Laser radar smoke and fire monitoring visual analysis method and system for forest protection

Abstract

The invention discloses a laser radar smoke and fire monitoring visual analysis method and system for forest protection, and relates to the technical field of smoke and fire monitoring visual analysis, and the method comprises the following steps: obtaining a dynamic point cloud region based on a continuous two-dimensional point cloud image, obtaining an image of the dynamic point cloud region in the projection direction of the two-dimensional point cloud image, marking the image as a detection image, and carrying out the detection of the detection image; obtaining a first number of historical flame images, carrying out graying processing on the historical flame images to obtain historical flame gray-scale images, obtaining a flame gray-scale threshold range based on the historical flame gray-scale images, judging whether the dynamic point cloud area is an abnormal area based on the flame gray-scale threshold range and the detection image, and if yes, sending out a flame generation signal; the method is used for solving the problem that the accuracy of judging whether smoke and fire occur or not based on the change of a laser radar point cloud picture is low due to the influence of environmental change in the existing laser radar smoke and fire monitoring technology.

Description

Technical Field

[0001] The present invention relates to the technical field of smoke and fire monitoring visualization analysis, and specifically to a laser radar smoke and fire monitoring visualization analysis method and system for forest protection. Background Art

[0002] Forest fires are one of the world's eight major natural disasters. They are sudden, destructive, dangerous, and difficult to handle and extinguish, seriously endangering people's lives, property, and forest resources. Traditional forest fire monitoring methods rely primarily on manual ground patrols and high-point inspections from watchtowers. These methods suffer from limited patrol range, low efficiency, and environmental constraints. With the development of technology, point cloud maps of forests can be obtained based on lidar. When a fire occurs, the obtained point cloud maps will change. Therefore, the changes in the point cloud maps can be monitored in real time to determine whether there is a fire. However, there are many factors that cause changes in the point cloud maps of forests, such as the movement of animals. This makes it not completely accurate to judge whether there are fireworks in the forest. That is, the existing lidar fireworks monitoring technology is affected by environmental changes, resulting in low accuracy in judging whether there are fireworks based on changes in the lidar point cloud maps. Summary of the Invention

[0003] The present invention aims to solve one of the technical problems in the prior art to at least a certain extent, by obtaining a dynamic point cloud area based on a continuous two-dimensional point cloud image, obtaining an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image, marking it as a detection image, obtaining a first number of historical flame images, graying the historical flame image to obtain a historical flame grayscale image, obtaining a flame grayscale threshold range based on the historical flame grayscale image, judging whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection image, and if so, issuing a flame signal, so as to solve the problem in the existing laser radar fireworks and fire monitoring technology that the accuracy of judging whether fireworks and fire occur based on changes in the laser radar point cloud image is low due to environmental changes.

[0004] To achieve the above objectives, this application provides a visual analysis method for laser radar smoke and fire monitoring for forest protection, comprising the following steps: Continuously acquire forest point cloud images through laser radar, obtain a two-dimensional projection image of the forest point cloud image, and mark it as a two-dimensional point cloud image; Obtain dynamic point cloud areas based on continuous two-dimensional point cloud images; Obtain an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image and mark it as a detection image; Obtain a first number of historical flame graphs, and grayscale the historical flame graphs to obtain historical flame grayscale graphs; Obtaining a flame grayscale threshold range based on a historical flame grayscale image; Based on the flame grayscale threshold range and the detection map, it is determined whether the dynamic point cloud area is an abnormal area. If so, a flame generation signal is issued.

[0005] Furthermore, obtaining a dynamic point cloud region based on a continuous two-dimensional point cloud image includes the following sub-steps: Mark the two-dimensional point cloud image acquired in real time as the first point cloud image, and mark the two-dimensional point cloud image acquired in the previous frame as the second point cloud image; Establish a plane rectangular coordinate system, marked as the cloud point coordinate system, place the first point cloud image in the first quadrant, and place the second point cloud image at the same position of the first point cloud image; Mark each cloud point in the first point cloud image as the first cloud point; Mark each cloud point of the second point cloud image as a second cloud point; In the cloud point coordinate system, mark the second cloud point closest to the first cloud point as an adjacent cloud point; Get the distance between each first cloud point and its corresponding adjacent cloud point, and mark it as the adjacent distance.

