Method and system for detecting the presence of a person

Abstract

A method for determining the presence of a person is provided, the method comprising: a) receiving IR sensor data (50) from a thermopile during a first time period, and using the IR sensor data to determine an IR background signal baseline (51) for the time period and to determine variability of the IR sensor data (50); b) using the IR background signal baseline (51) and the variability of the IR background signal level to determine a threshold (52) having a value higher than the background signal baseline (51), such that the greater the variability of the IR background signal (50), the higher the threshold (52); and c) receiving further IR sensor data (50) during a second time period after the first time period, and using the further IR sensor data (50) and the threshold (52) determined in step b) to determine that a person is present when the further IR sensor data (50) includes a value higher than the threshold.

Description

Technical Field

[0001] The present invention relates to a system and method for detecting the presence of a person using infrared rays.

Background Art

[0002] Presence detection is the ability of a device or system to detect whether a person is present. It is known to use an IR sensor for contactless presence detection by detecting body heat. Presence detection may be used, for example, to adjust the operation of a device, for example, to switch a device on or off. Examples of devices that use contactless presence detection include a computer that shifts to a power-saving mode when no person is present, a heating / air conditioning system that shifts to a power-saving mode, and a lighting device such as a lamp that automatically turns on when a person is present.

[0003] Thermopiles are IR sensors that can deliver output not only as a temperature change but also as an absolute temperature. However, when using thermopiles for presence detection, it has been found that it is difficult to adjust thermopiles with respect to background IR radiation in a room, especially in a "noisy" environment where room temperature fluctuates.

[0004] U.S. Patent Application Publication No. 2015 / 0185806 describes presence detection using a thermopile with a predetermined threshold.

Summary of the Invention

[0005] In a first aspect of the present invention, a method for determining the presence of a person, comprising: a) receiving IR sensor data during a first time period from an IR sensor, using the IR sensor data to determine an IR background signal baseline for the period, and determining the variability of the IR sensor data; b) using the variability of the IR background signal baseline and the IR background signal level to determine a threshold value having a value higher than the background signal baseline, such that the threshold value increases as the variability in the IR background signal increases; c) receiving further IR sensor data during a second time period after the first time period, and using the further IR sensor data and the threshold value determined in step b) to determine that a person is present when the further IR sensor data includes a value higher than the threshold value.

[0006] The method provides a dynamically self-adjusting and highly responsive system that adapts to changes in the environment in order to avoid false positives and false negatives during presence detection. As the variability of the data increases, the threshold moves away from the baseline, reducing the risk of an incorrect "present" state when the background room temperature varies.

[0007] The first and second time periods are preferably of equal length.

[0008] In a preferred embodiment, IR sensor data from the second time period is used to update the threshold value. In a preferred embodiment, steps a) to c) are repeatedly executed, preferably at least every 5 seconds.

[0009] In step b), a dampening filter can be applied. In one embodiment, the dampening filter dampens (suppresses) the movement of the threshold value towards a higher value. The dampening filter prevents the threshold value from "running away" from the IR sensor data as the IR sensor data increases.

[0010] In one embodiment, a top line signal is determined when a person is present, a second threshold value lower than the top line signal is determined using the top line signal, a transition from a non - present state to a present state is determined when the system is in the non - present state and a value higher than the first threshold is detected, and a transition from a present state to a non - present state is determined when the system is in the present state and a value lower than the second threshold is detected.

[0011] The problem when a person is sitting in front of a computer in a small room, especially a cold room, is that the temperature of the room tends to rise over time, which also raises the threshold. So when a person leaves the computer, the signal does not fully reach the threshold, and thus the signal does not trigger the threshold. One way to solve this is to use a second threshold for triggering the transition from present to non - present, and this threshold is higher than the first threshold.

[0012] A damping filter may be used to damp (suppress) movement below (downward) and above (upward) the second threshold.

