An intelligent smoke recognition and early warning method and system based on GIS

Abstract

The present invention relates to the technical field of fireworks recognition and early warning, and discloses an intelligent fireworks recognition and early warning method and system based on GIS. Real-time regional images are collected by an image acquisition device to obtain a sequence of gray values of the regional images; an initial fireworks analysis mark is generated according to the sequence of gray values of the regional images and the sequence of gray values of the comparison regional images; the real-time regional images are randomly divided into a plurality of sub-real-time regional images, and the sequence of gray values of the sub-region images is determined; the sub-fireworks recognition and early warning coefficient is calculated to determine a plurality of sub-fireworks recognition and early warning coefficient arrays; the sub-fireworks recognition and early warning coefficient arrays are analyzed, the fireworks recognition and early warning coefficient is calculated, and it is judged whether to issue an early warning. If so, a fire situation report information is generated according to the fireworks recognition and early warning coefficient and the GIS geographic information system, which can significantly improve the efficiency and accuracy of fireworks recognition, realize accurate recognition and instant early warning, and overcome the problems of low efficiency of manual visual recognition and insufficient accuracy of fire situation report.

Description

Technical Field

[0001] The present invention relates to the technical field of pyrotechnics recognition and early warning, and in particular, to an intelligent pyrotechnics recognition and early warning method and system based on GIS. Background Art

[0002] Ports belong to key areas and have inestimable value. However, the fire risk affects the safe development of ports at any time. Preventing fires, especially quickly detecting and warning in the initial stage of a fire, is particularly crucial.

[0003] In the prior art, for pyrotechnics recognition, it is mainly based on unmanned aerial vehicle (UAV) technology. The UAV is equipped with a high-definition camera, and through real-time image transmission, ground monitoring personnel can view the situation of the monitoring area in real time. However, this monitoring method has significant deficiencies: it solely relies on manual visual recognition of pyrotechnics, which is not only inefficient but also unable to guarantee accuracy; once pyrotechnics are detected, it requires monitoring personnel to manually report. Due to various factors such as the complex on-site environment, limited information acquisition means, and subjective human judgment, relevant personnel can only provide a rough descriptive report on the fire situation, making it difficult to accurately and comprehensively reflect the actual situation of the fire and missing the best fire extinguishing opportunity. Summary of the Invention

[0004] The embodiments of the present invention provide an intelligent pyrotechnics recognition and early warning method and system based on GIS. The present invention can significantly improve the efficiency and accuracy of pyrotechnics recognition, intelligently analyze the fire situation information in the real-time image, achieve accurate recognition and instant early warning, and overcome the problems of low efficiency of manual visual recognition and insufficient accuracy of reporting the fire situation.

[0005] To achieve the above object, the present invention provides an intelligent pyrotechnics recognition and early warning method based on GIS, including:

[0006] Pre-deploy image acquisition devices in the pyrotechnics recognition and early warning area, collect corresponding real-time area images based on the image acquisition devices, extract the image pixel values of the real-time area images, and perform gray conversion to obtain a sequence of gray values of the area images;

[0007] Determine a sequence of gray values of the comparison area image, and generate an initial pyrotechnics analysis mark for the pyrotechnics recognition and early warning area according to the sequence of gray values of the area image and the sequence of gray values of the comparison area image, wherein the initial pyrotechnics analysis mark includes an area without pyrotechnics mark and an area with pyrotechnics to be recognized mark;

[0008] When generating the area with pyrotechnics to be recognized mark, randomly divide the real-time area image into multiple sub-real-time area images, and determine the corresponding sequence of gray values of each sub-real-time area image according to the sequence of gray values of the area image;

[0009] Analyze the grayscale value sequence of the sub-region image, calculate the sub-smoke recognition and early warning coefficient of the sub-real-time region image based on the analysis result, and determine multiple sub-smoke recognition and early warning coefficient arrays according to all the sub-smoke recognition and early warning coefficients;

[0010] Analyze all the sub-smoke recognition and early warning coefficient arrays, calculate the smoke recognition and early warning coefficient of the smoke recognition and early warning region, and judge whether to issue an early warning according to the smoke recognition and early warning coefficient. If so, generate a fire situation report information according to the smoke recognition and early warning coefficient and the GIS geographic information system.

[0011] Further, when determining the grayscale value sequence of the comparison region image and generating an initial smoke analysis mark for the smoke recognition and early warning region according to the grayscale value sequence of the region image and the grayscale value sequence of the comparison region image, it includes:

[0012] Obtain a smoke-free region image corresponding to the smoke recognition and early warning region, perform grayscale conversion on the smoke-free region image to obtain the grayscale value sequence of the comparison region image;

[0013] Judge whether the grayscale values of the region images in the grayscale value sequence of the region image are all equal to the grayscale value sequence of the comparison region image. If so, generate the region smoke-free mark for the smoke recognition and early warning region;

[0014] If not, generate the region smoke to be recognized mark for the smoke recognition and early warning region.

[0015] Further, when analyzing the grayscale value sequence of the sub-region image and calculating the sub-smoke recognition and early warning coefficient of the sub-real-time region image based on the analysis result, it includes:

[0016] Compare the grayscale value sequence of the sub-region image with the grayscale value sequence of the comparison region image, extract the grayscale values of the sub-region images in the grayscale value sequence of the sub-region image that are equal to the grayscale value of the comparison region image, and generate an equal sub-region image grayscale value sequence;

[0017] Generate a recognition sub-region image grayscale value sequence according to the remaining grayscale values of the sub-region images in the grayscale value sequence of the sub-region image;

[0018] Calculate the image grayscale difference value between the recognition sub-region image grayscale value sequence and the grayscale value of the comparison region image, and generate a recognition sub-region image grayscale difference value sequence;

[0019] Preset a first preset image grayscale difference value and a second preset image grayscale difference value;

[0020] Map the first preset image gray difference value and the second preset image gray difference value to the recognition sub-region image gray difference value sequence;

[0021] Determine the minimum image gray difference value and the maximum image gray difference value in the recognition sub-region image gray difference value sequence, and count the number of the first image gray difference values between the minimum image gray difference value and the first preset image gray difference value;

[0022] Count the number of the second image gray difference values between the first preset image gray difference value and the second preset image gray difference value;

[0023] Count the number of the third image gray difference values between the maximum image gray difference value and the second preset image gray difference value;

[0024] Calculate the sub-smoke recognition warning coefficient of the sub-real-time region image according to the number of the first image gray difference values, the number of the second image gray difference values, and the number of the third image gray difference values.

