0096] The convolution signal can be the sum of the product of the current state and the previous state of the temperature signal, and it can be a time-weighted signal within the sample range. When human movement is detected within the field of view of the thermal sensor, the convolution program will show that there are high fluctuations in the convolution signal.
[0097] Figure 14 A flowchart of an embodiment of the above convolution procedure 133 is shown.
[0098] In the convolution procedure 133 executed by the personnel movement detection system, a number of samples are first obtained (step S141). The controller of the personnel movement detection system executes the following steps of the convolution procedure: first, it receives temperature signal values of the current state and the previous state from the thermal detection device (step S143); the controller calculates the time difference between the current state and the previous state (step S145); and by multiplying the current state, the previous state, and the time by interpolation, it obtains the convolution value (step S147). After repeating the above steps, the sum of multiple convolution values is calculated to obtain the object temperature signal after convolution (step S149).
[0099] Figure 15 A flowchart showing another embodiment of several scenarios for personnel movement detection based on convolutional signals, a first threshold, and a second threshold.
[0100] In the personnel movement detection system, the controller compares the convolutional signal with a first threshold and a second threshold (step S151). Specifically, if the convolutional signal is less than the first threshold but greater than the second threshold, meeting the first scenario, the controller determines that no personnel are moving (step S153). If the convolutional signal is greater than both the first and second thresholds, meeting the second scenario, the controller determines that personnel are moving (step S155). If the convolutional signal is less than both the first and second thresholds, meeting the third scenario, the controller determines that personnel are moving (step S157).
[0101] In another embodiment, the controller inputs the convolution signal to a low-pass filter to output a filtered convolution signal, wherein relevant process embodiments can be found in the embodiments. Figure 16 The flowcharts show several scenario implementation examples of personnel movement detection based on filtered convolutional signals, a first threshold, and a second threshold.
[0102] The controller first compares the filtered convolutional signal with the first threshold and the second threshold (step S161). If the filtered convolutional signal is less than the first threshold but greater than the second threshold, meeting the first scenario, the controller determines that no person has moved (step S163). If the filtered convolutional signal is greater than both the first and second thresholds, meeting the second scenario, the controller determines that someone has moved (step S165). If the filtered convolutional signal is less than both the first and second thresholds, meeting the third scenario, the controller determines that someone has moved (step S167).
[0103] Furthermore, according to some embodiments, the controller is used to count a third consecutive number of times the convolution signal is simultaneously greater than the first threshold and the second threshold, wherein when the third consecutive number reaches a third preset number, the controller determines that there is personnel movement.
[0104] Furthermore, the controller counts the fourth consecutive number of times the convolution signal is simultaneously less than the first threshold and the second threshold, wherein when the fourth consecutive number reaches a fourth preset number, the controller determines that there is personnel movement.
[0105] In summary, based on the personnel presence detection system and personnel movement detection system described in the above embodiments, a thermal sensor component is proposed to realize a personnel presence detector or a personnel movement detector. The system can effectively maximize the accuracy and speed of personnel presence detection and personnel movement detection, and the cost of the hardware system can be reduced through the algorithm executed by the controller and the hardware design with thermal sensors.
[0106] The content disclosed above is only a preferred and feasible embodiment of this application, and is not intended to limit the scope of the patent application. Therefore, all equivalent technical changes made based on the content of this application specification and drawings are included in the scope of the patent application.
Claims
1. A personnel presence detection system, characterized in that... The system includes: A thermal sensor; A lens mounted on the thermal sensor is used to converge thermal radiation incident on the thermal sensor from a spatial region, wherein the thermal sensor senses the thermal radiation from the spatial region and outputs a temperature signal; and A controller, coupled to the thermal sensor, is used to set a first threshold and a second threshold, wherein the first threshold is greater than the second threshold, and to perform personnel presence detection based on the temperature signal, the first threshold, and the second threshold; The controller includes a low-pass filter and a Kalman filter. The controller inputs the temperature signal to the low-pass filter, so that the low-pass filter outputs a filtered temperature signal. The controller further calculates a difference between the temperature signal and the filtered temperature signal to obtain a first difference signal, and inputs the first difference signal to the Kalman filter so that the Kalman filter outputs a second difference signal.
2. The personnel presence detection system according to claim 1, characterized in that: When the second difference signal is less than the first threshold or greater than the second threshold, the controller determines that no personnel are present. When the second difference signal is simultaneously greater than both the first threshold and the second threshold, the controller determines that a person is present; and When the second difference signal is simultaneously less than both the first threshold and the second threshold, the controller determines that a person is present.
3. The personnel presence detection system according to claim 1, characterized in that... The low-pass filter is a weighted low-pass filter.
