JP4597040B2 - Intrusion detection method and intrusion detection system - Google Patents

Description

  The present invention relates to an intrusion detection method and an intrusion detection system using a passive intrusion detection sensor such as a pyroelectric infrared sensor.

Conventionally, an intrusion detection apparatus using a pyroelectric infrared sensor element as a passive intrusion detection sensor can have a circuit configuration shown in FIG. 6, for example. That is, infrared rays collected by the infrared condensing lens 21 are input to the pyroelectric infrared sensor element 22, and an electric signal from the sensor element 22 is amplified by the amplifier 23 and input to the signal level determination circuit 24. The signal level determination circuit 24 determines whether the signal level is higher than the upper threshold value VH or lower than the lower threshold value VL. When it is larger than the upper threshold value VH or smaller than the lower threshold value VL, it is determined that the detected object has entered the detection area M of the infrared sensor element 22, and an intrusion detection signal is output. That is, when the signal level of the amplifier output signal is between the upper threshold value VH and the lower threshold value VL, there is no intruder in the detection area M, and the upper threshold value VH is greater than or less than the lower threshold value VL. In addition, it is configured to determine that there is an intruder in the detection area M.

  As shown in FIG. 7, the infrared sensor element 22 has a surface temperature of a back wall surface 32 as a back object existing in the detection area M when a detected object 31 (for example, an intruder) enters the detection area M. If the difference between the surface temperature of the detected object 31 is 2 to 3 degrees or more, it is detected and an electric signal is emitted.

  Therefore, in the system shown in FIG. 6, temperature compensation is performed for the amplifier 23 by linking the temperature sensor 25 that detects the ambient temperature (air temperature). As the temperature compensation method, the following (i) and (ii) are known.

(i) A method of controlling the gain of the amplifier 23 according to the ambient temperature of the installation location (environment) of the infrared sensor element 22 (see the line X in FIG. 8).

  (ii) A method of increasing the gain of the amplifier 23 when the infrared sensor element 22 is installed (environment) at a certain ambient temperature (for example, around 37 ° C. close to the human body temperature) (see line Y in FIG. 8).

  The above (i) and (ii) are determination methods based on the magnitude of the output signal of the amplifier 23 ((ii) is considered to be an improvement method of (i)). Therefore, the amplifier gain is increased to supplement the decrease in temperature difference.

In such a device, in order to prevent malfunction, there is known a device in which the upper and lower thresholds are changed according to the change in the output voltage when a human body or the like is not sensed (see, for example, Patent Document 1). ).
Japanese Patent Publication No. 06-016119

As described above, since the detection accuracy of the infrared sensor element varies depending on the ambient temperature that changes based on the season or whether it is daytime or nighttime, the above (i) (ii) simply supplements the decrease in the temperature difference.
This method and the technique described in Patent Document 1 in which the upper and lower thresholds are changed according to the change in the output voltage cannot be adequately dealt with.

  Accordingly, the inventor has conceived that substantially constant detection accuracy can be secured by correcting the upper and lower thresholds based on the (current) average temperature of the ambient temperature, and has led to the present invention. .

  By the way, the target object to be detected is generally an intruder, and the average human body temperature is typically 36 to 37 ° C. However, since clothes are worn, the human surface temperature is lower than the body temperature. In addition, since the number of clothes and the fabrics to be worn are different between summer and winter, winter is generally lower than summer. That is, in summer, clothes of people are light and the surface temperature approaches body temperature (about 37 ° C .: almost constant), while in winter, the clothes are thick and the surface temperature is lower than summer. As a result, the difference in surface temperature is smaller in summer than in winter. The average temperature increases from winter to spring and from spring to summer, and the average temperature decreases from summer to autumn and from autumn to winter. In other words, the average temperature is similar between spring and autumn, but the clothes (materials) of people are different between spring and autumn, which affects the surface temperature of the person being detected. Thus, the surface temperature of an object to be detected such as an intruder varies depending on the season. On the other hand, the surface temperature of the back object also changes depending on the season because it changes depending on the temperature (ambient temperature) of the installation place. In addition to the season, even on the same day, the ambient temperature changes between day and night, so the surface temperature of the back object also changes.

  From the above, the surface temperature of the object to be detected and the surface temperature of the back object change depending on the season, daytime or night, and the infrared sensor element 22 can detect the object 31 to be detected. The size or width is different. In other words, the detection accuracy of the infrared sensor element 22 varies depending on the season, daytime, or nighttime.

