WO2024083134A1

WO2024083134A1 - Fire determination method, system and apparatus, and storage medium

Info

Publication number: WO2024083134A1
Application number: PCT/CN2023/125066
Authority: WO
Inventors: 周璐琼
Original assignee: 浙江华感科技有限公司
Priority date: 2022-10-20
Filing date: 2023-10-17
Publication date: 2024-04-25
Also published as: CN115546727A

Abstract

Provided in the present invention are a fire determination method, system and apparatus, and a storage medium. The method comprises: determining a first fire location on the basis of a first image of a target area; determining a matched area of the first fire location in a second image of the target area, and extracting color information of the matched area, a difference in photographing time of the first image and the second image meeting a preset condition; and on the basis of the color information, determining whether a fire corresponding to the first fire location is a real fire. The method can be implemented by means of the fire determination apparatus. The method can also be implemented after computer instructions stored by a computer-readable storage medium are read. The technical solution provided by the embodiments of the present application can eliminate interferences of reflection points of sunlight, etc., static heat sources such as engine exhaust gas of static automobiles, and moving heat sources such as moving hot water cups, thereby reducing the false determination rate of fires.

Description

A fire situation judgment method, system, device and storage medium

cross reference

This application claims priority to Chinese application No. 202211289815.7 filed on October 20, 2022, the entire contents of which are incorporated herein by reference.

Technical Field

The present invention relates to the field of fire monitoring and image processing, and in particular to a fire determination method, system, device and storage medium.

Background technique

Under the influence of global warming and careless use of fire in daily life, natural fires and man-made fires are becoming more and more frequent, causing great harm to public safety and social economy. In the early stage of a fire, accurate fire judgment and fire alarm forecast can reduce the loss of life and property. Therefore, it is of great significance to accurately judge a fire.

At present, smoke, light, and temperature contact fire detectors that detect smoke concentration, light signals, and temperature signals are limited by space, have low detection sensitivity in open outdoor spaces, and have a high rate of misjudgment of fire detection. With the rapid development of digital image processing technology, fire judgment based on image processing can overcome the shortcomings of contact detectors such as strong dependence on the environment, and has gradually become the main research direction in the field of fire protection. Among them, infrared thermal imaging uses infrared detectors to detect the heat emitted by objects, and infrared thermal images can be obtained with the use of optical imaging objectives; by matching the image with the heat distribution field on the surface of the object one by one, high-temperature objects such as flames and the sun can be detected. This method is widely used in forest fire prevention, fire monitoring and other fields.

However, the method of fire detection based on temperature has a high false detection rate. For example, during the detection process, the sun or sunlight reflection points, as well as high-temperature objects such as car engines and exhaust pipes are sometimes mistaken as fire sources, resulting in false fire alarms.

Therefore, it is hoped to provide a fire situation judgment method, system, device and storage medium that can eliminate the interference of heat sources such as the sun, sunlight reflection points, exhaust pipes, etc. on fire situation judgment, and effectively reduce the misjudgment rate of fire situation.

Summary of the invention

This application provides a fire judgment method, system, device and storage device to solve the problem of false fire alarm caused by misjudging the sun or sunlight reflection point and the position of the car engine and exhaust pipe as the fire source. The specific implementation scheme is as follows:

In a first aspect, one of the contents of the present invention provides a fire judgment method, the method comprising: determining a first fire location based on a first image of a target area; determining a matching area of the first fire location in a second image of the target area, and extracting color information of the matching area, wherein the shooting time difference between the first image and the second image satisfies a preset condition; based on the color information, judging whether the fire corresponding to the first fire location is a real fire.

In some embodiments, the first image is an infrared thermal image, and the second image is a visible light image.

In some embodiments, the color information includes RGB values, and judging whether the fire corresponding to the first fire location is a real fire based on the color information includes: in response to the RGB value satisfying a first judgment condition, determining that the fire corresponding to the first fire location is the real fire; in response to the RGB value not satisfying the first judgment condition, determining that the fire corresponding to the first fire location is a non-real fire.

In some embodiments, the first judgment condition includes: the R value and the G value of the RGB value satisfy a first threshold range, and the B value satisfies a second threshold range.

In some embodiments, the method further comprises: in response to the fire corresponding to the first fire location being the A non-real fire situation is used to determine an interference category of the non-real fire situation.

In some embodiments, in response to the fire corresponding to the first fire location being the non-real fire, determining the interference category of the non-real fire includes: in response to the RGB value satisfying a second judgment condition, determining that the interference category is a reflective point heat source; in response to the RGB value not satisfying the second judgment condition, determining that the interference category is a non-reflective point heat source.

In some embodiments, the second judgment condition includes: the R value, the G value and the B value of the RGB value satisfy a third threshold range.

In some embodiments, the non-reflective point heat source includes a mobile heat source and a stationary heat source. After determining that the interference category of the non-real fire corresponding to the first fire location is the non-reflective point heat source, the method also includes: determining a second fire location based on the first image, the second fire location and the first fire location corresponding to the same fire area; determining whether the second fire location and the first fire location satisfy a third judgment condition; in response to satisfying the third judgment condition, determining that the interference category is the stationary heat source; in response to not satisfying the third judgment condition, determining that the interference category is the mobile heat source.

In some embodiments, the third judgment condition includes: the second fire location is consistent with the first fire location.

In a second aspect, one of the contents of the present invention provides a fire situation judgment system, the system comprising: a determination module, used to determine a first fire location based on a first image of a target area; an extraction module, used to determine a matching area of the first fire location in a second image of the target area, and extract color information of the matching area, and a shooting time difference between the first image and the second image satisfies a preset condition; a judgment module, used to judge whether the fire corresponding to the first fire location is a real fire based on the color information.

In a third aspect, one of the contents of the present invention provides a fire situation judgment device, which includes at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least part of the computer instructions to implement a fire situation judgment method.

In a fourth aspect, one of the contents of the present invention provides a computer-readable storage medium, wherein the storage medium stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a fire situation determination method.

The beneficial effects brought about by the above invention include but are not limited to: (1) through the relevant features of the visible light image (for example, color information, etc.), it can assist in judging the authenticity of the fire location that has been judged as a suspected fire by the infrared thermal image grayscale map, and can eliminate misjudgment in many situations, such as: strong sun reflection points, high temperature locations such as stationary car engines or exhaust pipes, moving high temperature objects, etc.; (2) The RGB information of the visible light image can solve the problem that the infrared thermal image grayscale map is easily affected by weather, season, and time, thereby improving the accuracy of fire judgment.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is a schematic diagram of an application scenario of a fire situation judgment system according to some embodiments of this specification;

FIG2 is an exemplary module diagram of a fire situation determination system according to some embodiments of this specification;

FIG3 is an exemplary flow chart of a fire situation determination method according to some embodiments of this specification;

FIG4A is an exemplary schematic diagram of an infrared thermal image grayscale map according to some embodiments of the present specification;

FIG4B is an exemplary schematic diagram of a grayscale image of a visible light image according to some embodiments of the present specification;

FIG5 is an exemplary schematic diagram of determining interference categories of non-real fire conditions according to some embodiments of the present specification;

FIG. 6 is an exemplary schematic diagram of a fire situation determination process according to some embodiments of the present specification.

Detailed ways

In order to more clearly illustrate the technical solutions of the embodiments of this specification, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some examples or embodiments of this specification. For ordinary technicians in this field, without paying creative work, this specification can also be applied to other similar scenarios based on these drawings. Unless it is obvious from the language environment or otherwise explained, the same reference numerals in the figures represent the same structure or operation.

The "system", "device", "unit" and/or "module" used herein are a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, if other words can achieve the same purpose, the words can be replaced by other expressions.

Unless the context clearly indicates an exception, the words "a", "an", "an" and/or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

Flowcharts are used in this specification to illustrate the operations performed by the system according to the embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in order. Instead, the steps may be processed in reverse order or simultaneously. At the same time, other operations may be added to these processes, or one or more operations may be removed from these processes.