[0006] Furthermore, obtaining a dynamic point cloud region based on a continuous two-dimensional point cloud image further includes the following sub-steps: Obtaining a second number of adjacent distances of historical flame occurrence portions, and marking them as historical flame distances; Get the range of historical flame distance, marked as flame movement range; Determine whether each adjacent distance is within the flame movement range. If so, mark the corresponding first cloud point as a point to be analyzed.

[0007] Furthermore, obtaining a dynamic point cloud region based on a continuous two-dimensional point cloud image further includes the following sub-steps: Get any point to be analyzed, mark it as the starting analysis point, establish an a×a analysis area with the starting analysis point as the center, and mark it as the starting division area; Determine whether the starting partition area contains a new point to be analyzed. If not, continue to search for the next point to be analyzed as the starting analysis point. If so, establish an a×a analysis area with the new point to be analyzed as the center, mark it as the search partition area, and again determine whether the search partition area contains a new point to be analyzed. Then repeat the establishment of a new search partition area and the judgment until the new search partition area does not contain a new point to be analyzed. Mark the continuously searched points to be analyzed as a coherent group analysis point, and use the coherent group analysis points of each search as a set of analysis data.

[0008] Furthermore, obtaining a dynamic point cloud region based on a continuous two-dimensional point cloud image further includes the following sub-steps: Under a set of analysis data, a coherent group of analysis points with a minimum value of the horizontal coordinate and a maximum value of the horizontal coordinate are obtained, and are marked as the first analysis point and the second analysis point respectively. A coherent group of analysis points with a minimum value of the vertical coordinate and a maximum value of the vertical coordinate are obtained, and are marked as the third analysis point and the fourth analysis point respectively. Then, a line segment passing through the first analysis point and parallel to the Y axis, a line segment passing through the second analysis point and parallel to the Y axis, a line segment passing through the third analysis point and parallel to the X axis, and a line segment passing through the fourth analysis point and parallel to the X axis are obtained, and are marked as the first line segment, the second line segment, the third line segment, and the fourth line segment respectively. The rectangular area formed by the first line segment, the second line segment, the third line segment, and the fourth line segment is marked as the dynamic point cloud area.

[0009] Furthermore, grayscale processing of the historical flame graph to obtain the historical flame grayscale graph includes the following sub-steps: Get the RGB value of each pixel in the historical flame graph and mark it as the historical flame RGB value; The grayscale conversion formula is used to convert all historical flame RGB values ​​in the historical flame graph into grayscale values ​​to obtain the historical flame grayscale graph.

[0010] Furthermore, obtaining the flame grayscale threshold range based on the historical flame grayscale image includes the following sub-steps: Mark the grayscale value of the pixel in the historical flame grayscale image as the flame grayscale value; Divide the range of all flame grayscale values ​​into b equal ranges, marked as flame grayscale division ranges; Count the frequency of each flame grayscale division range and mark it as the divided flame grayscale frequency; A histogram is drawn with the flame grayscale value as the X-axis, the flame grayscale frequency as the Y-axis, and the flame grayscale division range as the histogram interval, which is marked as the flame grayscale histogram.

[0011] Furthermore, obtaining the flame grayscale threshold range based on the historical flame grayscale image includes the following sub-steps: Get the sum of the divided grayscale frequencies and mark it as the historical total flame frequency; The first frequency threshold is calculated as: C1=0.1×D1 / b; where C1 is the first frequency threshold and D1 is the total historical flame frequency; In the flame grayscale histogram, starting from the leftmost divided flame grayscale frequency of the flame grayscale histogram and going rightward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the minimum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening minimum value; In the flame grayscale histogram, starting from the rightmost divided flame grayscale frequency of the flame grayscale histogram and going leftward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the maximum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening maximum value; The range from the screening minimum value to the screening maximum value is labeled as the flame grayscale threshold range.