Therefore, not only the upper threshold value VH and the lower threshold value VL are set based on the average temperature of the ambient temperature, but also the upper threshold value for detecting an object to be detected based on the season or whether it is daytime or nighttime. VH
If the lower threshold value VL is changed (corrected), it can be said that the detection accuracy is further improved.

  In addition, since the trend of changes in the average temperature of the season and the average temperature of the day or night varies depending on the area where the intrusion detection device is installed, the detection accuracy can also be improved by correcting based on the area where the intrusion detection device is installed. Can do.

  The present invention provides an intrusion detection method and an intrusion detection system capable of detecting an object (intruder) with a substantially constant detection accuracy.

According to the first aspect of the present invention, when it is determined whether or not an object to be detected has entered the detection area based on a signal from the passive intrusion detection sensor, the signal level is detected every unit time and the unit is based on the detected signal level. The amount of change in signal level per time is sequentially accumulated as a level difference, and when the number of accumulated level differences reaches the first set value L, the lower level of the accumulated level difference is below the upper threshold value. a whether the determined intrusion detection method intrusion of a detection object number of the level difference equal to or larger than the threshold value exceeds a second set value N, the past average temperature of day or night for ambient temperature, Prepare a data table showing the relationship between the season and region and the upper and lower thresholds, and use the data table to determine the upper and lower thresholds based on the current average temperature, season and region Characterized in that it constant. Here, the data table is created in advance by measuring the temperature for each region and for each season (or month) and using the data of the Japan Meteorological Agency to obtain the relationship between the upper and lower thresholds. Is done.

In this way, the upper and lower thresholds are set using the data table based on the current average temperature for the ambient temperature, so when determining intruders, it is related to the average temperature that varies depending on the season and region. Therefore, a certain detection accuracy can be ensured. Therefore, intrusion detection can be performed stably throughout the year.

The invention of claim 2 is a system for implementing this method, based on a signal from passive intrusion detection sensor, the signal level per unit based thereon time detects the signal level for each unit time The change amount is sequentially accumulated as a level difference, and when the number of accumulated level differences reaches the first set value L, the level difference that is less than the upper threshold and greater than or equal to the lower threshold among the accumulated level differences. An intrusion detection system that determines whether or not there is an intruder based on whether or not the number exceeds a second set value N, the past average daytime or nighttime ambient temperature, and the upper and lower sides A data table showing the relationship with the threshold value, an area selection switch for selecting the area to be installed, a temperature sensor for detecting the ambient temperature, and an average temperature for the ambient temperature by receiving a signal from the temperature sensor Temperature calculating means for calculating the average illuminance of the surroundings, illuminance sensor for detecting the average illuminance of the surroundings, day / night determination means for receiving the signal from the illuminance sensor to determine day and night based on the illuminance, the area selection switch and temperature calculating means, day / night determination And a threshold value setting means for setting the upper and lower threshold values using the data table based on the current average temperature, season, region and day and night with respect to the ambient temperature.

In this case, as described in claim 3 , the signal conversion means for A / D converting the analog signal from the intrusion detection sensor into a digital signal, and the signal from the signal conversion means are received at regular intervals. A sampling means for sampling the signal level of the digital signal, a level difference calculating means for receiving a signal from the sampling means and calculating a level difference between two consecutive sampling values, and a signal from the level difference calculating means. Memory means for sequentially storing the level differences in time series, a counter linked to the memory means for counting the number of level differences stored in the memory means, and receiving a signal from the counter by the counter When the first set value L is counted, the level within the set range among the stored level differences Number is a determining intrusion determining means for determining whether more than a second set value N, the memory means may be configured for storing the data table.

In this way, the object to be detected in the detection area (for example, an intruder) is focused on the rate of change in signal level (difference in signal level per short time) rather than simply determining the signal level. ), It is easy to reduce malfunctions. Therefore, there are few malfunctions even in the determination when installed outdoors, and the intrusion detection operation is stabilized.

In these cases, as described in claim 4 , the data table may be arranged with correction coefficients for multiplying the upper threshold value and the lower threshold value serving as a reference.

