A flame detection method, device, electronic equipment and readable storage medium

Abstract

The application discloses a flame detection method and device, electronic equipment and a readable storage medium. The method comprises the following steps: receiving a first alarm signal, wherein the first alarm signal is used for indicating that a flame is detected in a region; after receiving the first alarm signal, it is judged whether a second alarm signal is received, wherein the second alarm signal is used for indicating that a flame is detected in the region, and the second alarm signal is sent by other flame detection systems in the multiple flame detection systems after detecting the flame; in the case that at least one second alarm signal is received, it is determined that a fire occurs in the region, and a fire alarm is performed. The application solves the problem of waste of fire-fighting resources caused by misjudgment when using one flame detection system to detect in the prior art, thereby reducing the misjudgment rate of flame judgment and providing a technical guarantee for improving the accuracy of fire warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of flame detection, in particular, to a flame detection method and device, an electronic device and a readable storage medium. BACKGROUND

[0002] Fire is one of the reasons for property loss. In order to reduce the loss caused by fire, various fire detection systems have been developed in the prior art, which can detect fire and alarm after detecting fire, thereby helping to discover fire as soon as possible.

[0003] Commonly used fire detection systems include thermal imaging detection systems, visible light detection systems, etc. The thermal imaging detection system images by temperature and then determines whether a flame appears. The thermal imaging system relies on temperature sensors, and only when the temperature is higher than a predetermined threshold, an alarm will be generated. In actual situations, for some industrial equipment with high working temperature, the temperature may be higher than the predetermined threshold when it is working normally, and at this time, false alarm problems may occur, causing waste of fire fighting resources. For the visible light detection system, it mainly detects whether a flame will appear according to the image shot by visible light. When a color similar to the flame appears in the image (for example, a red flag, morning or evening glow), false judgment may also occur.

[0004] In summary, when using one kind of fire detection system to detect in the prior art, false judgment is prone to occur, thereby causing waste of fire fighting resources. SUMMARY

[0005] The embodiments of the present application provide a flame detection method, device, electronic device and readable storage medium to at least solve the problem of waste of fire fighting resources caused by false judgment when using one kind of fire detection system to detect in the prior art.

[0006] According to one aspect of the present application, a flame detection method is provided, comprising: receiving a first alarm signal, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent after a flame is detected by one of a plurality of flame detection systems, and the plurality of flame detection systems use different flame detection methods for flame detection; after receiving the first alarm signal, determining whether a second alarm signal is received, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent after a flame is detected by a flame detection system other than the flame detection system sending the first alarm signal among the plurality of flame detection systems; in the case that at least one second alarm signal is received, it is determined that a fire occurs in the region, and a fire alarm is performed.

[0007] Further, in the case that the plurality of flame detection systems comprises a dual light fusion flame detection system, the method further comprises: acquiring a first image captured by a thermal imaging lens of the dual light fusion flame detection system and a second image generated by a visible light lens of the dual light fusion flame detection system; acquiring a first coordinate of a flame detected in one of the first image and the second image in the image; converting the first coordinate of the flame in the one image into a second coordinate on the other one of the first image and the second image according to a straight line perpendicular distance from the thermal imaging lens and the visible light lens to the flame, wherein the straight line distance from the thermal imaging lens to the flame is equal to the straight line distance from the visible light lens to the flame; and determining whether the flame is detected at the second coordinate, and if the flame is detected, determining that the flame is detected by the dual light fusion flame detection system.

[0008] Further, converting the first coordinate in the one image into the second coordinate in the other one according to the straight line perpendicular distance comprises: acquiring a first deflection angle of the flame relative to a central axis of a first lens according to the first coordinate, a resolution of the one image and a field of view angle of the first lens; acquiring a second deflection angle of the flame relative to a central axis of a second lens according to the straight line perpendicular distance, the first deflection angle and a distance between the first lens and the second lens, wherein the second lens is a lens for capturing the other one of the first image and the second image; and acquiring the second coordinate according to the second deflection angle, a resolution of the other one of the first image and the second image and a field of view angle of the second lens.

[0009] Further, the first deflection angle d1 of the flame relative to the central axis of the first lens is acquired according to the following formula: d1 = ((first coordinate - first image size / 2) / first image size)*field of view angle of the first lens, wherein if a horizontal deflection angle is calculated, the first coordinate is a coordinate in a horizontal direction, the first image size is a width of the image, and the field of view angle of the first lens is a horizontal field of view angle; if a vertical deflection angle is calculated, the first coordinate is a coordinate in a vertical direction, the first image size is a height of the image, and the field of view angle of the first lens is a vertical field of view angle; and the second deflection angle d2 of the flame relative to the central axis of the second lens is acquired according to the following formula: d2 = arctan((XL X) / YL), wherein X is a distance between central axes of the two lenses, XL is a straight line parallel distance from a central axis of the thermal imaging lens to the flame, YL is a straight line perpendicular distance from the first lens and the second lens to the flame, X and YL are measured, XL = tan(d1)*YL, and the straight line parallel distance from the central axis of the visible light lens to the flame is XL X, the flame is on the right side of the center axis of the visible light lens, XL+X; if the flame is on the left side of the center axis of the visible light lens, XL-X; the value of the second deflection angle is substituted into the following equation to obtain the second coordinate: second deflection angle=((second coordinate-second image size / 2) / second image size)*second lens field of view angle, wherein if the horizontal direction coordinate in the second coordinate is calculated, the horizontal deflection angle is used, the second image size is the width of the image, and the second lens field of view angle is the horizontal field of view angle; if the vertical direction coordinate in the second coordinate is calculated, the vertical deflection angle is used, the second image size is the height of the image, and the second lens field of view angle is the vertical field of view angle.

[0010] Further, the first deflection angle of the position relative to the center axis of the first lens is obtained according to the first coordinate, the resolution of the image in which the first coordinate is located, and the field of view angle of the first lens for shooting the image, which includes: obtaining the deviation angle of the first lens, wherein the deviation angle is the angle deviation between the theoretical position and the actual position of the calibration object in the image of the first lens when the calibration object is in the center of the image of the second lens; obtaining the deflection angle of the flame relative to the center axis of the first lens according to the first coordinate, the resolution of the image in which the first coordinate is located, and the field of view angle of the first lens for shooting the image; and correcting the obtained deflection angle relative to the center axis of the first lens by using the deviation angle to obtain the first deflection angle.

[0011] Further, the first deflection angle is obtained by correcting the obtained deflection angle relative to the center axis of the first lens by using the deviation angle according to the following formula: the first deflection angle=((first coordinate-first image size / 2) / first image size)*first lens field of view angle-deviation angle.