Currently, when detecting fires, infrared thermal imaging uses infrared detectors to detect the heat emitted by objects, and infrared thermal images can be obtained by using optical imaging lenses; then the image is matched one-to-one with the heat distribution field on the surface of the object to detect high-temperature objects, such as fires. In this way, when a circular detection algorithm is used on the fire point, it is easy to misjudge the sun or sunlight reflection point and the position of the car engine and exhaust pipe as the fire source, resulting in false fire alarms and increasing the misjudgment rate of fire.

Therefore, an embodiment of the present application provides a fire judgment method, by which interference from reflective points such as sunlight, static heat sources such as engine exhaust of stationary cars, and mobile heat sources such as moving hot water cups can be eliminated, thereby reducing the misjudgment rate of fire.

FIG1 is a schematic diagram of an application scenario of a fire situation judgment system according to some embodiments of this specification.

As shown in Fig. 1, the application scenario 100 involved in the embodiments of this specification may include a thermal imaging device 110, a visible light camera 120, a processor 130, a network 140, a terminal device 150, and a storage device 160. In some embodiments, the fire situation determination system may implement the methods and/or processes disclosed in this specification.

The thermal imaging device 110 can be used to obtain infrared thermal images. In some embodiments, the thermal imaging device 110 can monitor and obtain infrared thermal images of the target area in real time. The thermal imaging device 110 can perform an all-round or fixed-point scan of the monitoring range to obtain an infrared thermal image of any area within the monitoring range. Among them, the thermal imaging device 110 can detect the infrared energy of the object in a non-contact manner, and convert the infrared energy into an electrical signal to form an infrared thermal image. In some embodiments, the infrared thermal image can be a grayscale image or a color image. In the infrared thermal image in the form of a grayscale image, the grayscale value of each pixel reflects the received thermal radiation energy or temperature. In the infrared thermal image in the form of a color image, different colors can be used to intuitively represent the temperature. For example, red can be used to represent a high temperature, and blue can be used to represent a low temperature. In some embodiments, the thermal imaging device 110 can send the acquired infrared thermal image to the processor 130 for processing. In some embodiments, the thermal imaging device 110 can send the acquired infrared thermal image to the storage device 160 for storage.

The visible light camera 120 can be used to obtain visible light images. For example, the visible light camera can be a security camera with an imaging system of an RGB sensor. In some embodiments, the visible light camera 120 can monitor and obtain visible light images of the target area in real time. The visible light camera 120 can perform an all-round or fixed-point scan of the monitoring range to obtain the monitoring range. The visible light camera 120 can acquire a visible light image of any area within the visible light spectrum. Among them, the visible light camera 120 can acquire a visible light image by capturing light within the visible light spectrum. In some embodiments, the visible light image can be a color image. In a visible light image in the form of a color image, each pixel has color information, such as RGB value, etc. For more explanation, see Figure 3 and its related description. In some embodiments, the visible light camera 120 can send the acquired visible light image to the processor 130 for processing. In some embodiments, the visible light camera 120 can send the acquired visible light image to the storage device 160 for storage.

The processor 130 may process data and/or information obtained from the thermal imaging device 110, the visible light camera 120, the terminal device 150, and/or the storage device 160. For example, the processor 130 may obtain a first image and a second image of a target area. For another example, the processor 130 may determine a first fire location based on the first image of the target area, determine a matching area of the first fire location in the second image of the target area, and extract color information of the matching area. For another example, the processor 130 may determine whether the fire corresponding to the first fire location is a real fire based on the color information.

In some embodiments, the processor 130 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 130 may be local or remote. For example, the processor 130 may access information and/or data from the thermal imaging device 110, the visible light camera 120, the terminal device 150, and/or the storage device 160 via the network 140. For another example, the processor 130 may be directly connected to the thermal imaging device 110, the visible light camera 120, the terminal device 150, and/or the storage device 160 to access information and/or data. In some embodiments, the processor 130 may be implemented on a cloud platform. For example, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an inter-cloud cloud, a multi-cloud, etc., or any combination thereof.

The network 140 may include any suitable network that facilitates information and/or data exchange. In some embodiments, one or more components of the application scenario 100 may exchange information and/or data through the network 140. For example, the processor 130 may receive an infrared thermal image captured by the thermal imaging device 110 and a visible light image captured by the visible light camera 120 through the network 140. The network 140 may include a local area network (LAN), a wide area network (WAN), a wired network, a wireless network, etc., or any combination thereof.

The terminal device 150 can communicate and/or connect with the thermal imaging device 110, the visible light camera 120 and/or the storage device 160. For example, the terminal device 150 can send one or more control instructions to the thermal imaging device 110 and/or the visible light camera 120 through the network 140 to control the thermal imaging device 110 and/or the visible light camera 120 to collect infrared thermal images and/or visible light images of the target area according to the instructions. In some embodiments, the terminal device 150 may include one or any combination of other devices with input and/or output functions such as a mobile device 150-1, a tablet computer 150-2, a laptop computer 150-3, a desktop computer 150-4, etc. In some embodiments, the terminal device 150 may include an input device, an output device, etc. The input device may include a keyboard, a touch screen, a mouse, a voice device, etc., or any combination thereof. The output device may include a display, a speaker, a printer, etc., or any combination thereof. In some embodiments, the terminal device 150 may be a part of the processor 130. In some embodiments, the terminal device 150 may be integrated with the processor 130 as an operating console of the fire judgment system.

The storage device 160 may store data, instructions, and/or any other information. In some embodiments, the storage device 160 may store data acquired from the thermal imaging device 110, the visible light camera 120, the processor 130, and/or the terminal device 150. For example, the storage device 160 may store infrared thermal images and/or visible light images acquired from the thermal imaging device 110 and/or the visible light camera 120. In some embodiments, the storage device 160 may store data and/or instructions used by the processor 130 to execute or use to complete the exemplary methods described in this specification. For example, the storage device 160 may store instructions for controlling the thermal imaging device 110 and the visible light camera 120 to acquire images.

In some embodiments, the storage device 160 may include a mass storage, a removable storage, a volatile read-write memory, a read-only memory (ROM), etc. or any combination thereof. In some embodiments, the storage device 160 may be implemented on a cloud platform. In some embodiments, the storage device 160 may be part of the processor 130.

It should be noted that the above description is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those of ordinary skill in the art, various changes and modifications may be made under the guidance of the contents of this specification. The features, structures, methods and other features of the exemplary embodiments described in this specification may be combined in various ways to obtain additional and/or alternative exemplary embodiments. For example, the thermal imaging device 110, the visible light camera 120, the processor 130 and the terminal device 150 may share a storage device 160, or may have their own storage devices. However, these changes and modifications do not deviate from the scope of this specification.

FIG2 is an exemplary module diagram of a fire situation determination system according to some embodiments of the present specification. In some embodiments, as shown in FIG2 , the fire situation determination system 200 may include a determination module 210, an extraction module 220, and a determination module 230. In some embodiments, the fire situation determination system 200 may be implemented by a processor 130.

In some embodiments, the determination module 210 may be used to determine the first fire location based on the first image of the target area. For more information about the first image and the first fire location, see FIG. 3 and its related description.

In some embodiments, the extraction module 220 can be used to determine the matching area of the first fire location in the second image of the target area, and extract the color information of the matching area, and the shooting time difference between the first image and the second image meets the preset condition. For more information about the second image, the matching area, the color information, etc., please refer to Figure 3 and its related description.

In some embodiments, the determination module 230 may be used to determine whether the fire corresponding to the first fire location is a real fire based on the color information of the matching area.