[0012] Furthermore, judging whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection map, if so, issuing a flame generation signal includes the following steps: Grayscale the detection image to obtain a detection grayscale image; obtain the grayscale value of each pixel in the detection grayscale image and mark it as the detection grayscale value; Calculate the mean of all detected grayscale values ​​and mark it as the detected grayscale mean; The detection grayscale mean is obtained to determine whether it is within the flame grayscale threshold range. If so, the dynamic point cloud area is determined to be an abnormal area, and a flame generation signal is issued.

[0013] The present application also provides a laser radar smoke and fire monitoring visualization analysis system for forest protection, comprising: a point cloud acquisition module, an area acquisition module, an image acquisition module, a grayscale processing module, a threshold acquisition module, and an anomaly judgment module; The point cloud image acquisition module is used to continuously acquire a forest point cloud image through a laser radar, and obtain a two-dimensional projection image of the forest point cloud image, which is marked as a two-dimensional point cloud image; The region acquisition module is used to acquire a dynamic point cloud region based on a continuous two-dimensional point cloud image; The image acquisition module is used to acquire an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image, which is marked as a detection image; The grayscale processing module is used to obtain a first number of historical flame graphs, and perform grayscale processing on the historical flame graphs to obtain historical flame grayscale graphs; The threshold acquisition module is used to obtain the flame grayscale threshold range based on the historical flame grayscale image; The abnormality judgment module is used to judge whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection map, and if so, send a flame generation signal.

[0014] Beneficial effects of the present invention: The present invention obtains a dynamic point cloud area based on a continuous two-dimensional point cloud image, obtains an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image, marks it as a detection image, obtains a first number of historical flame images, grayscales the historical flame images to obtain historical flame grayscale images, obtains a flame grayscale threshold range based on the historical flame grayscale images, and determines whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection image. If so, a flame signal is generated. The advantage of the present invention is that the laser radar detection and image detection are combined to perform flame judgment, so that the flame judgment is more accurate. The present invention obtains a dynamic point cloud area, which has the advantage of preliminarily identifying the dynamic point cloud area as a flame area and subsequently performing image processing on the dynamic point cloud area, thereby reducing the subsequent image processing amount and increasing the judgment efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a functional block diagram of the system of the present invention; Figure 2 A schematic diagram of adjacent distances of the present invention; Figure 3 A schematic diagram of the search area of ​​the present invention; Figure 4 is a schematic diagram of a dynamic point cloud area of ​​the present invention; Figure 5 is a schematic diagram of a flame grayscale histogram of the present invention; Figure 6 Schematic diagram of the screening minimum value and the screening maximum value of the present invention; Figure 7 Flow chart of the steps of the method of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0017] Example 1, please refer to Figure 1 As shown, this application provides a laser radar smoke and fire monitoring visualization analysis system for forest protection, including: The point cloud image acquisition module is used to continuously acquire forest point cloud images through laser radar, obtain a two-dimensional projection image of the forest point cloud image, and mark it as a two-dimensional point cloud image; The region acquisition module is used to obtain dynamic point cloud regions based on continuous two-dimensional point cloud images; The region acquisition module is configured with an adjacent distance acquisition strategy, which includes: Mark the two-dimensional point cloud image acquired in real time as the first point cloud image, and mark the two-dimensional point cloud image acquired in the previous frame as the second point cloud image; Establish a plane rectangular coordinate system, marked as the cloud point coordinate system, place the first point cloud image in the first quadrant, and place the second point cloud image at the same position of the first point cloud image; Mark each cloud point in the first point cloud image as the first cloud point; Mark each cloud point of the second point cloud image as a second cloud point; In the cloud point coordinate system, mark the second cloud point closest to the first cloud point as an adjacent cloud point; Obtain the distance between each first cloud point and its corresponding adjacent cloud point, marking it as adjacent distance; In practical applications, please refer to Figure 2 As shown, all adjacent distances are obtained, for example, the adjacent distances are 0.31 cm, 0.22 cm, and 0.51 cm.