[0012] Further, the deviation angle is calculated according to the following formula: deviation angle=((actual position coordinate-theoretical position coordinate) / first image size)*first lens field of view angle.

[0013] According to another aspect of the present application, a flame detection device is also provided, comprising: a receiving module configured to receive a first alarm signal, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent by one of a plurality of flame detection systems after detecting the flame, and the plurality of flame detection systems use different flame detection methods to detect the flame; a judging module configured to judge whether a second alarm signal is received after receiving the first alarm signal, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent by a flame detection system other than the flame detection system sending the first alarm signal among the plurality of flame detection systems after detecting the flame; and an alarming module configured to determine that a fire occurs in the region and perform a fire alarm in the case that at least one second alarm signal is received.

[0014] Further, in the case that the plurality of flame detection systems comprises a dual-optical fusion flame detection system, the device further comprises a flame detection module in the dual-optical fusion flame detection system; the flame detection module is configured to acquire a first image captured by a thermal imaging lens in the dual-optical fusion flame detection system and a second image generated by a visible light lens in the dual-optical fusion flame detection system; acquire a first coordinate of a flame detected in one of the first image and the second image in the image; convert the first coordinate of the flame in the one image into a second coordinate on another image of the first image and the second image according to a perpendicular distance of a straight line from the thermal imaging lens and the visible light lens to the flame, wherein the perpendicular distance of the straight line from the thermal imaging lens to the flame is equal to the perpendicular distance of the straight line from the visible light lens to the flame; judge whether the flame is detected at the second coordinate, and if the flame is detected, determine that the flame is detected by the dual-optical fusion flame detection system.

[0015] Further, the flame detection module is configured to: acquire a first deflection angle of the flame relative to a central axis of a first lens according to the first coordinate, a resolution of the one image in which the first coordinate is located, and a field of view angle of the first lens capturing the one image; acquire a second deflection angle of the flame relative to a central axis of a second lens according to the perpendicular distance of the straight line, the first deflection angle, and a distance between the first lens and the second lens, wherein the second lens is a lens capturing the other image; and acquire the second coordinate according to the second deflection angle, a resolution of the other image, and a field of view angle of the second lens.

[0016] Further, the flame detection module is configured to obtain a first deflection angle of the flame relative to a central axis of the first lens according to a formula: the first deflection angle d1 = ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens, wherein if a horizontal deflection angle is calculated, the first coordinate is a coordinate in a horizontal direction, the first image size is a width of the image, and the field of view angle of the first lens is a horizontal field of view angle; if a vertical deflection angle is calculated, the first coordinate is a coordinate in a vertical direction, the first image size is a height of the image, and the field of view angle of the first lens is a vertical field of view angle; and obtain a value of a second deflection angle of the flame relative to a central axis of the second lens according to a formula: the second deflection angle d2=arctan((XL X) / YL), wherein X is a distance between the central axes of the two lenses, XL is a parallel distance of a straight line from the central axis of the thermal imaging lens to the flame, and YL is a vertical distance of a straight line from the first lens and the second lens to the flame, X and YL are obtained by measurement, XL=tan(d1)*YL, and the parallel distance of a straight line from the central axis of the visible light lens to the flame is XL X, and when the flame is on the right side of the central axis of the visible light lens, XL+X; when the flame is on the left side of the central axis of the visible light lens, XL-X; and the value of the second deflection angle is substituted into an equation to obtain the second coordinate: the second deflection angle= ((second coordinate-second image size / 2) / second image size)*field of view angle of the second lens, wherein if a horizontal coordinate in the second coordinate is calculated, a horizontal deflection angle is used, the second image size is a width of the image, and the field of view angle of the second lens is a horizontal field of view angle; if a vertical coordinate in the second coordinate is calculated, a vertical deflection angle is used, the second image size is a height of the image, and the field of view angle of the second lens is a vertical field of view angle.

[0017] Further, the flame detection module is configured to obtain a deviation angle of the first lens, wherein the deviation angle is an angle deviation between a theoretical position of a calibration object in a picture of the first lens and an actual position of the calibration object in the picture of the first lens when the calibration object is in the center of a picture of the second lens; obtain a deflection angle of the flame relative to a central axis of the first lens according to the first coordinate, a resolution of an image in which the first coordinate is located, and a field of view angle of the first lens used to capture the image; and correct the obtained deflection angle relative to the central axis of the first lens using the deviation angle to obtain the first deflection angle.

[0018] Further, the flame detection module is configured to obtain the first deflection angle according to a formula: the first deflection angle = ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens-deviation angle.

[0019] Further, the flame detection module is configured to calculate the deviation angle using the following formula: deviation angle - ((actual position coordinate - theoretical position coordinate) / first image size) * field of view angle of the first lens.

[0020] According to another aspect of the present application, an electronic device is also provided, comprising a memory and a processor; wherein the memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method steps described above.

[0021] According to another aspect of the present application, a readable storage medium is also provided, having computer instructions stored thereon, wherein the computer instructions are executed by a processor to implement the method steps described above.

[0022] In the embodiments of the present application, a first alarm signal is received, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent by one of a plurality of flame detection systems after detecting a flame, wherein the plurality of flame detection systems use different flame detection methods for flame detection; after receiving the first alarm signal, it is determined whether a second alarm signal is received, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent by a flame detection system other than the flame detection system sending the first alarm signal among the plurality of flame detection systems after detecting a flame; in the case that at least one second alarm signal is received, it is determined that a fire occurs in the region, and a fire alarm is performed. The present application solves the problem of waste of fire-fighting resources caused by false judgment when using one flame detection system for detection in the prior art, thereby reducing the false judgment rate of flame judgment, and providing a technical guarantee for improving the accuracy of fire warning. BRIEF DESCRIPTION OF DRAWINGS

[0023] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and the illustrative embodiments of the present application and their description serve the purpose of explaining the present application. The accompanying drawings should not be construed as an inappropriate limitation on the present application. In the drawings:

[0024] Figure 1 is a flowchart of a flame detection method according to an embodiment of the present application;

[0025] Figure 2 is a flowchart of a post-revision of flame detection according to an embodiment of the present application;

[0026] Figure 3 is a schematic diagram of image coordinate conversion according to an embodiment of the present application;

[0027] Figure 4is a schematic diagram of lens deviation according to an embodiment of the present application;

[0028] Figure 5 is a schematic diagram of deviation angle calculation according to an embodiment of the present application; and,

[0029] Figure 6 is a schematic diagram of fire alarm according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0031] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0032] In the following embodiments, a flame detection system is involved, which can be a hardware-based flame detection system or a software system, which can be a flame detection algorithm, which is also referred to as a flame detection system in the following embodiments. In most cases, the flame detection system is a combination of hardware and software, in which relevant data can be obtained through hardware, and then the obtained data is processed through software to determine whether there is a flame. Considering that the prior art uses a flame detection system to determine whether a flame occurs, this determination method is prone to misjudgment, and after misjudgment, fire alarm will cause waste of fire fighting resources.