[0025] Further, when calculating the sub-smoke recognition warning coefficient of the sub-real-time region image according to the number of the first image gray difference values, the number of the second image gray difference values, and the number of the third image gray difference values, it includes:

[0026] Calculate the sub-smoke recognition warning coefficient of the sub-real-time region image according to the following formula:

[0027] ;

[0028] where q is the sub-smoke recognition warning coefficient of the sub-real-time region image, w is the number of the image gray difference values in the recognition sub-region image gray difference value sequence, r e is the e-th image gray difference value in the recognition sub-region image gray difference value sequence, t1 is the first preset image gray difference value, t2 is the second preset image gray difference value, y1 is the number of the first image gray difference values, y2 is the number of the second image gray difference values, y3 is the number of the third image gray difference values, u is the variance of the image gray difference values corresponding to the recognition sub-region image gray difference value sequence, and p is the adjustment coefficient.

[0029] Further, calculate the adjustment coefficient p according to the following method:

[0030] Perform normalization processing on the sub-region image gray value sequence, and construct a sub-region image gray value curve according to the normalization processing result;

[0031] Determine the inflection points of the sub-region image gray value curve, and determine the image gray values corresponding to each inflection point;

[0032] Count the number of equal image gray values in the equal sub-region image gray value sequence, and count the number of recognized image gray values in the recognized sub-region image gray value sequence;

[0033] Calculate the adjustment coefficient p according to the following formula:

[0034] ;

[0035] where d1 is the average value of all image gray values on the sub-region image gray value curve, a is the number of curve inflection points, s i is the inflection point slope corresponding to the i-th curve inflection point, f1 is the number of equal image gray values, f2 is the number of recognized image gray values, d2 is the average value of the image gray values corresponding to all curve inflection points, and e is a constant.

[0036] Further, when determining multiple sub-smoke recognition warning coefficient arrays according to all sub-smoke recognition warning coefficients, it includes:

[0037] Randomly extract the first sub-smoke recognition warning coefficient and the second sub-smoke recognition warning coefficient from all sub-smoke recognition warning coefficients, and use the first sub-smoke recognition warning coefficient and the second sub-smoke recognition warning coefficient as the initial sub-smoke recognition warning coefficient array;

[0038] Calculate the first coefficient sum value of the two sub-smoke recognition warning coefficients in the initial sub-smoke recognition warning coefficient array;

[0039] Randomly extract the third sub-smoke recognition warning coefficient, calculate the absolute value of the first difference between the third sub-smoke recognition warning coefficient and the first coefficient sum value, and determine whether the absolute value of the first difference is less than the preset absolute value of the difference. If so, update the third sub-smoke recognition warning coefficient to the initial sub-smoke recognition warning coefficient array;

[0040] If not, use the initial sub-smoke recognition warning coefficient array as a sub-smoke recognition warning coefficient array, randomly extract the fourth sub-smoke recognition warning coefficient, and use the fourth sub-smoke recognition warning coefficient and the third sub-smoke recognition warning coefficient as the second initial sub-smoke recognition warning coefficient array;

[0041] Randomly extract the fifth sub-smoke recognition warning coefficient, calculate the second coefficient sum value of the two sub-smoke recognition warning coefficients in the second initial sub-smoke recognition warning coefficient array, and calculate the absolute value of the second difference between the second coefficient sum value and the fifth sub-smoke recognition warning coefficient, and determine whether the absolute value of the second difference is less than the preset absolute value of the difference. If so, update the fifth sub-smoke recognition warning coefficient to the second initial sub-smoke recognition warning coefficient array;

[0042] Otherwise, use the second initial sub-smoke recognition and warning coefficient array as a sub-smoke recognition and warning coefficient array;

[0043] Repeat the above steps to obtain multiple sub-smoke recognition and warning coefficient arrays.

[0044] Further, when analyzing all the sub-smoke recognition and warning coefficient arrays and calculating the smoke recognition and warning coefficient of the smoke recognition and warning area, it includes:

[0045] Calculate the sum value of the sub-smoke recognition and warning coefficients corresponding to each sub-smoke recognition and warning coefficient array;

[0046] Determine the average value of the sum values of the sub-smoke recognition and warning coefficients corresponding to all the sub-smoke recognition and warning coefficient sum values;

[0047] Allocate all the sub-smoke recognition and warning coefficient sum values less than the average value of the sub-smoke recognition and warning coefficient sum values to the first sub-smoke recognition and warning coefficient sum value array, and allocate all the sub-smoke recognition and warning coefficient sum values greater than or equal to the average value of the sub-smoke recognition and warning coefficient sum values to the second sub-smoke recognition and warning coefficient sum value array;

[0048] Sort the first sub-smoke recognition and warning coefficient sum value array and the second sub-smoke recognition and warning coefficient sum value array from small to large respectively;

[0049] Determine the first starting sub-smoke recognition and warning coefficient sum value and the first ending sub-smoke recognition and warning coefficient sum value of the first sub-smoke recognition and warning coefficient sum value array;

[0050] Determine the second starting sub-smoke recognition and warning coefficient sum value and the second ending sub-smoke recognition and warning coefficient sum value of the second sub-smoke recognition and warning coefficient sum value array;

[0051] Calculate the first average value of the first sub-smoke recognition and warning coefficient sum value array, and determine the number g1 of the sub-smoke recognition and warning coefficient sum values between the first average value and the first starting sub-smoke recognition and warning coefficient sum value;

[0052] Calculate the second average value of the second sub-smoke recognition and warning coefficient sum value array, and determine the number g2 of the sub-smoke recognition and warning coefficient sum values between the second average value and the second ending sub-smoke recognition and warning coefficient sum value;

[0053] Calculate the first quantity ratio k1 based on the number g1 of the first sub-smoke recognition and warning coefficient sum values and the number g2 of the second sub-smoke recognition and warning coefficient sum values, and use it as the first calculation coefficient;

[0054] ;

[0055] Determine the number g3 of the third sub-smoke recognition and warning coefficient sums between the first mean value and the sum of the first last-sub-smoke recognition and warning coefficients;

[0056] Determine the number g4 of the fourth sub-smoke recognition and warning coefficient sums between the second mean value and the sum of the first start-sub-smoke recognition and warning coefficients;

[0057] Calculate the second quantity ratio k2 based on the number g3 of the third sub-smoke recognition and warning coefficient sums and the number g4 of the fourth sub-smoke recognition and warning coefficient sums, and use it as the second calculation coefficient;

[0058] ;

[0059] Calculate the smoke recognition and warning coefficient of the smoke recognition and warning area based on the first calculation coefficient and the second calculation coefficient.

[0060] Further, when calculating the smoke recognition and warning coefficient of the smoke recognition and warning area based on the first calculation coefficient and the second calculation coefficient, it includes:

[0061] Calculate the absolute value of the coefficient difference between the first calculation coefficient and the second calculation coefficient, and calculate the sum of the coefficients of the first calculation coefficient and the second calculation coefficient;

[0062] Use the coefficient ratio of the absolute value of the coefficient difference to the sum of the coefficients as the smoke recognition and warning coefficient of the smoke recognition and warning area.