In some embodiments, the color information includes RGB values, and the judgment module 230 can be further used to: in response to the RGB value satisfying the first judgment condition, determine that the fire corresponding to the first fire location is a real fire; in response to the RGB value not satisfying the first judgment condition, determine that the fire corresponding to the first fire location is a non-real fire.

In some embodiments, the judgment module 230 may be further used to: in response to the fire corresponding to the first fire position being a non-real fire, determine the interference category of the non-real fire. In some embodiments, the judgment module 230 may be further used to: in response to the RGB value satisfying the second judgment condition, determine the interference category as a reflective point heat source; in response to the RGB value not satisfying the second judgment condition, determine the interference category as a non-reflective point heat source.

In some embodiments, the non-reflective point heat source includes a mobile heat source and a stationary heat source. After determining that the interference category of the non-real fire corresponding to the first fire location is the non-reflective point heat source, the judgment module 230 can be further used to: determine a second fire location based on the first image, the second fire location and the first fire location corresponding to the same fire area; determine whether the second fire location and the first fire location meet a third judgment condition; in response to meeting the third judgment condition, determine that the interference category is a stationary heat source; in response to not meeting the third judgment condition, determine that the interference category is a mobile heat source.

For more information on determining whether the fire corresponding to the first fire location is a real fire, see Figures 3 and 5 and their related descriptions.

It should be noted that the above description of the candidate display, determination system and its modules is only for convenience of description and cannot limit this specification to the scope of the embodiments. It can be understood that for those skilled in the art, after understanding the principle of the system, it is possible to arbitrarily combine the various modules without deviating from this principle, or to form a subsystem connected with other modules. In some embodiments, the determination module 210, extraction module 220 and judgment module 230 disclosed in Figure 2 can be different modules in a system, or a module can realize the functions of two or more modules mentioned above. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the scope of protection of this specification.

FIG. 3 is an exemplary flow chart of a fire situation determination method according to some embodiments of the present specification.

In some embodiments, the fire condition determination method may be executed by the processor 130 or the fire condition determination system 200. For example, the process 300 may be stored in a storage device (eg, the storage device 160) in the form of a program or instruction. When the fire situation judgment system 200 executes the program or instruction, the process 300 can be implemented. The operation diagram of the process 300 presented below is illustrative. In some embodiments, the process can be completed using one or more additional operations not described and/or one or more operations not discussed. In addition, the order of the operations of the process 300 shown in Figure 3 and described below is not restrictive.

Step 301 : determining a first fire location based on a first image of a target area. In some embodiments, step 310 may be performed by the processor 130 or the determination module 210 .

The target area refers to the area where the fire situation is to be determined, wherein the fire situation may include the location where the fire is suspected to have occurred, the actual situation of the fire, etc.

The target area can be in various forms. For example, the target area can be in various forms such as an outdoor square, an outdoor parking lot, an indoor square, a shopping mall, etc. The description of the target area is only for illustrative purposes and does not constitute a limitation on the implementation method.

The first image refers to image data used to determine the fire situation.

In some embodiments, the first image is an infrared thermal image. For example, the first image may be an infrared thermal image in the form of a color image.

In some embodiments, the first image can be acquired by a monitoring device deployed at a monitoring position. The monitoring position can be the location where the monitoring device is located. In some embodiments, the monitoring device can include a thermal imaging device. In an optional application scenario, a plurality of monitoring columns are dispersedly arranged in an outdoor square, a pan-tilt is arranged at the top of the plurality of monitoring columns, and the monitoring device is mounted on the pan-tilt, which can rotate so that a plurality of monitoring devices can collect infrared thermal images of the outdoor square, etc. Therefore, in the above application scenario, the monitoring position can be the location of the pan-tilt. For a plurality of monitoring devices, they can be installed on pan-tilts located at different locations, and a plurality of monitoring devices located on different pan-tilts can exchange information with a monitoring center through a network. For more information about the thermal imaging device, see FIG. 1 and its related description.

The first fire location refers to the location area where the fire is suspected to have occurred.

In some embodiments, the processor 130 may process the first image of the target area in a variety of ways to determine the first fire location. In some embodiments, the processor 130 may process the original infrared thermal image (e.g., a color infrared thermal image) in a grayscale processing manner to obtain an infrared thermal image in a grayscale image (hereinafter referred to as an infrared thermal image grayscale image), and determine the first fire location in the infrared thermal image grayscale image according to the fire location determination method.

In some embodiments, the method for determining the location of a fire includes: when there are one or more pixels in the grayscale image of the infrared thermal image with a grayscale value greater than a preset grayscale threshold, the location area composed of these pixels is determined as the first fire location. In some embodiments, the method for determining the location of a fire includes: determining a location area composed of multiple pixels whose grayscale values are greater than a preset grayscale threshold; determining the temperature statistics of the location area (for example, standard deviation, mean difference, etc.); determining the fluctuation characteristics of the location area based on the temperature statistics (for example, determining the ratio of the temperature standard deviation to the temperature mean difference as the fluctuation characteristics); when the fluctuation characteristics (for example, the ratio of the aforementioned temperature standard deviation to the temperature mean difference) is greater than the preset ratio, determining the location area as the first fire location. Among them, the preset grayscale threshold and the preset ratio can be system default values, experience values, artificial pre-set values, etc. or any combination thereof, and can be set according to actual needs, and this specification does not limit this.

It should be noted that the embodiments of this specification do not have any special limitations on the grayscale processing method and the fire location determination method, and operations familiar to those skilled in the art may be used.

In some embodiments, after the first fire location is determined, the processor 130 can determine the reference coordinate point in the infrared thermal image grayscale map by a reference coordinate point determination method, determine the first fire location coordinates of the first fire location based on the reference coordinate point and a rectangular coordinate system established based on the reference coordinate point, and determine the fire area spreading outward with the first fire location as the center according to the fire area division method.

In some embodiments, the fire area division method includes: taking the fire location as the center and dividing the fire area by a preset distance value. The rectangular area obtained by the external diffusion is the fire area. In some embodiments, the fire area is divided by: the area obtained by splicing multiple pixel points whose grayscale difference with the grayscale value at the fire location is within a preset difference range is determined as the fire area. The preset distance value and the preset difference range can be system default values, experience values, artificial preset values, etc. or any combination thereof, and can be set according to actual needs, and this specification does not limit this.

It should be noted that the embodiments of this specification do not have any special restrictions on the method of dividing the fire area, and operations familiar to those skilled in the art can be used.

In some embodiments, the reference coordinate point determination method includes: determining the reference coordinate point based on the upper left corner of the infrared thermal image grayscale image. In some embodiments, the reference coordinate point determination method also includes: determining the reference coordinate point based on the lower left corner of the infrared thermal image grayscale image. The specific reference coordinate point determination method is not limited in the embodiments of this specification.

For example, Figure 4A is an infrared thermal image grayscale image after grayscale processing of the original infrared thermal image of the target area taken by the thermal imaging device. Through the judgment condition of the fire location, a certain position W in the infrared thermal image grayscale image is determined as the first fire location. A rectangular coordinate system is established with the lower left corner of the infrared thermal image grayscale image as the origin (0, 0), and the coordinates of the first fire location (i.e., position W) are determined to be (5, 5). This coordinate is the first fire location coordinate, and the rectangular area radiating outward from position W is the fire area Q.

Step 302 , determining a matching area of the first fire location in the second image of the target area, and extracting color information of the matching area. In some embodiments, step 320 may be performed by the processor 130 or the extraction module 220 .

The second image refers to image data used to determine the fire situation. The second image is different from the first image.

In some embodiments, the second image is a visible light image. For example, the second image can be a visible light image in the form of a color image.

In some embodiments, the second image may be acquired after the first fire location is determined. In some embodiments, the shooting time difference between the first image and the second image satisfies a preset condition. In some embodiments, the preset condition may be that the shooting time difference between the first image and the second image is less than a preset time difference threshold. The preset condition and the preset time difference may be set based on actual needs and are not limited here.

