Novel intelligent fire alarm device

Abstract

The invention relates to a fire alarm device based on voice interaction and local area network communication, and belongs to the technical field of fire safety. The device comprises a central processing module, a fire detection module, a voice interaction module, an image recognition module and a local area network communication module. Fire behavior is detected in a multi-mode mode, information intercommunication and linkage of alarms of all floors in a building are achieved through a local area network, and real-time and accurate personalized escape guidance is provided for residents at different positions through an intelligent voice interaction system. The problems that a traditional alarm system is isolated in information and single in warning content are effectively solved, and the fire emergency response efficiency and the personnel survival probability are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of fire safety technology, in particular to a kind of intelligent fire alarm device integrated voice interaction, image recognition and local area network communication function. BACKGROUND

[0002] Traditional fire alarm system is mostly used independent smoke detector or temperature sensor, and alarm is sent through buzzer.This way has obvious defects: alarm information is single, cannot provide specific fire location and escape guide;There is lack of effective communication between each alarm unit, especially in high-rise building, it is easy to appear " lower layer fire, upper layer unaware " dangerous situation, leading to personnel miss the best escape opportunity.In the prior art, although there are alarm systems with voice broadcast function or local area network communication capability, but mostly isolated application, unable to form integrated solution of intelligent sensing, information intercommunication and personalized voice guidance, difficult to meet the security needs of modern complex building environment. SUMMARY

[0003] The purpose of the present application is to overcome the deficiencies of the prior art, provide a kind of fire alarm device based on voice interaction and local area network communication, it can realize multi-storey information linkage, and provide accurate escape guidance for different positions of the household through intelligent voice.

[0004] To achieve the above object, the technical scheme adopted by the present application is as follows:

[0005] A kind of fire alarm device based on voice interaction and local area network communication, including central processing module, fire detection module, voice interaction module, image recognition module and local area network communication module.

[0006] The central processing module is used as the control core of system, for coordinating and processing the data of each module, executing fire determination logic, and generating control instruction and feedback information.

[0007] The fire detection module is connected with central processing module, for collecting environmental fire signal, it can include at least one of smoke sensor and temperature sensor.

[0008] The voice interaction module is connected with central processing module, including voice recognition unit and voice synthesis unit;The voice recognition unit is used to receive and identify the voice input of user, judges whether it contains preset keyword;The voice synthesis unit is used to synthesize and play voice feedback content according to the instruction of central processing module.

[0009] The image recognition module is connected with central processing module, for collecting environmental video data, and carries out fire identification through analyzing the color characteristics and dynamic morphological characteristics of flame.

[0010] The local area network communication module is connected with the central processing module, and is used for constructing a local communication network and realizing real-time transmission and sharing of fire information, floor data and control instructions between multiple terminal devices.

[0011] Further, the central processing module pre-stores differentiated escape plans for different floors and fire locations.

[0012] Further, the device can be further connected with external equipment such as a toxic gas detector and a water immersion sensor to expand the disaster identification range.

[0013] The present application has the advantages that: the alarm reliability is improved through multi-modal fire identification (sensor + image); the alarm linkage of the whole building is realized through local area network communication, and the information island is broken; real-time and accurate personalized escape guidance is provided through intelligent voice interaction, and the fire emergency response efficiency and the survival probability of personnel are significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is the overall system structure diagram of the fire alarm device. Figure 2 is the working flow chart of the voice interaction module in the embodiment of the present application. Figure 3 is the network topology diagram of the local area network communication module in the embodiment of the present application. Figure 4 is the flow chart of the flame determination of the image recognition module in the embodiment of the present application. DETAILED DESCRIPTION

[0015] The technical solutions of the present application will be described in detail below with reference to the drawings and embodiments. It should be understood that the following embodiments are only used to explain the present application, but not limited.

[0016] Reference Figure 1 , which shows the overall structure of the system of the present application. The hardware core of the device is a central processing module, preferably an embedded development board such as Raspberry Pi. The fire detection module, the image recognition module and the voice interaction module are directly connected with the central processing module. The central processing module is connected with multiple terminal devices deployed on each floor (such as floor A, floor B, floor C and floor D) in the building through the local area network communication module, forming a complete distributed alarm network.

[0017] At the software level, the control program based on Python language is running on the central processing module. The working process of the system is as follows:

[0018] Voice interaction process: reference Figure 2After the system is started, the voice interaction module continuously collects voice input. After a voice signal is collected, the keyword recognition analysis stage is entered. If a preset fire alarm keyword such as "help" or "on fire" is recognized, the system extracts floor information in the voice or combines the position information of the alarm terminal itself to generate an escape guidance scheme that conforms to the actual situation of the floor, and finally outputs through voice synthesis. If a non-fire alarm keyword is recognized, the system returns to the listening state.

[0019] Fire identification process: see Figure 4 The image recognition module collects environment video in real time through the camera. First, the image is preprocessed and color space is converted, and then color feature analysis (such as HSV threshold judgment, RGB analysis), morphological feature analysis (such as contour detection, area calculation), and dynamic feature analysis (such as inter-frame difference, floating detection) are performed in turn. If it is confirmed that there is a flame, the alarm is triggered and the central processing module 1 is notified; if not, it continues to monitor.

[0020] Information intercommunication process: see Figure 3 The system adopts a star network topology. The central server is the core of the proxy service. When any terminal device confirms a fire, it uploads the information to the central server through the local area network communication module. The central server immediately distributes the alarm information and evacuation instructions to all other terminal devices in the network through the local area network communication module, realizing the linkage alarm of the whole building.

[0021] In specific work, once the image recognition module or the fire detection module confirms a fire, the central processing module will immediately broadcast alarm information to all other terminals in the network through the local area network communication module. At the same time, the local voice interaction module is activated and enters the listening state. When the preset keyword is recognized, the system will broadcast the most targeted evacuation guide through the voice synthesis unit, such as "XX floor has a fire, please evacuate downward through the east safety passage". After receiving the alarm, the terminals on other floors will broadcast a preventive notice to remind residents to be vigilant and prepare for evacuation.

[0022] The above is only a preferred embodiment of the present application, and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A fire alarm device based on voice interaction and local area network communication, characterized in that, include: The central processing module serves as the control core of the system; the fire detection module is connected to the central processing module and is used to collect environmental fire signals; the voice interaction module is connected to the central processing module and is used to receive and recognize user voice input and synthesize and play voice feedback. An image recognition module, connected to the central processing module, is used to identify flames through video images; The local area network communication module is connected to the central processing module and is used to realize data communication between multiple terminal devices.

2. The fire alarm device according to claim 1, characterized in that, The voice interaction module includes a voice recognition unit and a voice synthesis unit, and the voice recognition unit is preset with a keyword triggering mechanism.

3. The fire alarm device according to claim 2, characterized in that, The keywords include "help", "on fire", or other similar words.

4. The fire alarm device according to claim 1, characterized in that, The local area network communication module is implemented based on the Socket communication protocol.

5. The fire alarm device according to claim 1, characterized in that, The image recognition module uses the OpenCV library to analyze the color and dynamic morphological features of the flame.

6. The fire alarm device according to claim 1, characterized in that, The central processing module uses a Raspberry Pi embedded development board.

7. The fire alarm device according to claim 1, characterized in that, The central processing module contains pre-stored differentiated escape plans for different floors.

8. The fire alarm device according to claim 1, characterized in that, It also includes at least one of a smoke sensor and a temperature sensor.

9. The fire alarm device according to claim 1, characterized in that, It also includes toxic gas detectors.

10. The fire alarm device according to claim 1, characterized in that, The device supports triggering other disaster emergency response branches via voice.

Images