426 results about "Fire smoke" patented technology

To provide a security camera system for displaying a picture so as to easily know a place where an abnormal situation occurs. The security camera system includes a security camera 61 having at least more than one rotation axis and a controller 70 for controlling the security camera 61. The controller 70 is provided with a moving picture detector 80 for detecting a motion from pictures taken by the security camera during an automatic monitoring operation. The controller switches an operation of the security camera from automatic monitoring to still monitoring in the case when the moving picture detector detects a motion from monitored pictures. Therefore, in the case when a suspicious person, fire smoke, or the like is detected in a subject building at night, for example, during an automatic monitoring operation for shooting target monitoring places sequentially, a shooting point of the security camera is fixed there so as to shoot and display an abnormal situation on a monitor screen.

A monitoring system has integrated exit signs, which preferably have electroluminescent elements. The signs connect to internal wiring of a building, but local power sources can provide emergency power if building power is lost. Controllers on the signs communicate with a central workstation using building wiring and / or wireless communication. The controllers have one or more automated features for monitoring operation of the exit signs and the surrounding environment. These automated components, including an intensity sensor, an ambient light sensor, a temperature sensor, a smoke detector, a camera, a speaker, a microphone, a motion detector, an RFID detector, and the like. Because the signs are widely distributed throughout the building, operators can get detailed information of the building environment, security, fire, smoke, temperature, etc. The exit signs can store this information locally and can communicate useful information using a communication link to the central workstation.

The invention provides a real-time video field fire smoke detection method based on a convolutional neural network. A smoke image data set is collected through an experimental simulation mode, and a training set, a test set and a verification set are created; the training set, the test set and the verification set are subjected to automatic annotation, and in combination of manual adjustment, thetraining set, the test set and the verification set with a real label are obtained respectively; the training set and the verification set with the real label are subjected to image rotation processing, color channel color addition and subtraction processing and scaling processing to obtain the processed training set and the processed verification set with the real label; the parameters of the convolutional neural network are initialized, and according to the training set with the real label after scaling processing, a well-built convolutional neural network model is trained; a to-be-detectedfield monitoring image is acquired in real time, and through the trained convolutional neural network model, a smoke target detection frame is predicted and optimized; and inter-frame confidence enhancement and relocation are carried out on the target detection result given by the trained convolutional neural network model.

The invention discloses a method for recognizing video fire smoke image based on optical flow, which is characterized in that: optical flow of corners and points in smoke motion regions is extracted by a computer based on the video image obtained by a surveillance camera; whether the smoke is fire smoke is judged based on the set threshold by calculating the average and variance of the array consisting all corner and point optical flow speed in the foreground image and the average and variance of the array consisting optical flow speed directions; if a fire smoke is identified, the computer sends orders which controls the alarm to send alarm signals and controls the interlocked fire extinguisher to start the fire extinguishing function quickly; if a fire smoke is not identified, the first step is returned to. The method provided by the invention can not only accurately reflect the motion characteristics of smoke, reducing light interference and impact of similar pixels on the results, but also reduce computation and improve practicality. The method can greatly reduce false alarm rate and quickly and accurately realize fire smoke detection function.

The invention relates to an aerosol grain size sensing method based on dual-wavelength scattered signals and an application of the method to fire smoke detection, and belongs to the technical field of fire alarms. The method comprises steps as follows: calculating the ratio R of the blue light scattered power to the infrared light scattered power after receiving corresponding scattered signals reflected by aerosol with the blue light scattered power PBL and the infrared light scattered power PIR; determining the median grain size dmed according to the relation between the ratio R of the blue light scattered power to the infrared light scattered power and the median grain size dmed of aerosol; comparing the blue light scattered power PBL and the infrared light scattered power PIR with set corresponding thresholds PBLth and PIRth, and sending corresponding interference prompt signals or corresponding firm alarm signals. With the adoption of the method, fire type alarm signals can be judged and sent, so that targeted reasonable measures can be taken, and meanwhile, false alarms caused by aerosol under non-fire conditions can be avoided.