4. The personnel presence detection system according to claim 1, characterized in that... The first threshold is greater than zero and the second threshold is less than zero.
5. The personnel presence detection system according to claim 1, characterized in that... The controller counts a first consecutive number of times the second difference signal is simultaneously greater than the first threshold and the second threshold, wherein when the first consecutive number reaches a first preset number, the controller determines that a person is present.
6. The personnel presence detection system according to claim 1, characterized in that... The controller counts a second consecutive number of times the second difference signal is simultaneously less than the first threshold and the second threshold, wherein when the second consecutive number reaches a second preset number, the controller determines that a person is present.
7. A personnel movement detection system, characterized in that... The system includes: A thermal sensor; A lens mounted on the thermal sensor is used to converge thermal radiation incident on the thermal sensor from a spatial region, wherein the thermal sensor senses the thermal radiation from the spatial region and outputs a temperature signal; and A controller, coupled to the thermal sensor, is used to set a first threshold and a second threshold, wherein the first threshold is greater than the second threshold, and to perform personnel movement detection based on the temperature signal, the first threshold, and the second threshold; The controller includes a differentiator and a median filter. The controller inputs the temperature signal to the differentiator, causing the differentiator to output a rate of change of the temperature signal. The controller inputs the rate of change of the temperature signal to the median filter, so that the median filter outputs a filtered rate of change of the temperature signal.
8. The personnel movement detection system according to claim 7, characterized in that: When the rate of change of the filtered temperature signal is less than the first threshold and greater than the second threshold, the controller determines that no personnel are moving. When the rate of change of the filtered temperature signal is simultaneously greater than both the first threshold and the second threshold, the controller determines that there is personnel movement; and When the rate of change of the filtered temperature signal is less than both the first threshold and the second threshold, the controller determines that there is personnel movement.
9. The personnel movement detection system according to claim 7, characterized in that... The first threshold is greater than zero and the second threshold is less than zero.
10. A personnel movement detection system, characterized in that... The system includes: A thermal sensor; A lens mounted on the thermal sensor is used to converge thermal radiation incident on the thermal sensor from a spatial region, wherein the thermal sensor senses the thermal radiation from the spatial region and outputs a temperature signal; and A controller, coupled to the thermal sensor, is used to set a first threshold and a second threshold, wherein the first threshold is greater than the second threshold, and to perform personnel movement detection based on the temperature signal, the first threshold, and the second threshold; The controller performs a convolution procedure on the temperature signal to obtain a convolution signal.
11. The personnel movement detection system according to claim 10, characterized in that: When the convolution signal is less than the first threshold and greater than the second threshold, the controller determines that no personnel are moving. When the convolutional signal is simultaneously greater than both the first threshold and the second threshold, the controller determines that there is personnel movement; and When the convolutional signal is simultaneously less than both the first threshold and the second threshold, the controller determines that there is personnel movement.
12. The personnel movement detection system according to claim 10, characterized in that... The convolution procedure includes: Obtain a number of samples to perform convolution; The following steps are performed on each sample to be convolved, including: Obtain the values of the current state and the previous state of the temperature signal; Calculate the time difference between the current state and the previous state; Multiply the current state value, the previous state value, and the time difference to obtain multiple convolution values; and The sum of these multiple convolution values is obtained to derive the convolution signal.
13. The personnel movement detection system according to claim 10, characterized in that... The controller includes a low-pass filter. The controller inputs the convolution signal to the low-pass filter, causing the low-pass filter to output a filtered convolution signal, wherein: When the filtered convolutional signal is less than the first threshold and greater than the second threshold, the controller determines that no personnel are moving. When the filtered convolutional signal is simultaneously greater than both the first threshold and the second threshold, the controller determines that there is personnel movement; and When the filtered convolutional signal is simultaneously less than both the first threshold and the second threshold, the controller determines that there is personnel movement.
14. The personnel movement detection system according to claim 13, characterized in that... The low-pass filter is a weighted average filter.
15. The personnel movement detection system according to claim 10, characterized in that... The controller is used to count a third consecutive number of times the convolutional signal is simultaneously greater than the first threshold and the second threshold, wherein when the third consecutive number reaches a third preset number, the controller determines that there is personnel movement.
16. The personnel movement detection system according to claim 10, characterized in that... The controller is used to count a fourth consecutive number of times the convolutional signal is simultaneously less than the first threshold and the second threshold, wherein when the fourth consecutive number reaches a fourth preset number, the controller determines that there is personnel movement.
17. The personnel movement detection system according to claim 10, characterized in that... The first threshold is greater than zero and the second threshold is less than zero.