In some embodiments, the second image can be acquired by a monitoring device deployed at the monitoring location. In some embodiments, the monitoring device may also include a visible light camera. For example, a security camera based on an imaging system with an RGB sensor may capture a raw image corresponding to the target area; the raw image is processed by a raw image processing method to obtain a corresponding visible light image. For more information about the visible light camera, see FIG. 1 and its related description.

It should be noted that the embodiments of this specification do not specifically limit the raw image processing method, and operations well known to those skilled in the art may be used.

The matching area refers to the partial image area corresponding to the fire area corresponding to the first fire location in the second image.

In some embodiments, the processor 130 can determine the matching area of the first fire location in the second image of the target area in a variety of ways. For example, the processor 130 can configure the first image and the second image through an image registration algorithm, and then determine the matching area of the first fire location in the second image based on the fire area of the first fire location in the first image. Among them, the image registration algorithm includes but is not limited to a template-based matching algorithm, a grayscale-based matching algorithm, a feature-based matching algorithm, etc. For another example, the processor 130 can determine the reference coordinate points of the second image through a reference coordinate point determination method, and construct a rectangular coordinate system based on the reference coordinate points; mark the fire area in the first image in the second image based on the rectangular coordinate system to obtain a matching area.

For example, as shown in FIG. 4B , a visible light image grayscale image of the target area (i.e., the second image in grayscale form, obtained by a visible light camera) is used to establish a rectangular coordinate system with the lower left corner of the visible light image grayscale image as the origin (0, 0). The fire area determined in step 310 is marked with a rectangle in the visible light image grayscale image, i.e., the infrared thermal image in FIG. 4A The rectangular area Q in the grayscale image is obtained to obtain the matching area Q'. It should be noted here that FIG4A and FIG4B are respectively the infrared thermal image grayscale image and the visible light image grayscale image of the same target area. The grayscale values of different pixels in the infrared thermal image grayscale image are related to the temperature, while there is no specific corresponding relationship between the grayscale values of different pixels in the visible light image grayscale image and the temperature.

The color information refers to information on the color included in the matching area.

In some embodiments, the color information may include RGB values. The RGB values of the matching area may be obtained in a variety of ways. The embodiments of this specification do not specifically limit the method for extracting the RGB values, and operations familiar to those skilled in the art may be used.

In some embodiments, the color information of the matching area may include the color information of each pixel in the matching area. In some embodiments, the color information of the matching area may be a statistical value of the color information of each pixel in the matching area, such as a mean, an extreme value, a median, or the like.

Step 303 : Based on the color information, determine whether the fire corresponding to the first fire position is a real fire. In some embodiments, step 330 may be executed by the processor 130 or the determination module 230 .

In some embodiments, the processor 130 can determine whether the fire corresponding to the first fire position is a real fire based on the RGB value in the color information. In some embodiments, the processor 130 can determine whether the fire corresponding to the first fire position is a real fire based on the RGB value of each pixel in the matching area. For example, in response to the RGB values of a preset number of pixels in the matching area satisfying a specific judgment condition, the fire corresponding to the first fire position is judged to be a real fire, otherwise it is an unreal fire. For another example, in response to the average of the RGB values of each pixel satisfying a specific judgment condition, the fire corresponding to the first fire position is judged to be a real fire, otherwise it is an unreal fire.

In some embodiments, the processor 130 can determine that the fire corresponding to the first fire location is a real fire in response to the RGB value satisfying the first judgment condition; in response to the RGB value not satisfying the first judgment condition, determine that the fire corresponding to the first fire location is a non-real fire.

The first judgment condition refers to a judgment condition used to judge whether the fire corresponding to the first fire location is a real fire.

In some embodiments, the first judgment condition includes: the R value and the G value in the RGB value satisfy the first threshold range, and the B value satisfies the second threshold range. The first threshold range may be [a, b], and the second threshold range may be [c, d], where d < a, a, b, c, d∈(0,255). As an example only, the first threshold range may be [250,255], and the second threshold range may be [210,230]. Preferably, the second threshold range may be [215,225]. In some embodiments, a and b may be the same. As an example only, the first threshold range may be 255.

For example, the R value, G value, and B value of the rectangular area Q' in Figure 4B are processed by averaging to obtain R value, G value, and B value of 255, 254, and 252, respectively. At this time, the R value and G value are within the first preset threshold range, and the B value is not within the second threshold range. It can be determined that the R, G, and B values of the matching area do not meet the first judgment condition. For another example, the R value, G value, and B value of the rectangular area Q' in Figure 4B are processed by averaging to obtain R value, G value, and B value of 255, 254, and 220, respectively. At this time, the R value and G value are within the first preset threshold range, and the B value is within the second threshold range. It can be determined that the R, G, and B values of the matching area meet the first judgment condition.

By extracting the R, G, and B values of the matching area in the above manner and comparing them with the first judgment condition, the accuracy of fire location judgment is improved and the misjudgment rate of fire judgment is reduced. At the same time, only the RGB values of the suspicious fire points are calculated, which effectively reduces the amount of calculation.

In some embodiments, the first threshold range and the second threshold range may be predetermined. For example, the first threshold range and the second threshold range may be system default values, empirical values, artificially preset values, or any combination thereof, and may be set according to actual needs, and this specification does not limit this.

In some embodiments, the processor 130 may also dynamically determine the corresponding first Threshold range and second threshold range. Different actual situations correspond to different first threshold ranges and/or different second threshold ranges, and the corresponding first threshold range and second threshold range can be dynamically determined according to the actual situation.

In some embodiments, the processor 130 may acquire a visible light image sequence of the target area, and determine a first threshold range and a second threshold range based on the visible light image sequence.

In some embodiments, the process of acquiring the visible light image sequence may be triggered after acquiring the second image. After acquiring the second image, the processor 130 may control the visible light camera to capture the visible light image of the target area to acquire the visible light image sequence.

In some embodiments, the processor 130 may control the visible light camera to capture the target area multiple times to obtain a visible light image sequence. For example, the processor 130 may control the visible light camera to capture an image of the target area once at a preset time interval to obtain a visible light image, thereby obtaining a visible light image sequence.

In some embodiments, the visible light image sequence may be captured at a first capture frequency. The processor 130 may control the visible light camera to capture the target area multiple times at the first capture frequency to acquire the visible light image sequence.

The first capture frequency is a capture frequency used to acquire a visible light image sequence. For example, the first capture frequency may be x images/min, etc., where x is a positive integer.

In some embodiments, the first snapshot frequency may be preset based on prior knowledge or historical data. For example, the snapshot frequency that is used most times in historical data may be determined as the first snapshot frequency.

In some embodiments, the processor 130 may determine the first capture frequency according to the actual situation of the target area. In some embodiments, the processor 130 may acquire a preset number of initial visible light images at a second capture frequency; determine an ambient light intensity sequence and a difference distribution sequence based on the preset number of initial visible light images; and determine the first capture frequency based on the ambient light intensity sequence and the difference distribution sequence.

The second capture frequency is a preliminarily determined capture frequency. The second capture frequency can be used to determine the first capture frequency. For example, firstly, image acquisition is performed at the second capture frequency, and analysis and processing are performed based on the acquired image to determine the first capture frequency. The second capture frequency can be the same as or different from the first capture frequency.

In some embodiments, the second snapshot frequency can be preset based on prior knowledge or historical data. For example, the snapshot frequency that is used most times in historical data can be determined as the second snapshot frequency.

In some embodiments, the preset number may be a system default value, an experience value, a manually preset value, or any combination thereof, and may be set according to actual needs, and this specification does not impose any restrictions on this.

The initial visible light image refers to a visible light image acquired according to the second capture frequency.

The ambient light intensity sequence refers to a sequence consisting of the ambient light intensity of a preset number of initial visible light images. In some embodiments, the processor 130 may obtain the ambient light intensity of each of the preset number of initial visible light images to obtain the ambient light intensity sequence. For more information on obtaining the ambient light intensity, see below.