The invention discloses a fire scene multi-parameter laser wavelength modulation spectrum detection method and a device thereof, wherein, fire smoke products are treated through a smoke pre-treatment device and then pumped into a white cell through an air extracting pump; a multi-channel laser controller scans and modulates the wavelengths of a plurality of DFB lasers according to the time-sharing and multiplexing working mode, and all waves of laser share an optical fiber and are outputted sequentially through a wave combiner, then are aligned by a fiber collimator and sent to the white cell to detect the fire smoke products; a photoelectric detector converts light intensity signals of laser repeatedly reflected and absorbed by the white cell into electrical signals which are sent to two phase-locking magnifier modules for frequency selecting magnification so as to obtain a fundamental frequency component and a second harmonic component; a multi-channel data acquisition card converts output voltage signals of the phase-locking magnifiers into digital signals which are sent to a micro-computer for real-time data processing, so that the fire scene multi-parameter information containing the oxygen concentration of the fire scene, concentrations of various toxic gases and smoke concentration is obtained. The detection method and device can realize real-time on-line monitoring to fire scene multi-parameters, and have the advantages of being real-time, multi-component and highly sensitive, good selectivity of gas, high reliability and strong environmental disturbance resistance ability.

The invention discloses a method for detecting fire disaster smog by video on the basis of color saturation and motion mode, which comprises the following steps: firstly, extracting a foreground motion block from a video image acquired by a monitoring camera through a difference method; secondly, detecting the color saturation of the foreground motion block, estimating the direction of the motion block, and calculating the cumulant and the main moving direction of the motion block; and thirdly, calculating the color saturation detection percentage of each region, the average cumulant and the main moving direction ratio to form a characteristic vector, and adopting a Bayesian classifier to judge whether the smog is the fire disaster smog. The detection of the color saturation reflects a rough color distribution of the smog and eliminates a large amount of foreground interfering objects with glowing colors. The color detection in a blocking mode further improves the anti-interference performance of a system. The cumulant shows the characteristic of continuous movement nearby a smoldering point of the smog, and has very good anti-interference performance for a non-reciprocating object. The comprehensive utilization of the color saturation detection percentage, the average cumulant and the main movement ratio can greatly reduce the rate of false alarm for the system.

The labyrinth includes probing chamber composed of labyrinth cover and labyrinth base, as well as forward scattering light path and back scattering light path constituted from transmitting tube (3), transmitting tube (6) and receiving tube. Dual scattering light paths including two transmitting paths and one receiving path are adopted in the invention. Using characteristics of different responses on particles in different colors from forward or back scatterings, the method recognizes smog in different colors and grain sizes in order to distinguish not fire alarm factors such as water spray, dust etc so as to raise responsibility of detector, and lower rate of false alarm. Comparing prior art, the invention possesses advantages of superior orientation responsibility, reasonable design, balanced responses for smog in different colors and grain sizes, and lower cost. The invention is applicable to spot type optical detector of fire smoke in auto fire alarm system.

A forest fire smoke detection method using random forest classification is provided. In the method, a first reference value is set. For consecutively captured frames, images between the frames are compared, each block, in which a number of pixels, motions of which have been identified, is equal to or greater than the first reference value, is set as a candidate block, and a keyframe is selected. The selected keyframe is compared with at least one frame previous to the keyframe and then a plurality of feature vectors are extracted from the candidate blocks. The extracted feature vectors are learned using different random forest algorithms. Probabilities output to terminal nodes for classes are accumulated, and two first cumulative probability histograms are generated. The two first cumulative probability histograms are averaged, and then a second cumulative probability histogram is generated. A detected state of each candidate block is determined.