[0033] In the following embodiments, multiple flame detection systems are introduced, which use different flame detection methods for flame detection. Considering that if a real fire occurs, multiple flame detection systems can detect the flame and alarm; if a false alarm occurs in a certain flame detection system, since different detection systems use different flame detection methods, the probability of false alarm in other flame detection systems is not high. Therefore, after using multiple flame detection systems, if at least two flame detection systems identify a flame, fire alarm can be performed, which reduces the false alarm rate and also accurately alarms when a real fire occurs, so that fire fighting resources are fully utilized.

[0034] In the following embodiments, a flame detection method is provided, Figure 1 is a flowchart of a flame detection method according to an embodiment of the present application, as Figure 1The following will be explained in detail with reference to the steps involved in the method. Figure 1

[0035] In step S102, a first alarm signal is received, wherein the first alarm signal is used to indicate that a flame is detected in the area, and the first alarm signal is sent by one of a plurality of flame detection systems after detecting the flame, wherein the plurality of flame detection systems use different flame detection methods to detect the flame.

[0036] In step S104, after receiving the first alarm signal, it is determined whether a second alarm signal is received, wherein the second alarm signal is used to indicate that a flame is detected in the area, and the second alarm signal is sent by one of the plurality of flame detection systems other than the flame detection system sending the first alarm signal after detecting the flame.

[0037] As an optional embodiment, in this step, after receiving the first alarm signal, a time period (e.g., 2 minutes or 100 seconds, etc.) can be set, and if the second alarm signal is not received within the time period, it means that the first alarm signal is a false alarm, and no fire alarm is performed. If the alarm signal is received after the time period, the alarm signal is still regarded as the first alarm signal, and it is still necessary to determine whether the alarm signal from other flame detection systems is received. The time period can be set according to experience, or can be set according to the detection speed of the plurality of flame detection systems, and the time period is greater than the detection time required by the slowest detection system of the plurality of flame detection systems.

[0038] In another case, if the alarm signal from the same flame detection system is identified as a false alarm more than a threshold number of times within a time period, the alarm signal from the flame detection system identified as a false alarm is sent to the user. This processing manner can on one hand allow the user to determine whether the alarm signal from the flame detection system is a false alarm, and can prevent false negatives to a certain extent; on the other hand, it can also allow the user to accumulate data, and the accumulated data is used to analyze the cause of the false alarm of the flame detection system, and provides data support for further improvement of the flame detection system.

[0039] In step S106, in the case where at least one second alarm signal is received, it is determined that a fire occurs in the area, and a fire alarm is performed.

[0040] In this step, as an optional embodiment, when the fire alarm is performed, it can also be prompted which detection results of the fire detection systems are used to perform the alarm, so that more information can be provided to the staff, which is beneficial for further processing. ​

[0041] In the above steps, after receiving the first alarm signal, it is indicated that the flame detection system has detected the emergence of flame, at this time, it is judged whether the second alarm signal can be received, if the second alarm signal emitted by the other flame detection system in detecting the flame is received, it is determined that the first alarm signal is not a false alarm, in this case, it is determined that the fire occurs in the region and the fire alarm is performed. Therefore, the problem of waste of fire-fighting resources caused by false judgment when using one flame detection system for detection in the prior art is solved by the above steps, thereby reducing the false judgment rate of flame judgment, and providing a technical guarantee for improving the accuracy of fire warning.

[0042] In an optional embodiment, the flame detection systems of different detection modes can be selected as needed, and considering that in order to reduce the false alarm rate, the selected multiple flame detection systems can be systems based on different physical properties of the flame for detection. For example, an image detection system based on the color of the flame can be selected, which identifies the region of a predetermined color as a flame by means of image processing according to the color attribute of the flame; for example, a thermal imaging detection system based on the temperature of the flame can also be selected, which identifies a region above a temperature threshold and takes the region as a flame; for example, a detection system based on the optical properties of the flame can also be selected, which determines the region emitting a specific frequency of light (for example, ultraviolet light and / or infrared light) as a flame after detecting the emission of the specific frequency of light. For example, a system for detecting the posture of the flame can also be selected, in which the change of the shape of the color block is determined by recording the video to determine whether the color block is a flame. Of course, systems for detecting the flame according to other physical properties of the flame can also be selected, which will not be described again.

[0043] Or the selected flame detection systems can also be flame detection systems based on the same physical properties but using different detection algorithms. For example, two systems for detecting the flame according to the color of the flame can be included in the multiple flame detection systems, the difference being that the algorithms used by the two flame detection systems are different, since different algorithms are used, the results detected by the two systems will be different in some cases (one of which must be a false judgment), at this time, the fire alarm is performed only when the two systems both detect the flame, which can reduce the false alarm rate. Therefore, the flame detection systems based on the same physical properties of the flame but using different detection algorithms can also be used as different flame detection systems.

[0044] In an optional embodiment, in order to further reduce the false alarm rate, the visible light-based image detection and the thermal imaging-based image detection can be integrated into a flame detection system, for the convenience of description, the flame detection system integrating visible light and thermal imaging is referred to as a dual-light fusion flame detection system, wherein the dual-light fusion flame detection system is used for flame detection according to a first image generated based on thermal imaging and a second image generated based on visible light. The dual-light fusion flame detection system detects based on the temperature feature and the color feature of the flame, and then a system based on another feature of the flame other than the temperature and the color can be selected to be combined with the dual-light fusion flame detection system, and in the optional embodiment, a system based on the frequency of light is selected to be used for flame detection, for example, a point-type flame detection system, wherein the point-type flame detection system is used for flame detection according to the frequency of light emitted when the flame burns. The two selected flame detection systems are based on different physical properties of light for detection, and in addition, in the optional embodiment, a flame detection system using a different algorithm can be introduced, for example, a machine learning-based flame detection system is introduced, wherein the machine learning-based flame detection system is used for flame detection based on a trained neural network model, and the neural network model can be obtained through supervised training, and in the supervised training, a plurality of sets of training data are used, and each set of training data can be a labeled image, wherein the label indicates whether the flame exists in the image, and after training, the neural network model can be used. Alternatively, a plurality of key frames can be extracted from a video, and the flame target is tracked and labeled in the plurality of key frames, and the color feature, the edge feature, the circularity and the texture feature of the flame target in the plurality of key frames are extracted, and the features are input into the neural network model as training data, and after training, the neural network model can be used.