[0018] The region acquisition module is configured with strategies for acquiring points to be analyzed. These strategies include: Obtaining a second number of adjacent distances of historical flame occurrence portions, and marking them as historical flame distances; Get the range of historical flame distance, marked as flame movement range; Determine whether each adjacent distance is within the flame movement range. If so, mark the corresponding first cloud point as the point to be analyzed; In actual applications, for example, if the historical flame distances are 0.62cm, 0.11cm, and 0.32cm, the flame movement range is 0.21cm to 0.62cm. When the adjacent distance is between 0.21cm and 0.62cm, the principle is that if the static point cloud images before and after are obtained, the position of each cloud point remains unchanged, that is, the adjacent distance is 0. If a flame is generated, the point cloud image obtained at this time changes, so the adjacent distance becomes larger.

[0019] The region acquisition module is configured with a coherent group analysis point acquisition strategy, which includes: Get any point to be analyzed, mark it as the starting analysis point, establish an a×a analysis area with the starting analysis point as the center, and mark it as the starting division area; Determine whether the starting partitioned area contains a new point to be analyzed. If not, continue to search for the next point to be analyzed as the starting analysis point. If so, establish an a×a analysis area with the new point to be analyzed as the center, mark it as the search partitioned area, and again determine whether the search partitioned area contains a new point to be analyzed. Repeat the establishment and determination of new search partitioned areas until the new search partitioned area does not contain a new point to be analyzed. Mark the continuously searched points to be analyzed as a coherent group of analysis points, and use the coherent group of analysis points searched each time as a set of analysis data; In practical applications, please refer to Figure 3 As shown, the setting a can be set based on the distribution of cloud points. For example, if the average interval of cloud points in the point cloud is 0.5 cm, a can be set to twice the average interval of cloud points in the point cloud, that is, 1 cm. This search method can exclude individual noise points and obtain each changed fire area. A set of analysis data represents a changed area.

[0020] The region acquisition module is configured with a dynamic point cloud region acquisition strategy, which includes: Under a set of analysis data, a coherent group of analysis points with a minimum abscissa value and a maximum abscissa value are obtained, and are marked as the first analysis point and the second analysis point respectively. A coherent group of analysis points with a minimum ordinate value and a maximum ordinate value are obtained, and are marked as the third analysis point and the fourth analysis point respectively. Then, a line segment passing through the first analysis point and parallel to the Y axis, a line segment passing through the second analysis point and parallel to the Y axis, a line segment passing through the third analysis point and parallel to the X axis, and a line segment passing through the fourth analysis point and parallel to the X axis are obtained, and are marked as the first line segment, the second line segment, the third line segment, and the fourth line segment respectively. The rectangular area formed by the first line segment, the second line segment, the third line segment, and the fourth line segment is marked as the dynamic point cloud area. In practical applications, please refer to Figure 4 As shown, a dynamic point cloud area is obtained, which may be a flame generating area.

[0021] The image acquisition module is used to acquire an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image, which is marked as a detection image; The grayscale processing module is used to obtain a first number of historical flame graphs, and grayscale the historical flame graphs to obtain historical flame grayscale graphs; The grayscale processing module is configured with grayscale processing strategies, which include: Get the RGB value of each pixel in the historical flame graph and mark it as the historical flame RGB value; The grayscale conversion formula is used to convert all historical flame RGB values ​​in the historical flame graph into grayscale values ​​to obtain the historical flame grayscale graph.

[0022] The threshold acquisition module is used to obtain the flame grayscale threshold range based on the historical flame grayscale image; The threshold acquisition module is configured with a flame grayscale histogram establishment strategy, which includes: Mark the grayscale value of the pixel in the historical flame grayscale image as the flame grayscale value; Divide the range of all flame grayscale values ​​into b equal ranges, marked as flame grayscale division ranges; Count the frequency of each flame grayscale division range and mark it as the divided flame grayscale frequency; The flame gray value is the X-axis, the flame gray frequency is the Y-axis, and the flame gray division range is the histogram interval to draw a histogram, which is marked as the flame gray histogram; In practical applications, please refer to Figure 5 As shown, the range of all flame grayscale values ​​is evenly divided into 7 equal ranges to obtain a flame grayscale histogram.

