High-reliability fire detection method and system

Abstract

The invention discloses a high-reliability fire detection method and system, and relates to the technical field of fire alarm, and the detection method comprises the following steps: S1, setting a threshold value; s2, initializing a background value, updating the detection value at regular time, and updating the background value to initialize the background value; s3, identifying the aerosol type; s4, identifying the movement condition of the personnel; s5, identifying a high-humidity environment; s6, identifying a high-temperature environment; and S7, judging a fire alarm. The detection system comprises a threshold setting module, an initialization module, a smoke detection module, a radar system, a humidity detection module, a temperature detection module, a background value updating module and an alarm module. According to the invention, the radar system is used for judging and detecting the motion state of environmental personnel, detection is carried out in combination with factors such as smoke, temperature and humidity, and different levels of alarm thresholds are set for judgment, so that the identification capability of the fire detector on non-fire smoke is enhanced, and the false alarm probability under the non-fire smoke condition is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fire alarm, and in particular to a high-reliability fire detection method and system. Background Art

[0002] At present, the fire alarm system is an indispensable system for building facilities to ensure the fire safety of people in the building facilities. When a fire or other situations occurs in the building facilities, the supporting fire alarm system will promptly issue a fire alarm to facilitate the timely escape of people in the building facilities and reduce the risk of casualties.

[0003] In the existing fire alarm system, photoelectric smoke detectors are one of the commonly used fire triggering devices. Existing photoelectric smoke detectors mostly use dual optical path detection. By calculating the changes in the detection values ​​on different optical paths, interfering substances such as water vapor are distinguished from fire smoke, and a larger alarm threshold is set for interfering substances to improve the detector's ability to prevent false alarms for interfering substances.

[0004] However, the existing fire detection alarm system still has shortcomings: the photoelectric smoke detectors used are easily contaminated by dust, oil smoke, etc. over time. Once the pollution reaches a certain level, parameters such as the sensitivity of the detector will change, causing the detector to misjudge the fire smoke as an interfering substance, and thus fail to warn of the fire in time. In addition, factors such as smoke, temperature and humidity are easily affected by changes in environmental factors. For example, the ambient temperature and humidity will change significantly with the seasons, or due to factors such as turning on the air conditioner indoors. If the fire alarm judgment is based solely on these factors, it is easy to result in inaccurate detection and judgment results and false alarms. Summary of the invention

[0005] The purpose of the present invention is to provide a high-reliability fire detection method and system to solve the technical problem that the fire detection alarm identification method in the prior art only detects and judges factors such as smoke, temperature and humidity, which is prone to inaccurate detection and judgment results and false alarms.

[0006] The technical problem to be solved by the present invention can be achieved by the following technical solutions:

[0007] A high reliability fire detection method, applied to a dual optical path photoelectric smoke fire detector, is characterized by comprising the following steps:

[0008] S1. Set an aerosol type threshold, a humidity threshold Hg and a temperature threshold Tg; obtain a smoke ratio r, a humidity value H and a temperature value T; the aerosol type threshold is a scattering power ratio threshold rg for distinguishing aerosol categories;

[0009] S2, initialize the background value, regularly update the detection value and update the background value;

[0010] S3, determine whether the smoke ratio r is greater than the scattering power ratio threshold rg; if yes, set the alarm threshold to Th1, and then enter S7; if no, enter S4;

[0011] S4, using the radar system to detect whether there are people in abnormal motion in the environment; if yes, set the alarm threshold to Th3 and then go to S7; if no, go to S5;

[0012] S5, determine whether the humidity value H is greater than the humidity threshold Hg; if yes, set the alarm threshold to Th3, and then enter S7; if no, enter S6;

[0013] S6, judging whether the temperature value T is less than the temperature threshold value Tg, if yes, setting the alarm threshold value to Th2, and then entering S7; if no, setting the alarm threshold value to Th1, and then entering S7;

[0014] S7, determine whether the smoke detection value increment Δ is greater than the current alarm threshold, if yes, alarm, otherwise, jump to S2.

[0015] As a further solution of the present invention: the dual-path photoelectric smoke fire detector includes a smoke detection light path A and a smoke detection light path B, and the initialization of the background value in step S2 includes initializing the background value B of the smoke detection light path A. A And the background value B of the smoke detection optical path B B .