[0063] Further, when judging whether to issue a warning according to the smoke recognition and warning coefficient, it includes:

[0064] Judge whether to issue a warning according to the relationship between the smoke recognition and warning coefficient and the preset smoke recognition and warning coefficient;

[0065] When the smoke recognition and warning coefficient is less than the preset smoke recognition and warning coefficient, it is judged that there is no smoke in the smoke recognition and warning area, and no warning is issued;

[0066] When the smoke recognition and warning coefficient is greater than or equal to the preset smoke recognition and warning coefficient, it is judged that there is smoke in the smoke recognition and warning area, and a warning is issued.

[0067] To achieve the above object, the present invention also provides a GIS-based intelligent smoke recognition and warning system, including:

[0068] An image acquisition module, which is used to pre-deploy image acquisition devices in the smoke recognition and warning area, collect corresponding real-time area images based on the image acquisition devices, extract the image pixel values of the real-time area images, and perform gray-scale conversion to obtain a sequence of area image gray-scale values;

[0069] A marking generation module, configured to determine a sequence of gray values of the comparison region image, and generate an initial firework analysis mark for the firework recognition and early warning region according to the sequence of gray values of the region image and the sequence of gray values of the comparison region image, wherein the initial firework analysis mark includes a region without firework mark and a region with firework to be recognized mark;

[0070] An image division module, configured to randomly divide the real-time region image into a plurality of sub-real-time region images when generating the region with firework to be recognized mark, and determine a sequence of sub-region image gray values corresponding to each sub-real-time region image according to the sequence of gray values of the region image;

[0071] A data calculation module, configured to analyze the sequence of sub-region image gray values, calculate a sub-firework recognition and early warning coefficient of the sub-real-time region image based on the analysis result, and determine a plurality of sub-firework recognition and early warning coefficient arrays according to all the sub-firework recognition and early warning coefficients;

[0072] A recognition and early warning module, configured to analyze all the sub-firework recognition and early warning coefficient arrays, calculate a firework recognition and early warning coefficient of the firework recognition and early warning region, and determine whether to issue an early warning according to the firework recognition and early warning coefficient. If so, generate a fire situation report information according to the firework recognition and early warning coefficient and the GIS geographic information system.

[0073] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0074] The present invention discloses an intelligent firework recognition and early warning method and system based on GIS. Based on an image acquisition device to acquire a real-time region image, a sequence of gray values of the region image is obtained; an initial firework analysis mark is generated according to the sequence of gray values of the region image and the sequence of gray values of the comparison region image; the real-time region image is randomly divided into a plurality of sub-real-time region images, and a sequence of sub-region image gray values is determined; a sub-firework recognition and early warning coefficient is calculated, and a plurality of sub-firework recognition and early warning coefficient arrays are determined; the sub-firework recognition and early warning coefficient arrays are analyzed, a firework recognition and early warning coefficient is calculated, and it is determined whether to issue an early warning. If so, fire situation report information is generated according to the firework recognition and early warning coefficient and the GIS geographic information system, which can significantly improve the efficiency and accuracy of firework recognition, achieve accurate recognition and instant early warning, and overcome the problems of low efficiency of manual visual recognition and insufficient accuracy of reporting the fire situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] By reading the detailed description of the preferred embodiments below, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0076] Figure 1 shows a schematic flowchart of a GIS-based intelligent smoke and fire recognition and early warning method in an embodiment of the present invention;

[0077] Figure 2 shows a schematic structural diagram of a GIS-based intelligent smoke and fire recognition and early warning system in an embodiment of the present invention. Detailed implementation manners

[0078] The following will further describe in detail the specific implementation manners of the present invention in conjunction with the accompanying drawings and embodiments. The following embodiments are used to illustrate the present invention, but are not used to limit the scope of the present invention.

[0079] The terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0080] The following is a description of the preferred embodiments of the present invention in conjunction with the accompanying drawings.

[0081] As Figure 1 shown, an embodiment of the present invention discloses a GIS-based intelligent smoke and fire recognition and early warning method, including:

[0082] S110: Pre-deploy image acquisition devices in the smoke and fire recognition and early warning area, collect corresponding real-time area images based on the image acquisition devices, extract the image pixel values of the real-time area images, and perform gray conversion to obtain a sequence of area image gray values;

[0083] In this embodiment, the image acquisition devices include drones and high-definition cameras, and the high-definition cameras are carried by the drones to collect real-time area images of the smoke and fire recognition and early warning area.

[0084] In this embodiment, gray conversion is performed on the real-time area images to obtain corresponding area image gray values, and a sequence of area image gray values is constructed. The gray conversion method of the images will not be introduced in detail.

[0085] S120: Determine a sequence of comparison area image gray values, and generate an initial smoke and fire analysis mark for the smoke and fire recognition and early warning area according to the sequence of area image gray values and the sequence of comparison area image gray values, where the initial smoke and fire analysis mark includes an area without smoke mark and an area with smoke to be recognized mark;

[0086] In some embodiments of the present application, when determining the gray value sequence of the regional image and generating an initial firework analysis mark for the firework recognition and warning area according to the regional image gray value sequence and the gray value sequence of the comparison regional image, it includes:

[0087] Obtain a smokeless area image corresponding to the firework recognition and warning area, perform gray conversion on the smokeless area image to obtain the gray value sequence of the comparison regional image;

[0088] Judge whether the gray values of the regional images in the regional image gray value sequence are all equal to the gray value sequence of the comparison regional image. If so, generate the regional smokeless mark for the firework recognition and warning area;

[0089] If not, generate the regional firework to-be-recognized mark for the firework recognition and warning area.

[0090] In this embodiment, the smokeless area image refers to the image taken when there is no firework in the firework recognition and warning area.

[0091] In this embodiment, the gray value sequence of the comparison regional image corresponds one-to-one with the regional image gray value sequence.

[0092] The beneficial effects of the above technical solutions are as follows: The present invention can realize the initial judgment of the firework recognition and warning area. When the gray values of the regional images in the regional image gray value sequence are all equal to the gray value sequence of the comparison regional image, it is directly judged that there is no firework in the firework recognition and warning area, improving the judgment efficiency.

[0093] S130: When generating the regional firework to-be-recognized mark, randomly divide the real-time regional image into multiple sub-real-time regional images, and determine the sub-regional image gray value sequence corresponding to each sub-real-time regional image according to the regional image gray value sequence;

[0094] In this embodiment, the number of divisions of the sub-real-time regional images is preferably 20, and it can also be adjusted according to the actual situation.

[0095] In this embodiment, analyze the regional image gray value sequence to obtain the sub-regional image gray value sequence corresponding to each sub-real-time regional image.