The invention discloses an active infrared video based fire smoke detection method in zero-illumination environment. The method comprises the steps of: acquiring active infrared video image data; extracting foreground images from the infrared video image data by a difference method and carrying out intensity value analysis; extracting and analyzing the speed characteristics of clustering centers in the foreground images; calculating an average value and a variance of the speed characteristics of the clustering centers, and comparing a preset threshold value with the calculated variance of the speed characteristics of the clustering centers to judge whether fire-disaster smoke exists in the monitored and limited zero-illumination space or not; if so, giving alarming signals; and if not, returning to the former step to acquire infrared video image data of next frame continuously. In the method, the influence of the background noise of infrared video images is reduced and the active infrared video smoke detection is realized in the zero-illumination environment.

The invention provides an experiment platform for simulation of control over fire smoke at key connection positions of a subway station. The experiment platform for simulation of control over the fire smoke at the key connection positions of the subway station comprises a simulated station hall layer, a simulated platform layer, a simulated tunnel rail traveling layer, a simulated subway train, a ventilation and smoke discharge system, an air curtain system and a fire parameter measurement system. By the adoption of the experiment platform, study on a series of basic theories such as the flowing characteristics of fire smoke, generated when a fire disaster occurs on a subway station, at the key positions where a platform, a station hall and a tunnel rail traveling area are connected and control mode optimization and engineering practical problems can be conducted; a complete solution to study on the evolvement rule of the characteristic parameters such as the effectiveness of a technology for controlling the fire smoke at the key connection positions of the subway station, the temperature field and the poisonous gas concentration field in the scene that fire disaster occur on subway train fire-prevention parking stations and a subway station platforms is provided.

The invention provides a multifunctional patrol robot with an independent map building function and an independent wireless charging function, and belongs to the technical field of robots. The robot comprises a robot body and a movable chassis, wherein the robot body is arranged on the movable chassis; a Kinect sensor, a horn, a combustible gas sensor and a fire smoke sensor are arranged on the outer part of the robot body; the movable chassis is moved by dual motors arranged in the movable chassis; and the robot body internally comprises a main control module, a network and wireless communication module, a motor drive module, an independent wireless charging module, a voice playing module, an intelligent identifying module, a combustible gas alarm module, a fire smoke alarm module and the like. The robot has the independent map building function and the self-charging function, and can independently go on patrol and work for a long time in an unfamiliar environment; and the robot is combined with internet of things, so that the service of the robot is increased; and the robot is more intelligent and user-friendly by virtue of a navigation module and the independent wireless charging module.

The invention discloses a forest fire smoke detect method based on colorful CCD image analysis, which comprises that based on some front frames of obtained image, finding the horizon line in the image to represent the spatial precision information in the image, based on Mallat method, analyzing the high-frequency information loss character, using fusion algorism to fuse the high-frequency information loss character, smoke low-frequency motion character, and color character, to be accumulated on time axis. Based on the position information and accumulated result of horizon line, the invention extracts smoke area under spatial precision compensation, to mark the smoke source area and smoke area. The invention can improve the reliability of smoke image detection, and reduce the undetected rate and the error rate of smoke detection, with significant benefits for forest fire automatic detection.

The invention discloses a ventilation, smoke discharge and evacuation system for an underwater single-tube double-layer shield traffic tunnel. The ventilation, smoke discharge and evacuation system comprises an upper-layer lane, a lower-layer lane, a jet fan and a wind tower. An axial flow fan is arranged in the wind tower. A sector space on one side face of a round shield section of the tunnel is an independent smoke discharge channel. An escape channel is arranged in a sector space on the other side of the round shield section of the tunnel. A staircase connected with the upper layer and the lower layer is arranged in the escape channel. An upper-layer front room is arranged between the staircase and a lower-layer escape door. A lower-layer front room is arranged between the staircase and an upper-layer escape door. Positive pressure fans are installed on the wall shared by the upper-layer front room and the staircase and the wall shared by the lower-layer front room and the staircase respectively. The upper end of the side wall of the tunnel is provided with an upper-layer smoke outlet and a lower-layer smoke outlet. The upper-layer smoke outlet and the lower-layer smoke outlet are respectively provided with an automatic smoke discharge fire damper. By means of the ventilation, smoke discharge and evacuation system, fire smoke can not diffuse between the upper layer and the lower layer, the visibility of the escape environment is good, the escape environment is not invaded by the smoke, personnel evacuation is efficient and quick, escape time is greatly shortened, personnel casualties are reduced, construction is safe, and later-period maintenance is easy.