[0016] As a further solution of the present invention: the step S2 of regularly updating the detection value and updating the background value includes:

[0017] According to the real-time smoke detection value V corresponding to the smoke detection optical path A and the smoke detection optical path B A 、V B As well as the air humidity value H and the ambient temperature value T, the corresponding data are overwritten and updated to obtain the real-time detection value;

[0018] The real-time detection value and the historical background value are weighted and summed.

[0019] As a further solution of the present invention: the calculation formula of the smoke ratio r is:

[0020] r=(V A -B A ) / (V B -B B ).

[0021] As a further solution of the present invention: the calculation formula of the smoke detection value increment Δ is as follows:

[0022] Δ=V A -B A .

[0023] As a further solution of the present invention: the radar system in step S4 includes a human body recognition module, a position detection module and a counting module.

[0024] As a further solution of the present invention: the method of detecting whether there are people in abnormal motion in the environment by using a radar system comprises the following steps:

[0025] Step 1: Use the human body recognition module to determine whether there is a human body in the detection area; if not, mark the detection area as a person with no abnormal movement; if yes, jump to step 2;

[0026] Step 2: Periodically detect the position of the human body through the position detection module, and calculate the displacement difference S according to the position of the human body in adjacent items t , based on the displacement difference S t and interval time t to calculate the instantaneous speed V of the target person p , determine the instantaneous movement speed V of the target person p Is it 0? If not, jump to step 3; if yes, then calculate the instantaneous motion speed V p The static continuous time of 0 determines whether there are people in abnormal motion;

[0027] Step 3: Set the total detection time to t 总 , counting threshold X, personnel movement speed threshold V m ; Set some instantaneous motion speed V p With speed threshold V m Compare and calculate the instantaneous motion speed V p Greater than the speed threshold V m The total number of times a; when the detection time of the position detection module reaches the total detection time t 总 Finally, compare the total number of times a with the counting threshold X; if the total number of times a is greater than the counting threshold X, it is determined that the movement state of the person in the detection environment is abnormal; otherwise, it is marked as a person with no abnormal movement state.

[0028] As a further solution of the present invention: the instantaneous motion speed V p The continuous static time of 0 determines whether there are people in abnormal motion, including:

[0029] Statistics instantaneous movement speed V p The static continuous time t1 when it is 0, and the timing threshold t m ; Compare the static continuous time t1 with the timing threshold t m Compare; when the static continuous time t1 is equal to the timing threshold t m, start the high-definition cameras distributed in the detection environment to capture the person's face, perform facial feature recognition, and determine whether the person is in a conscious state. If so, mark him as a person with no abnormal movement; if not, mark him as a person with abnormal movement.

[0030] As a further solution of the present invention: the high-definition camera performs facial feature recognition of a person based on a face recognition module, and the method of judging whether a person is awake is:

[0031] Determine whether the face recognition module recognizes the eye features of the face; if not, mark the person's status as unconscious;

[0032] If yes, then further determine whether the eye feature is in a closed eyes state; if yes, then mark the person's state as unconscious; if no, then mark the person's state as awake.

[0033] A high reliability fire detection system, comprising:

[0034] A threshold setting module, used to set an alarm threshold, an aerosol type threshold, a humidity threshold, and a temperature threshold;

[0035] Initialization module, used to initialize the background value;

[0036] Smoke detection module, used to regularly update the smoke detection value;

[0037] Radar system, used to regularly update the movement status of people in the detection environment;

[0038] Humidity detection module, used to regularly update humidity detection values;

[0039] Temperature detection module, used to regularly update the temperature detection value;

[0040] Background value updating module, used for updating background value;

[0041] Alarm module, used to issue fire and smoke alarms.

[0042] Beneficial effects of the present invention:

[0043] 1. The present invention regularly updates the background value and detection value of various fire influencing factors such as smoke, temperature and humidity, and detects the movement status of people in the environment in real time through the radar system. By tracking and analyzing the positions of people in the detection environment, the movement speed of people in the detection environment can be effectively calculated, and compared with the set speed threshold, so as to facilitate the judgment whether the movement state is abnormal. When the abnormal movement state of people is detected, the alarm threshold of the corresponding level is set, and a fire alarm judgment is made. Otherwise, the temperature and humidity in the detection environment are detected, and the alarm threshold of the corresponding level is adjusted and set for judgment to avoid false alarms caused by factors such as water vapor and dust. That is, the radar system is relied on to judge the movement status of people in the detection environment, combined with the detection of factors such as smoke, temperature and humidity, and different levels of alarm thresholds are set for judgment, so as to facilitate the auxiliary enhancement of the fire detector's recognition ability for non-fire smoke, reduce the probability of false alarms in the case of non-fire smoke, and improve the product's anti-interference ability for non-fire smoke.