[0013] In one embodiment, the following method is executed at startup (start): : i) Determine an initial IR signal during a start - up period and calculate an initial signal baseline for the initial IR signal. ii) Use the initial IR signal baseline and the calculated variability of the initial IR signal during the start - up period to determine an upper - limit start - up threshold and a lower - limit start - up threshold. iii) Receive further IR sensor data in a later period. 1. Determine that the further IR sensor data exceeds the upper - limit start - up threshold, and then use the initial signal baseline as the IR background signal baseline in step b). Or 2. Determine that further IR sensor data is less than a lower start-up threshold, then determine the level of a new IR signal baseline and use that baseline as the IR background signal baseline in step b).

[0014] The IR sensor is preferably a thermopile. Since thermopiles are small, highly sensitive, relatively low-cost, and can measure absolute temperature non-contact, they are very useful for IR-based non-contact presence detection.

[0015] In a second aspect of the present invention, there is provided a system including an IR sensor, a processor, and a memory, the system including a signal processing module, threshold determination logic, and presence determination logic, the system being configured to use the thermopile during a first period to determine the IR sensor data and supply the IR sensor data to the signal processing module, the signal processing module being configured to use the IR sensor data to determine an IR background signal baseline for the period and to determine the variability of the IR sensor data, the threshold determination logic being configured to use the IR background signal baseline and the variability of the IR background signal to determine a threshold having a value higher than the background signal baseline such that the higher the variability of the IR background signal, the higher the threshold, and the presence determination logic being configured to use the threshold to determine (decide) that a person is present when further R sensor data is received and the IR sensor data includes a value higher than the threshold.

[0016] In a third aspect of the present invention, there is provided a device for receiving information from the system according to the second aspect of the present invention or a device configured as such, the device interpreting information from the system that no person is present and then configuring the device to enter a power-saving mode, or interpreting information from the system that a person is present and then waking up the device from the power-saving mode. The device may be a portable computer or a display for a computer.

Brief Description of the Drawings

[0017] The accompanying drawings form part of this specification, schematically illustrate preferred embodiments of the present invention, and are useful for exemplifying the principles of the present invention.

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Embodiments for Carrying Out the Invention

[0018] Figure 1 shows an embodiment of a presence determination system 1 comprising an IR sensor 2, preferably a thermopile 2, which can preferably measure absolute temperature values. A useful thermopile is shown in WO 2004 / 0968256, but other types of thermopiles can also be used. However, any suitable type of IR sensor, such as a bolometer, can be used. The thermopile 2 can detect IR radiation and provide IR sensor data 8 (see Figure 2), 50 (see Figures 6, 10, 11) to the subsystem 3. The subsystem 3 may be implemented in hardware or software, or a combination thereof, and Figure 1 shows an embodiment implemented in hardware and software. The subsystem 3 comprises an input interface 4 for receiving signals from the IR sensor 2. The subsystem 3 further comprises a processor 5, a memory 6, and an output interface 7. The system 1 may also include one or more of the signal filters, amplifiers, A / D converters, and similar devices known in the art of signal processing, particularly for processing signals from the IR sensor 2. The system 1 is powered from a power source. The subsystem 3 can be mounted within the same device as the IR sensor 2 or can be separate from the IR sensor 2. In one embodiment, all steps of the method described herein are performed by the same processor 5.

[0019] Referring to Figure 2, the memory 6 can store IR sensor data 8 and threshold data 9. The memory has threshold determination logic 10 and presence determination logic 11. The memory 6 also has presence status data 12 and a signal processing module 13. The memory 6 can have startup logic 14, which will be described in more detail below.

[0020] System 1 can typically output information regarding at least two states, namely "a person is present" or "a person is not present", or transitions from "present" to "absent state" and back again. The presence state of System 1 is stored in memory 6 as presence state data 12.

[0021] Referring to FIGS. 3 - 4, IR sensor 2 has a field of view 15 in which IR sensor 2 detects IR radiation and System 1 determines whether person 16 is present. Person 16 typically emits stronger IR radiation than background IR radiation (caused by the ambient room temperature). Thus, the IR radiation detected within field of view 15 can be used to determine whether person 16 is present within field of view 15. In some embodiments, System 1 can have multiple thermopiles 2 each having a different field of view 15. The multiple IR sensors 2 can use the same subsystem 3, and then subsystem 3 can provide data storage, signal processing, and threshold determination for each of the IR sensors.