The present invention detects a signal level every unit time based on a signal from a passive intrusion detection sensor, and sequentially accumulates a change amount of the signal level per unit time as a level difference based on the signal level, and accumulates the level difference. When the number reaches the first set value L, the intrusion depends on whether the number of level differences that are less than the upper threshold and greater than or equal to the lower threshold among the accumulated level differences exceeds the second set value N When determining whether or not there is a person, using a data table showing the relationship between the average temperature, season and region in the past day or night of the ambient temperature, and the upper and lower thresholds, Since the upper and lower thresholds are set based on the temperature, a constant detection accuracy can be ensured regardless of the average temperature that changes depending on the season.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a schematic configuration of an intrusion detection system according to the present invention.

  As shown in FIG. 1, the intrusion detection system 11 inputs infrared rays collected through the infrared condenser lens 1 to a pyroelectric infrared sensor element 2 (intrusion detection sensor) that converts the amount of change of infrared rays into an electrical signal. Then, the analog signal from the infrared sensor element 2 is input to the control means 4 (CPU) for determining whether there is an intruder through the amplifier 3 that amplifies the analog signal.

  The control unit 4 includes a signal conversion unit 4A (A / D conversion unit), a sampling unit 4B, a level difference calculation unit 4C, a memory unit 4D, a counter 4E, and an intrusion determination unit 4F. Further, after the determination, there is also provided a clearing unit 4G that receives a signal from the intrusion determining unit 4F and clears the level difference stored in the memory unit 4D. The signal conversion means 4A (A / D conversion means) and the memory means 4D do not have to be built in as a part of the control means 4.

The signal conversion means 4A A / D converts the output signal (analog signal) from the amplifier 3 (amplifies the analog signal of the pyroelectric infrared sensor element 2), quantizes it with R (bit), and converts it into a digital signal. Is. Assuming that the frequency component of the output signal of the infrared sensor element 2 is fL (Hz) to fH (Hz), the sampling frequency fs for quantization is fs ≧ 2 × fH (Hz) according to Shannon's sampling theorem.

The sampling means 4B receives the signal from the signal conversion means 4A and samples the signal level of the digital signal at regular intervals. Note that the sampling frequency fs2 at this time has a relationship of fs2 ≦ fs, and the number of level difference determinations is reduced, thereby saving labor.
We are trying to save energy and speed up judgment.

  The level difference calculation means 4C receives a signal from the sampling means 4B and calculates a level difference between two consecutive sampling values. That is, in the digital signal, if the signal levels of sampling times T1, T2,..., Tn are P1, P2,..., Pn, the level differences are Δ12, Δ23,. 1) n... Is calculated as the difference between P1 and P2, P2 and P3,... Pn-1 and Pn, respectively.

  The memory means 4D receives the signals from the level difference calculating means 4C and sequentially stores the level differences Δ12, Δ23,... In time series. In addition, data tables # 1 to # 3 indicating the relationship between the average daytime or night temperature and the season / month and region are stored in advance. Data table # 1 shows the relationship between quantized data and temperature (see Table 1), data table # 2 shows the relationship between quantized data and illuminance (see Table 2), and data table # 3 shows the average temperature, season The relationship between the month and the correction coefficients α and β (see Table 3) is shown.

カウンタ４Ｅは、メモリ手段４Ｄに連係されメモリ手段４Ｄに記憶されるレベル差Δ１２，Δ２３，・・・・の個数をカウントするものである。

The counter 4E counts the number of level differences Δ12, Δ23,... Linked to the memory means 4D and stored in the memory means 4D.

  The intrusion determining means 4F receives a signal from the counter 4E, and when the first set value L is counted by the counter 4E, that is, when the number of level differences Δ12, Δ23,. The L level differences Δ12, Δ23,... Are read from the memory means 4D, and the number of level differences Δ12 satisfying A ≦ Δ ≦ B among the L level differences Δ12, Δ23,. , Δ23,... Are determined to be N or more. When there are N or more, it is determined that there is an intruder in the detection area, and an intrusion detection signal is output. Here, A and B are lower and upper threshold values predetermined by experience or experiment, respectively, and are corrected by multiplication by correction coefficients α and β selected based on the average temperature or the like.

The clearing unit 4G receives a signal from the intrusion determining unit 4F and clears the level difference stored in the memory unit 4D after the determination. That is, if there are N or more, it is determined that there is an intruder in the detection area, an intrusion detection signal is output, a clear signal is sent, and the level differences Δ12, Δ23,. .. Is cleared, but if there are not more than N, a clear signal is immediately sent to the memory means 4D. Thereafter, the determination operation by the intrusion determination unit 4F is repeated each time L level differences Δ12, Δ23,... Are stored in the memory unit 4D in the same manner as described above. Since the signal level difference may be either positive or negative, it is stored as an absolute value in the memory means 4D, and the setting values A and B are also absolute values. L and N are integer values having a relationship of L ≧ N> 0.

