A universal intelligent fire host information processing system and method

Abstract

The present invention proposes a universal intelligent fire host information processing system and method, compatible with various fire hosts to obtain warning information and notify the corresponding fire safety person in charge. The system includes: a camera for acquiring the warning information image on the fire host display screen; a processing module connected to the camera to acquire the warning information image, which is preset with a neural network to identify the warning information contained in the warning information image; and a wireless communication module, controlled by the processing module to send the acquired warning information to a preset application terminal. The method is to aim the camera at the fire host display screen, capture the warning information on the fire host display screen, and upload the warning information image to the processing module; the processing module processes the warning information image, extracts the screen text, and analyzes and processes it; the processing module converts the text content containing the warning information into warning instructions, and sends the warning instructions to the application terminal via the wireless communication module.

Description

Technical Field

[0001] The present invention relates to fire safety management and control technology, and in particular to a universal intelligent fire host information processing system and method. Background Art

[0002] Traditional urban fire remote monitoring technology has played an important role in social fire safety management. However, with the increasingly sound national fire laws and regulations and the rapid development of fire technology, new and higher requirements have been put forward for the current urban fire remote monitoring construction, "smart fire protection", which has brought great expectations to many companies engaged in the smart fire protection industry and related industries.

[0003] IoT platform operations based on "IoT + Smart Firefighting" will become the fundamental path for industry development. However, mature IoT technology must be utilized to transmit vast amounts of fire monitoring information to the cloud. Through information analysis and judgment, program response, fire disposal, and information sharing, remote monitoring and service objectives can be achieved. Closely aligned with the requirements of fire management work under the new situation, existing monitoring technology systems will be further upgraded and transformed. Modern sensing technology will be used to address a series of issues such as the limited functionality of traditional fire monitoring, high equipment false alarm rates, and high operating and maintenance costs. However, many fire protection systems are currently not open to the public, and existing technologies must be integrated with the original fire protection system manufacturers to complete related linkage operations, which creates an obstacle to technological innovation. Summary of the Invention

[0004] Based on the requirements mentioned in the background technology, the present invention proposes a universal intelligent fire host information processing system and method, which is compatible with various fire hosts to obtain alarm information and notify the corresponding fire safety person in charge. The specific technical content is as follows:

[0005] A universal intelligent fire host information processing system, characterized by comprising:

[0006] Camera, used to obtain warning information images on the fire host display screen;

[0007] a processing module connected to the camera to obtain the warning information image, wherein the processing module is pre-set with a neural network to identify the warning information contained in the warning information image; and

[0008] The wireless communication module is controlled by the processing module to send the acquired alarm information to a preset application terminal.

[0009] In one or more embodiments of the present invention, a deep learning module is further included, which is connected to the processing module and performs two-way information transmission. The application terminal returns the processed information to the processing module through the wireless communication module.

[0010] In one or more embodiments of the present invention, there are multiple application terminals.

[0011] In one or more embodiments of the present invention, the camera is an infrared camera.

[0012] A general intelligent fire host information processing method includes the following steps:

[0013] A. Aim the camera at the fire host display screen to capture the warning information on the fire host display screen;

[0014] B. The camera uploads the captured warning information image to the processing module;

[0015] C. The processing module processes the warning information image through a neural network and extracts the screen text;

[0016] D. The processing module analyzes and processes the extracted screen text;

[0017] E. The processing module converts the text content containing the alarm information into an alarm instruction, and sends the alarm instruction to the application terminal through the wireless communication module.

[0018] In one or more embodiments of the present invention, the camera is an infrared camera. In steps A and B, the camera uses optical flow capture technology to determine whether there is warning information on the fire host display screen. If the camera determines that there is warning information, it will take a picture and upload the warning information image to the processing module.

[0019] In one or more embodiments of the present invention, the processing module performs image analysis through a neural network and extracts text from the warning information image through optical character recognition technology.

[0020] In one or more embodiments of the present invention, step F is further included, wherein the application terminal returns the alarm processing result to the processing module through the wireless communication module.

[0021] In one or more embodiments of the present invention, there are a plurality of application terminals, and they can synchronously receive the alarm instructions sent by the wireless communication module.

[0022] In one or more embodiments of the present invention, one of the application terminals returns the processing result to the processing module through the wireless communication module. After receiving the processing result instruction, the processing module no longer sends the same alarm instruction to the wireless communication module.