[0022] System 1 can use output interface 7 to provide information about whether a person is present to a second device 17. Output interface 7 can be any suitable interface through which System 1 can provide data to the second device 17, and output interface 7 can be implemented in hardware and / or software. The second device 17 can be a computer system. The information provided from System 1 to the second device 17 can be information regarding the state (present / absent) or information regarding state transitions from one state to another.

[0023] The second device 17 can use the output from system 1 in various ways. The presence and absence states may, in some cases, be used to switch the second device 17 on or off. The second device 17 can be, for example, any type of device that can benefit from non-contact presence detection, such as a personal computer like a laptop, a heating system, an air conditioning system or a ventilation system, for example, to ensure that such a system shuts down or is put into a power-saving mode when no person is present for a set period of time, or the like. Thus, the presence or absence state can trigger a timer. Also, the second device 17 may be a device that provides light, such as an indoor or outdoor lamp. Also, the second device 17 may be switched on depending on the presence state so that a computer, display, lamp, or ventilation system is switched on when a person is present. The second device 17 may also be an alarm device such as an intruder alarm. Generally, all of these devices can benefit from being switched on, switched off, put into a power-saving mode, or woken up from a power-saving mode by the methods and systems described herein.

[0024] For example, the output from system 1 can be used to put the second device 17 into a power-saving mode and / or to shut down the display 18. Thus, when no person is present, the second device 17 may be put into a power-saving mode, perhaps after being detected for a minimum period of time when the absence state is likely. The "presence state" provided by system 1 can be used to wake up the second device 17.

[0025] For example, portions of subsystem 3 such as processor 5 and / or memory 6 may be part of the second device 17, especially when the second device 17 includes or consists of a computer. And the software described herein may be installed on the hard drive of a computer and used by the computer's CPU. Generally, portions of system 1 may be integrated with portions of the second device 17 such that processor 5 or memory 6 may be part of the second device 17. The entire system 1 may be fully integrated into the second device 17 as its subparts (sub - parts) implemented with an appropriate combination of hardware and software.

[0026] That is, system 1 may be attached within the second device 17 or may be separate from the second device 17. In one embodiment (shown in FIG. 4), the IR sensor 2 can be attached adjacent to a computer display 18, which can be a display 18 for a laptop computer, a tablet computer, a smartphone (such as an iPhone (registered trademark) or an Android phone), or a stand - alone display 18 for a stationary computer. The field of view 15 is preferably directed towards the intended position of the person 16 in front of the display 18. The display 18 may be an LCD display. In a preferred embodiment, system 1 is used in a laptop. A portable computer such as a laptop moves around between different rooms, indoors and outdoors, thereby being frequently exposed to a usage environment with a "noisy" thermal pattern, so it is particularly difficult to adjust (calibrate) the IR sensor.

[0027] IR sensor data is supplied from the IR sensor 2 to the subsystem 3. The subsystem 3 analyzes the sensor data and provides a presence / absence output to the second device 17. The system 1 can sample the thermal data within the field of view 15 using any suitable sampling interval. Preferably, the sampling frequency is from once every 5 seconds to 100 times per second. Sampling of the IR sensor data to obtain the IR sensor signal 50 is performed by the signal processing module 13 and stored as the IR sensor data 8. This can be done continuously and in real time. Usually, a thermopile sends the output as an output voltage. Preferably, all IR sensor data (threshold, baseline, etc.) here are absolute temperature values or can be converted to absolute temperature values rather than relative temperature values. Therefore, preferably, the thermopile is used as the IR sensor.

[0028] Next, with reference to FIG. 5, a method for determining the presence of a person will be described. It should be understood that the system 1 is configured to execute such a method.