[0045] Through the above optional embodiment, the dual-light fusion flame detection system, the point-type flame detection system and the machine learning-based flame detection system are selected, and the three systems not only include systems using different physical properties of the flame for detection, but also include detection systems using different algorithms, and in the three systems, if a system detects the flame, it does not alarm temporarily, at this time, it is judged whether the remaining systems also detect the alarm, which is equivalent to a review, and after the other systems detect the alarm, the authenticity of the flame is determined, that is, any at least two systems in the three systems detect the flame and then perform fire alarm, which can further reduce the false alarm rate.

[0046] Figure 2 is a flowchart of the flame detection and review according to the embodiment of the application, as Figure 2As shown, the flames detected by the three flame detection systems are respectively referred to as A-type targets, B-type targets and C-type targets (A, B and C are used to distinguish the flames detected by the three flame detection systems), after receiving the A-type target, it is determined whether the B-type target is detected, if the B-type target does not appear, it means that the A-type target is a false alarm. If the B-type target appears, it is further determined whether the regions where the B-type target and the A-type target are located are consistent, so as to determine whether the A-type target and the B-type target are the same flame, at this time, the B-type target is traversed to determine whether there is an intersecting target (i.e. the A-type target and the B-type target are in the same region), if there is, it is considered that the A-type target is rechecked and passed, and fire alarm is performed; if there is no B-type target intersecting with the A-type target, the C-type target is traversed to determine whether there is a target intersecting with the A-type target, if there is not, it is considered that the A-type target is a false alarm, if there is, it is considered that the A-type target is rechecked and passed, and fire alarm is performed.

[0047] The above three flame detection systems can be configured to alarm after any two flame detection systems detect a flame. Alternatively, considering that the dual-light fusion flame detection system has higher accuracy, it can also be configured to receive an alarm signal sent by the dual-light fusion flame detection system after detecting a flame, and then determine whether an alarm signal from the point-type flame detection system and / or the machine learning-based flame detection system is received, i.e. the first alarm signal is from the dual-light fusion flame detection system, and the second alarm signal is from the point-type flame detection system and / or the machine learning-based flame detection system. If the alarm signal from the point-type flame detection system and / or the machine learning-based flame detection system is received after receiving the alarm signal from the dual-light fusion flame detection system, fire alarm is performed.

[0048] In the dual-light fusion flame detection system, two types of lenses are used, one type of lens is used for thermal imaging and photographing (the image captured by this type of lens is referred to as a first image), and the other type of lens is used for photographing using visible light (the image captured by this type of lens is referred to as a second image). There are many ways to detect flames in the first image and the second image, which are illustrated as follows.

[0049] In the first image, the manner of detecting the flame can pre-configure a temperature threshold (for example, 80 degrees or 100 degrees), and points higher than the temperature threshold are regarded as points where the flame exists. A plurality of points where the flame exists are connected to form a region, and the position of the region is regarded as the position of the flame. That is, in this example, obtaining the position of the detected flame in the first image can include the following steps: screening a plurality of temperature points greater than the temperature threshold according to the temperature threshold; obtaining a region including the plurality of temperature points according to the plurality of temperature points, wherein the region is a region where the flame is detected in the first image; and obtaining the position of the region, wherein the position of the region is the position of the flame in the first image. In this example, after a plurality of temperature points greater than the temperature threshold are detected, a binaryzation process can be performed, that is, the pixel value of the temperature point greater than the temperature threshold can be configured as 1, and the pixel value of other points can be configured as 0. After binaryzation, the points with a pixel value of 1 are connected to obtain the region. Through the algorithm in this example, not only the region where the flame is located can be obtained, but also the area covered by the flame (which can be determined according to the distance from the lens to the flame measured by the range finder and the proportion of the flame region in the first image) can be obtained. If fire alarm is needed, different levels of alarm signals can be sent according to the area covered by the flame. The larger the area covered by the flame, the higher the level of the alarm signal sent.

[0050] In the second image, the manner of detecting the flame can be to determine whether the flame appears in the second image according to a pre-obtained flame color model. The flame color model includes color characteristics of the flame. Pixel points meeting the color characteristics of the flame are searched for in the second image, and a region where the flame is located can be obtained by connecting a plurality of pixel points meeting the color characteristics of the flame. The position of the region in the second image is the position of the flame in the second image. In the previous example, the area of the flame in the first image and the real area of the flame can be obtained. Similarly, in this example, the area of the flame in the second image and the real area of the flame calculated based on the area in the second image can also be obtained.

[0051] In the above two examples, after the region where the flame is located is obtained, a rectangular frame can be used to cover the region where the flame is located in the rectangular frame, and the rectangular frame is used to represent the position of the flame in the first image and the second image.

[0052] In the dual light fusion flame detection system, if the alarm signal is sent after the flame is detected according to the image captured by one of the two types of lenses in order to reduce the false negative rate, the false positive rate is relatively high. In an optional embodiment, the flame positions in the first image and the second image captured by the two lenses can be used for verification to improve the accuracy of flame identification. That is, the flame detection according to the first image and the second image can include the following steps: obtaining the position of the flame detected in one of the first image and the second image in the image; determining whether the flame is detected at the same position in the other image of the first image and the second image; if the flame is detected at the same position, it is determined that the flame is detected by the dual light fusion flame detection system. It should be noted that since the flame is a region in the first image and the second image (assuming the flame is a first region in the first image and a second region in the second image), if the overlapping part of the first region and the second region exceeds a threshold, it is considered that the flame is detected at the same position in the first image and the second image. For example, the first region covers the second region or the second region covers the first region, in which case it is considered that the overlapping part of the first region and the second region exceeds the threshold; for another example, although the first region and the second region do not cover each other, the overlapping part of the first region and the second region exceeds a predetermined percentage (such as 60%) of the area of the first region or the area of the second region, and it is considered that the flame is detected at the same position in the first image and the second image.