[0096] S140: Analyze the sub-regional image gray value sequence, calculate the sub-firework recognition and warning coefficient of the sub-real-time regional image based on the analysis result, and determine multiple sub-firework recognition and warning coefficient arrays according to all the sub-firework recognition and warning coefficients;

[0097] In some embodiments of the present application, when analyzing the sub-regional image gray value sequence and calculating the sub-firework recognition and warning coefficient of the sub-real-time regional image based on the analysis result, it includes:

[0098] Compare the sub-region image gray value sequence with the comparison region image gray value sequence, extract the sub-region image gray values in the sub-region image gray value sequence that are equal to the comparison region image gray values, and generate an equal sub-region image gray value sequence;

[0099] Generate a recognition sub-region image gray value sequence based on the remaining sub-region image gray values in the sub-region image gray value sequence;

[0100] Calculate the image gray value difference between the recognition sub-region image gray value sequence and the comparison region image gray value, and generate a recognition sub-region image gray value difference sequence;

[0101] Preset a first preset image gray value difference and a second preset image gray value difference in advance;

[0102] Map the first preset image gray value difference and the second preset image gray value difference to the recognition sub-region image gray value difference sequence;

[0103] Determine the minimum image gray value difference and the maximum image gray value difference in the recognition sub-region image gray value difference sequence, and count the number of the first image gray value differences between the minimum image gray value difference and the first preset image gray value difference;

[0104] Count the number of the second image gray value differences between the first preset image gray value difference and the second preset image gray value difference;

[0105] Count the number of the third image gray value differences between the maximum image gray value difference and the second preset image gray value difference;

[0106] Calculate the sub-firework recognition warning coefficient of the sub-real-time region image according to the number of the first image gray value differences, the number of the second image gray value differences, and the number of the third image gray value differences.

[0107] In this embodiment, the values in the sub-region image gray value sequence are named sub-region image gray values, and the values in the comparison region image gray value sequence are named comparison region image gray values.

[0108] In this embodiment, the values in the recognition sub-region image gray value difference sequence are named image gray value differences.

[0109] In this embodiment, the first preset image gray value difference is less than the second preset image gray value difference. Here, the first preset image gray value difference is preferably 40, and the second preset image gray value difference is preferably 60. Specifically, it can also be adjusted according to the actual situation.

[0110] In this embodiment, before mapping the first preset image gray difference value and the second preset image gray difference value to the recognition sub-region image gray difference value sequence, the image gray difference values in the recognition sub-region image gray difference value sequence are sorted from small to large.

[0111] In this embodiment, when counting the number of the first image gray difference values, the first preset image gray difference value is not counted; when counting the number of the second image gray difference values, the first preset image gray difference value and the second preset image gray difference value are not counted; when counting the number of the third image gray difference values, the second preset image gray difference value is not counted. If there are image gray difference values equal to the first preset image gray difference value and the second preset image gray difference value, these image gray difference values are also not counted.

[0112] The beneficial effects of the above technical solution are as follows: The present invention calculates the sub-firework recognition warning coefficient of the sub-real-time region image according to the number of the first image gray difference values, the number of the second image gray difference values, and the number of the third image gray difference values. The present invention can accurately calculate the sub-firework recognition warning coefficient, obtain the sub-firework recognition warning coefficient of each sub-real-time region image, and then reflect the presence of fireworks in each sub-real-time region image, realizing refined calculation and ensuring the accuracy of subsequent firework recognition and warning.

[0113] In some embodiments of the present application, when calculating the sub-firework recognition warning coefficient of the sub-real-time region image according to the number of the first image gray difference values, the number of the second image gray difference values, and the number of the third image gray difference values, it includes:

[0114] Calculating the sub-firework recognition warning coefficient of the sub-real-time region image according to the following formula:

[0115] ;

[0116] where q is the sub-firework recognition warning coefficient of the sub-real-time region image, w is the number of image gray difference values in the recognition sub-region image gray difference value sequence, r e is the e-th image gray difference value in the recognition sub-region image gray difference value sequence, t1 is the first preset image gray difference value, t2 is the second preset image gray difference value, y1 is the number of the first image gray difference values, y2 is the number of the second image gray difference values, y3 is the number of the third image gray difference values, u is the variance of the image gray difference values corresponding to the recognition sub-region image gray difference value sequence, and p is the adjustment coefficient.

[0117] In some embodiments of the present application, the adjustment coefficient p is calculated according to the following method:

[0118] Performing normalization processing on the sub-region image gray value sequence, and constructing a sub-region image gray value curve according to the normalization processing result;

[0119] Determine the curve inflection points on the gray value curve of the sub-region image, and determine the image gray value corresponding to each curve inflection point;

[0120] Count the number of equal image gray values in the equal sub-region image gray value sequence, and count the number of recognized image gray values in the recognized sub-region image gray value sequence;

[0121] Calculate the adjustment coefficient p according to the following formula:

[0122] ;

[0123] where d1 is the average value of all image gray values on the gray value curve of the sub-region image, a is the number of curve inflection points, s i is the inflection point slope corresponding to the i-th curve inflection point, f1 is the number of equal image gray values, f2 is the number of recognized image gray values, d2 is the average value of the image gray values corresponding to all curve inflection points, and e is a constant.

[0124] In this embodiment, the normalization processing method will not be introduced in detail here.

[0125] In this embodiment, when determining the gray value curve of the sub-region image, the normalized image gray value is used as the ordinate, and the abscissa is defaulted to 1, 2, 3, 4, 5,..., m, where m is the number of image gray values.

[0126] In this embodiment, the curve inflection point is the point where the concavity and convexity of the curve change, which can be judged in combination with the gray value curve of the sub-region image.

[0127] The beneficial effects of the above technical solution are: According to the number of equal image gray values and the number of recognized image gray values, the present invention calculates the adjustment coefficient p, realizes the dynamic adjustment of the sub-smoke recognition and warning coefficient, further ensures the calculation accuracy of the sub-smoke recognition and warning coefficient, eliminates calculation errors, and comprehensively considers the equal sub-region image gray value sequence and the recognized sub-region image gray value sequence, which can further ensure the calculation accuracy of the sub-smoke recognition and warning coefficient.