[0029] In step 100, an IR background signal baseline 51 (the "baseline", see FIG. 8) is determined by the signal processing module 13. The baseline 51 represents historical IR sensor data 50 (also referred to herein as "IR signal 50") that has been processed by a (preferably digital) smoothing method that removes extreme values to show a trend. Examples of useful smoothing methods include moving average or median. In some embodiments, outliers in the data are removed. The baseline 51 is determined for a first period. The IR background signal baseline 51 reflects the background temperature of a room or other environment. The IR background signal baseline 51 preferably reflects the IR sensor data 50 when no person is present within the field of view 15. The baseline 51 can vary due to ventilation, sunlight, the number of people in the room, window openings, etc. The baseline 51 may be determined in different ways. The IR background signal baseline 51 may be determined, for example, as the average of a number of data points collected from the IR sensor data 50 during the first period. The appropriate lengths of the first period and the second period may be, for example, from 0.01 seconds to 5 seconds, more preferably from 0.01 seconds to 1 second. For the moving average, for example, a moving average over the last 1 to 15 seconds can be used. Thus, the baseline 51 can be updated at least every 5 seconds, more preferably at least every 1 second, more preferably at least every 2 seconds, and most preferably at least every 10 seconds.

[0030] The baseline 51 is determined when no person is within the field of view 15. The fact that no person is within the field of view 13 can be determined in different ways, for example, when the IR signal 50 falls below a threshold as described herein. One method of selecting the correct baseline 51 at startup will be described below with reference to FIGS. 9 - 11. The baseline 51 is preferably repeatedly updated using the IR signal 50 when the system is in a non - occupied state and is updated whenever the system returns to the non - occupied state.

[0031] In step 101, the variability of the IR data 50 for the first period is determined. This determination is made by the signal processing module 13. Any useful dispersion parameter or variability measurement can be used and applied to the IR signal 50. For example, the standard deviation, absolute variation, mean absolute deviation, or variance of the IR sensor data 50 can be used. Thus, in one embodiment, the standard deviation of the IR sensor data 50 is used to determine the variability of the IR background signal baseline 51. Higher background noise of the IR signal 50 results in higher variability and thus can result in a higher standard deviation over the measured time interval. Alternatively, the variability may be determined as the difference between the maximum and minimum values for the first period. In one embodiment, the variability of the background baseline 51 is determined (for example, the sampling interval for the baseline 51 is shorter than when determining the threshold 52). Step 101 can be executed before, after, or simultaneously with step 100.

[0032] In step 102, a threshold 52 for determining presence (also referred to herein as the "first threshold 52"; see FIG. 8) is determined by the threshold determination logic 10. The threshold determination logic 10 uses the data variability for the baseline 51 and the IR sensor data 50, which were used to determine the baseline 51 determined in steps 100 and 101. A factor n can be multiplied by the data variability parameter. Further, a predetermined offset can be used. Generally, the threshold is determined as follows: Threshold = Baseline + n * (Variability) + Offset

[0033] Here, n can be a dimensionless parameter having a constant value selected from 0.1 to 10, more preferably from 0.1 to 5, especially when the standard deviation is used to determine variability. The offset is selected depending on the configuration of the IR sensor and the selection of the amplifier and A / D converter, and may be determined using a specific configuration of System 1. As a rule of thumb, the offset can be about 10% to 70% of the difference between the typical differences in the measured signal intensities in the present and absent states. The offset can be determined for a particular application and the IR sensor used. The use of the offset is optional, and thus the offset may be zero. Higher variability results in a higher threshold 52, and as a result, a threshold 52 where variations caused by external factors do not trigger "present". The determined threshold 52 is stored as threshold data 9 in the memory 6. The purpose of the offset is to ensure that the threshold 52 is sufficiently far from the baseline 51 so that false positives are not detected by System 1.

[0034] Steps 100 to 102 can be repeatedly executed, thereby dynamically changing the baseline 51 and the threshold 52. In a preferred embodiment, the system determines the baseline 51 during successive periods, and the baseline 51 is used to repeatedly update the threshold 52. The successive periods may be discrete or may partially overlap. The successive periods can have the same length. As described above, the time can be from 0.01 seconds to 5 seconds, more preferably from 0.01 seconds to 1 second. Thus, the threshold can be updated at least every 5 seconds, more preferably at least every 1 second, more preferably at least every 2 seconds, and most preferably at least every 10 seconds. The threshold can be updated in real time or near real time.