The difference distribution sequence refers to a sequence consisting of difference distributions of visible light images captured at adjacent moments in a preset number of initial visible light images. In some embodiments, the difference distribution sequence includes multiple difference distributions, each of which is determined based on image information of initial visible light images captured at adjacent moments in a preset number of initial visible light images.

In some embodiments, the processor 130 may perform semantic segmentation on two initial visible light images taken at adjacent moments to obtain multiple groups of semantic block pairs; calculate the pixel difference matrix of each semantic block pair in the multiple groups of semantic block pairs; perform statistical feature extraction on the pixel difference matrix to obtain difference statistics; obtain multiple difference statistics based on multiple pixel difference matrices corresponding to the multiple groups of semantic block pairs, and the multiple difference statistics constitute a difference distribution. By performing the above processing on each adjacent two visible light images in a preset number of initial visible light images, a difference distribution sequence can be obtained.

A semantic block pair consists of two semantic blocks, which come from different visible light images. There is a correspondence between the blocks. Among them, the correspondence between semantic blocks means that two semantic blocks correspond to the same object. For example, a semantic block (such as a car) is recognized in visible light image A, and a semantic block (the same car as the previous one) is recognized in visible light image B, then the two semantic blocks correspond to each other. Just as an example, two adjacent visible light images include visible light image A and visible light image B, visible light image A is divided into semantic blocks a1, a2, a3, and visible light image B is divided into semantic blocks b1, b2, b3, where a1 corresponds to b1, a2 corresponds to b2, and a3 corresponds to b3.

In some embodiments, the processor 130 can perform semantic segmentation on two adjacent visible light images respectively by various feasible methods such as semantic segmentation algorithms or semantic segmentation models (e.g., machine learning models, etc.), to obtain multiple groups of semantic block pairs. The embodiments of this specification do not specifically limit the semantic segmentation algorithms or semantic segmentation models, and operations familiar to those skilled in the art can be used.

In some embodiments, when there is a correspondence between two semantic blocks, there may also be a correspondence between the pixels in the two semantic blocks. In some embodiments, the correspondence between two pixels may refer to the two pixels having the same relative position. The relative position refers to the relative position of a pixel among multiple pixels contained in a semantic block. For example, there is a correspondence between pixel R in semantic block X of visible light image A and pixel R' in semantic block X' of visible light image B, which means that pixel R and pixel R' are in the same relative position, for example, pixel R is the pixel in the third row and third column of visible light image A, and pixel R' is also the pixel in the third row and third column of visible light image B.

The pixel difference matrix refers to the matrix composed of the pixel differences between every two corresponding pixels in a semantic block pair.

In some embodiments, the processor 130 may calculate the difference between two pixel points at corresponding pixel positions in the semantic block pair, and obtain a pixel difference matrix according to the difference of each pixel position. In some embodiments, the difference between two pixel points may be calculated and determined by a distance function or the like. For example, the distance function includes but is not limited to Euclidean distance, cosine distance, and the like.

The difference statistic refers to a statistic obtained by extracting statistical features from the pixel difference matrix. For example, the difference statistic may be a mean value, a variance value, etc. of the differences between each pixel in the pixel difference matrix.

In some embodiments, the processor 130 may extract statistical features from the pixel difference matrix using a statistical feature extraction algorithm to obtain a difference statistical value. For example, the statistical feature extraction algorithm includes an algorithm for extracting a mean, an algorithm for extracting a variance, etc., which are not limited here.

In some embodiments, the processor 130 may determine the first snapshot frequency based on the ambient light intensity sequence and the difference distribution sequence in a variety of ways.

In some embodiments, the processor 130 may determine the first snapshot frequency by vector retrieval based on the ambient light intensity sequence and the difference distribution sequence. The processor 130 may construct a vector to be matched based on the ambient light intensity sequence and the difference distribution sequence; perform vector matching in a vector database based on the vector to be matched to determine an associated vector; and determine the first snapshot frequency based on the associated vector.

In some embodiments, the vector database may include multiple reference vectors and their corresponding reference capture frequencies. In some embodiments, the reference vectors may be constructed based on historical data. For example, multiple reference vectors are obtained by vector construction of multiple historical ambient light intensity sequences and multiple historical difference distribution sequences. The reference capture frequencies corresponding to the reference vectors may be manually labeled based on prior knowledge and actual historical capture frequencies.

In some embodiments, the processor 130 may determine a reference vector that meets a preset matching condition in a vector database based on the vector to be matched, determine the reference vector that meets the preset matching condition as an associated vector, and determine the reference snapshot frequency corresponding to the associated vector as the first snapshot frequency corresponding to the vector to be matched. The preset matching condition may refer to a judgment condition for determining the associated vector. In some embodiments, the preset matching condition may include a vector distance less than a distance threshold, a vector distance minimum, etc.

In some embodiments, the processor 130 may be configured to generate a frequency signal based on the ambient light intensity sequence and the difference distribution sequence. The rate determination model determines the first snapshot frequency.

In some embodiments, the frequency determination model may be a machine learning model, such as a deep neural network (DNN) model. In some embodiments, the input of the frequency determination model includes an ambient light intensity sequence and a difference distribution sequence, and the output includes the first capture frequency.

In some embodiments, the frequency determination model can be trained by various methods based on multiple first training samples with first labels to update model parameters. For example, the training can be based on the gradient descent method. For example, the training can be based on the gradient descent method. As an example only, multiple first training samples with first labels can be input into the initial frequency determination model, and a loss function can be constructed by the first label and the output result of the initial frequency determination model, and the parameters of the initial frequency determination model can be iteratively updated based on the loss function. When the loss function of the initial frequency determination model meets the preset conditions, the model training is completed, and a trained image recognition model is obtained. Among them, the preset conditions can be that the loss function converges, the number of iterations reaches a threshold, etc.

In some embodiments, the first training sample includes a sample ambient light intensity sequence and a sample difference distribution sequence corresponding to a sample initial visible light image sequence of the sample area (the sample initial visible light image sequence includes multiple initial visible light images acquired at a sample second capture frequency), and the first label includes a first capture frequency for capturing visible light images of the sample area. In some embodiments, the sample area may be an area where the first fire location is determined. In some embodiments, the first training sample and the first label may be determined based on historical data. For example, the area where the first fire location is determined may be used as the sample area, and the ambient light intensity sequence and the difference distribution sequence of the historical initial visible light image sequence of the sample area may be used as the first training sample. In some embodiments, the processor 130 may determine the lowest capture frequency that can identify a real fire under the first training sample as the first label corresponding to the first training sample.

Some embodiments of this specification can effectively analyze whether the ambient light intensity of the target area is in a relatively stable state or a frequently changing state (i.e., the changing state of the ambient light), and can effectively analyze whether the objects contained in the target area are stationary objects or moving objects (i.e., the dynamic and static state of the objects) by first taking multiple initial visible light images of the target area, and then determining the ambient light intensity sequence and difference distribution sequence of the multiple initial visible light images. By analyzing the changing state of the ambient light and the dynamic and static state of the objects in the target area, the influence of these factors on the judgment of the real fire situation can be fully considered, and then a reasonable first capture frequency can be determined to obtain a reasonable visible light image sequence, so that the first threshold range and the second threshold range set based on the visible light image sequence are more accurate and reasonable.

The processor 130 can process the visible light image sequence in a variety of ways to determine the first threshold range and the second threshold range. In some embodiments, the processor 130 can identify the ambient light intensity of each visible light image in the visible light image sequence and calculate the average ambient light intensity; based on the average ambient light intensity, the first threshold range and the second threshold range are determined by querying a first preset comparison table.