[0044] 2. When the present invention detects and calculates the movement speed of people in the environment through the radar system, it calculates the instantaneous movement speed for multiple times in a set time and calculates the number of times the speed threshold is exceeded. Only when the number of times the instantaneous speed of people in the environment exceeds the speed threshold reaches the set number within the set time can it be determined as an abnormal escape state, thereby avoiding short-term falls without movement speed during the escape process of people in the detection environment, resulting in inaccurate results of the judgment of the movement state of the people and affecting the fire alarm judgment result.

[0045] 3. When the radar system of the present invention detects that there is no movement of people in the environment, it will recognize and analyze the facial and eye features to determine whether the people in the environment are awake. The corresponding alarm threshold is set according to whether the non-moving people in the environment are awake, so as to effectively make fire alarm judgments and reduce the possibility of misjudgment while ensuring the safety of people. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The present invention will be further described below in conjunction with the accompanying drawings.

[0047] Figure 1 The present invention is a logic principle block diagram of a high reliability fire detection method. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] like Figure 1As shown, a high reliability fire detection method is applied to a dual-path photoelectric smoke fire detector, and can also be applied to a multi-path photoelectric smoke fire detector. Here, the dual-path is mainly used as an example for explanation. The dual-path photoelectric smoke fire detector includes a smoke sensing module. The detection method specifically includes the following steps:

[0050] S1. Set the fire alarm threshold Th, aerosol type threshold, humidity threshold Hg and temperature threshold Tg; the alarm threshold Th is divided into three levels of thresholds, namely, the first threshold Th1, the second threshold Th2, and the third threshold Th3, and the three thresholds are from large to small, Th1>Th2>Th3; and obtain the smoke ratio r, humidity value H, and temperature value T; the aerosol type threshold is the scattering power ratio threshold rg used to distinguish the aerosol category, where aerosol scattering refers to the scattering effect of aerosol particles on light. When light waves encounter aerosol particles, scattering will occur, that is, the direction and intensity of the light waves will change, and the scattering power ratio refers to the relative size of the scattered light power of aerosol particles in different directions or different categories; the humidity threshold Hg is the threshold used to judge whether the ambient humidity reaches a high humidity environment; the temperature threshold Tg is the threshold used to judge whether the ambient temperature reaches a high temperature environment.

[0051] S2. Initialize background value. It should be noted that the dual-path photoelectric smoke detector includes a smoke detection path A and a smoke detection path B. The initialized background value includes the background value of the initialized smoke detection path A and the background value of the smoke detection path B. The background value of the initialized smoke detection path A is marked as B. A Initialize the background value of smoke detection optical path B as B B ; The background value refers to the basic or stable state value of a system or device under specific conditions when there is no external interference or specific target objects. Therefore, the initialized background value here refers to the detection value detected by the smoke module of the dual-path photoelectric smoke detector in a smoke-free state, that is, the most basic original value.

[0052] S3, regularly update the detection value; the smoke sensing module of the dual-path photoelectric smoke fire detector regularly collects the photoelectric signal strength of the smoke detection light path A and the smoke detection light path B, and converts it into a real-time smoke detection value, which is marked as V A , V B ; The radar system is used to regularly detect the movement of people, and the radar system detects the signal R; the humidity module regularly collects the resistance of the humidity sensor, and converts the air humidity value H according to the resistance-humidity relationship curve; the temperature module regularly collects the resistance of the thermistor and converts it into the temperature value T, and regularly updates the detection value, that is, the value V is regularly updated. A 、V B , R, H, T, and obtain real-time detection values.

[0053] S4, regularly update the background value; the background value here includes the background value B of the smoke detection optical path A A and the background value B of the smoke detection optical path B B , periodically obtain the background value B of the smoke detection optical path A A and the background value B of the smoke detection optical path B B The update method is once every 4 hours. The real-time detection value and the historical background value are weighted and summed. The weight of the real-time detection value is 1 / 8, and the weight of the historical background value is 7 / 8. The historical background value is the last background value obtained when the background value is updated regularly.