[0035] In step 103, IR data 50 for a second period, which is after the first period, is supplied by sensor 2 to subsystem 3 and processed by presence determination logic 11 that accesses threshold 52 in the form of threshold data 9 in memory 6. The IR data 50 for the second period may be supplied from signal processing module 13 to presence determination logic 11. If the IR data 50 for the second period is higher than the threshold, it is determined that a person is present; if the further IR data is not higher than the threshold, it is determined that no person is present. One single measurement (measurement value) higher than the threshold may be sufficient, but it may also be required to exceed the threshold over a minimum duration, such as during a continuous or repeated period. The IR data 50 or baseline 51 for the second period can be used for comparison with threshold 52. The determination of presence is preferably made at the same frequency as the update of the threshold.

[0036] The IR data for the second period may in particular be used by threshold determination logic 11 to update the threshold as described in steps 100 to 102 if it is determined that no person is present.

[0037] Consecutive non - overlapping or overlapping periods can be used to repeatedly check whether the threshold is exceeded. The periods are preferably of equal length.

[0038] FIG. 6 shows how the baseline 51a from the first period 60a is used for the threshold determination logic 10 to determine the threshold 52a. The signal 50 in the subsequent period 60b is more variable and results in a higher threshold 52b. Thus, the presence determination logic 11 can check whether the threshold has been exceeded over a continuous period, and if it is determined that the threshold has not been exceeded, the signal is used to update the threshold 52 to the value indicated as 52b. The threshold is exceeded at time T1 during the period 60c. The transition to the present state occurs at time T1, but may also occur after the end of the sampling interval 60c, such as after the end of the time interval 60c like time T2 (see FIG. 6). The latter approach is more useful for short sampling intervals. Thus, in one embodiment, the IR signal 50 is checked against the threshold 52 at the same interval as the baseline and the threshold is updated.

[0039] In a preferred embodiment, when determining the threshold 52 in step 102, a (preferably digital) dampening filter is applied. For example, when person 16 slowly approaches the field of view 15 of the sensor, the threshold 52 gradually increases so that the signal 50 does not reach the threshold 52, and thus the presence state is not triggered, which is undesirable. The dampening filter preferably suppresses (dampens) the movement of the threshold 52, at least when the baseline 51 moves upward. For suppression, any suitable type of low-pass filter can be used. One way to suppress the movement of the threshold 52 is to allow movement only by a percentage of the change in the baseline. The percentage may be selected by those skilled in the art and may be, for example, 0.5 to 10%. The amount of change can be selected according to the sampling frequency. In this case, a higher percentage is used for a lower sampling frequency. In particular, in order to avoid the "slow approach effect" described above, the dampening filter should suppress the movement of the threshold 52 when the baseline moves upward. However, the threshold 52 may be able to move "downward" (towards a lower value) without being suppressed (damped). The dampening filter may be used by the threshold determination logic 10.

[0040] The above-mentioned threshold 52 can be used to determine the transition from the non-presence state to the presence state, and the transition from the presence state to the non-presence state, such that the state is switched each time the signal crosses the threshold 52.

[0041] However, in a preferred embodiment, a second threshold value 53 is used to switch from an existing state to a non-existing state. That is, the hysteresis principle is applied to the threshold value. Thus, there may be a first threshold value 52 that causes a switch from the (above-mentioned) non-existing state to the existing state, and a second threshold value 53 that causes a switch from the existing state to the non-existing state. The first threshold value 52 triggers the existing state only when the signal 50 crosses the first threshold value 52 from a lower value to a higher value, and the second threshold value 53 triggers the non-existing state only when the signal 50 crosses the second threshold value 53 from a higher value to a lower value. The second threshold value 53 preferably has a higher value than the first threshold value at any given time, but this is not a necessary condition.

[0042] The second threshold value 53 is determined in relation to a topline signal 54 that is determined when the person 24 is present in front of the computer. Otherwise, similar to how the IR background signal baseline 51 is determined as described above, the second threshold value 53 is determined with the important difference that it is below the topline signal 54. Second threshold value = Topline signal - (n * (variability) + offset)

[0043] The topline signal 54 may also be referred to as the "existing state baseline 54". Thus, the second threshold value 53 is preferably determined in the same way as described above for the first threshold value 52, but is determined by subtracting the variability and possibly an offset from the topline signal. The constant n and the offset may be different from or the same as in the case of the calculation of the first threshold value 52. Also, note that the "baseline 51" and the "topline 54" are determined in essentially the same way and may collectively be referred to as "smoothed IR signal data".