The invention discloses a mine fire escape route cellular automaton optimizing and guiding method, and belongs to mine fire safe escape methods. An underground monitoring subsystem, a smoke plume migration simulation subsystem, an escape route selection subsystem and a rescue and escape command guiding subsystem are included, and the monitoring system is composed of a ground central station and various underground sensors and monitors temperature, gas and flue gas parameters in real time; the smoke plume migration simulation system simulates dynamic evolution rules of underground fire smoke through FDS software; the escape route selection subsystem applies a cellular automaton for building a mathematical physical model under the mine ventilation network structure actual condition, dynamically quantizes the complex roadway environment and the smoke pollution range, calculates the escape efficiency and determines the best escape route; the rescue and escape command guiding subsystem provides guidance for underground escape persons on the basis of the escape route calculation result, and the best escape route is fed back to underground workers by means of an upper computer and an underground voice alarm system, the escape efficiency is comprehensively improved, and casualties are reduced to the maximum extent.

The invention discloses an intelligent household control system and method. The system comprises an infrared receiving part, an infrared signal emitting control air conditioner, a fire smoke sensor, a combustible gas sensor, a brightness and humidity and the like analog signal expanding acquirer, a GSM (Global System for Mobile Communications) network signal command, a TC35 module, an amplifying converter, an amplifying shaper, a voltage converter, an RS232 serial communicator, a PC, an LED liquid crystal display system, a photoelectric isolator, a switching device, a window motor, a smoke exhaust ventilator, a hall map, a stair lamp, a sound system controller, a power controller, an electric water heater, an electric cooker executing device and an MCU (Microprogrammed Control Unit) minimum master control system; the infrared receiving part is used for receiving a 38KH infrared signal from a demodulating remote controller and transmitting to the MCU controller by the external interrupting mode, then the 38KH infrared signal is decoded through the software and used for recognizing the key value of a remote controller. The intelligent household control system has the characteristics that the whole structure is small and convenient to use; the intelligent household control system is applicable to popularization and application.

The invention provides a forest fire smoke recognizing method and device. The method includes the steps of recognizing smoke pixels and non-smoke pixels by means of a multi-channel threshold method and the reflective rates or the brightness temperature values of 36 spectrum channels, obtaining input characteristics of a neural network classifier from the reflective rates or the brightness temperature values of the 36 spectrum channels, training the neural network classifier by means of smoke input characteristics of the smoke pixels and non-smoke input characteristics of the non-smoke pixels, obtaining a smoke recognizing and classifying unit, and precisely recognizing the smoke pixels and the non-smoke pixels by means of the smoke recognizing and classifying unit. The method achieves precise classification of the smoke pixels, and the probability that cloud pixels are mistaken for the smoke pixels is greatly reduced due to the fact that the non-smoke pixels include the cloud pixels. Besides, due to the fact that the smoke pixels include thin-smoke pixels, the thin-smoke pixels can be precisely recognized by means of the smoke recognizing and classifying unit, and precision of thin-smoke area recognition is further improved.

