Fire anti-interference detection method and device based on flame jitter characteristics

Abstract

The application provides a fire anti-interference detection method and device based on flame jitter characteristics, and the method comprises the following steps: acquiring a video to be detected; reading two adjacent frames of images of the video to be detected in sequence, and reducing the two adjacent frames of images to obtain a matrix array of the reduced two adjacent frames of images; segmenting the matrix array of the reduced two adjacent frames of images to obtain a matrix array of the segmented images; performing HSV color detection on the matrix array of the segmented images to obtain a matrix array containing suspected flame images; performing boundary extraction on the matrix array containing suspected flame images to obtain a matrix array of a boundary binary image; calculating a matrix array of a flame edge change image according to the matrix array of the boundary binary image; calculating an edge change rate of the matrix array of the flame edge change image; and judging whether to send a fire warning signal according to the edge change rate. The application can quickly detect fire flames, is simple in operation, has strong anti-interference performance, and has high accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fire monitoring, in particular to a fire flame detection method based on machine vision and a fire flame detection device based on machine vision. BACKGROUND

[0002] With the improvement of living standards, the use of various fuels in people's work and life also increases, and the frequency of fire occurring every year also increases year by year. Therefore, minimizing the loss caused by fire is the key research content in the field of fire prevention technology, and the automatic fire detection system plays a crucial role.

[0003] The current fire detection system is still a traditional fire detector, which uses delay, temperature and other sensors to detect at the top of the room. However, the flame is generally small in the early stage of fire, and the smoke and temperature generated are difficult to reach the top of the room, and are easily affected by ventilation speed, on-site interference and noise, etc., so that timely fire monitoring and alarm cannot be performed. Due to the single criterion of current fire image recognition, the false alarm and missed alarm rates are high, the system is poor, and the real-time performance is poor, so it is difficult to realize timely early warning of fire. SUMMARY

[0004] The present application provides a fire anti-interference detection method and device based on flame jitter characteristics to solve the above technical problems, which can quickly detect fire flames, is not only simple to operate, but also has strong anti-interference and high accuracy.

[0005] The technical solution adopted by the present application is as follows:

[0006] A fire anti-interference detection method based on flame jitter characteristics, comprising the following steps: acquiring a to-be-detected video; reading adjacent two frames of images of the to-be-detected video in turn, and reducing the adjacent two frames of images to obtain a matrix array of the reduced adjacent two frames of images; segmenting the matrix array of the reduced adjacent two frames of images to obtain a matrix array of the segmented images; performing HSV color detection on the matrix array of the segmented images to obtain a matrix array containing suspected flame images; performing boundary extraction on the matrix array containing suspected flame images to obtain a matrix array of a boundary binary image; calculating a matrix array of a flame edge change image according to the matrix array of the boundary binary image; calculating an edge change rate of the matrix array of the flame edge change image; and determining whether to issue a fire warning signal according to the edge change rate.

[0007] The to-be-tested video is a fire video and an interference video, wherein the fire video is a candle lighting video, the interference video is a lamp with a color similar to the flame and a person wearing clothes with a color similar to the flame, a candle in the candle lighting video is lit at the a-th frame, and the candle is extinguished at the b-th frame in the candle lighting video, wherein b is greater than a.

[0008] The adjacent two frame images of the to-be-tested video are respectively reduced to a resolution size of 480*320 pixels.

[0009] The matrix array of the reduced adjacent two frame images is segmented, specifically including: the matrix array of the reduced adjacent two frame images is subjected to grayscale processing to obtain a matrix array of a grayscale image; the matrix array of the grayscale image is subjected to binaryzation processing by using a binaryzation formula to obtain a binaryzation mask image; and the binaryzation mask image is multiplied by the matrix array of the reduced adjacent two frame images to obtain a matrix array of the segmented image.

[0010] The binaryzation formula is:

[0011]

[0012] Wherein, m is a variable of an image matrix row vector, the change range of m is [1, 480], n is a variable of an image matrix column vector, the change range of n is [1, 320], and f is a variable of an image matrix.