[0044] The second threshold 53 is determined by the threshold determination logic 10. An attenuation filter may be applied when determining the second threshold 53. The attenuation filter can suppress (attenuate) the movement of the second threshold 53, but can also suppress upward movement. When the sensor is incorporated into the second device 17 having a display 18 such as a laptop computer, the upward suppression prevents an overly high second threshold 53 from occurring when the person 16 leans forward toward the display 18 and the topline signal 54 instantaneously moves upward. Otherwise, in certain situations, it may trigger a non - existent state when the person 16 leans backward again in the system 1.

[0045] Referring to FIG. 7, method steps 200 - 204 are implemented in a manner similar to steps 100 - 104 of FIG. 5. In step 200, the topline signal 54 is determined. This is done in the same way as the baseline 51 and is done only in the "existent" state. The existent state may be triggered by step 104, or may be determined at startup by the method described below with reference to FIGS. 9 - 11, or by any other suitable method. For example, the existent state may be determined (judged) by detecting a keystroke or mouse movement within a minimum time interval. In step 201, the variability of the IR signal 50 is used (determined) to determine the topline 54 in step 200. In step 202, the second threshold 53 is determined. In step 203, IR sensor data for a second period that is below the second threshold 53 is received. Thereby, in step 204, a transition to a non - existent state is triggered. Preferably, the determination frequency and the update frequency are the same for the first and second thresholds.

[0046] In one embodiment, the method of FIG. 7 is used without the method of FIG. 5, that is, the method of determining a non - existent state is used alone without the method of determining an existent state.

[0047] System 1 can vary from a non - existent state to an existent state and return to the non - existent state. Each time System 1 returns to the non - existent state, the baseline 51 may be updated using measurement data from the corresponding period in the non - existent state. Each time System 1 returns to the existent state, the topline signal 54 may be updated using measurement data from the corresponding period in the existent state. The threshold values 52, 53 are updated when the baseline 51 and the topline 54 are updated, and thus, the threshold values 52, 53 can sometimes be left unchanged (i.e., the first threshold value 52 can be left unchanged in the existent state, and the second threshold value 53 can be left unchanged in the non - existent state). An example of IR data that varies System 1 with the first and second threshold values 52, 53 between the existent state and the non - existent state is shown in FIG. 8. The graph shows actual data from a thermopile that detects the presence of a person at a distance of about 1 meter, and System 1 determines and uses the first threshold value 52 and the second threshold value 53. The positions of the baseline 51, the topline 54, and the threshold values 52, 53 are shown schematically. The threshold values 52, 53 in FIG. 8 have been adjusted multiple times at short time intervals and appear to be changing gradually. Suppression (attenuation) is used as follows: when the baseline 51 moves upward, the movement of the first threshold value 52 is suppressed upward, and when the topline moves downward, the movement of the second threshold value 53 is suppressed downward. The "spikes" in the first threshold value 52 at 8.5 seconds and 14 seconds are caused by the high variability of the baseline 51. Similarly, the second threshold value 53 moves slightly downward at 7 seconds and 12 seconds, which is also caused by the high variability of the topline 54.

[0048] When the system 1 is starting up, the system 1 does not know whether the person 16 is present within the field of view 15. For example, when the laptop 17 equipped with the system 1 is starting up, the person 16 may be sitting in front of the laptop 17 or doing something else, such as fetching coffee. Therefore, when the system 1 starts detecting the IR signal, the system 1 does not know whether it is detecting the baseline 51 or the top line 54. The system 1 can use the input from the device 17 to detect, for example, whether the keyboard is being tapped to detect this. Another way to solve this problem is the method shown in FIGS. 9 to 11 using the startup logic 14. At startup, the system 1 determines the initial signal 55 and determines the lower threshold 56 and the upper threshold 57. The thresholds 56, 57 may be determined as described above with reference to FIGS. 5 to 7, where the lower threshold 56 is determined as described for the second threshold 53, and the upper threshold 57 is determined as described for the first threshold 52.