The invention provides a tunnel fire extinguishing system based on water curtain shielding effects. The tunnel fire extinguishing system comprises a fire detection system; the fire detection system includes a fire detector and a fire alarm controller; the fire detector is in communication connection with the fire alarm controller. A water curtain fire extinguishing system comprises a water supply system and multiple sets of water curtain nozzle sections arranged along the longitudinal interval of the tunnel; the water supply system consists of deluge valves units corresponding to all of the sets of water curtain nozzle sections. The deluge valves units are in communication connection with the fire alarm controller. The sets of water curtain nozzle sections comprise multi-row water curtain nozzles installed along the cross section of the tunnel. The fire detector is arranged on the vault or the side wall of the tunnel which is between the adjacent sets of water curtain nozzle sections. According to the tunnel fire extinguishing system, the water curtain nozzle sections produce a full section dense water curtain with a certain thickness on both sides of the fire source and form a water curtain partition section to prevent the spread of fire smoke and to block the air supply on both sides of the water curtain, and finally to achieve the effects of controlling the fire smoke in the tunnel and extinguishing the fire.

The invention relates to a video detection method for early smoke of field fire, and the method comprises the following steps: 1, carrying out the decompression of a video shot, and obtaining a framesequence; 2, carrying out the filtering of an original image through a 5*5 mean filter, and removing a noise point through a morphological method; 3, extracting fire smoke candidate regions through the movement characteristics of fire smoke, and extracting a movement foreground through a Gaussian hybrid model; 4, extracting fire smoke candidate regions for the second time through the static characteristics of the early smoke color of the field fire; 5, extracting fire smoke candidate regions for the third time through the direction characteristics of movement in a smoke diffusion process; 6, carrying out the recognition of the fire smoke candidate regions through a convolution neural network, wherein the convolution neural network is designed to have one input layer, three convolution layers, three pooling layers, one full-connection neural network layer and one output layer. The method is high in recognition rate, and facilitates the popularization and application.

The invention discloses a remote monitoring system for a wireless fire smoke sensing alarm, which is characterized by comprising a monitoring device (1) at a monitoring center, a fire alarm monitoring host machine (2), a wireless aural and visual alarm device (3), a wireless manual alarm button (4) and a wireless smoke-sensing fire detection alarm (5). The remote monitoring system has the advantages of flexible construction, convenience for installation and debugging, small construction quantity, great shortening in construction period, convenience for maintenance and service, reduction in maintenance cost, simple operation, easiness and convenience for installation, high sensitivity, safety and reliability.

The invention discloses an intelligent smoke stove linkage control system. The system comprises a range hood and a cooker, the cooker comprises a position sensor and an ignition control unit, the range hood comprises a motor, a smoke sensor and a drive control unit, the motor is connected with the drive control unit, and the drive control unit and the ignition control unit are connected by wireless. The driving control unit obtains the firepower value and the smoke concentration value, determines the change of the firepower value, and selectively adjusts the smoke concentration value accordingto the change of the firepower value, and sets the smoke extraction gear according to the change of the firepower value and the smoke concentration value. The linkage control system of the inventionadjusts the state of the range hood according to the firepower of the cooker and the smoke, and realizes the linkage control of the smoke extraction gear, the firepower of the cooker and the concentration of the smoke, and solves the small fire and heavy smoke problem in the actual operation. The problem that the fire smoke and smoke change concentrations of the cooker in the actual operation aredifferent, so that the adjustment of the smoke extraction gear is not match the actual smoke concentration can be solved.

The invention discloses an infrared light beam type fire smoke detector and a detection method thereof. The detector comprises a MCU core circuit, a power supply circuit, a clock circuit, a keyboard, a display interface, a local communication circuit, a remote communication circuit, a switching value input circuit, a switching value output circuit, an acousto-optical alarm circuit, an analog signal acquisition module and an infrared transmitting and receiving circuit. The detection method is a fuzzy neural network fire detection algorithm, and a fuzzy neural network model adopts a four-layer structure which respectively represents an input layer, division of a fuzzy subspace of the input variable, division of a fuzzy subspace of the fire occurrence probability U, and an output layer; and the generalization ability of the network is improved through the learning and training of the fuzzy neural network, and the trained fuzzy neural network can accuracy judge the current input. The invention can reduce the failure rate when improving the flexibility of the detector simultaneously, has strong anti-interference ability, and greatly improves the reliability of the fire detection system.