[0013] The matrix array of the image containing the suspected flame is subjected to boundary extraction, specifically including: the matrix array of the image containing the suspected flame is subjected to binaryzation processing to obtain a matrix array of a binaryzation image; and the matrix array of the binaryzation image is subjected to eight-neighbor boundary extraction to obtain a matrix array of a boundary binaryzation image.

[0014] The calculation formula of the matrix array of the flame edge change image is:

[0015]

[0016] Wherein, represents XOR, represents logical AND, X represents a matrix array of a previous frame image, Y represents a matrix array of a next frame image, and K represents a matrix array of an edge change image.

[0017] The calculation formula of the edge change rate is:

[0018]

[0019] Wherein, i is a loop variable and the change range of i is [1, N], N is the total frame number of the to-be-tested video, and PL is the edge change rate.

[0020] The method comprises the following steps: determining whether to send a fire warning signal, specifically comprising: taking N-1 numbers of the edge change rate array, each A column as a group, wherein A is less than N-1; each time the edge change rate in each A column is greater than or equal to B, the frequency variable is added by one, wherein the change range of B is [0, 1]; if the value of the frequency variable in the A column is greater than or equal to A / 2, it is determined that there is a fire in the A column, and a fire warning signal is sent.

[0021] A fire anti-interference detection device based on flame shaking characteristics comprises: an acquisition unit configured to acquire a video to be detected; a reduction unit configured to reduce the adjacent two frames of images to obtain a matrix array of the reduced adjacent two frames of images; a segmentation unit configured to segment the matrix array of the reduced adjacent two frames of images to obtain a matrix array of the segmented images; a detection unit configured to perform HSV color detection on the matrix array of the segmented images to obtain a matrix array of images containing suspected flames; an image processing unit configured to perform boundary extraction on the matrix array of the images containing suspected flames to obtain a matrix array of a binary boundary image; a first calculation unit configured to calculate a matrix array of a flame edge change image according to the matrix array of the binary boundary image; a second calculation unit configured to calculate an edge change rate of the matrix array of the flame edge change image; and a judgment unit configured to determine whether to send a fire warning signal according to the edge change rate.

[0022] The present application has the following advantages:

[0023] In the present application, the matrix array of the adjacent two frames of images of the video to be detected is segmented, the matrix array of the segmented images is subjected to HSV color detection, the suspected flame area is determined through boundary extraction, the matrix array of the flame edge change image is calculated through the suspected flame area, and finally, it is determined whether to send a fire warning signal through the calculation of the flame edge change rate. Thus, the fire flame can be quickly detected, the operation is simple, the anti-interference performance is strong, and the accuracy is high. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 The flowchart of the fire anti-interference detection method based on flame shaking characteristics of the embodiment of the present application;

[0025] Figure 2 The state diagram of a candle being lit in a fire video of an embodiment of the present application;

[0026] Figure 3 The lamp similar to the color of the flame in an interference video of an embodiment of the present application;

[0027] Figure 4 a person wearing clothes similar to the color of the flame in the interference video;

[0028] Figure 5 a suspected flame area graph according to an embodiment of the present application;

[0029] Figure 6 a block schematic diagram of a flame anti-interference detection device based on flame jitter characteristics according to an embodiment of the present application. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0031] Figure 1 a flow chart of a flame anti-interference detection method based on flame jitter characteristics according to an embodiment of the present application.

[0032] As shown in Figure 1 , the flame anti-interference detection method based on flame jitter characteristics according to an embodiment of the present application includes the following steps:

[0033] S1, acquiring a to-be-tested video.

[0034] In an embodiment of the present application, the to-be-tested video can be downloaded from a network, and the to-be-tested video is a fire video and an interference video respectively. The fire video can be a candle lighting video, and the interference video can be a lamp similar to the color of the flame and a person wearing clothes similar to the color of the flame respectively. In the candle lighting video, a candle is lit at the ath frame, and the candle is extinguished at the bth frame, where b is greater than a.