  Further, the control means 4 includes an area selection switch 12 for selecting the area where the installation is performed so that the thresholds A and B can be corrected based on the current average temperature, and a temperature sensor 13 for detecting the temperature around the installation place The illuminance sensor 14 for detecting the illuminance around the installation location is linked. In addition to the means 4A to 4G, the control means 4 receives a signal from the temperature sensor 13 and calculates an average temperature for the ambient temperature, and whether it is a daytime signal from the illuminance sensor. Memory means based on average daytime or nighttime temperature, season / month, and region in response to signals from day / night judgment means 4J for judging whether it is night, area selection switch 12, temperature calculation means 4H and day / night judgment means 4J Threshold setting means 4K is provided for setting the upper and lower thresholds based on a 4D table. Note that the memory unit 4D may not be incorporated as a part of the control unit 4.

  By the way, in general, a passive infrared sensor element outputs a signal when an object to be detected enters a detection area near or far from the installation, and the output signal has a signal output time as the detection position is further away. The amount of change becomes smaller. That is, as shown in FIG. 2, the signal waveform W2 has an amplitude when it is far away, and a signal waveform W1 with a large amplitude when it is near.

  Even if it enters the same location within the detection area (for example, a distant area where the installation location is the same) depending on the season / month, the signal waveform W2 will be displayed in winter, but the surface temperature in spring, summer and autumn. The difference is large, the amplitude of the signal waveform W1 is large, and there is a tendency to differ depending on the season and month.

  Further, when there is no object to be detected, the signal waveform is a fluctuation signal waveform W3 due to sunlight, electric light, and various noises, and the amount of change within the same time is extremely small compared to the waveforms W1 and W2. Become.

  The intrusion detection system 11 does not have a timer function or a function for determining the current date and time by receiving a time signal such as a radio clock, but has a function for determining the season / month. That is, the average temperature of the day or night is calculated by quantizing the output signal of the temperature sensor 13 and the output signal of the illuminance sensor 14, and the past average temperature and season / month previously stored in the memory means 4D are calculated. The current season and month can be determined from the causal relationship data. Further, the user may periodically input the season / month to the control means 4 with a manual switch. However, since the average temperature of the season and the month varies depending on the area, the user selects the area to be used by the area selection switch 12 (DIP SW) (for example, in the case of Kobe City, Hyogo Prefecture, for example, the southern Hyogo Prefecture 1010 in 4-bit display). Must be entered. Thereby, prior to the start of intrusion detection, a data table corresponding to the use region is selected from a large number of data tables created for each fixed region.

  A correction coefficient is selected based on the current average temperature using the data table thus selected. Then, the above-mentioned A and B are multiplied and corrected by the correction coefficients α and β to determine the presence or absence of intruders. Even if the average temperature is the same, the season determined as described above is used.・ Because the month is taken into account, the value of the correction factor will be different if the season and month are different.

  Next, the flow of intrusion determination processing in the control unit 4 will be described with reference to FIG.

  When starting, first, the power is turned on (step S1), initialized (initialized) (step S2), it is determined whether one minute has passed after the power is turned on (step S3), and in order to stabilize the system, The determination in step S3 is repeated until 1 minute has elapsed after the power is turned on.

When one minute has passed since the power was turned on, the output signal of the illuminance sensor 14 is read, and as shown in FIG. 4 described later, the day / night determination means 4J performs quantization (digitization) and day / night determination (step S4). The determination data for day and night is stored in the memory # 1 in the memory means 4D (step S5). In this step S5, if the currently stored data is blank or different from the current determination data, it is overwritten.

  Next, the output signal of the temperature sensor 13 is read, and as shown in FIG. 5 described later, the temperature calculation means 4H performs quantization (digitization) and season / month determination (step S6). Are stored in the memory # 1 in the memory means 4D (step S7). Even in step S7, if the currently stored data is blank or different from the current determination data, it is overwritten. As described above, the season / month can be determined based on the causal relationship data between the past average temperature and the season / month previously stored in the memory unit 4D.