[0054] S5, identify the aerosol type; calculate the smoke ratio r, and determine whether the smoke ratio r is greater than rg. If yes, set the alarm threshold to Th1, and then enter S9; if no, enter S6; the calculation formula of r is:

[0055] r=(V A -B A ) / (V B -B B ).

[0056] It should be noted that, since the dust concentration in the environment is prone to be high and is relatively common, a maximum level alarm threshold Th1 is set to prevent misjudgment.

[0057] S6, identifying the movement of people; using the radar system to detect whether there are people in abnormal movement status in the environment. Here, people in abnormal movement status refer to those whose movement speed in the detection environment is greater than a certain value within the set detection time and are in a state of fleeing. If yes, set the alarm threshold to Th3 and then enter S9; if not, enter S7; the radar system detects the movement of people by sending and receiving electromagnetic waves.

[0058] The radar system includes a human body recognition module, a position detection module and a counting module. The radar system detects whether there are people in abnormal motion in the environment in the following manner:

[0059] Step 1: Identify whether there is anyone in the detection environment through the human body recognition module; if not, mark the detection area as a person without abnormal motion; if so, the radar system continuously tracks the dynamic position of the target person based on the target tracking algorithm, and establishes a three-dimensional coordinate system based on the space where the detection environment is located, and projects the position of the target person in the three-dimensional coordinate system. Based on the coordinate position of the target person and through the phased array technology, the direction of the electromagnetic beam emitted by the position detection module is adjusted in real time, so that the target person is always within the coverage of the beam. Here, the phased array technology controls the phase and amplitude of each electromagnetic wave radiation unit through a computer, so that these units can synthesize main beams with different phases and perform phase changes in both axes, thereby realizing precise control of the signal space beam; then jump to step 2;

[0060] Step 2: The position detection module periodically transmits electromagnetic waves to the target person to be tracked, then receives and analyzes the reflected electromagnetic waves, and detects the location of the target person through FMCW waveform processing technology. The electromagnetic waves are sent and received once every interval t to detect the location of the target person. Here, the interval t can be set to seconds, such as 2 seconds or 3 seconds, and the total detection time is t 总 , t 总 It can be 30 seconds; and the displacement difference S is calculated based on the two consecutive locations of the target personnel t , then based on the displacement difference S t and interval time t to calculate the instantaneous speed V of the target person p =S t / t;

[0061] The above FMCW waveform processing technology is: frequency modulating radar, so that it transmits a continuous wave with a linearly changing frequency, and receives the signal reflected back. By measuring the frequency difference between the transmitted signal and the reflected signal, the distance and speed of the target can be calculated;

[0062] Step 3: The radar system also includes a comparison module and a counting module to set the speed threshold V of the person moving. m , the instantaneous motion speed V obtained each time p With V m By comparing the module, the instantaneous motion speed V is calculated. p Greater than the speed threshold V m The total number of times a, and set the counting threshold X; if V p Greater than V m , the counting module adds 1 to the total number of times a; V p Less than V m When the total detection time t is reached, no count change occurs. 总After that, the accumulated total number of times a is compared with the counting threshold X through the comparison module. If a is greater than X, it is determined that the state of the personnel in the detection environment is an abnormal escape state, and the environment is very likely to have a fire, so the alarm threshold is set to Th3, and then enter S9; if a is less than X, it means that the personnel in the environment only occasionally run, then it is determined that the movement state of the personnel in the detection environment is normal, there is no abnormal movement state personnel, and the possibility of fire in the environment is small, and enter S7. The counting method is used here mainly to avoid the situation that when a fire occurs, the personnel fall down due to panic during the escape process, and the system misjudgments due to a short period of no movement.