[0035] In a specific embodiment of the present application, the fire video can be a candle lighting video with a time length of 20 seconds and a frame rate of 25 frames per second. As shown in Figure 2 , the candle is lit at the 21st frame, and the candle is extinguished at the 333rd frame. Figure 3 and Figure 4 are interference videos of a lamp similar to the color of the flame and a person wearing clothes similar to the color of the flame respectively.

[0036] S2, sequentially reading adjacent two frames of images of the to-be-tested video, and reducing the adjacent two frames of images to obtain a matrix array of the reduced adjacent two frames of images.

[0037] In one embodiment of the present application, the two adjacent frames of the video to be tested are reduced to 480*320 pixel resolution size respectively, so as to reduce the subsequent operation amount and improve the operation speed.

[0038] S3, the matrix array of the reduced adjacent two frames of images is segmented to obtain the matrix array of the segmented image.

[0039] Specifically, first, the matrix array of the reduced adjacent two frames of images is subjected to grayscale processing to obtain the matrix array of the grayscale image.

[0040] Further, the matrix array of the grayscale image is subjected to binaryzation processing by using a binaryzation formula to obtain a binaryzation mask image, wherein the binaryzation formula is:

[0041]

[0042] Wherein, m is a variable of the image matrix row vector, the change range of m is [1, 480], n is a variable of the image matrix column vector, the change range of n is [1, 320], and f is a variable of the image matrix.

[0043] Finally, the binaryzation mask image is multiplied with the matrix array of the reduced adjacent two frames of images to obtain the matrix array of the segmented image.

[0044] S4, the matrix array of the segmented image is subjected to HSV color detection to obtain the matrix array of the image containing suspected flame.

[0045] In one embodiment of the present application, the flame image in the video to be tested can be segmented by using the independent component analysis method. Specifically, first, the image sequence of the video to be tested is subjected to whitening processing, then the region containing independent information in the image is separated by using the independent component analysis, and finally the region where the flame is located is determined by using the HSV color model. By using the independent component analysis method, the suspected flame region can be effectively segmented, and is not easily affected by noise and light changes.

[0046] S5, the matrix array of the image containing suspected flame is subjected to boundary extraction to obtain the matrix array of the boundary binaryzation image.

[0047] Specifically, first, the matrix array of the image containing suspected flame is subjected to binaryzation processing to obtain the matrix array of the binaryzation image, and then the matrix array of the binaryzation image is subjected to eight-neighbor boundary extraction to obtain the matrix array of the boundary binaryzation image. Figure 5 As shown in Figure 2 The suspected flame region image after image segmentation, HSV color detection and boundary extraction.

[0048] S6, calculate a matrix array of a flame edge change image according to the matrix array of the boundary binary image.

[0049] In one embodiment of the present application, the formula for calculating the matrix array of the flame edge change image is:

[0050]

[0051] wherein, ^ represents XOR, · represents logical AND, X represents the matrix array of the previous frame image, Y represents the matrix array of the next frame image, and K represents the matrix array of the edge change image.

[0052] S7, calculate an edge change rate of the matrix array of the flame edge change image.

[0053] In one embodiment of the present application, the formula for calculating the edge change rate is:

[0054]

[0055] wherein, i is a loop variable and the change range of i is [1, N], N is the total frame number of the video to be tested, and PL is the edge change rate.

[0056] S8, determine whether to send a fire warning signal according to the edge change rate.

[0057] In one specific embodiment of the present application, first, for the N-1 numbers of the edge change rate array, every 10 columns can be set as a group. Then, every time the edge change rate in every 10 columns is greater than or equal to 0.2, the frequency variable is added by one. Finally, if the value of the frequency variable in the 10 columns is greater than or equal to 6, it is determined that there is a fire in the 10 columns, and a fire warning signal is sent.