[0128] In some embodiments of the present application, when determining multiple sub-smoke recognition and warning coefficient arrays according to all sub-smoke recognition and warning coefficients, it includes:

[0129] Randomly extract the first sub-smoke recognition and warning coefficient and the second sub-smoke recognition and warning coefficient from all sub-smoke recognition and warning coefficients, and use the first sub-smoke recognition and warning coefficient and the second sub-smoke recognition and warning coefficient as the initial sub-smoke recognition and warning coefficient array;

[0130] Calculate the first coefficient sum value of the two sub-smoke recognition and warning coefficients in the initial sub-smoke recognition and warning coefficient array;

[0131] Randomly extract the third sub-smoke recognition warning coefficient, calculate the absolute value of the first difference between the third sub-smoke recognition warning coefficient and the first coefficient sum value, and determine whether the absolute value of the first difference is less than the preset absolute difference value. If so, update the third sub-smoke recognition warning coefficient to the initial sub-smoke recognition warning coefficient array;

[0132] If not, take the initial sub-smoke recognition warning coefficient array as a sub-smoke recognition warning coefficient array, randomly extract the fourth sub-smoke recognition warning coefficient, and use the fourth sub-smoke recognition warning coefficient and the third sub-smoke recognition warning coefficient as the second initial sub-smoke recognition warning coefficient array;

[0133] Randomly extract the fifth sub-smoke recognition warning coefficient, calculate the second coefficient sum value of the two sub-smoke recognition warning coefficients in the second initial sub-smoke recognition warning coefficient array, and calculate the absolute value of the second difference between the second coefficient sum value and the fifth sub-smoke recognition warning coefficient. Determine whether the absolute value of the second difference is less than the preset absolute difference value. If so, update the fifth sub-smoke recognition warning coefficient to the second initial sub-smoke recognition warning coefficient array;

[0134] If not, take the second initial sub-smoke recognition warning coefficient array as a sub-smoke recognition warning coefficient array;

[0135] Repeat the above steps to obtain multiple sub-smoke recognition warning coefficient arrays.

[0136] In this embodiment, the preset absolute difference value is preferably 5, and it can be specifically adjusted according to the actual situation.

[0137] In this embodiment, if the absolute value of the first difference is less than the preset absolute difference value, update the third sub-smoke recognition warning coefficient to the initial sub-smoke recognition warning coefficient array, randomly extract the fourth sub-smoke recognition warning coefficient, and repeat the above steps until the obtained absolute value of the difference is greater than or equal to the preset absolute difference value.

[0138] The beneficial effects of the above technical solutions are as follows: The present invention determines multiple sub-smoke recognition warning coefficient arrays based on all sub-smoke recognition warning coefficients, realizes the division of all sub-smoke recognition warning coefficients, and provides a basis for calculating the smoke recognition warning coefficient by obtaining multiple sub-smoke recognition warning coefficient arrays.

[0139] S150: Analyze all sub-smoke recognition warning coefficient arrays, calculate the smoke recognition warning coefficient of the smoke recognition warning area, and determine whether to issue a warning according to the smoke recognition warning coefficient. If so, generate a fire situation report information according to the smoke recognition warning coefficient and the GIS geographic information system.

[0140] In this embodiment, when the firework recognition and early warning coefficient is larger, it means the scale of the firework is larger; conversely, when the firework recognition and early warning coefficient is smaller, it indicates that the scale of the firework is smaller or there is even no firework. The firework recognition and early warning coefficient can provide an accurate numerical basis for fire department dispatch and fire extinguishing strategies, ensuring the accuracy of response measures. Combining with the GIS geographic information system, the specific location of the firework can be quickly located, thus guiding the fire extinguishing operation more efficiently.

[0141] In this embodiment, the fire situation report information refers to the firework recognition and early warning coefficient and the specific location of the firework.

[0142] In some embodiments of the present application, when analyzing all sub-firework recognition and early warning coefficient arrays and calculating the firework recognition and early warning coefficient of the firework recognition and early warning area, it includes:

[0143] Calculating the sum value of the sub-firework recognition and early warning coefficients corresponding to each sub-firework recognition and early warning coefficient array;

[0144] Determining the average value of the sum values of the sub-firework recognition and early warning coefficients corresponding to all sub-firework recognition and early warning coefficient sum values;

[0145] Allocating all sub-firework recognition and early warning coefficient sum values less than the average value of the sub-firework recognition and early warning coefficient sum values to the first sub-firework recognition and early warning coefficient sum value array, and allocating all sub-firework recognition and early warning coefficient sum values greater than or equal to the average value of the sub-firework recognition and early warning coefficient sum values to the second sub-firework recognition and early warning coefficient sum value array;

[0146] Sorting the first sub-firework recognition and early warning coefficient sum value array and the second sub-firework recognition and early warning coefficient sum value array from small to large respectively;

[0147] Determining the first starting sub-firework recognition and early warning coefficient sum value and the first ending sub-firework recognition and early warning coefficient sum value of the first sub-firework recognition and early warning coefficient sum value array;

[0148] Determining the second starting sub-firework recognition and early warning coefficient sum value and the second ending sub-firework recognition and early warning coefficient sum value of the second sub-firework recognition and early warning coefficient sum value array;

[0149] Calculating the first average value of the first sub-firework recognition and early warning coefficient sum value array and determining the number g1 of sub-firework recognition and early warning coefficient sum values between the first average value and the first starting sub-firework recognition and early warning coefficient sum value;

[0150] Calculating the second average value of the second sub-firework recognition and early warning coefficient sum value array and determining the number g2 of sub-firework recognition and early warning coefficient sum values between the second average value and the second ending sub-firework recognition and early warning coefficient sum value;

[0151] Calculate a first quantity ratio k1 based on the first sub-smoke recognition and early warning coefficient sum value quantity g1 and the second sub-smoke recognition and early warning coefficient sum value quantity g2, and use it as the first calculation coefficient;

[0152] ;

[0153] Determine the third sub-smoke recognition and early warning coefficient sum value quantity g3 between the first mean value and the sum value of the last sub-smoke recognition and early warning coefficients;

[0154] Determine the fourth sub-smoke recognition and early warning coefficient sum value quantity g4 between the second mean value and the sum value of the first sub-smoke recognition and early warning coefficients;

[0155] Calculate a second quantity ratio k2 based on the third sub-smoke recognition and early warning coefficient sum value quantity g3 and the fourth sub-smoke recognition and early warning coefficient sum value quantity g4, and use it as the second calculation coefficient;

[0156] ;

[0157] Calculate the smoke recognition and early warning coefficient of the smoke recognition and early warning area based on the first calculation coefficient and the second calculation coefficient.

[0158] In this embodiment, the sum value of the first initial sub-smoke recognition and early warning coefficients refers to the minimum sum value of the sub-smoke recognition and early warning coefficients in the first sub-smoke recognition and early warning coefficient sum value array, and the sum value of the first final sub-smoke recognition and early warning coefficients refers to the maximum sum value of the sub-smoke recognition and early warning coefficients in the first sub-smoke recognition and early warning coefficient sum value array.

[0159] In this embodiment, the sum value of the second initial sub-smoke recognition and early warning coefficients refers to the minimum sum value of the sub-smoke recognition and early warning coefficients in the second sub-smoke recognition and early warning coefficient sum value array, and the sum value of the second final sub-smoke recognition and early warning coefficients refers to the maximum sum value of the sub-smoke recognition and early warning coefficients in the second sub-smoke recognition and early warning coefficient sum value array.