[0023] The beneficial effects of the present invention are: by using the camera as an interface, it can match all existing fire host computers, without the need to modify the original fire host equipment, or the trouble of having to obtain authorization from the fire equipment manufacturer to connect, so that the fire protection system can be upgraded and modified on the original basis, and the cost is greatly reduced; in addition, a remote alarm mechanism and a reporting mechanism are set up, which can enable the logistics processing personnel on patrol to discover and handle the alarm in time, and minimize the impact of the alarm. The system can greatly improve the time utilization of logistics personnel. They do not need to be on standby in front of the fire host at any time, and can handle the alarm while patrolling, improve the efficiency of property personnel management, and reduce the cost of logistics personnel. The universal intelligent fire host information processing system of this application does not require the fire host to be modified or the communication interface to be connected. It is applicable to any brand of fire host and can also perform deep learning to continuously improve the recognition precision and accuracy. This technical solution is the first of its kind in the industry and has the characteristics of simple application and low cost. It can be widely used in the fire protection industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a principle block diagram of the universal intelligent fire host information processing system of the present invention. DETAILED DESCRIPTION

[0025] The following is combined with Figure 1 , further describe this application plan:

[0026] A universal intelligent fire host information processing system, comprising:

[0027] Infrared camera 1, used to obtain the warning information image on the display screen of the fire host 2;

[0028] A processing module 3 is connected to the infrared camera 1 to obtain the above-mentioned warning information image, and is pre-set with a neural network to identify the warning information contained in the warning information image; and

[0029] The wireless communication module 4 is controlled by the processing module 3 to send the acquired alarm information to the preset application terminal 5, such as a mobile phone, tablet computer, etc. There are several application terminals 5 (51, 52...5n); it also includes

[0030] The deep learning module 6 is connected to the processing module 3 and performs two-way information transmission. The deep learning module 6 receives the warning information image and alarm information uploaded by the processing module 3, and the application terminal 5 returns the processed information to the processing module 3 through the wireless communication module 4.

[0031] The fire host 2 is connected to and controls several terminal monitoring devices 7 to obtain on-site monitoring data. It has a preset alarm recognition model, which processes the input on-site monitoring data and generates corresponding early warning information; the terminal monitoring device 7 is used to obtain on-site monitoring data, including cameras, radio microphones, temperature sensors, gas concentration sensors, etc. The on-site monitoring data includes video data, audio data, temperature data or gas concentration data.

[0032] A general intelligent fire host information processing method includes the following steps:

[0033] A. Aim the infrared camera 1 at the display screen of the fire host 2 to capture the warning information on the display screen of the fire host 2;

[0034] B. Infrared camera 1 uploads the captured warning information image to processing module 2;

[0035] C. Processing module 3 processes the warning information image through a neural network and extracts the screen text;

[0036] D. Processing module 4 analyzes and processes the extracted screen text;

[0037] E. The processing module 5 converts the text content containing the alarm information into an alarm instruction, and sends the alarm instruction to the application terminal through the wireless communication module 4.

[0038] In a further optimization, infrared camera 1 uses optical flow capture technology to determine whether the display screen of fire control unit 2 has a warning message. If the infrared camera 1 determines that a warning message exists, it takes a picture and uploads the warning message image to the processing module. Processing module 3 performs image analysis using a neural network and extracts text from the warning message photo using optical character recognition technology.

[0039] The system further includes step F, in which the application terminal 5 returns the alarm processing result to the processing module 3 via the wireless communication module 4. Furthermore, there are multiple application terminals 5 (51, 52, ..., 5n), which can synchronously receive alarm instructions sent by the wireless communication module. One of the application terminals returns the processing result to the processing module 3 via the wireless communication module 4. After receiving the processing result instruction, the processing module 3 no longer sends the same alarm instruction to the wireless communication module 4.

[0040] When the fire control unit 2 displays an alarm message, such as "XX Community, Room 503, Temperature Sensor and Gas Concentration Sensor Abnormality," infrared camera 1 captures the alarm message on the fire control unit 2's display. Processing module 3 extracts the text from the alarm message image and transmits the extracted text as an alarm instruction via wireless communication terminal 4 to application terminal 5. Application terminal 5 displays the message "XX Community, Room 503, Temperature Sensor and Gas Concentration Sensor Abnormality" in some form, such as text or an image, to the processing personnel. The processing personnel then rush to the scene to handle the alarm based on the alarm message displayed on application terminal 5. To further ensure timely handling of the alarm, application terminal 5 includes several terminals (51, 52, ..., 5n) that can synchronously receive the alarm instructions sent by wireless communication module 4. When the processing personnel of the application terminal 51 receives the alarm instruction, he clicks the processing button on the application terminal 51, and the processing information will be returned to the processing module 3 through the wireless communication module 4. After receiving the processing information instruction, the processing module 3 will no longer send the same alarm instruction to the wireless communication module 4. In this way, the application terminal 5 will not receive the alarm information, so as to avoid the processing personnel of other application terminals rushing to the scene and wasting too much manpower, thereby enabling the logistics processing personnel to reasonably allocate time and improve personnel management efficiency. At the same time, this method effectively reduces the possibility of larger accidents caused by individual processing personnel failing to discover the alarm instruction in time and failing to take timely action. The application terminal 5 is also provided with a reporting mechanism. When the processing personnel of the first application terminal 51 arrive at the scene of the alarm, they can report the actual situation to the processing module 3 through the first application terminal 51. For example, if the fire has been handled, the abnormality of the temperature sensor and gas concentration sensor is an equipment abnormality and there is no actual alarm, and the absence of any abnormality at the scene is a false alarm, etc., after reporting the processing results to the processing module 3, the deep learning module 6 will proofread the returned processing information with the early uploaded warning information image and the text content of the alarm information, and continuously correct the processing accuracy, speed and accuracy of the processing module 3. Of course, the processing module 3 can return the final processing results of the first application terminal 51 to each application terminal 5, so that each processing personnel can understand the alarm handling status in a timely manner.