In some embodiments, the processor 130 may use a trained image recognition model to recognize the ambient light intensity of the visible light image. The image recognition model may be a machine learning model, such as a convolutional neural network (CNN) model. In some embodiments, the input of the image recognition model includes the visible light image, and the output includes the ambient light intensity of the visible light image.

In some embodiments, the image recognition model can be trained by various methods based on multiple second training samples with second labels to update model parameters. The training process of the image recognition model is similar to the training process of the frequency determination model. For more details, please refer to the relevant description above, which will not be repeated here.

In some embodiments, the second training sample includes a sample visible light image of the sample area, and the second label includes the ambient light intensity of the sample visible light image. In some embodiments, the second training sample can be obtained based on historical image data. The second label can be obtained based on historical detection data. The historical image data includes historical visible light images of the sample area taken by a visible light camera, and the historical detection data includes historical ambient light intensity of the sample area collected by a light intensity detection instrument.

In some embodiments, the first preset comparison table includes correspondences between different reference ambient light intensity means and different reference first threshold ranges and reference second threshold ranges. In some embodiments, the correspondences between different reference ambient light intensity means and different reference first threshold ranges and reference second threshold ranges can be constructed by preset construction rules based on prior knowledge or historical data to obtain the first preset comparison table. In some embodiments, the preset construction rules include: under different historical ambient light intensity means, according to historical judgment situations, the first threshold range and the second threshold range corresponding to the lowest misjudgment rate are determined as the first threshold range and the second threshold range corresponding to the historical ambient light intensity mean.

In some embodiments, the historical judgment situation includes the correspondence between the judgment result of whether the fire at the first fire location is a real fire and the actual situation. For example, the historical judgment situation includes wrong judgment and accurate judgment. When the judgment result does not match the actual situation, it is a wrong judgment, otherwise it is an accurate judgment. For example, when the fire at the first fire location is judged to be a real fire, but in fact the fire at the first fire location is not a real fire, it is a wrong judgment; when the fire at the first fire location is judged to be a real fire, but in fact the fire at the first fire location is a real fire, it is an accurate judgment. The misjudgment rate is the ratio of the number of wrong judgments in multiple judgments to the total number of judgments.

In some embodiments, the processor 130 can also match the same or similar historical ambient light intensity mean values in the historical data based on the current ambient light intensity mean value, and determine the historical first threshold range and the historical second threshold range corresponding to the same or similar historical ambient light intensity mean values as the currently corresponding first threshold range and the second threshold range.

In some embodiments of the present specification, after determining that the target area includes the first fire location, an actual visible light image sequence of the target area can be obtained, and then the first threshold range and the second threshold range can be determined based on the visible light image sequence. This method is conducive to adaptively and dynamically determining the first threshold range and the second threshold range based on the actual situation of the target area, so that the subsequent result of judging the authenticity of the fire based on color information is more accurate.

In some embodiments, the processor 130 may determine the interference category of the unreal fire in response to the fire corresponding to the first fire location being an unreal fire. For more information about this embodiment, see FIG. 5 and its related description.

In some embodiments of this specification, the relevant features of the visible light image (e.g., color information, etc.) are used to assist in determining the authenticity of the fire location that has been determined as a suspected fire by the infrared thermal image grayscale image, and misjudgment in various situations can be eliminated, such as: strong sun reflection points, high-temperature locations such as stationary car engines or exhaust pipes, moving high-temperature objects, etc. At the same time, the RGB information of the visible light image can solve the problem that the infrared thermal image grayscale image is easily affected by weather, season, and time, and improve the accuracy of fire judgment.

It should be noted that the above description of process 300 is only for example and illustration, and does not limit the scope of application of this specification. For those skilled in the art, various modifications and changes can be made to process 300 under the guidance of this specification. However, these modifications and changes are still within the scope of this specification.

FIG. 5 is an exemplary schematic diagram of determining interference categories of non-real fire conditions according to some embodiments of the present specification.

In some embodiments, the processor 130 may determine an interference category of the unreal fire in response to the fire corresponding to the first fire location being an unreal fire.

In some embodiments, the interference category may refer to a heat source category that is not a real fire. In some embodiments, the interference category includes a reflective point heat source and a non-reflective point heat source.

Reflective point heat sources refer to points or objects where the temperature rises due to reflection, for example, the sun's reflection points.

Non-reflective point heat sources refer to points or objects whose temperature rises due to reasons other than reflection, such as car engines, exhaust vents, hot water cups, etc.

The processor 130 may determine the interference category of the non-real fire in a variety of ways. In some embodiments, the processor 130 may determine the interference category of the non-real fire based on the shape of the fire area corresponding to the first fire location. For example, when the shape of the fire area is circular or quasi-circular, the interference category of the non-real fire at the first fire location is determined to be a reflective point heat source. In some embodiments, the processor 130 may determine the interference category of the non-real fire based on the color information (eg, RGB value, etc.) corresponding to the matching area of the first fire location in the second image of the target area.

5 , in some embodiments, the processor 130 may determine that the interference category is a reflective point heat source in response to the RGB value satisfying the second judgment condition; and determine that the interference category is a non-reflective point heat source in response to the RGB value not satisfying the second judgment condition.

The second judgment condition is a judgment condition for judging the interference category that is not a real fire situation.

In some embodiments, the second judgment condition includes: the R value, the G value, and the B value in the RGB value satisfy the third threshold range. The third threshold range may be [e, f], a＜e, b≤f, e, f∈(0,255). As an example only, the first threshold range may be [250,255], the second threshold range may be [215,225], and the third threshold range may be [254,255]. In some embodiments, e and f may be the same. As an example only, the third threshold range may be 255.

For example, the R value, G value, and B value of the rectangular area Q' in Figure 4B are all 255 after mean processing, and the R value, G value, and B value are all within the third threshold range. At this time, the R value, G value, and B value meet the second judgment condition, and it is determined that the non-real fire corresponding to the position W in the rectangular area Q in Figure 4A is a reflective point heat source.

By comparing the extracted R value, G value, and B value with the second judgment condition in the above manner, it can be accurately determined that the non-real fire comes from reflective points such as sunlight.

In some embodiments, the third threshold range may be predetermined. For example, the third threshold range may be a system default value, an empirical value, a manually preset value, or any combination thereof, and may be set according to actual needs, which is not limited in this specification.

In some embodiments, the processor 130 may also dynamically determine a corresponding third threshold range according to the actual situation of the target area.

In some embodiments, the processor 130 may obtain temperature distribution data of the target area from the first image, obtain color distribution data of the target area from the second image, and determine a third threshold range based on the temperature distribution data and the color distribution data.

The temperature distribution data refers to the temperature conditions of different sub-regions after the first image is divided into multiple sub-regions, wherein the sub-region refers to a partial region of the first image.

In some embodiments, the processor 130 may divide the first image into a plurality of sub-regions in a variety of ways. For example, the processor 130 may divide the first image into a plurality of sub-regions according to a grid of a preset width. For another example, the processor 130 may divide the first image into a plurality of sub-regions by semantic segmentation. The method of semantically segmenting the first image (e.g., an infrared thermal image) is similar to the method of semantically segmenting the visible light image. For more explanation, see FIG. 3 and its related description.

In some embodiments, the temperature data of the sub-region may include one or more of the temperature mean, temperature variance, and temperature median of the sub-region. In some embodiments, the processor 130 may directly read the temperature value of each pixel from the first image, and then obtain the temperature data of the sub-region through statistical analysis. The embodiments of this specification do not specifically limit the method of statistical analysis, and operations familiar to those skilled in the art may be used.

The color distribution data refers to the color conditions of different sub-regions after the second image is divided into multiple sub-regions. The method of dividing the second image is similar to the method of dividing the first image, and will not be repeated here.

In some embodiments, the color information of the sub-region may include the mean R value, the mean G value, the mean B value, etc. of the sub-region. In some embodiments, the processor 130 may directly read the RGB value of each pixel from the second image, and then obtain the color information of the sub-region through statistical analysis.