[0049] In step 300 of FIG. 9, the initial signal 55 baseline is detected and determined. This is done in the same way as determining the above-mentioned background baseline 51 and top line signal 54, and since the state is unknown, it can only not be designated as the baseline 51 or the top line signal 54. Therefore, the initial signal baseline 55 is determined using the IR sensor data 50. Therefore, the initial signal baseline 55 may be an average signal. In step 301, the variability of the initial sensor data 50 is determined in the same way as in steps 101 and 201 above. In step 302, the lower startup threshold 56 and the upper startup threshold 57 are determined. The lower startup threshold 56 and the upper startup threshold 57 are determined as follows: Lower startup threshold = Initial signal baseline - n * (variability) - offset Upper start-up threshold = Initial signal baseline + n*(Variability) + Offset

[0050] Thresholds 56 and 57 may be at the same or substantially the same distance from the initial signal baseline 55 as shown in FIGS. 10-11, but this is not essential. The constant n and / or the offset may or may not be the same for the lower and upper start-up thresholds 56 and 57.

[0051] In step 303, IR data for at least a second period is received from IR sensor 2 until it exceeds either the lower start-up threshold 56 or the upper start-up threshold 57. If the upper start-up threshold 57 is exceeded, in step 304, it is determined that the initial signal 55 is to be used as the background baseline 51. Next, system 1 can proceed to determine the top line 54 (FIG. 10), and if the lower start-up threshold 56 is exceeded (passed), the initial signal baseline 55 is not used as the background baseline 51. Instead, the background baseline 51 is determined in step 305 (FIG. 11). Then, the initial signal baseline 55 determined in step 300 may optionally be used as the top line 54 (the data may also be discarded).

[0052] This process can be used whenever system 1 needs to define the baseline 51 or the top line 54, for example, at startup of system 1 or when system 1 needs to be recalibrated.

[0053] Here, how system 1 is used to detect the presence of person 16 is described. System 1 may also be used to detect the presence of an animal. The animal is preferably a warm-blooded animal such as a mammal or a bird.

[0054] Although the present invention has been described with reference to specific exemplary embodiments, this description is generally intended to illustrate only the concepts of the present invention and should not be construed as limiting the scope of the present invention. The scope is generally defined by the claims.