[0023] The threshold acquisition module is configured with a first frequency threshold acquisition strategy, which includes: Get the sum of the divided grayscale frequencies and mark it as the historical total flame frequency; The first frequency threshold is calculated as: C1=0.1×D1 / b; where C1 is the first frequency threshold, and D1 is the total historical flame frequency; it is set to 0.1 because 0.1 is a small proportion, which means it is considered to be less distributed; In practical applications, please refer to Figure 5 As shown, the total historical flame frequency is 4.05 million, and the first frequency threshold is: C1=0.1×405 / 7=58,000, and the result is rounded to one decimal place; In the flame grayscale histogram, starting from the leftmost divided flame grayscale frequency of the flame grayscale histogram and going rightward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the minimum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening minimum value; In the flame grayscale histogram, starting from the rightmost divided flame grayscale frequency of the flame grayscale histogram and going leftward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the maximum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening maximum value; The range from the screening minimum value to the screening maximum value is marked as the flame grayscale threshold range; In practical applications, please refer to Figure 5 and Figure 6As shown, in the flame grayscale histogram, starting from the leftmost divided flame grayscale frequency of 10,000, and judging to the right that the divided flame grayscale frequency is less than 58,000, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency of 110,000, and stop until it is greater than 58,000. The minimum value of the horizontal coordinate of the flame grayscale histogram at this time is 225, so the screening minimum value is 225, and starting from the rightmost divided flame grayscale frequency of 750,000, and judging to the left that the divided flame grayscale frequency is greater than 58,000, stop judging, and the maximum value of the horizontal coordinate of the flame grayscale histogram at this time is 255, so the screening maximum value is 255. By screening less distributed data, a more accurate flame grayscale threshold range is obtained.

[0024] The abnormality judgment module is used to judge whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection map. If so, it sends a flame generation signal; The abnormality judgment module is configured with an abnormality judgment strategy, which includes: Grayscale the detection image to obtain a detection grayscale image; obtain the grayscale value of each pixel in the detection grayscale image and mark it as the detection grayscale value; Calculate the mean of all detected grayscale values ​​and mark it as the detected grayscale mean; Obtain whether the detected grayscale mean is within the flame grayscale threshold range. If so, the dynamic point cloud area is considered to be an abnormal area, and a flame generation signal is issued; In practical applications, if the detected grayscale mean value is 234 and is within the flame grayscale threshold range of 225 to 255, the dynamic point cloud area is considered to be an abnormal area, and a flame signal is issued, indicating that the dynamic point cloud area has changed, and the grayscale value is close to the flame during the change, so a flame signal is issued.

[0025] Example 2, please refer to Figure 7 As shown, the present application provides a laser radar smoke and fire monitoring visualization analysis method for forest protection, comprising the following steps: Step S1: continuously acquire a forest point cloud image through a laser radar, obtain a two-dimensional projection image of the forest point cloud image, and mark it as a two-dimensional point cloud image.

[0026] Step S2, obtaining a dynamic point cloud area based on the continuous two-dimensional point cloud image; Step S2 includes the following sub-steps: Step S201: Mark the two-dimensional point cloud image acquired in real time as the first point cloud image, and mark the two-dimensional point cloud image acquired in the previous frame as the second point cloud image; Step S202: Establish a plane rectangular coordinate system, marked as a cloud point coordinate system, place the first point cloud image in the first quadrant, and place the second point cloud image at the same position as the first point cloud image; Step S203, marking each cloud point of the first point cloud image as a first cloud point; marking each cloud point of the second point cloud image as a second cloud point; Step S204: Mark the second cloud point closest to the first cloud point as an adjacent cloud point in the cloud point coordinate system; Step S205: Obtain the distance between each first cloud point and the corresponding adjacent cloud point, and mark it as the adjacent distance.