[0053] For the thermal imaging lens and the visible light lens, if the thermal imaging lens and the visible light lens adopt the same resolution and field of view angle, and the distance between the thermal imaging lens and the visible light lens is small (can be ignored), the coordinate systems in the first image and the second image are the same, and when comparing the positions of the first region and the second region, the coordinates in the first image and the second image can be directly compared. In actual application, considering that the resolutions and field of view angles adopted by the thermal imaging lens and the visible light lens are generally different, in an optional embodiment, the position in the first image or the second image can be converted in coordinates before being judged. That is, judging whether the same position as the position on the other image of the first image and the second image detects a flame can include the following steps: obtaining a first coordinate of the position in the one image; converting the first coordinate into a second coordinate in the other image; and judging whether the second coordinate detects a flame. In this optional embodiment, the first image photographed by the thermal imaging lens in the dual-light fusion flame detection system and the second image generated by the visible light lens in the dual-light fusion flame detection system are obtained; a first coordinate of a flame detected in one of the first image and the second image in the image is obtained; the first coordinate of the flame in the one image is converted into a second coordinate on the other image of the first image and the second image according to the straight line vertical distance from the thermal imaging lens to the visible light lens to the flame, wherein the straight line distance from the thermal imaging lens to the flame is equal to the straight line distance from the visible light lens to the flame; and it is judged whether the second coordinate detects a flame, and if a flame is detected, it is determined that a flame is detected by the dual-light fusion flame detection system. Through this optional embodiment, the coordinate of the flame position in the first image can be converted into the coordinate in the second image, and then the converted coordinate is compared with the coordinate of the flame position in the second image to determine whether a flame appears at the same position; or the coordinate of the flame position in the second image can be converted into the coordinate in the first image, and then the converted coordinate is compared with the coordinate of the flame position in the first image to determine whether a flame appears at the same position.

[0054] When performing coordinate comparison, considering that the straight-line vertical distance from the flame to the thermal imaging lens and the visible light lens (in the following text, if the thermal imaging lens is the first lens, the visible light lens is the second lens, and if the thermal imaging lens is the second lens, the visible light lens is the first lens) to the flame is the same (this distance can be measured by a rangefinder), the coordinate transformation between the first and second images can be performed using the straight-line vertical distance from the lens to the flame, as well as the resolution and field of view of the thermal imaging lens and the visible light lens. Converting the first coordinate in one of the first and second images into the second coordinate in the other image includes: obtaining a first deflection angle of the flame relative to the central axis of the first lens based on the first coordinate, the resolution of the image containing the first coordinate (the image resolution indicates the size of the image, for example, an 800*600 resolution means the image width is 800 and the height is 600), and the field of view of the first lens that captured the image; obtaining a second deflection angle of the flame relative to the central axis of the second lens based on the straight-line vertical distance from the dual-light fusion detection system to the flame, the first deflection angle, and the distance between the second lens and the first lens, wherein the second lens is the lens that captured the other image; and obtaining the second coordinate based on the second deflection angle, the resolution of the other image, and the field of view of the second lens.

[0055] Among the above optional coordinate calculation methods, the deflection angle is used for calculation. Figure 3 This is a schematic diagram of image coordinate transformation according to an embodiment of this application. Figure 3 The calculation method for the horizontal direction is shown, while the calculation method for the vertical direction is the same as that for the horizontal direction. Therefore, the coordinates are not distinguished into horizontal and vertical coordinates. Figure 3 As shown, assume the first lens is a thermal imaging lens, the first image is a thermal image, the second lens is a visible light lens, and the second image is a visible light image. Figure 3 Lieutenant General Flame ( Figure 3The dark gray block in the first image represents the flame. The coordinates of the flame in the first image are converted into the coordinates of the second image. The distance WL of the flame in the image from the center axis of the thermal imaging image picture is (first coordinate-first image size / 2), and thus the first deflection angle d1 is = ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens. It should be noted that in the formula, if the horizontal deflection angle is calculated, the first coordinate is the coordinate in the horizontal direction, the first image size is the width of the image, and the field of view angle of the first lens is the horizontal field of view angle; if the vertical deflection angle is calculated, the first coordinate is the coordinate in the vertical direction, the first image size is the height of the image, and the field of view angle of the first lens is the vertical field of view angle. In another calculation method, the distance WL can also be uniformly normalized, for example, considering the width of the image as 1, and the planning value of the first coordinate is the first coordinate / width of the first image when the first coordinate is the coordinate in the horizontal direction, and thus the above formula can also be described as: the distance WL (normalized to 0~1) of the flame in the image from the actual center axis of the thermal imaging image picture is (first coordinate / first image width)-0.5), and thus the first deflection angle d1 is = (first coordinate / first image width-0.5)*field of view angle of the first lens (in the calculation of the horizontal deflection angle, the same as Figure 3 The horizontal field of view angle a1 and the first image width and the first coordinate in the horizontal direction are used in the calculation of the horizontal deflection angle; the vertical field of view angle and the height of the first image and the first coordinate in the vertical direction are used in the calculation of the vertical deflection angle).

[0056] The straight line vertical distance (or straight line distance vertical component) of the first lens and the second lens to the flame is the same, that is, Figure 3 YL, according to the right triangle formula, the straight line parallel distance (or straight line distance horizontal component) of the thermal imaging lens center axis to the flame is XL=tan(d1)*YL, and thus the straight line parallel distance of the visible light lens center axis to the flame is XL+X (in Figure 2 XL+X if the flame is on the right side of the visible light lens center axis; if the flame is on the left side of the visible light lens center axis, it is XL-X), where X is the distance between the two lens center axes. The second deflection angle d2=arctan((XL+X) / YL), and thus the value of the second deflection angle d2 can be obtained through the formula, and the second deflection angle = ((second coordinate-second image size / 2) / second image size)*field of view angle of the second lens (in the calculation of the horizontal deflection angle, the same as Figure 3The horizontal field of view b1 shown is used when calculating the vertical deflection angle. That is, if calculating the horizontal coordinates in the second coordinate system, the horizontal deflection angle is used, the second image size is the image width, and the second lens's field of view is the horizontal field of view; if calculating the vertical coordinates in the second coordinate system, the vertical deflection angle is used, the second image size is the image height, and the second lens's field of view is the vertical field of view. With the second deflection angle, the second lens's field of view, and the second image size (i.e., the second image resolution) all known, the second coordinate system can be obtained, thus completing the transformation from the first coordinate system to the second coordinate system.