The processor 130 can process the temperature distribution data and the color distribution data in a variety of ways to determine the third threshold value. Value range.

In some embodiments, the processor 130 may determine the third threshold range by querying a second preset comparison table based on the temperature distribution data and the color distribution data.

In some embodiments, the second preset comparison table includes correspondences between different reference temperature distribution data, different reference color distribution data, and different reference third threshold ranges. The second preset comparison table is constructed in a similar manner to the first preset comparison table, and will not be described in detail here. For more information, see step 330 and its related description.

In some embodiments, the processor 130 can also match the same or similar historical temperature distribution data and historical color distribution data in the historical data based on the current temperature distribution data and color distribution data, and determine the historical third threshold range corresponding to the same or similar historical temperature distribution data and historical color distribution data as the current corresponding third threshold range.

In some embodiments, the processor 130 may determine material distribution data within a preset range of the first fire location based on the second image; and determine a third threshold range based on the temperature distribution data, the color distribution data, and the material distribution data.

The preset range refers to a certain range around the first fire location. For example, the preset range may be a 100-meter range around the first fire location. For another example, the preset range may be the fire area of the first fire location. The preset range may be a system default value, an experience value, a manually preset value, or any combination thereof, and may be set according to actual needs, and this specification does not limit this.

Material distribution data refers to the material conditions at different locations within a preset range of the first fire location. For example, when the first fire location is a car exhaust outlet, the material conditions at different locations within the preset range may include, but are not limited to, one or more of iron, aluminum, and aluminum alloy.

In some embodiments, the processor 130 may determine the material distribution data within a preset range of the first fire location based on the second image by means of a material detection algorithm or the like. Exemplary material detection algorithms include material detection algorithms based on visual features, etc. The embodiments of this specification do not specifically limit the material detection algorithm, and operations familiar to those skilled in the art may be used.

In some embodiments, the processor 130 may determine the third threshold range by querying a third preset comparison table based on the temperature distribution data, the color distribution data, and the material distribution data. In some embodiments, the third preset comparison table includes correspondences between different reference temperature distribution data, different reference color distribution data, different reference material distribution terms, and different reference third threshold ranges. The construction method of the third preset comparison table is similar to the construction method of the first preset comparison table, and will not be repeated here. For more explanation, see step 330 and its related description.

In some embodiments, the processor 130 may also match relevant historical data in the historical data based on the temperature distribution data, the color distribution data, and the material distribution data, thereby determining the third threshold range. For more information, refer to the above description of matching the historical data based on the temperature distribution data and the color distribution data.

In some embodiments, the non-reflective point heat source may include a mobile heat source and a stationary heat source. A mobile heat source refers to a non-reflective point heat source in a moving state. For example, an engine of a moving car, an exhaust outlet, etc. A stationary heat source refers to a non-reflective point heat source in a stationary state. For example, an air conditioner outdoor unit, a fixed hot water cup, etc.

5 , in some embodiments, after determining that the interference category of the non-real fire corresponding to the first fire location is a non-reflective point heat source, the processor 130 may further determine whether the interference category of the non-reflective point heat source is a moving heat source or a stationary heat source.

5 , in some embodiments, the processor 130 may determine a second fire location based on a third image of the target area; determine whether the second fire location and the first fire location satisfy a third determination condition; and in response to satisfying the third determination condition, The interference category is determined to be a stationary heat source; in response to not satisfying the third judgment condition, the interference category is determined to be a moving heat source.

The third image refers to image data used to determine the fire situation. In some embodiments, the third image may be an infrared thermal image. In some embodiments, the processor 130 may acquire the third image of the target area at a time point after determining the first fire location. The acquisition method of the third image is the same as the acquisition method of the first image, which will not be repeated here.

The second fire location refers to the location area where a fire is suspected to have occurred.

In some embodiments, the processor 130 may determine the second fire location in the third image in the same manner as the first fire location. In some embodiments, the processor 130 may determine the second fire location coordinates of the second fire location based on a rectangular coordinate system established by the reference coordinate points.

In some embodiments, after the second fire location is determined, the processor 130 may align the third image with the first image, and then determine the second fire location coordinates of the second fire location based on the rectangular coordinate system established in the first image, and determine the fire area spreading outward with the second fire location as the center according to the fire area division method.

For more information on determining the first fire location, establishing a rectangular coordinate system, and dividing the fire area, refer to step 310 and its related description.

In some embodiments, the second fire location corresponds to the same fire area as the first fire location. In some embodiments, corresponding to the same fire area may mean that the fire area corresponding to the first fire location completely overlaps or partially overlaps the fire area corresponding to the second fire location. The area of the partially overlapping area may meet a preset area threshold. The preset area threshold may be a system default value, an experience value, a manually preset value, or any combination thereof, and may be set according to actual needs, and this specification does not limit this.

The third judgment condition is a judgment condition for further judging whether the interference category is a moving heat source or a stationary heat source.

In some embodiments, the third judgment condition includes: the second fire location is consistent with the first fire location. In some embodiments, the second fire location is consistent with the first fire location, which may mean that the second fire location coordinates are consistent with the first fire location coordinates. Accordingly, the processor 130 may determine whether the second fire location and the first fire location meet the third judgment condition based on the second fire location coordinates of the second fire location and the first fire location coordinates of the first fire location.

If they are consistent, it is determined that the non-real fire at the first fire position corresponding to the above-mentioned first fire position coordinates belongs to a static heat source, such as the engine or exhaust pipe position of a stationary car; if they are inconsistent, it is determined that the non-real fire at the first fire position corresponding to the above-mentioned first fire position coordinates belongs to a mobile heat source, such as the engine of a moving car, a mobile hot water cup, etc.

In some embodiments of this specification, by comparing the first fire location coordinates with the second fire location coordinates in the above manner, it is possible to determine whether the unreal fire source comes from a stationary heat source or a mobile heat source. At the same time, the above manner only calculates the RGB value for the suspicious location, which can greatly reduce the amount of calculation and is convenient and fast.

FIG6 is an exemplary schematic diagram of a fire condition judgment processing process according to some embodiments of this specification. Referring to FIG6, in some embodiments, the fire condition judgment processing process includes:

S1. Acquire infrared thermal images and visible light images of the target area.

S2. Determine the first fire location based on the infrared thermal image.

S3. Obtain the first fire location coordinates of the first fire location.

S4. Determine a matching area of the first fire position in the visible light image. For example, the first fire position in the infrared thermal image may be mapped to the visible light image to determine a matching area.

S5. Get the RGB value of the matching area.

Next, determining the authenticity of the fire corresponding to the first fire location includes:

S6. Determine whether the R value and the G value in the RGB value meet the first threshold range, and whether the B value meets the second threshold range.

S7. In response to the R value and the G value satisfying the first threshold range and the B value satisfying the second threshold range, determining that the fire corresponding to the first fire location is a real fire.

S8. In response to the R value and/or the G value not satisfying the first threshold range, and/or the B value not satisfying the second threshold range, determining that the fire corresponding to the first fire location is a non-real fire.

After determining that the fire corresponding to the first fire location is an unreal fire, the next step of judging operation may be performed to determine the interference category of the unreal fire, including:

S9. Determine whether the R value, the G value, and the B value in the RGB value meet a third threshold range.

S10 , in response to the R value, the G value, and the B value satisfying a third threshold range, determining that the interference category of the non-real fire situation is a reflective point heat source.

S11 , in response to the R value, the G value and/or the B value not satisfying the third threshold range, determining that the interference category of the non-real fire situation is a non-reflective point heat source.

After determining that the interference category of the non-real fire corresponding to the first fire location is a non-reflective point heat source, it is also possible to further determine whether the interference category of the non-reflective point heat source is a moving heat source or a stationary heat source, including:

S12: Determine whether the second fire location is consistent with the first fire location.