[0160] In this embodiment, when determining the first sub-smoke recognition and early warning coefficient sum value quantity g1, judge whether there is a sub-smoke recognition and early warning coefficient sum value equal to the first mean value in the first sub-smoke recognition and early warning coefficient sum value array. If so, include the equal sub-smoke recognition and early warning coefficient sum value in the first sub-smoke recognition and early warning coefficient sum value quantity g1.

[0161] In this embodiment, when determining the second sub-smoke recognition and early warning coefficient sum value quantity g2, judge whether there is a sub-smoke recognition and early warning coefficient sum value equal to the second mean value in the second sub-smoke recognition and early warning coefficient sum value array. If so, include the equal sub-smoke recognition and early warning coefficient sum value in the second sub-smoke recognition and early warning coefficient sum value quantity g2.

[0162] In this embodiment, when determining the sum value of the third sub-smoke recognition warning coefficients and the number g3 of values, it is judged whether there is a sub-smoke recognition warning coefficient sum value equal to the first average value in the first sub-smoke recognition warning coefficient sum value array. If so, the equal sub-smoke recognition warning coefficient sum values are repeatedly counted into the sum value of the third sub-smoke recognition warning coefficients and the number g3.

[0163] In this embodiment, when determining the sum value of the fourth sub-smoke recognition warning coefficients and the number g4 of values, it is judged whether there is a sub-smoke recognition warning coefficient sum value equal to the second average value in the second sub-smoke recognition warning coefficient sum value array. If so, the equal sub-smoke recognition warning coefficient sum values are repeatedly counted into the sum value of the fourth sub-smoke recognition warning coefficients and the number g4.

[0164] The beneficial effects of the above technical solution are as follows: The present invention determines the sum value of the first sub-smoke recognition warning coefficients and the number g2 of values, and then calculates the first calculation coefficient, determines the sum value of the third sub-smoke recognition warning coefficients and the number g3 of values, and the sum value of the fourth sub-smoke recognition warning coefficients and the number g4 of values, and then determines the second calculation coefficient, realizing the division and quantity judgment of all sub-smoke recognition warning coefficient sum values, and realizing cross-array calculation, performing cross-analysis on the first sub-smoke recognition warning coefficient sum value array and the second sub-smoke recognition warning coefficient sum value array, ensuring the sufficiency of the analysis, avoiding errors, and laying a foundation for the calculation of the smoke recognition warning coefficient through the first calculation coefficient and the second calculation coefficient.

[0165] In some embodiments of the present application, when calculating the smoke recognition warning coefficient of the smoke recognition warning area according to the first calculation coefficient and the second calculation coefficient, it includes:

[0166] Calculate the absolute value of the coefficient difference between the first calculation coefficient and the second calculation coefficient, and calculate the sum value of the coefficients of the first calculation coefficient and the second calculation coefficient;

[0167] Take the coefficient ratio of the absolute value of the coefficient difference to the sum value of the coefficients as the smoke recognition warning coefficient of the smoke recognition warning area.

[0168] The beneficial effects of the above technical solution are as follows: The present invention calculates the smoke recognition warning coefficient of the smoke recognition warning area according to the first calculation coefficient and the second calculation coefficient. Through the smoke recognition warning coefficient, it can be reflected whether there is smoke in the smoke recognition warning area and the specific fire situation, without manual participation in judgment, ensuring the smoke recognition efficiency and recognition accuracy, and eliminating subjectivity.

[0169] In some embodiments of the present application, when judging whether to issue a warning according to the smoke recognition warning coefficient, it includes:

[0170] Judge whether to issue an early warning according to the relationship between the fireworks recognition early warning coefficient and the preset fireworks recognition early warning coefficient;

[0171] When the fireworks recognition early warning coefficient is less than the preset fireworks recognition early warning coefficient, it is judged that there is no fireworks in the fireworks recognition early warning area, and no early warning is issued;

[0172] When the fireworks recognition early warning coefficient is greater than or equal to the preset fireworks recognition early warning coefficient, it is judged that there is fireworks in the fireworks recognition early warning area, and an early warning is issued.

[0173] In this embodiment, the preset fireworks recognition early warning coefficient is preferably 0.4, and can be specifically adjusted according to actual needs. The preset fireworks recognition early warning coefficient can judge whether there is fireworks in the fireworks recognition early warning area.

[0174] The beneficial effects of the above technical solution are: The present invention judges whether to issue an early warning according to the relationship between the fireworks recognition early warning coefficient and the preset fireworks recognition early warning coefficient, which can significantly improve the efficiency and accuracy of fireworks recognition, achieve accurate recognition and instant early warning, and overcome the problems of low efficiency of manual visual recognition and insufficient accuracy of fire situation reporting.

[0175] In order to further elaborate the technical idea of the present invention, the technical solution of the present invention will be described in combination with specific application scenarios.

[0176] Correspondingly, as Figure 2 shown, the present application also provides a GIS-based intelligent fireworks recognition early warning system, including:

[0177] An image acquisition module, used to pre-deploy image acquisition devices in the fireworks recognition early warning area, collect corresponding real-time area images based on the image acquisition devices, extract the image pixel values of the real-time area images, and perform gray conversion to obtain a sequence of gray values of the area images;

[0178] A marker generation module, used to determine a sequence of gray values of a comparison area image, and generate an initial fireworks analysis marker for the fireworks recognition early warning area according to the sequence of gray values of the area image and the sequence of gray values of the comparison area image, wherein the initial fireworks analysis marker includes a marker for no fireworks in the area and a marker for pending fireworks recognition in the area;

[0179] An image division module, used to randomly divide the real-time area image into multiple sub-real-time area images when the marker for pending fireworks recognition in the area is generated, and determine the corresponding sequence of gray values of the sub-area images for each sub-real-time area image according to the sequence of gray values of the area image;

[0180] A data calculation module is configured to analyze the grayscale value sequence of the sub-region image, calculate the sub-smoke recognition and early warning coefficient of the sub-real-time region image based on the analysis result, and determine multiple sub-smoke recognition and early warning coefficient arrays according to all the sub-smoke recognition and early warning coefficients;

[0181] An identification and early warning module is configured to analyze all the sub-smoke recognition and early warning coefficient arrays, calculate the smoke recognition and early warning coefficient of the smoke recognition and early warning region, and determine whether to issue an early warning according to the smoke recognition and early warning coefficient. If so, a fire situation report information is generated according to the smoke recognition and early warning coefficient and the GIS geographic information system.