[0057] In practical applications, lens deviation may occur. Figure 4 This is a schematic diagram illustrating lens deviation according to an embodiment of this application, such as... Figure 4 As shown, flame ( Figure 4 (The dark color block in the image represents the flame.) When the flame is in the center of the second lens's frame, the calculated theoretical position and actual position of the flame in the first lens are different. This deviation is caused by an angular deviation in the central axis of the first lens, and is therefore called the deviation angle of the first lens. After obtaining the deviation angle, the first deflection angle can be corrected using the deviation angle. That is, obtaining the first deflection angle of the position relative to the central axis of the first lens based on the first coordinate, the resolution of the image containing the first coordinate, and the field of view of the first lens capturing the image can include the following steps: obtaining the deviation angle of the first lens, wherein the deviation angle is the angular deviation between the theoretical and actual positions of the calibration object in the first lens's frame when the calibration object is in the center of the second lens's frame; obtaining the deflection angle of the flame relative to the central axis of the first lens based on the first coordinate, the resolution of the image containing the first coordinate, and the field of view of the first lens capturing the image; and correcting the obtained deflection angle relative to the central axis of the first lens using the deviation angle to obtain the first deflection angle. At this point, the first deflection angle d1 is = ((first coordinate - first image size / 2) / first image size) * the field of view angle of the first lens b1 - deviation angle. That is, the distance WL (normalized to 0~1) between the flame's position in the image and the actual central axis of the thermal imaging image is (first coordinate / first image size - (0.5 + deviation angle / field of view angle of the first lens)). Therefore, the first deflection angle d1 is = (first coordinate / first image size - (0.5 + (deviation angle / field of view angle of the first lens))) * the field of view angle of the first lens (when calculating the horizontal deflection angle, the same...). Figure 3The horizontal field of view angle a1 and the first image width in the first image are used to calculate the horizontal deviation angle, and the vertical field of view angle and the first image height are used to calculate the vertical deviation angle. The modification of the deviation angle can make the coordinate conversion between the first image and the second image more accurate. It should be noted that in actual application, if the deviation angle is small, the deviation angle can also be ignored.

[0058] There are many methods for calculating the deviation angle. For example, the deviation angle can be equal to ((actual coordinate-theoretical coordinate) / first image size)*field of view angle of the first lens. It should be noted that the deviation angle calculation formula can calculate the vertical deviation angle and the horizontal deviation angle, and the vertical coordinate and the horizontal coordinate are respectively brought into the formula.

[0059] Figure 5 The deviation angle calculated according to the embodiment of the present application is shown in the schematic diagram as shown in Figure 5 The target (also the heat source) is placed in the center of the second lens and within the shooting range of the first lens. The deviation of the actual center axis and the theoretical center axis of the first lens leads to the appearance of the deviation angle. In combination with Figure 4 , it is assumed that the horizontal field of view angle of the second lens is b1, and the vertical field of view angle is b2. The distance between the double light fusion detection system and the target is y, and the distance between the first lens and the second lens is x. The horizontal deviation angle and the vertical deviation angle can be calculated using the following formula respectively:

[0060] Theoretical horizontal coordinate=(arctan(x / y) / horizontal field of view angle b1 of the second lens+0.5)*width of the first image;

[0061] Horizontal deviation angle= ((actual horizontal coordinate-theoretical horizontal coordinate) / width of the first image)*horizontal field of view angle b1 of the second lens;

[0062] Theoretical vertical coordinate=0.5 (the same normalized value)

[0063] Vertical deviation angle= ((actual vertical coordinate-theoretical vertical coordinate) / height of the first image)*vertical field of view angle b2 of the second lens.

[0064] Through the above optional embodiment, the double light fusion flame detection system is more accurate when detecting the flame, thereby reducing the false alarm rate. In an optional embodiment, the detection results of the double light fusion flame detection system, the point type flame detection system and the detection system based on machine learning (which can also be a deep learning algorithm) are verified with each other. When at least two of the three detection systems detect the flame, the fire alarm is performed. Figure 6This is a schematic diagram of a fire alarm process according to an embodiment of this application, such as... Figure 6 As shown, after detecting a flame, a point flame detector sends an alarm signal (also called an alarm marker). Upon receiving the alarm signal, a full-screen alarm target is generated for the point flame detector. It should be noted that after sending the alarm signal, the location of the flame is determined based on the location of the point flame detector; that is, the alarm signal from a point flame detector does not specify the location of the flame. For a dual-light fusion flame detection system, the location of the flame can be marked in the image. In this system, only a portion of the image shows the flame, while the alarm signal from a point flame detector does not specify the flame's location. Therefore, when displayed on the screen, the entire screen is considered to be filled with flame, hence the term "full-screen alarm target." The dual-light fusion flame detection system first measures the temperature using a thermal imaging lens, then generates a first image based on a pre-set temperature threshold (also known as an alarm threshold). This first image includes the flame (the alarm target). Dual-light fusion is then performed, converting the flame coordinates from the first image into coordinates in a second image captured by a visible light lens. The system then determines whether a flame is detected at these coordinates in the second image. If a flame is detected, it's confirmed that a flame also exists in the visible light image, indicating that the alarm target has been detected by both visible light and temperature, thus triggering an alarm. Alternatively, visible light images can also be used to identify flame targets in deep learning algorithms. For example, this deep learning algorithm can train a neural network module. This neural network model can be obtained through supervised training, using multiple sets of training data. Each set of training data can be labeled images, indicating the presence of a flame. After training, the neural network model is ready to use. If a flame target is detected by the deep learning algorithm (also called an AI flame target because it's detected by AI), an alarm is triggered.

[0065] The review module connects the three flame detection systems, and the flames detected by the three flame detection systems are respectively referred to as A-type targets, B-type targets and C-type targets (A, B and C are used to distinguish the flames detected by the three flame detection systems). After receiving the A-type target, it is determined whether the B-type target is detected. If the B-type target does not appear, it is indicated that the A-type target is a false alarm. If the B-type target appears, it is further determined whether the regions of the B-type target and the A-type target are consistent, so that it is determined whether the A-type target and the B-type target are the same flame. At this time, the B-type target is traversed to determine whether there is an intersecting target (i.e., the A-type target and the B-type target are in the same region). If there is, it is considered that the A-type target is reviewed and passed, and fire alarm is performed. If there is no B-type target intersecting with the A-type target, the C-type target is traversed to determine whether there is a target intersecting with the A-type target. If there is no, it is considered that the A-type target is a false alarm. If there is, it is considered that the A-type target is passed, and fire alarm is performed.

[0066] The above embodiment solves the problem of waste of fire fighting resources caused by false judgment when using one flame detection system for detection in the prior art, thereby reducing the false judgment rate of flame judgment and providing a technical guarantee for improving the accuracy of fire warning.

[0067] In the embodiment, an electronic device is provided, including a memory and a processor, the memory stores a computer program, and the processor is configured to run the computer program to execute the method in the above embodiment.

[0068] The above program can be run in the processor, or can also be stored in the memory (or called computer readable medium), the computer readable medium includes permanent and non-permanent, removable and non-removable media, and can be realized by any method or technology to store information. The information can be computer readable instructions, data structure, program module or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tape, magnetic tape disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information accessible by a computing device. According to the definition in this paper, the computer readable medium does not include the transitory computer readable medium (transitory medium), such as modulated data signal and carrier wave.