[0058] According to the fire anti-interference detection method based on flame jitter characteristics provided by the embodiments of the present application, the matrix arrays of the adjacent two frame images of the video to be tested are segmented, the matrix arrays of the segmented images are subjected to HSV color detection, then the suspected flame area is determined through boundary extraction, the matrix array of the flame edge change image is calculated through the suspected flame area, and finally it is determined whether to send a fire warning signal through the calculation of the flame edge change rate. Thus, the fire flame can be quickly detected, the operation is simple, the anti-interference performance is strong, and the accuracy is high.

[0059] To realize the fire anti-interference detection method based on flame jitter characteristics of the above-mentioned embodiments, the present application further provides a fire anti-interference detection device based on flame jitter characteristics.

[0060] As Figure 6 ​As shown, the fire anti-interference detection device based on flame shaking characteristics in the embodiment of the present application comprises: an acquisition unit 10, a reduction unit 20, a segmentation unit 30, a detection unit 40, an image processing unit 50, a first calculation unit 60, a second calculation unit 70 and a judgment unit 80. Wherein, the acquisition unit 10 is used for acquiring the video to be detected; the reduction unit 20 is used for reducing the adjacent two frames of images to obtain the matrix array of the reduced adjacent two frames of images; the segmentation unit 30 is used for segmenting the matrix array of the reduced adjacent two frames of images to obtain the matrix array of the segmented images; the detection unit 40 is used for performing HSV color detection on the matrix array of the segmented images to obtain the matrix array containing the suspected flame images; the image processing unit 50 is used for performing boundary extraction on the matrix array containing the suspected flame images to obtain the matrix array of the boundary binary image; the first calculation unit 60 is used for calculating the matrix array of the flame edge change image according to the matrix array of the boundary binary image; the second calculation unit 70 is used for calculating the edge change rate of the matrix array of the flame edge change image; and the judgment unit 80 is used for judging whether to send the fire warning signal according to the edge change rate.

[0061] In an embodiment of the present application, the acquisition unit 10 can download the video to be detected from the network, and the video to be detected is the fire video and the interference video respectively, wherein the fire video can be the candle lighting video, and the interference video can be the lamp with the color similar to the flame and the person wearing the color similar to the flame respectively. In the candle lighting video, the candle is lit in the a-th frame, and the candle is extinguished in the b-th frame, wherein b is greater than a.

[0062] In a specific embodiment of the present application, the fire video can be the candle lighting video with the time length of 20 seconds and the frame rate of 25 frames / second. Wherein, the candle is lit in the 21st frame, and the candle is extinguished in the 333rd frame.

[0063] In an embodiment of the present application, the reduction unit 20 reduces the adjacent two frames of images of the video to be detected to 480*320 pixel resolution size respectively, so as to reduce the subsequent operation amount and improve the operation speed.

[0064] In an embodiment of the present application, the segmentation unit 30 segments the matrix array of the reduced adjacent two frames of images by the following steps:

[0065] Firstly, the matrix array of the reduced adjacent two frames of images is subjected to gray processing to obtain the matrix array of the gray image; further, the matrix array of the gray image is subjected to binary processing by using the binary formula to obtain the binary mask image, wherein the binary formula is:

[0066]

[0067] Wherein, m is the variable of the image matrix row vector, the change range of m is [1, 480], n is the variable of the image matrix column vector, the change range of n is [1, 320], f is the variable of the image matrix; finally, the binary mask image is multiplied with the matrix array of the reduced adjacent two frames of images, and the matrix array of the segmented image is obtained.

[0068] In an embodiment of the present application, the detection unit 40 can use the independent component analysis method to segment the flame image in the video to be detected. Specifically, first, the image sequence of the video to be detected is whitened, then the region containing independent information in the image is separated using independent component analysis, and finally the region where the flame is located is determined through the HSV color model. By using the independent component analysis method, the suspected flame region can be effectively segmented, and is not easily affected by noise and light changes.

[0069] In an embodiment of the present application, the image processing unit 60 first performs binaryzation processing on the matrix array containing the suspected flame image to obtain the matrix array of the binaryzation image, and then performs eight-neighbor boundary extraction on the matrix array of the binaryzation image to obtain the matrix array of the boundary binaryzation image.