Claims

1. A method for determining the presence of a person, the method comprising a system (1) that can be in a first state indicating the absence of a person and a second state indicating the presence of a person, the method being repeatedly executed, a) receiving IR sensor data (50) from a thermopile (2) during a first period, using the IR sensor data to determine a smoothed IR signal (51, 54) for the period, and determining the variability of the IR sensor data (50); and then, b) using the smoothed IR signal (51, 54) and the variability of the IR background signal level to determine a first threshold (52) having a value higher than the smoothed IR signal (51, 54) such that the higher the variability of the IR background signal when the system is in the first state determined by the system (1), the higher the threshold value, and determining a second threshold (53) having a value lower than the smoothed IR signal such that the higher the variability of the IR background signal when the system is in the second state determined by the system (1), the lower the threshold value; and then, c) receiving further IR sensor data (50) during a second period after the first period, and using the further IR sensor data (50) to switch the state to the second state when the further IR sensor data (50) includes a value higher than the first threshold when the system is in the first state, not switching the state to the second state when the further IR sensor data (50) does not include a value higher than the first threshold, switching the state to the first state when the further IR sensor data (50) includes a value lower than the second threshold (53) when the system is in the second state, and not switching the state to the first state when the further IR sensor data (50) does not include a value lower than the second threshold (53), comprising, In step b), a damping filter that suppresses the movement of the first threshold to a higher value is used, and a damping filter that damps the second threshold downward is used, The first threshold is, the smoothed IR signal (51, 54) + n * (variability parameter) + offset determined by, The second threshold is, The smoothed IR signal (51, 54) - (n * (volatility parameter) + offset) is determined, where n is a constant value selected from 0.1 to 10, the offset is a non - negative numerical value, and the volatility parameter is the standard deviation, absolute variation, mean absolute deviation, or variance of the IR sensor data (50). A method. **Claim 2** The method according to claim 1, wherein the lengths of the first period and the second period are equal. A method. **Claim 3** The method according to claim 1 or 2, wherein the IR sensor data from the second period is used to update the first threshold. A method. **Claim 4** The method according to any one of claims 1 to 3, wherein the system has a single IR sensor. A method. **Claim 5** The method according to any one of claims 1 to 4, wherein the method is executed at least every 5 seconds. A method. **Claim 6** The method according to any one of claims 1 to 5, wherein the system (1) is used in a laptop computer having a display (18), the thermopile (2) is attached adjacent to the display (18), the thermopile (2) has a field of view (15) directed at the intended position of the person (16) in front of the display (18), and the absent state is used to put the laptop computer into a power - saving mode or to shut down the display (18). A method. **Claim 7** The method according to any one of claims 1 to 6, wherein at startup, the following process: i) Determine an initial IR signal during a start - up period and calculate an initial signal baseline for the initial IR signal, then ii) Use the initial IR signal baseline and the calculated volatility of the initial IR signal during the start - up period to determine an upper start - up threshold and a lower start - up threshold, then iii) Receive further IR sensor data in a subsequent period, 1. Determine that the further IR sensor data exceeds the upper start - up threshold, and then use the initial signal baseline as the baseline of the IR background signal in step b), or 2. Determine that the additional IR sensor data is less than the lower start-up threshold, then determine a new IR signal baseline level, and use that baseline as the IR background signal baseline in step b). A method that is executed. **Claim 8** A system (1) comprising an IR sensor (2) that is a thermopile (2), a processor (5), and a memory (6), wherein the system (1) comprises a signal processing module (13), a threshold determination logic (10), and a presence determination logic (11), and the system (1) is configured to store a first state or a second state in the memory (6) as presence state data (12), the first state indicating the absence of a person, and the second state indicating the presence of a person. The system (1) is further configured to use the IR sensor (2) to determine IR sensor data (50) during a first period and provide the IR sensor data (50) to the signal processing module (13). The signal processing module (13) is configured to use the IR sensor data (50) to determine a smoothed IR signal (51, 54) for the period and determine the variability of the IR sensor data (50). The threshold determination logic (10) uses the smoothed IR signal (51, 54) and the variability of the IR background signal level to determine a first threshold (52) having a value higher than the smoothed IR signal (51, 54) such that the higher the variability of the IR background signal, the higher the threshold (52) when the system is in the first state determined by the system (1), and a second threshold (53) having a value lower than the smoothed IR signal (51, 54) such that the higher the variability of the IR background signal, the lower the threshold (53) when the system is in the second state determined by the system (1). The presence determination logic (11) uses the threshold value (52) to determine that when further IR sensor data (50) is received and the system is in the first state, if the further IR sensor data (50) includes a value higher than the first threshold value, the state is switched to the second state; if the further IR sensor data (50) does not include a value higher than the first threshold value, the state is not switched to the second state; when the system is in the second state, if the further IR sensor data (50) includes a value lower than the second threshold value (53), the state is switched to the first state; if the further IR sensor data (50) does not include a value lower than the second threshold value (53), the state is not switched to the first state, and the system is configured to repeatedly determine the state. An attenuation filter that suppresses movement to a higher value of the first threshold value is used, and an attenuation filter that attenuates the second threshold value downward is used. The first threshold value is determined by the smoothed IR signal (51, 54) + n * (variability parameter) + offset The second threshold value is determined by the smoothed IR signal (51, 54) - (n * (variability parameter) + offset), where n is a constant value selected from 0.1 to 10, the offset is a non - negative numerical value, and the variability parameter is the standard deviation, absolute variation, mean absolute deviation, or variance of the IR sensor data (50).

9. A device that constitutes the system (1) according to claim 8 or a device configured to receive information from the system (1) according to claim 8, wherein when receiving information from the system (1) that no person is present, the device is put into a power - saving mode, or when receiving information from the system (1) that a person is present, the device is configured to wake up from the power - saving mode.

10. The device according to claim 9, which is a portable computer or a display for a computer.

11. The device according to claim 9, wherein the device is a laptop computer having a display (18), the thermopile (2) is attached next to the display (18), the thermopile (2) has a field of view (15) directed at the intended position of a person (16) in front of the display (18), and the absent state is used to put the laptop computer into a power-saving mode or to shut down the display (18).

Images