[0069] These computer programs (also known as programs, software, software applications programs, applications, components, program components, or code) include any electronic component for providing Figure 1 The steps of a method or algorithm described in connection with the Figure 1 The steps of a method or algorithm described in connection with the

[0070] The present embodiment provides such a device. The device is referred to as a flame detection device, comprising: a receiving module, configured to receive a first alarm signal, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent by one of a plurality of flame detection systems after detecting the flame, and the plurality of flame detection systems use different flame detection methods for flame detection; a judging module, configured to judge whether a second alarm signal is received after receiving the first alarm signal, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent by a flame detection system other than the flame detection system sending the first alarm signal among the plurality of flame detection systems after detecting the flame; and an alarm module, configured to determine that a fire occurs in the region and perform fire alarm in the case that at least one second alarm signal is received.

[0071] The device is used to implement the functions of the method in the above-mentioned embodiments, and each module in the system or device corresponds to each step in the method, which has been described in the method and will not be repeated here.

[0072] In an optional embodiment, the plurality of flame detection systems comprises systems for detecting based on different physical properties of the flame, and / or systems for detecting based on the same physical properties of the flame and using different algorithms. For example, the plurality of flame detection systems can comprise at least one of: a dual-light fusion flame detection system, a point-type flame detection system, and a machine learning-based flame detection system, wherein the dual-light fusion flame detection system is used to detect a flame according to a first image generated based on thermal imaging and a second image generated based on visible light; the point-type flame detection system is used to detect a flame according to the frequency of light emitted when the flame burns; and the machine learning-based flame detection system is used to detect a flame based on a trained neural network model.

[0073] Optionally, the apparatus can further comprise a flame detection module located in the dual-optical fusion flame detection system, configured to acquire a position of a flame detected in one of the first image and the second image in the image; determine whether the flame is detected at the same position in the other of the first image and the second image; and determine that the flame is detected by the dual-optical fusion flame detection system if the flame is detected at the same position.

[0074] Optionally, the flame detection module is configured to filter out a plurality of temperature points greater than a temperature threshold; acquire a region including the plurality of temperature points according to the plurality of temperature points, wherein the region is a region in which the flame is detected in the first image; and acquire a position of the region, wherein the position of the region is the position of the flame in the first image. That is, the flame detection module is configured to acquire a first image captured by a thermal imaging lens in the dual-optical fusion flame detection system and a second image generated by a visible light lens in the dual-optical fusion flame detection system; acquire a first coordinate of a flame detected in one of the first image and the second image in the image; convert the first coordinate of the flame in the one image into a second coordinate on the other of the first image and the second image according to a perpendicular distance of a straight line from the thermal imaging lens and the visible light lens to the flame, wherein the perpendicular distance of the straight line from the thermal imaging lens to the flame is equal to the perpendicular distance of the straight line from the visible light lens to the flame; and determine whether the flame is detected at the second coordinate, and determine that the flame is detected by the dual-optical fusion flame detection system if the flame is detected.

[0075] Optionally, the flame detection module is configured to acquire a first deflection angle of the flame relative to a central axis of a first lens according to the first coordinate, a resolution of the one image in which the first coordinate is located, and a field of view angle of the first lens; acquire a second deflection angle of the flame relative to a central axis of a second lens according to the perpendicular distance of the straight line, the first deflection angle, and a distance between the first lens and the second lens, wherein the second lens is a lens for capturing the other of the first image and the second image; and acquire the second coordinate according to the second deflection angle, a resolution of the other of the first image and the second image, and a field of view angle of the second lens.

[0076] Optionally, the flame detection module is configured to obtain a first deflection angle of the flame relative to a central axis of the first lens according to a formula: the first deflection angle d1 = ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens, wherein if a horizontal deflection angle is calculated, the first coordinate is a coordinate in a horizontal direction, the first image size is a width of the image, and the field of view angle of the first lens is a horizontal field of view angle; if a vertical deflection angle is calculated, the first coordinate is a coordinate in a vertical direction, the first image size is a height of the image, and the field of view angle of the first lens is a vertical field of view angle; and obtain a value of a second deflection angle of the flame relative to a central axis of the second lens according to a formula: the second deflection angle d2=arctan((XL X) / YL), wherein X is a distance between the central axes of the two lenses, XL is a parallel distance from the central axis of the thermal imaging lens to the flame, YL is a vertical distance from the central axes of the first lens and the second lens to the flame, X and YL are obtained by measurement, XL=tan(d1)*YL, a parallel distance from the central axis of the visible light lens to the flame is XL X, and when the flame is on the right side of the central axis of the visible light lens, XL+X; when the flame is on the left side of the central axis of the visible light lens, XL-X; and the value of the second deflection angle is substituted into a formula: the second deflection angle= ((second coordinate-second image size / 2) / second image size)*field of view angle of the second lens, wherein if a horizontal coordinate in the second coordinate is calculated, a horizontal deflection angle is used, the second image size is a width of the image, and the field of view angle of the second lens is a horizontal field of view angle; if a vertical coordinate in the second coordinate is calculated, a vertical deflection angle is used, the second image size is a height of the image, and the field of view angle of the second lens is a vertical field of view angle.

[0077] Optionally, the flame detection module is configured to obtain a deviation angle of the first lens, wherein the deviation angle is an angle deviation between a theoretical position and an actual position of a calibration object in a picture of the first lens when the calibration object is in the picture center of the second lens; obtain a deflection angle of the flame relative to a central axis of the first lens according to the first coordinate, a resolution of an image in which the first coordinate is located, and a field of view angle of the first lens used to capture the image; and correct the obtained deflection angle relative to the central axis of the first lens using the deviation angle to obtain the first deflection angle.

[0078] Optionally, the flame detection module is configured to obtain the first deflection angle according to a formula: the first deflection angle = ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens-deviation angle.

[0079] Optionally, the flame detection module is configured to calculate the deviation angle using the following formula: deviation angle = ((actual position coordinate - theoretical position coordinate) / first image size) * field of view angle of the first lens.

[0080] Optionally, the first alarm signal is from the dual-light fusion flame detection system, and the second alarm signal is from the point-type flame detection system and / or the machine learning-based flame detection system.

[0081] The above embodiments solve the problem of waste of fire-fighting resources caused by misjudgment when using one flame detection system for detection in the prior art, thereby reducing the misjudgment rate of flame judgment and providing a technical guarantee for improving the accuracy of fire warning.