[0070] In an embodiment of the present application, the calculation formula of the matrix array of the flame edge change image is:

[0071]

[0072] Wherein, represents XOR, represents logical AND, X represents the matrix array of the previous frame image, Y represents the matrix array of the next frame image, and K represents the matrix array of the edge change image.

[0073] In an embodiment of the present application, the calculation formula of the edge change rate is:

[0074]

[0075] Wherein, i is a loop variable and the change range of i is [1, N], N is the total frame number of the video to be detected, and PL is the edge change rate.

[0076] In a specific embodiment of the present application, the judgment unit 80 judges whether to issue a fire warning signal specifically includes: first, the N-1 numbers of the edge change rate array can be set as a group every 10 columns. Then, every time the edge change rate in every 10 columns is greater than or equal to 0.2, the frequency variable is added by one. Finally, if the value of the frequency variable in the 10 columns is greater than or equal to 6, it is judged that there is a fire in the 10 columns, and a fire warning signal is issued.

[0077] The fire flame anti-interference detection device based on the flame jitter feature according to the embodiment of the application can quickly detect the fire flame, is simple in operation, has strong anti-interference and high accuracy.

[0078] In the description of the present application, the terms "first", "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. The meaning of "a plurality of" is two or more, unless otherwise specifically limited.

[0079] In the present application, unless otherwise specifically defined and limited, the terms "mounting", "connection", "connecting", "fixing" and the like should be understood in a broad sense, for example, can be fixed connection, can also be detachable connection, or integral; can be mechanical connection, can also be electrical connection; can be directly connected, can also be indirectly connected through an intermediate medium; can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0080] In the present application, unless otherwise specifically defined and limited, the first feature is "on" or "under" the second feature, which can be direct contact between the first and second features, or indirect contact between the first and second features through an intermediate medium. Moreover, the first feature "above", "over" and "on" the second feature can be directly above or obliquely above the first feature, or only indicate that the horizontal height of the first feature is higher than that of the second feature. The first feature "below", "under" and "under" the second feature can be directly below or obliquely below the first feature, or only indicate that the horizontal height of the first feature is less than that of the second feature.

[0081] In the description of the specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any suitable manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the specification and the features of the different embodiments or examples, without contradiction.

[0082] Any process or method descriptions or descriptions of the flow diagrams in the specification can be understood as representing code modules, segments, or portions of code which include one or more executable instructions for performing specific logic functions or steps in the process, and the various embodiments of the application include additional or different modules, segments, or portions of code when implemented in software or firmware in combination with the process described. The various embodiments of the application can be implemented in a computer program product tangibly embodied in a machine-readable storage medium (e.g., magnetic disk, optical disk, memory, etc.) including one or more code modules or portions of code.

[0083] The logic and / or steps represented in the flow diagrams or otherwise described herein, for example, can be embodied in computer-readable instructions, modules, or portions of code, which can be executed by one or more processing units, cores, or processors of one or more computing devices or systems. The various embodiments of the application include additional or different modules, segments, or portions of code when implemented in software or firmware in combination with the process described. The various embodiments of the application can be implemented in a computer program product tangibly embodied in a machine-readable storage medium (e.g., magnetic disk, optical disk, memory, etc.) including one or more code modules or portions of code.

[0084] It should be understood that parts of the present application can be implemented in hardware, software, firmware or a combination thereof. In the above embodiments, a plurality of steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and as in another embodiment, it can be implemented using any or a combination of the following technologies, which are known in the art: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0085] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment methods can be completed by programs instructing relevant hardware, and the programs can be stored in a computer readable storage medium, and when the programs are executed, one or a combination of the steps of the method embodiments is included.

[0086] In addition, each functional unit in each embodiment of the present application can be integrated into one processing module, or each unit can exist physically, or two or more units can be integrated into one module. The above integrated module can be realized in the form of hardware or in the form of a software functional module. The integrated module, if realized in the form of a software functional module and sold or used as an independent product, can also be stored in a computer readable storage medium.