[0063] It should be noted that if the instantaneous speed V of the target person is calculated in step 2 p =0, the radar system uses the set timing module to independently time the static continuous time t1, and sets the timing threshold t m , calculate V several times in a row p =O, the timing module keeps timing status. If V p ≠O, t1 returns to zero. If t1=t m , start the high-definition cameras distributed in the detection environment to capture the faces of the people and perform facial feature recognition. In order to ensure the effect of capturing the faces of the people, multiple high-definition cameras can be installed at multiple angles in the detection environment to determine whether the people are in a sober state. If so, it means that there are no people in abnormal motion in the detection environment, and the possibility of fire in the environment is small, and enter S7; No, it means that there are people in abnormal motion in the detection environment, and the alarm threshold TH1 is set, and enter S9. Here, when it is determined that the person is in a non-conscious state, because the person may be in a resting state, such as a lunch break, or may be in a coma due to the occurrence of harmful substances caused by the burning of the fire, etc., but for the sake of life safety, in either case, to prevent fire from occurring when the person is in a non-conscious state and causing serious casualties, directly enter S9 to determine the fire alarm, but in order to reduce the probability of misjudgment, the alarm threshold is set to TH1, that is, the judgment level is increased.

[0064] It should also be noted that the high-definition camera recognizes the facial features of a person based on the face recognition module. When judging whether a person is in a conscious state, the specific method is: the face recognition module is used to recognize the facial image of the person taken by the high-definition camera and extract the eye features of the face. If the eye features are extracted, the eye state is identified to determine whether the eyes are open or closed. If the eyes are open, the person is judged to be conscious; if the eyes are closed, the person is judged to be unconscious; if the eye features are not extracted, the person is also judged to be unconscious.

[0065] S7, identify high humidity environment; the humidity module regularly collects the resistance value of the humidity sensitive resistor, and converts the air humidity value H according to the resistance value and humidity relationship curve, and determines whether the humidity value H is greater than Hg. If so, the alarm threshold is set to Th3, and then enters S9. It should be noted that the humidity value H exceeds the humidity threshold Hg, which means that the detection environment is very likely to trigger the indoor fire sprinkler equipment due to fire. When the humidity threshold Hg is set high, the general environmental humidity is difficult to exceed the humidity threshold Hg. Therefore, if the humidity threshold Hg is exceeded at this time, the possibility of fire is relatively high, so the lower level alarm threshold is set to Th3, which is convenient for judging fire alarms; if H is less than Hg, enter S8.

[0066] S8, identify high temperature environment; regularly collect thermistor resistance value, and convert the temperature value T according to thermistor relationship curve, determine whether the temperature value T is less than Tg, if yes, set the alarm threshold to Th2, and then enter S9; otherwise, set the alarm threshold to Th1, and then enter S9.

[0067] S9, judge fire alarm; judge whether the smoke detection value increment Δ is greater than the current alarm threshold, if yes, alarm, otherwise return to S3. The calculation formula of Δ is as follows:

[0068] Δ=V A -B A .

[0069] The invention also discloses a high-reliability fire detection system, comprising: a threshold setting module, an initialization module, a smoke detection module, a radar system, a humidity detection module, a temperature detection module, a background value update module and an alarm module.

[0070] A threshold setting module, used to set an alarm threshold, an aerosol type threshold, a humidity threshold, and a temperature threshold;

[0071] Initialization module, used to initialize the background value;

[0072] Smoke detection module, used to regularly update the smoke detection value;

[0073] Radar system, used to regularly update the movement status of people in the detection environment;

[0074] Humidity detection module, used to regularly update humidity detection values;

[0075] Temperature detection module, used to regularly update the temperature detection value;

[0076] A background value updating module, which is used to update the background value; and

[0077] Alarm module, used to issue fire and smoke alarms.

[0078] The embodiments of the present invention are described in detail above, but the embodiments of the present invention are not limited thereto and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A high reliability fire detection method, applied to a dual optical path photoelectric smoke fire detector, characterized in that: The following steps are involved: S1. Set an aerosol type threshold, a humidity threshold Hg and a temperature threshold Tg; obtain a smoke ratio r, a humidity value H and a temperature value T; the aerosol type threshold is a scattering power ratio threshold rg for distinguishing aerosol categories; S2, initialize the background value, regularly update the detection value and update the background value; S3, judging whether the smoke ratio r is greater than the scattering power ratio threshold rg; If yes, set the alarm threshold to Th1 and then go to S7; If not, then go to S4; S4, using the radar system to detect whether there are people in abnormal motion in the environment; if yes, setting the alarm threshold to Th3, and then entering S7; If not, then go to S5; S5, determine whether the humidity value H is greater than the humidity threshold Hg; if yes, set the alarm threshold to Th3, and then enter S7; If not, then go to S6; S6, judging whether the temperature value T is less than the temperature threshold value Tg, if yes, setting the alarm threshold value to Th2, and then entering S7; if no, setting the alarm threshold value to Th1, and then entering S7; S7, determine whether the smoke detection value increment Δ is greater than the current alarm threshold, if yes, alarm, otherwise, jump to S2.