The invention discloses an effective smoke-stop of fire hazard air screen system with applied power in order to prevent smoke or set up smoke-stop partition which utilizes the properties of air screen effectively preventing smoke and integrates the characters of building provided with fan drive. It includes fan drive, ducts, valves, jet equipment. The fan drive is installed in the front of building duct entrance and the jet equipment is installed on smoke-stop point; the jet equipment includes jet box, the inner backflow box with static pressure and guide plate, the jet orifice lies on the bottom of jet box and is connected with the outlet of the backflow box with static pressure, the valves are installed on entrance of the backflow box and connected with ducts; the fan drive blows out to the ducts, after the wind from valves to the backflow box with static pressure is equably pressed, it is ejected over the guide plate through the jet entrance at 15-20m / s velocity at 25-35o angle with droop line and inclining smoke zone at 2.5-4.5cm thickness from the top down, then forms air screen so as to prevent fire hazard smoke from diffusing to safety zone.

The invention discloses a fireproof door core plate sheet material. The fireproof door core plate sheet material is prepared from the following components in parts by weight: 5-30 parts of reinforced fibers, 2-10 parts of bentonite, 6-30 parts of an adhesive, 10-25 parts of inorganic gel material, 1-12 parts of an adhesion agent, 1-10 parts of fireproof expansion blocking material, 0.2-2 parts of a waterproof agent and 150-260 parts of water, wherein the adhesion agent has the effect of promoting the compatibility and the foaming effect of all the raw materials and the adhesive and the inorganic gel material are used for keeping generated foam; and the reinforced fibers have the effect of endowing the materials with certain flexibility and good mechanical strength. When the fireproof expansion blocking material suffers from a fire, the fireproof expansion blocking material expands to form a ceramic carbon layer to effectively block cracks which are formed when the sheet material suffers from the fire, and thus the fireproof door core plate sheet material has good heat insulation and fire smoke blocking performances. A preparation method of the fireproof door core plate sheet material is simple and convenient, is easy to operate and can be easily put into industrial production. The fireproof door core plate sheet material has the characteristics of small volume-weight, long fireproof time, no moisture absorption, no halogenation and the like.

The invention provides a fire smoke flow field tracer method which comprises a fire source analog system, a laser sheet optical system, an image pickup system, a smoke particle analog method and a tracer grain adding method; wherein the fire source analog system adopts a propane gas burner with controllable gas flow rate; smoke particles generated by burning sandalwood are adopted to simulate firesmoke particles; the grain diameter of the tracer grain is 1-5mu m of boron nitride grain; the boron nitride grains are evenly and stably transmitted into a flue gas layer by an air compressor; the method can be widely applied to small-scale and full-scale tests of fire, so as to lead the process of fire test to be safe, controllable and environment-friendly; by adopting the method, the key parameters such as the thickness, the form, the flow field structure and the like of the fire flue gas layer can be obtained, thus having important significance for guiding fire smoke control.

The invention discloses a forest fire smoke video target detection method based on characteristic roots and hydromechanics. The method comprises: firstly, extracting a dynamic region from a video by extracting continuous frames of image from the video; secondly, using a morphological algorithm to calculate the connected region skeleton image of the dynamic region; then extracting suspicious smokeroot characteristic candidate points in the endpoints of continuous frames of skeleton image; subjecting the smoke root characteristic candidate points to determination based on a Navier-Stokes equation to obtain the predicted smoke image of the smoke root characteristic; and finally fusing and determining the predicted image and a current actual image to obtain a smoke area. Because the algorithmhas certain predictability and can continuously affect the decision information of continuous frames, a determined result is stable and accidental calculation errors caused by an image processing stage can be avoided.