[0182] In the description of the above embodiments, the specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0183] Although the present invention has been described above with reference to the embodiments, various improvements can be made to it and components therein can be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed by the present invention can be combined with each other in any way, and only for the sake of saving space and resources, the situations of these combinations are not all described in this specification.

[0184] Those of ordinary skill in the art can understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A GIS-based intelligent fireworks recognition and early warning method, characterized in that: include: An image acquisition device is pre-deployed in the fireworks recognition and warning area, and the corresponding real-time regional image is acquired based on the image acquisition device, and the image pixel value of the real-time regional image is extracted and grayscale conversion is performed to obtain a grayscale value sequence of the regional image; Determine a gray value sequence of the contrasting regional image, and generate an initial fireworks analysis mark for the fireworks identification and warning region according to the regional image gray value sequence and the contrasting regional image gray value sequence, wherein the initial fireworks analysis mark includes a regional no-smoke and fire mark and a regional fireworks to-be-identified mark, and the contrasting regional image gray value sequence corresponds one-to-one to the regional image gray value sequence; When generating the regional fireworks to-be-recognized mark, the real-time regional image is randomly divided into a plurality of sub-real-time regional images, and a sub-region image grayscale value sequence corresponding to each sub-real-time regional image is determined according to the regional image grayscale value sequence; Analyzing the gray value sequence of the sub-region image, and calculating the sub-fireworks recognition warning coefficient of the sub-real-time region image based on the analysis result, and determining a plurality of sub-fireworks recognition warning coefficient arrays according to all the sub-fireworks recognition warning coefficients; Analyze all the sub-fireworks recognition and warning coefficient arrays, calculate the firework recognition and warning coefficient of the firework recognition and warning area, and determine whether to issue a warning based on the firework recognition and warning coefficient. If so, generate fire report information based on the firework recognition and warning coefficient and the GIS geographic information system; When analyzing the gray value sequence of the sub-region image and calculating the sub-fireworks recognition warning coefficient of the sub-real-time region image based on the analysis result, it includes: Compare the sub-region image grayscale value sequence with the contrast region image grayscale value sequence one by one, extract the sub-region image grayscale value that is equal to the corresponding contrast region image grayscale value in the sub-region image grayscale value sequence, and generate an equal sub-region image grayscale value sequence; Generating an identification sub-region image grayscale value sequence according to the remaining sub-region image grayscale values ​​in the sub-region image grayscale value sequence; Calculating the image grayscale difference between the grayscale value sequence of the identified sub-region image and the corresponding grayscale value of the contrast region image to generate the grayscale difference sequence of the identified sub-region image; Presetting a first preset image grayscale difference value and a second preset image grayscale difference value; Mapping the first preset image grayscale difference and the second preset image grayscale difference to the recognition sub-region image grayscale difference sequence; Determine a minimum image grayscale difference value and a maximum image grayscale difference value in the identified sub-region image grayscale difference value sequence, and count the number of first image grayscale difference values ​​between the minimum image grayscale difference value and the first preset image grayscale difference value; Counting the number of second image grayscale difference values ​​between the first preset image grayscale difference value and the second preset image grayscale difference value; Counting the number of third image grayscale difference values ​​between the maximum image grayscale difference value and the second preset image grayscale difference value; Calculating a sub-fireworks recognition warning coefficient of the sub-real-time area image according to the grayscale difference quantity of the first image, the grayscale difference quantity of the second image, and the grayscale difference quantity of the third image; When calculating the sub-fireworks recognition warning coefficient of the sub-real-time area image according to the grayscale difference quantity of the first image, the grayscale difference quantity of the second image, and the grayscale difference quantity of the third image, the method includes: The sub-fireworks recognition warning coefficient of the sub-real-time area image is calculated according to the following formula: ； Among them, q is the sub-fireworks recognition warning coefficient of the sub-real-time area image, w is the number of image grayscale differences in the grayscale difference sequence of the identified sub-area image, and r e To identify the e-th image grayscale difference in the sub-region image grayscale difference sequence, t1 is the first preset image grayscale difference, t2 is the second preset image grayscale difference, y1 is the number of first image grayscale differences, y2 is the number of second image grayscale differences, y3 is the number of third image grayscale differences, u is the image grayscale difference variance corresponding to the identified sub-region image grayscale difference sequence, and p is the adjustment coefficient.

2. The GIS-based intelligent fireworks recognition and early warning method according to claim 1 is characterized in that: When determining the gray value sequence of the contrasting region image, and generating an initial fireworks analysis mark for the fireworks recognition and warning region according to the region image gray value sequence and the contrasting region image gray value sequence, the method includes: Acquire a smoke and fire free area image corresponding to the smoke and fire identification and warning area, and perform grayscale conversion on the smoke and fire free area image to obtain a grayscale value sequence of the comparison area image; Determine whether the regional image grayscale values ​​in the regional image grayscale value sequence are all equal to the comparison regional image grayscale value sequence, and if so, generate a no-smoke and fire mark for the fireworks recognition and warning area; If not, a regional fireworks to be identified mark is generated for the fireworks identification warning area.

3. The GIS-based intelligent fireworks recognition and early warning method according to claim 1 is characterized in that: The adjustment factor p is calculated according to the following method: Normalizing the sub-region image grayscale value sequence, and constructing a sub-region image grayscale value curve according to the normalization result; Determine the inflection points of the sub-region image grayscale value curve, and determine the image grayscale value corresponding to each inflection point of the curve; Counting the number of equal image grayscale values ​​of the equal sub-region image grayscale value sequence, and counting the number of identified image grayscale values ​​of the identified sub-region image grayscale value sequence; The adjustment coefficient p is calculated according to the following formula: ； Among them, d1 is the average value of all image gray values ​​on the sub-region image gray value curve, a is the number of inflection points of the curve, and s i is the inflection point slope corresponding to the i-th curve inflection point, f1 is the number of equal image gray values, f2 is the number of identified image gray values, d2 is the average of the image gray values ​​corresponding to all curve inflection points, and e is a constant.