  Next, correction coefficients α and β are selected from the correction coefficient list (table # 3) shown in Table 3 stored in the data table 3 based on different data of season / month and day / night, and the memory # 1. (Step S8). If the currently stored correction coefficient is different from the current correction coefficient, it is overwritten.

After these processes, the signal from the amplifier 3 is sampled at the sampling frequency fs in the A / D conversion means 4A, quantized (digitized) with R (bit) (step S9), and the level difference Δ between the sampling values. (n-1) n = Sn-Sn-1 is calculated and sequentially stored in the memory # 2 in the memory means 4D (step S10). This level difference measurement data is stored in the memory # 2 in the memory means 4D together with the daytime or nighttime temperature measurement data.

  Then, it is determined whether or not the number of level differences Δ (n−1) n stored in the memory means 4D has become L (step S11), and if it has become L, the process proceeds to step S12. If not L, the process returns to step S9, and the processes in steps S9 to S11 are repeated until L.

  In step S12, the thresholds A and B and the correction coefficients α and β are taken out from the memory # 1 of the memory means 4D and multiplied by each. Therefore, the threshold values for multiplying the threshold values A and B by the correction coefficients α and β are set every time it is determined in step S11 that the number of level differences is L, and the threshold values A and B are updated. Become.

  Then, it is determined whether or not there are N or more of Δ (n−1) n satisfying αA ≦ Δ (n−1) n ≦ αB (step S13). Since it is presumed that there was an intruder in the area, it is determined that there is an intruder (person) in the detection area, an alarm operation (output of an intrusion detection signal) is performed (step S14), and then in the memory means 4D Memory # 2 (data of level difference Δ (n−1) n) is cleared (step S15), and the process proceeds to step S16, while Δ (n−1) n is changed to αA ≦ Δ (in step S13. n-1) If the number of objects satisfying n ≦ αB is not N or more, it is determined that there is no intruder (step S17), the process proceeds to step S15, and all the memory # 2 in the memory means 4D is cleared. Then, the process proceeds to step S16.

In step S16, it is determined whether or not 30 minutes have passed since the determination of the season / month and day / night. If 30 minutes have passed, the process returns to step S3, and if 30 minutes have not passed, the process returns to step S9. .
(Quantization (digitization) of illumination signal output signal and judgment of day and night)
As shown in FIG. 4, when starting, first, the output of the illuminance sensor 14 is input (step S21), sampled at the sampling period TB, and quantized (digitized) with DB (bit) (step S22).

Then, from Table 2 (data table # 2), the temperature EB (Lx
) Is determined (step S23).

Then, it is determined whether or not the illuminance E satisfies E <1000 (Lx) (step S24), and if satisfied, it is determined as day (step S25), while if not satisfied, it is determined as night ( Step S26), the process ends.
(Quantization (digitalization) of temperature sensor output signal and judgment of season / month)
As shown in FIG. 5, when starting, first, the output of the temperature sensor 13 is input (step S31), sampled at a sampling period TT, and quantized (digitized) at DT (bit) (step S32).

Then, from Table 2 (Data Table # 2), using the quantized numerical value, the temperature ET (° C.
) Is determined (step S33), and the temperature ET (° C.) is stored in the memory # 2 (step S).
34).

  Then, in the day / night determination, it is determined whether or not a change has been made from noon to night or from night to noon (step S35). Average temperature (average temperature) for determination or night determination is calculated, stored in memory # 1 (step S36), and season / month is calculated using the past average temperature in Table 3 (data table # 3) (Step S37), the process ends.

  Here, the season is based on the average temperature, if the temperature range is in the range of 5-25 ° C, it is spring or autumn, if it is 25 ° C or more, it is summer, if it is less than 5 ° C, it is winter. If it is, it is determined seamlessly. Further, the moon is further subdivided so that it can be discriminated seamlessly in the same way as the season.

  The embodiment described above can be modified as follows.

(i) In the above embodiment, as shown in Table 3, in addition to the average temperature and season, day and night and the moon are used as the determination elements. It is also possible to make a determination based on only the above. In this way, there is an advantage that the memory capacity can be reduced. Although the accuracy of correction is low, it is considered that there is no practical problem.