The invention requests for a system and a method for predicting fire fume characteristics to protect buildings and relates to predicting of fire fume development and evolution laws in high-rise buildings and multi-layer and multi-room buildings. The invention constitutes a complete building fire fume characteristic predicting system and a fire fume characteristic predicting method. A building is transformed into a building ventilation tree composed of nodes and branches so as to construct a junction matrix and a closed-loop matrix for ventilation circuits of the building; a fume characteristic equation is composed and solved to plot the relationship of time and concentration changes of CO2 and CO, then the time to reach the dangerous concentration is predicted. The system can predict the structure changes of the building during the development of the fire so as to make out forced ventilation proposals and smoke control and exhaust proposals for each part in the building and realize performance-based design of smoke prevention and exhaust systems and safety evacuation of the building with the maintenance of the fire safety.

The invention discloses a fire ventilation and smoke exhaust experimental simulation device based on an urban underground traffic linkage tunnel system. The fire ventilation and smoke exhaust experimental simulation device comprises a simulation fire source device, a tunnel longitudinal smoke exhaust control optimization device, a fire detection device, a multi-feature parameter measurement device and a tunnel slope adjustment device. The fire ventilation and smoke exhaust experimental simulation device consists of four single hole tunnels, the fire ventilation and smoke exhaust experimental simulation device is placed in directions of up-north, lower-south, left-west and right-east, and an underground parking garage is arranged in the center; the fire ventilation and smoke exhaust experimental simulation device is provided with an emergency escape door, model vehicles, an inlet jet blower, an outlet jet blower, axial flow fans, jet flow fans and smoke exhaust ports; a plurality of the axial flow fans and a plurality of the jet flow fans are respectively arranged on walls and upper ends of the tunnels of the simulation device; and the slope of the tunnel device can be adjusted. The fire ventilation and smoke exhaust experimental simulation device can simulate fire smoke characteristics and fire detection performance of the urban underground traffic linkage tunnels, and provides a research platform for acquiring reasonable fire smoke control schemes and smoke exhaust effect of the urban underground traffic linkage tunnels.

The invention discloses an evacuation route arrangement structure for a combined highway and metro shield tunnel. The structure comprises a lateral evacuation platform and an evacuation route, wherein a first fireproof door is arranged between the lateral evacuation platform and the evacuation route; the evacuation route comprises a walking plane and stairs; the walking plane comprises a low walking plane on the same elevation with the lateral evacuation platform, a high walking plane at the same elevation of an entrance of the stairs and a gentle slope for connecting the low and high walking planes; the stairs are positioned in a stair well, and the stair well consists of a stair well partition wall and a shield tunnel pipe piece; a second fireproof door is arranged on the stair well partition wall between the high walking plane and the stairs; a pressurization fan is arranged another stair well partition wall opposite to the second fireproof door. According to the structure, the width of the evacuation route is greatly increased, the evacuation time is shortened, and the evacuation efficiency is improved; in addition, the evacuation route is prevented from being invaded by fire smoke on a highway layer, so that the evacuation safety is improved.

The present invention discloses a visible light intelligent identification algorithm. A visible light intelligent identification system employs a fire smoke detection module to perform motion detection for multi-frame images and segment frames of the motion; video images for a time are extracted, preprocessing of the video images is performed according to the form, the color, the texture and the motion form of fire smoke, and areas of the time area and the space area of the fire smoke are determined; secondary identification of features of the fire smoke is performed, the memory learning interference is eliminated, the multiple probability calculation is performed; and real data is obtained, and alarm prompting is given. The algorithm concretely comprises the following steps: (1) environment learning; (2) moving object identification; (3) area division and numbering; (4) smog identification; (5) flame identification; (6) probability analysis; and (7) body simulation and memory storage. The visible light intelligent identification algorithm is mainly applied to forest fire monitoring identification, agriculture moorburn identification and inflammable open fire identification, wherein the state comprises smog, flame form and the like, and the visible light intelligent identification algorithm can identify whether fire behavior is generated or not based on the image technology.