S13: In response to the second fire location being consistent with the first fire location, determining that the interference category of the non-real fire is a stationary heat source.

S14: In response to the second fire location being inconsistent with the first fire location, determining that the interference category of the non-real fire is a moving heat source.

After confirming that the fire corresponding to the first fire location is a real fire, report the fire.

After determining that the fire corresponding to the first fire location is not actually acquired and further determining the interference category of the non-real fire source, the fire is reported. For more information, see the related description above.

Visible light images can assist in determining the authenticity of suspicious points that have been identified as fires by infrared thermal imaging grayscale images, eliminating misjudgments in many situations, such as strong sun reflections, high-temperature locations such as stationary car engines or exhaust pipes, and moving high-temperature objects, which interfere with the judgment results and improve the accuracy of fire judgments.

Based on the same inventive concept, an embodiment of this specification also provides a fire situation judgment device, which includes at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute at least part of the computer instructions to implement the fire situation judgment method discussed above.

Based on the same inventive concept, an embodiment of this specification also provides a storage medium, which stores computer instructions. When the computer instructions are executed on a computer, the computer executes the fire situation judgment method discussed above.

In some possible implementations, various aspects of the fire situation determination method provided in the embodiments of this specification may also be implemented in the form of a program product, which includes a program code. When the program product is run on the device, the program code is used to enable the control device to execute the steps of the fire situation determination method according to various exemplary embodiments of the present application described above in this specification.

When the operations performed in the embodiments of this specification are described in steps, unless otherwise specified, the order of the steps is interchangeable, steps may be omitted, and other steps may be included in the operation process.

The description of the system and its modules in the embodiments of this specification is only for convenience of description and is not limited to the scope of the embodiments. It is possible to combine the modules arbitrarily or form a subsystem connected with other modules without departing from the principle of the system.

The embodiments in this specification are only for illustration and description, and do not limit the scope of application of this specification. For those skilled in the art, various modifications and changes that can be made under the guidance of this specification are still within the scope of this specification. Inside.

Certain features, structures or characteristics of one or more embodiments of this specification may be appropriately combined.

Various aspects of this specification may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may be referred to as "data blocks", "modules", "engines", "units", "components" or "systems", etc. In addition, various aspects of this specification may be expressed as a computer product located in one or more computer-readable media, which includes computer-readable program code.

Computer storage media can be any computer-readable media that can be connected to an instruction execution system, device or apparatus to communicate, propagate or transport the program for use. The program code on the computer storage medium can be transmitted via any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

The computer program code required for the operation of the various parts of this specification can be written in any one or more programming languages. The program code can be run entirely on the user's computer, or run on the user's computer as a separate software package, or run partially on the user's computer and partially on a remote computer, or run entirely on a remote computer or processing device. In the latter case, the remote computer can be connected to the user's computer through any network form, such as a local area network (LAN) or a wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as software as a service (SaaS).

In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used to describe the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise specified, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may change according to the required features of individual embodiments. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, such numerical values are set as accurately as possible within the feasible range.

Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, as an example and not a limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

A method for judging a fire situation, characterized in that the method comprises:

determining a first fire location based on the first image of the target area;

Determine a matching area of the first fire location in the second image of the target area, and extract color information of the matching area, wherein a shooting time difference between the first image and the second image satisfies a preset condition;

Based on the color information, determine whether the fire corresponding to the first fire location is a real fire.
The method according to claim 1, characterized in that the first image is an infrared thermal image and the second image is a visible light image.
The method according to claim 1, wherein the color information includes an RGB value, and the determining, based on the color information, whether the fire corresponding to the first fire location is a real fire comprises:

In response to the RGB value satisfying a first judgment condition, determining that the fire corresponding to the first fire position is the real fire;

In response to the RGB value not satisfying the first judgment condition, it is determined that the fire corresponding to the first fire location is not a real fire.
The method according to claim 3 is characterized in that the first judgment condition includes: the R value and the G value of the RGB value meet the first threshold range, and the B value meets the second threshold range.
The method according to claim 3, characterized in that the method further comprises:

In response to the fire corresponding to the first fire location being the unreal fire, an interference category of the unreal fire is determined.
The method according to claim 5, wherein the fire corresponding to the first fire position is the unreal fire, and determining the interference category of the unreal fire comprises:

In response to the RGB value satisfying a second judgment condition, determining that the interference category is a reflective point heat source;

In response to the RGB value not satisfying the second judgment condition, the interference category is determined to be a non-reflective point heat source.
The method according to claim 6, characterized in that the second judgment condition includes: the R value, the G value and the B value in the RGB value meet a third threshold range.
The method according to claim 6, characterized in that the non-reflective point heat source includes a moving heat source and a stationary heat source, and after determining that the interference category of the non-real fire corresponding to the first fire position is the non-reflective point heat source, the method further includes:

determining a second fire location based on a third image of the target area, wherein the second fire location corresponds to the same fire area as the first fire location;

Determining whether the second fire location and the first fire location satisfy a third determination condition;

In response to satisfying the third judgment condition, determining that the interference category is the stationary heat source;

In response to the third judgment condition not being satisfied, the interference category is determined to be the mobile heat source.
The method according to claim 8 is characterized in that the third judgment condition includes: the second fire location is consistent with the first fire location.
A fire situation judgment system, characterized in that the system comprises:

A determination module, configured to determine a first fire location based on a first image of a target area;

an extraction module, configured to determine a matching area of the first fire location in the second image of the target area, and extract color information of the matching area, wherein a shooting time difference between the first image and the second image satisfies a preset condition;

A judgment module is used to judge whether the fire corresponding to the first fire position is a real fire based on the color information.
The system according to claim 10, characterized in that the first image is an infrared thermal image and the second image is a visible light image.
The system of claim 10, wherein the color information comprises RGB values, and the determination module is further configured to:

In response to the RGB value satisfying a first judgment condition, determining that the fire corresponding to the first fire position is the real fire;

In response to the RGB value not satisfying the first judgment condition, it is determined that the fire corresponding to the first fire location is not a real fire.
The system as described in claim 12 is characterized in that the first judgment condition includes: the R value and the G value of the RGB value meet the first threshold range, and the B value meets the second threshold range.
The system according to claim 12, wherein the determination module is further configured to:

In response to the fire corresponding to the first fire location being the unreal fire, an interference category of the unreal fire is determined.
The system according to claim 14, characterized in that the determination module is further used to:

In response to the RGB value satisfying a second judgment condition, determining that the interference category is a reflective point heat source;

In response to the RGB value not satisfying the second judgment condition, the interference category is determined to be a non-reflective point heat source.
The system as claimed in claim 15, characterized in that the second judgment condition includes: the R value, the G value and the B value in the RGB value meet a third threshold range.
The system according to claim 15, characterized in that the non-reflective point heat source includes a moving heat source and a stationary heat source, and after determining that the interference category of the non-real fire corresponding to the first fire position is the non-reflective point heat source, the judgment module is further used to:

determining a second fire location based on a third image of the target area, wherein the second fire location corresponds to the same fire area as the first fire location;

Determining whether the second fire location and the first fire location satisfy a third determination condition;

In response to satisfying the third judgment condition, determining that the interference category is the stationary heat source;

In response to the third judgment condition not being satisfied, the interference category is determined to be the mobile heat source.
The system as described in claim 17 is characterized in that the third judgment condition includes: the second fire location is consistent with the first fire location.
A fire situation judgment device, characterized in that the device comprises at least one processor and at least one memory;

The at least one memory is used to store computer instructions;

The at least one processor is used to execute at least part of the computer instructions to implement the fire situation judgment method according to any one of claims 1-9.
A computer-readable storage medium storing computer instructions, characterized in that after a computer reads the computer instructions in the storage medium, the computer executes a fire situation determination method as described in any one of claims 1 to 9.