[0087] Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and cannot be understood as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A flame flicker characteristic-based fire anti-interference detection method, characterized in that, The method comprises the following steps: acquiring a video to be tested; reading adjacent two frames of images of the video to be tested in sequence, and reducing the adjacent two frames of images to obtain a matrix array of the reduced adjacent two frames of images; segmenting the matrix array of the reduced adjacent two frames of images to obtain a matrix array of segmented images; performing HSV color detection on the matrix array of the segmented images to obtain a matrix array of images containing suspected flames; performing boundary extraction on the matrix array of the images containing suspected flames to obtain a matrix array of boundary binarization images; calculating a matrix array of flame edge change images according to the matrix array of the boundary binarization images; calculating an edge change rate of the matrix array of the flame edge change images; judging whether to issue a fire warning signal according to the edge change rate, segmenting the matrix array of the reduced adjacent two frames of images, specifically comprising: performing grayscale processing on the matrix array of the reduced adjacent two frames of images to obtain a matrix array of grayscale images; performing binarization processing on the matrix array of the grayscale images by using a binarization formula to obtain a binarization mask image; and multiplying the binarization mask image and the matrix array of the reduced adjacent two frames of images to obtain the matrix array of the segmented images, the binarization formula is: ， wherein m is a variable of a row vector of an image matrix, the variation range of m is [1, 480], n is a variable of a column vector of an image matrix, the variation range of n is [1, 320], f and g are variables of image matrices before and after binarization respectively, performing boundary extraction on the matrix array of the images containing suspected flames, specifically comprising: performing binarization processing on the matrix array of the images containing suspected flames to obtain a matrix array of binarization images; and performing eight-neighbor boundary extraction on the matrix array of the binarization images to obtain the matrix array of the boundary binarization images, the calculation formula of the matrix array of the flame edge change images is: K = X · (X XOR Y), wherein XOR represents exclusive OR, · represents logical AND, X represents a matrix array of a previous frame of image, Y represents a matrix array of a next frame of image, and K represents a matrix array of an edge change image.

2. The flame flicker feature based fire anti-tamper detection method according to claim 1, characterized in that, The video to be tested is a fire video and an interference video, wherein the fire video is a candle lighting video, the interference video is a lamp with a color similar to that of a flame and a person wearing clothes with a color similar to that of the flame, a candle in the candle lighting video is lit at a first frame, and the candle is extinguished at a second frame, wherein b is greater than a.

3. The flame flicker feature based fire anti-tamper detection method according to claim 1, wherein, The adjacent two frames of images of the video to be tested are reduced to 480*320 pixel resolution size respectively.

4. The flame flicker feature based fire anti-tamper detection method according to claim 1, wherein, judging whether to issue a fire warning signal, specifically comprising: N-1 numbers of the edge change rate array are taken as a group every A columns, wherein A is less than N-1; the frequency variable is added by one every time the edge change rate in each A column is greater than or equal to B, wherein the variation range of B is [0, 1]; if the value of the frequency variable in the A column is greater than A / 2, it is judged that there is a fire in the A column, and a fire warning signal is issued.

5. A flame-flicker characteristic based fire-immune detection apparatus implementing the method of any one of claims 1-4, characterized by, comprising: an acquisition unit, configured to acquire a video to be tested; a reducing unit, configured to reduce two adjacent frames of the video to be detected to obtain a matrix array of the reduced two adjacent frames; a dividing unit, configured to divide the matrix array of the reduced two adjacent frames to obtain a matrix array of divided images; a detecting unit, configured to perform HSV color detection on the matrix array of the divided images to obtain a matrix array of images containing suspected flames; an image processing unit, configured to perform boundary extraction on the matrix array of the images containing suspected flames to obtain a matrix array of boundary binary images; a first calculating unit, configured to calculate a matrix array of a flame edge change image according to the matrix array of the boundary binary images; a second calculating unit, configured to calculate an edge change rate of the matrix array of the flame edge change image; a judging unit, configured to judge whether to send a fire warning signal according to the edge change rate.

Images