4. The GIS-based intelligent fireworks recognition and early warning method according to claim 1 is characterized in that: When multiple sub-fireworks identification warning coefficient arrays are determined according to all sub-fireworks identification warning coefficients, it includes: Randomly extracting a first sub-fireworks recognition warning coefficient and a second sub-fireworks recognition warning coefficient from all sub-fireworks recognition warning coefficients, and using the first sub-fireworks recognition warning coefficient and the second sub-fireworks recognition warning coefficient as an initial sub-fireworks recognition warning coefficient array; Calculate the first coefficient and value of two sub-fireworks recognition warning coefficients in the initial sub-fireworks recognition warning coefficient array; Randomly extracting a third sub-fireworks recognition warning coefficient, calculating a first difference absolute value between the third sub-fireworks recognition warning coefficient and the sum of the first coefficients, determining whether the first difference absolute value is less than a preset difference absolute value, and if so, updating the third sub-fireworks recognition warning coefficient to the initial sub-fireworks recognition warning coefficient array; If not, taking the initial sub-fireworks recognition warning coefficient array as a sub-fireworks recognition warning coefficient array, randomly extracting the fourth sub-fireworks recognition warning coefficient, and taking the fourth sub-fireworks recognition warning coefficient and the third sub-fireworks recognition warning coefficient as the second initial sub-fireworks recognition warning coefficient array; Randomly extract the fifth sub-fireworks recognition warning coefficient, calculate the second coefficient sum of the two sub-fireworks recognition warning coefficients in the second initial sub-fireworks recognition warning coefficient array, and calculate the second absolute value of the difference between the second coefficient sum and the fifth sub-fireworks recognition warning coefficient, determine whether the second absolute value of the difference is less than a preset absolute value of the difference, and if so, update the fifth sub-fireworks recognition warning coefficient to the second initial sub-fireworks recognition warning coefficient array; If not, taking the second initial sub-fireworks recognition warning coefficient array as a sub-fireworks recognition warning coefficient array; Repeat the above steps to obtain multiple sub-fireworks recognition warning coefficient arrays.

5. The GIS-based intelligent fireworks recognition and early warning method according to claim 1 is characterized in that: When analyzing all the sub-fireworks recognition warning coefficient arrays and calculating the firework recognition warning coefficient of the firework recognition warning area, it includes: Calculate the sub-fireworks recognition warning coefficient and value corresponding to each sub-fireworks recognition warning coefficient array; Determine the mean of the sub-fireworks identification warning coefficients and values ​​corresponding to all the sub-fireworks identification warning coefficients and values; Allocate all sub-fireworks recognition warning coefficients and values ​​that are less than the mean value of the sub-fireworks recognition warning coefficients and values ​​to a first sub-fireworks recognition warning coefficient and value array, and all sub-fireworks recognition warning coefficients and values ​​that are greater than or equal to the mean value of the sub-fireworks recognition warning coefficients and values ​​to a second sub-fireworks recognition warning coefficient and value array; respectively sorting the first sub-pyrotechnics recognition warning coefficient and value array and the second sub-pyrotechnics recognition warning coefficient and value array from small to large; Determine a first initial sub-fireworks recognition warning coefficient and value and a first final sub-fireworks recognition warning coefficient and value of the first sub-fireworks recognition warning coefficient and value array; Determine a second initial sub-fireworks recognition warning coefficient and value and a second final sub-fireworks recognition warning coefficient and value of the second sub-fireworks recognition warning coefficient and value array; Calculate a first mean value of the first sub-fireworks recognition warning coefficient and value array, and determine the number g1 of first sub-fireworks recognition warning coefficient and value between the first mean value and the first initial sub-fireworks recognition warning coefficient and value; Calculate a second mean value of the second sub-fireworks identification warning coefficient and value array, and determine the number g2 of second sub-fireworks identification warning coefficient and value between the second mean value and the second last sub-fireworks identification warning coefficient and value; Calculate a first quantity ratio k1 according to the first sub-fireworks identification warning coefficient and value quantity g1 and the second sub-fireworks identification warning coefficient and value quantity g2, and use it as a first calculation coefficient; ； Determine a third sub-fireworks identification warning coefficient and value quantity g3 between the first mean value and the first last sub-fireworks identification warning coefficient and value; Determine a fourth sub-fireworks identification warning coefficient and value quantity g4 between the second mean value and the first initial sub-fireworks identification warning coefficient and value; Calculate a second quantity ratio k2 according to the third sub-fireworks identification warning coefficient and value quantity g3 and the fourth sub-fireworks identification warning coefficient and value quantity g4, and use it as a second calculation coefficient; ； A fireworks recognition and warning coefficient of the fireworks recognition and warning area is calculated according to the first calculation coefficient and the second calculation coefficient.

6. The GIS-based intelligent fireworks recognition and early warning method according to claim 5 is characterized in that: When calculating the fireworks recognition and warning coefficient of the fireworks recognition and warning area according to the first calculation coefficient and the second calculation coefficient, the method includes: Calculating an absolute value of a coefficient difference between the first calculation coefficient and the second calculation coefficient, and calculating a coefficient sum of the first calculation coefficient and the second calculation coefficient; The coefficient ratio of the absolute value of the coefficient difference to the coefficient sum is used as the fireworks recognition and warning coefficient of the fireworks recognition and warning area.

7. The GIS-based intelligent fireworks recognition and early warning method according to claim 1 is characterized in that: When judging whether to issue an early warning according to the fireworks recognition early warning coefficient, it includes: Determining whether to issue a warning based on the relationship between the fireworks recognition warning coefficient and a preset fireworks recognition warning coefficient; When the fireworks recognition warning coefficient is less than the preset fireworks recognition warning coefficient, it is determined that there are no fireworks in the fireworks recognition warning area, and no warning is issued; When the fireworks recognition warning coefficient is greater than or equal to the preset fireworks recognition warning coefficient, it is determined that fireworks exist in the fireworks recognition warning area and a warning is issued.

8. A GIS-based intelligent fire and smoke identification and early warning system, applied to the GIS-based intelligent fire and smoke identification and early warning method according to any one of claims 1 to 7, characterized in that: include: An image acquisition module is used to pre-deploy an image acquisition device in the fireworks recognition and warning area, acquire a corresponding real-time regional image based on the image acquisition device, extract image pixel values ​​of the real-time regional image, and perform grayscale conversion to obtain a regional image grayscale value sequence; a mark generation module, used to determine a gray value sequence of a comparison region image, and generate an initial fireworks analysis mark for the fireworks identification and warning region according to the region image gray value sequence and the comparison region image gray value sequence, wherein the initial fireworks analysis mark includes a region no fireworks mark and a region fireworks to be identified mark, and the comparison region image gray value sequence corresponds to the region image gray value sequence one by one; An image division module, used for randomly dividing the real-time regional image into a plurality of sub-real-time regional images when generating the regional fireworks to-be-recognized mark, and determining a sub-region image grayscale value sequence corresponding to each sub-real-time regional image according to the regional image grayscale value sequence; A data calculation module, used for analyzing the gray value sequence of the sub-region image, and calculating the sub-fireworks recognition warning coefficient of the sub-real-time region image based on the analysis result, and determining a plurality of sub-fireworks recognition warning coefficient arrays according to all the sub-fireworks recognition warning coefficients; The identification and warning module is used to analyze all the sub-fireworks identification and warning coefficient arrays, calculate the firework identification and warning coefficient of the firework identification and warning area, and determine whether to issue a warning based on the firework identification and warning coefficient. If so, generate fire report information based on the firework identification and warning coefficient and the GIS geographic information system.

Images