[0041] In actual applications, it is sufficient to install the infrared camera 1 on the client, and other components (except the application terminal 5) are set up on the remote server. In this way, it is possible to connect to multiple fire hosts 2 through only one remote server control, and it can also be managed in a group and unified manner, with low transformation cost and high efficiency.

[0042] This application is to conduct product research and development around the intelligent fire protection field. There are many brands of old fire host machines, and the cost for the original owners to replace them with smart fire host machines is high. There are many different fire host signal modules on the market. Although they are universal, they can only alarm for old fire host equipment, but cannot locate the location of the fire, which has little practical significance. For old fire host machines, our company will use AI technology to accurately monitor fire host equipment. There is no need to modify the tasks of the fire host machine or connect the communication interface. It is suitable for any brand of fire host machine. The AI ​​smart fire host can use the monitoring system to create an AI fire host adaptation technology that can learn quickly. This technical solution will be an industry first and has the characteristics of simple application and low cost. It can be widely used in the fire protection industry.

[0043] This application requires a large number of fire host samples as materials for AI basic learning. Due to the large number of fire host brands and the complex on-site application environment, our company's important shareholders have served in the Zhuhai fire protection field for many years, completed more than 60 large-scale fire protection projects, and maintained more than 160 customers in total, including the Hong Kong-Zhuhai-Macao Bridge Border Inspection Building, Sinopec, CNOOC, Zhuhai Chimelong Ocean Kingdom, Honggang Banshan No. 1, Hengqin World Trade Center, Hengqin Base, Hengqin Chimelong Hotel, Gongbei Lai Meili Hotel, etc.

[0044] This application is an AIT intelligent fire host early warning product that can monitor any brand of fire host on the market. When the fire host issues an early warning, it will notify all staff through mobile phone APP, text messages, etc., effectively reducing the impact of fire.

[0045] The above preferred embodiments should be regarded as examples of the implementation methods of the present application scheme. Any technical deductions, replacements, improvements, etc. that are identical or similar to the present application scheme or made based on it should be regarded as within the scope of protection of this patent.

Claims

1. A universal intelligent fire host information processing system, characterized in that: include: Camera, used to obtain warning information images on the fire host display screen; a processing module connected to the camera to obtain the warning information image, which is preset with a neural network to identify the warning information contained in the warning information image; as well as A wireless communication module, controlled by the processing module, sends the acquired alarm information to a preset application terminal; It also includes a deep learning module, the deep learning module is connected to the processing module and performs two-way information transmission, and the application terminal returns the processed information to the processing module through the wireless communication module; There are several application terminals; the camera is an infrared camera.

2. A universal intelligent fire host information processing method, used in the universal intelligent fire host information processing system according to claim 1, characterized in that: The following steps are involved: A. Aim the camera at the fire host display screen to capture the warning information on the fire host display screen; B. The camera uploads the captured warning information image to the processing module; C. The processing module processes the warning information image through a neural network and extracts the screen text; D. The processing module analyzes and processes the extracted screen text; E. The processing module converts the text content containing the alarm information into an alarm instruction, and sends the alarm instruction to the application terminal through the wireless communication module.

3. The universal intelligent fire host information processing method according to claim 2, characterized in that: The camera uses an infrared camera. In steps A and B, the camera uses optical flow capture technology to determine whether there is warning information on the fire host display screen. If the camera determines that there is warning information, it will take a picture and upload the warning information image to the processing module.

4. The method for processing information of a universal intelligent fire host according to claim 3, characterized in that: The processing module performs image analysis through a neural network and extracts text from the text portion of the warning information image through optical character recognition technology.

5. The universal intelligent fire host information processing method according to claim 4, characterized in that: It also includes step F, in which the application terminal returns the alarm processing result to the processing module through the wireless communication module.

6. The universal intelligent fire host information processing method according to claim 5, characterized in that: There are several application terminals, and they can synchronously receive the alarm instructions sent by the wireless communication module.

7. The universal intelligent fire host information processing method according to claim 6, characterized in that: One of the application terminals returns the processing result to the processing module through the wireless communication module. After receiving the processing result instruction, the processing module no longer sends the same alarm instruction to the wireless communication module.

Images