2. A high reliability fire detection method according to claim 1, characterized in that: The dual-path photoelectric smoke fire detector includes a smoke detection light path A and a smoke detection light path B. Initializing the background value in step S2 includes initializing the background value B of the smoke detection light path A. A And the background value B of the smoke detection optical path B B .

3. A high reliability fire detection method according to claim 2, characterized in that: The step S2 of regularly updating the detection value and updating the background value includes: According to the real-time smoke detection value V corresponding to the smoke detection optical path A and the smoke detection optical path B A 、V B As well as the air humidity value H and the ambient temperature value T, the corresponding data are overwritten and updated to obtain the real-time detection value; The real-time detection value and the historical background value are weighted and summed.

4. A high reliability fire detection method according to claim 3, characterized in that: The calculation formula of the smoke ratio r is: r=(V A -B A ) / (V B -B B )。 5. A high reliability fire detection method according to claim 3, characterized in that: The calculation formula of the smoke detection value increment Δ is as follows: Δ=V A -B A 。 6. A high reliability fire detection method according to claim 1, characterized in that: The radar system in step S4 includes a human body recognition module, a position detection module and a counting module.

7. A high reliability fire detection method according to claim 6, characterized in that: The method of detecting whether there is a person in abnormal motion in the environment by using a radar system comprises the following steps: Step 1: Use the human body recognition module to determine whether there is a human body in the detection area; if not, mark the detection area as a person with no abnormal movement; if yes, jump to step 2; Step 2: Periodically detect the position of the human body through the position detection module, and calculate the displacement difference S according to the position of the human body in adjacent items t , based on the displacement difference S t and interval time t to calculate the instantaneous speed V of the target person p , determine the instantaneous movement speed V of the target person p Is it 0? If not, jump to step 3; if yes, then calculate the instantaneous motion speed V p The static continuous time of 0 determines whether there are people in abnormal motion; Step 3: Set the total detection time to t 总 , counting threshold X, personnel movement speed threshold V m ; Set some instantaneous motion speed V p With speed threshold V m Compare and calculate the instantaneous motion speed V p Greater than the speed threshold V m The total number of times a; when the detection time of the position detection module reaches the total detection time t 总 Finally, compare the total number of times a with the counting threshold X; if the total number of times a is greater than the counting threshold X, it is determined that the movement state of the person in the detection environment is abnormal; otherwise, it is marked as a person with no abnormal movement state.

8. A high reliability fire detection method according to claim 7, characterized in that: According to the instantaneous motion speed V p The continuous static time of 0 determines whether there are people in abnormal motion, including: Statistics instantaneous movement speed V p The static continuous time t1 when it is 0, and the timing threshold t m ; Compare the static continuous time t1 with the timing threshold t m Compare; when the static continuous time t1 is equal to the timing threshold t m , start the high-definition cameras distributed in the detection environment to capture the person's face, perform facial feature recognition, and determine whether the person is in a conscious state. If so, mark him as a person with no abnormal movement; if not, mark him as a person with abnormal movement.

9. A high reliability fire detection method according to claim 8, characterized in that: The high-definition camera recognizes the facial features of a person based on the face recognition module, and the method for determining whether a person is awake is as follows: Determine whether the face recognition module recognizes the eye features of the face; If not, the person's status is marked as non-conscious; If yes, then further determine whether the eye feature is in a closed-eye state; if yes, then mark the person's state as unconscious; If not, the person's status is marked as awake.

10. A high reliability fire detection system, used to implement the high reliability fire detection method according to claim 1, characterized in that: include: A threshold setting module, used to set an alarm threshold, an aerosol type threshold, a humidity threshold, and a temperature threshold; Initialization module, used to initialize the background value; Smoke detection module, used to regularly update the smoke detection value; Radar system, used to regularly update the movement status of people in the detection environment; Humidity detection module, used to regularly update humidity detection values; Temperature detection module, used to regularly update the temperature detection value; Background value updating module, used for updating background value; Alarm module, used to issue fire and smoke alarms.

Images