[0082] The above is only an embodiment of the present application and is not intended to limit the present application. Various modifications and changes can be made to the present application by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the scope of the claims of the present application.

Claims

1. A flame detection method, characterized by, The method comprises: receiving a first alarm signal, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent by one of a plurality of flame detection systems after the flame is detected, and the plurality of flame detection systems use different flame detection methods for flame detection; after receiving the first alarm signal, determining whether a second alarm signal is received, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent by a flame detection system other than the flame detection system sending the first alarm signal in the plurality of flame detection systems after the flame is detected; in the case where at least one second alarm signal is received, it is determined that a fire occurs in the region, and a fire alarm is given; in the case where the plurality of flame detection systems includes a dual-light fusion flame detection system, the method further comprises: obtaining a first image photographed by a thermal imaging lens in the dual-light fusion flame detection system and a second image generated by a visible light lens of the dual-light fusion flame detection system; obtaining a first coordinate of a flame detected in one of the first image and the second image in the image; obtaining a deviation angle of a first lens photographing the image, wherein the deviation angle is an angle deviation between a theoretical position and an actual position of a calibration object in a picture of the first lens when the calibration object is in a center of a picture of a second lens, and the second lens is a lens photographing the other one of the first image and the second image; obtaining a deflection angle of the flame relative to a central axis of the first lens according to the first coordinate, a resolution of the image in which the first coordinate is located, and a field of view angle of the first lens photographing the image; correcting the obtained deflection angle relative to the central axis of the first lens using the deviation angle to obtain a first deflection angle; obtaining a second deflection angle of the flame relative to a central axis of the second lens according to a straight-line perpendicular distance from the thermal imaging lens and the visible light lens to the flame, the first deflection angle, and a distance between the second lens and the first lens, wherein the straight-line distance from the thermal imaging lens to the flame is equal to the straight-line distance from the visible light lens to the flame; obtaining a second coordinate on the other image according to the second deflection angle, a resolution of the other image, and a field of view angle of the second lens; determining whether the flame is detected at the second coordinate, and if the flame is detected, it is determined that the flame is detected by the dual-light fusion flame detection system.

2. The method of claim 1, wherein the first deflection angle of the flame relative to the central axis of the first lens is obtained according to the following formula: ​ The first deflection angle d1 is equal to ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens, wherein if a horizontal deflection angle is calculated, the first coordinate is a coordinate in a horizontal direction, the first image size is a width of the image, and the field of view angle of the first lens is a horizontal field of view angle; if a vertical deflection angle is calculated, the first coordinate is a coordinate in a vertical direction, the first image size is a height of the image, and the field of view angle of the first lens is a vertical field of view angle. A value of a second deflection angle of the flame relative to a central axis of the second lens is obtained according to a formula: second deflection angle d2 = arctan((XL X) / YL), where X is a distance between the central axes of the two lenses, XL is a straight-line parallel distance of the central axis of the thermal imaging lens to the flame, YL is a straight-line vertical distance of the first lens and the second lens to the flame, X and YL are obtained by measurement, XL = tan(d1)*YL, and the straight-line parallel distance of the central axis of the visible light lens to the flame is XL X, XL+X when the flame is on the right side of the central axis of the visible light lens, and XL-X when the flame is on the left side of the central axis of the visible light lens; the value of the second deflection angle is substituted into an equation: second deflection angle = ((second coordinate-second image size / 2) / second image size)*field angle of the second lens, where if a coordinate in a horizontal direction of the second coordinate is calculated, a horizontal deflection angle is used, the second image size is a width of the image, and the field angle of the second lens is a horizontal field angle; if a coordinate in a vertical direction of the second coordinate is calculated, a vertical deflection angle is used, the second image size is a height of the image, and the field angle of the second lens is a vertical field angle.

3. The method of claim 1, wherein, The first deflection angle is obtained by correcting the deflection angle of the first lens using the deviation angle according to the following formula: The first deflection angle is equal to ((first coordinate-first image size / 2) / first image size)*field of view angle of the first lens-deviation angle.

4. The method according to claim 1 or 3, characterized in that, The deviation angle is calculated according to the following formula: Deviation angle-((actual position coordinate-theoretical position coordinate) / first image size)*field of view angle of the first lens.

5. A flame detection apparatus, characterized by, The method comprises the following steps: A receiving module is configured to receive a first alarm signal, wherein the first alarm signal is used to indicate that a flame is detected in a region, and the first alarm signal is sent by one of a plurality of flame detection systems after a flame is detected, wherein the plurality of flame detection systems use different flame detection methods to detect a flame. A judging module is configured to judge whether a second alarm signal is received after the first alarm signal is received, wherein the second alarm signal is used to indicate that a flame is detected in the region, and the second alarm signal is sent by a flame detection system other than the flame detection system that sends the first alarm signal among the plurality of flame detection systems after a flame is detected. An alarm module is configured to determine that a fire occurs in the region and give a fire alarm when at least one second alarm signal is received. The dual-light fusion flame detection system comprises a flame detection module; the flame detection module is used for acquiring a first image captured by a thermal imaging lens in the dual-light fusion flame detection system and a second image generated by a visible light lens of the dual-light fusion flame detection system; acquiring a first coordinate of a flame detected in one of the first image and the second image in the image; acquiring a deviation angle of a first lens for capturing the one image, wherein the deviation angle is an angle deviation between a theoretical position and an actual position of a calibration object in a picture of the first lens, which is calculated when the calibration object is in a picture center of a second lens, the second lens being a lens for capturing the other one of the first image and the second image; acquiring a deflection angle of the flame relative to a central axis of the first lens according to the first coordinate, a resolution of the one image in which the first coordinate is located, and a field of view angle of the first lens for capturing the one image; correcting the acquired deflection angle relative to the central axis of the first lens using the deviation angle to obtain a first deflection angle; acquiring a second deflection angle of the flame relative to a central axis of the second lens according to a straight-line perpendicular distance from the thermal imaging lens and the visible light lens to the flame, the first deflection angle, and a distance between the second lens and the first lens, wherein the straight-line distance from the thermal imaging lens to the flame is equal to the straight-line distance from the visible light lens to the flame; acquiring a second coordinate on the other image according to the second deflection angle, the resolution of the other image, and the field of view angle of the second lens; and judging whether the flame is detected at the second coordinate, and if the flame is detected, determining that the flame is detected by the dual-light fusion flame detection system.

6. An electronic device comprising a memory and a processor; wherein, The memory is configured to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method steps of any one of claims 1 to 4.

7. A readable storage medium having stored thereon computer instructions, wherein, The computer instructions are executed by the processor to implement the method steps of any one of claims 1 to 4.

Images