(ii)前記実施の形態においては、１つの検知センサで１つの検知エリアを持つように構成しているが、複数の検知センサを用いて複数の検知エリアについて検知動作をするようにて、いずれかの検知エリアあるいは複数の検知エリアから侵入検知信号があった場合に、侵入者があったと判断する公正とすることも可能である。

(ii) In the above embodiment, one detection sensor is configured to have one detection area, but a plurality of detection sensors are used to perform detection operations for a plurality of detection areas. If there is an intrusion detection signal from such a detection area or a plurality of detection areas, it may be fair to determine that there is an intruder.

(iii) It is determined every time L pieces of sampling data for the level difference Δ are collected.
Once L pieces of sampling data have been collected, the sampling data can be updated sequentially, and determination can be made every time the sampling data is updated M times.

  (iv) In addition to an apparatus using a pyroelectric infrared sensor, the present invention can be similarly applied to an apparatus using a Doppler ultrasonic sensor or a microwave sensor as a human detection sensor.

It is a figure which shows schematic structure of the intrusion detection apparatus which is one embodiment which concerns on this invention. It is explanatory drawing of the waveform of a signal. It is a flowchart figure which shows the flow of a process. It is a flowchart figure which shows the flow of a process about the output signal of a temperature sensor. It is a flowchart figure which shows the flow of a process about the output signal of an illumination intensity sensor. It is explanatory drawing of the conventional intrusion detection apparatus. It is explanatory drawing of the detection principle of a passive infrared sensor element. It is explanatory drawing of the conventional intrusion determination method.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Infrared condensing lens 2 Infrared sensor element 3 Amplifier 4 Control means 4A Signal conversion means 4B Sampling means 4C Level difference calculation means 4D Memory means 4E Counter 4F Intrusion determination means 4G Clear means 4H Temperature calculation means 4J Day / night determination means 4K Threshold setting means 11 Intrusion detection device 12 Area selection switch 13 Temperature sensor 14 Illuminance sensor

Claims (4)

When determining whether or not an object has entered the detection area based on a signal from a passive intrusion detection sensor, the signal level is detected every unit time and the signal level changes per unit time based on it. When the amount of level differences is accumulated to the first set value L, the level difference that is less than the upper threshold and greater than or equal to the lower threshold is accumulated. An intrusion detection method for determining the intrusion of an object to be detected based on whether or not the number of objects exceeds a second set value N,
Prepare a data table showing the relationship between the average temperature of the past day or night, the season and the region, and the upper and lower thresholds for the ambient temperature,
An intrusion detection method comprising setting the upper and lower thresholds using the data table based on the current average temperature, season and region . When determining whether or not an object has entered the detection area based on a signal from a passive intrusion detection sensor, the signal level is detected every unit time and the signal level changes per unit time based on it. When the amount of level differences is accumulated to the first set value L, the level difference that is less than the upper threshold and greater than or equal to the lower threshold is accumulated. An intrusion detection system that determines the intrusion of an object to be detected based on whether or not the number of objects exceeds a second set value N,
A data table indicating the relationship between the average temperature in the day or night for the ambient temperature and the upper and lower thresholds;
An area selection switch for selecting the area to be installed;
A temperature sensor that detects the ambient temperature;
Temperature calculation means for receiving a signal from the temperature sensor and calculating an average temperature for the ambient temperature;
An illuminance sensor that detects the average illuminance of the surroundings;
Day and night determination means for receiving a signal from the illuminance sensor and determining day and night based on illuminance;
The upper and lower thresholds are set using the data table based on the current average temperature, season, area, and day / night for the ambient temperature in response to signals from the area selection switch, temperature calculation means, and day / night determination means. An intrusion detection system comprising a threshold setting unit. Signal conversion means for A / D converting the analog signal from the intrusion detection sensor into a digital signal;
Sampling means for receiving a signal from the signal converting means and sampling the signal level of the digital signal at regular intervals;
Level difference calculating means for receiving a signal from the sampling means and calculating a level difference between two consecutive sampling values;
Memory means for receiving a signal from the level difference calculation means and sequentially storing the level differences in time series;
A counter linked to the memory means for counting the number of level differences stored in the memory means;
When the signal from this counter is received and the first setting device L is counted by the counter, the number of the level differences within the setting range among the stored level differences is the second set value N. An intrusion determination means for determining whether or not
The intrusion detection system according to claim 2 , wherein the memory means stores the data table. The intrusion detection system according to claim 2 or 3, wherein the data table is arranged with correction coefficients for multiplying the upper threshold value and the lower threshold value as a reference.

Images