[0027] Step S206, obtaining a second number of adjacent distances of historical flame occurrence portions, and marking them as historical flame distances; Step S207, obtaining the range of historical flame distances, and marking it as the flame movement range; Step S208, determining whether each adjacent distance is within the flame movement range, and if so, marking the corresponding first cloud point as a point to be analyzed; Step S209: obtain any point to be analyzed, mark it as the starting analysis point, establish an a×a analysis area with the starting analysis point as the center, and mark it as the starting division area; Step S210: Determine whether the starting partitioned area contains a new point to be analyzed. If not, continue searching for the next point to be analyzed as the starting analysis point. If so, establish an a×a analysis area with the new point to be analyzed as the center, mark it as the search partitioned area, and again determine whether the search partitioned area contains a new point to be analyzed. Repeat the establishment and determination of new search partitioned areas until the new search partitioned area contains no new point to be analyzed. Mark the continuously searched points to be analyzed as a coherent group of analysis points, and use the coherent group of analysis points from each search as a set of analysis data. In step S211, for a set of analysis data, a coherent group of analysis points with a minimum value of the horizontal coordinate and a maximum value of the horizontal coordinate are obtained, and are marked as the first analysis point and the second analysis point, respectively. A coherent group of analysis points with a minimum value of the vertical coordinate and a maximum value of the vertical coordinate are obtained, and are marked as the third analysis point and the fourth analysis point, respectively. Then, a line segment passing through the first analysis point and parallel to the Y-axis, a line segment passing through the second analysis point and parallel to the Y-axis, a line segment passing through the third analysis point and parallel to the X-axis, and a line segment passing through the fourth analysis point and parallel to the X-axis are obtained, and are marked as the first line segment, the second line segment, the third line segment, and the fourth line segment, respectively. The rectangular area formed by the first line segment, the second line segment, the third line segment, and the fourth line segment is marked as the dynamic point cloud area.

[0028] Step S3: Acquire an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image and mark it as a detection image.

[0029] Step S4, obtaining a first number of historical flame graphs, and gray-scaling the historical flame graphs to obtain historical flame grayscale graphs; Step S4 includes the following sub-steps: Step S401, obtaining the RGB value of each pixel in the historical flame graph and marking it as the historical flame RGB value; Step S402: using a grayscale conversion formula to convert all historical flame RGB values ​​in the historical flame graph into grayscale values ​​to obtain a historical flame grayscale graph.

[0030] Step S5, obtaining a flame grayscale threshold range based on the historical flame grayscale image; Step S5 includes the following sub-steps: Step S501, marking the grayscale value of the pixel in the historical flame grayscale image as the flame grayscale value; Step S502: Divide the range of all flame grayscale values ​​into b equal ranges, marked as flame grayscale division ranges; Step S503, counting the frequency of each flame grayscale division range, and marking it as the divided flame grayscale frequency; Step S504: draw a histogram with the flame grayscale value as the X-axis, the flame grayscale frequency as the Y-axis, and the flame grayscale division range as the histogram interval, and mark it as the flame grayscale histogram; Step S505, obtaining the sum of the divided grayscale frequencies, and marking it as the historical total flame frequency; Step S506 , calculating the first frequency threshold as: C1=0.1×D1 / b; where C1 is the first frequency threshold, and D1 is the historical total flame frequency; Step S507: Starting from the leftmost divided flame grayscale frequency in the flame grayscale histogram and proceeding to the right, determine whether the divided flame grayscale frequency is less than a first frequency threshold. If so, delete the portion of the flame grayscale histogram corresponding to the divided flame grayscale frequency. This process stops when it is not, and obtains the minimum value of the horizontal coordinate of the flame grayscale histogram after deletion, marking it as the screening minimum value. Step S508: Starting from the rightmost divided flame grayscale frequency in the flame grayscale histogram and moving leftward, determine whether the divided flame grayscale frequency is less than a first frequency threshold. If so, delete the portion of the flame grayscale histogram corresponding to the divided flame grayscale frequency. This process stops when it is not, and obtains the maximum value of the horizontal coordinate of the flame grayscale histogram after deletion, marking it as the screened maximum value. Step S509 : Mark the range from the minimum screening value to the maximum screening value as the flame grayscale threshold range.

[0031] Step S6: judging whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection map; if so, issuing a flame generation signal. Step S6 includes the following sub-steps: Step S601: grayscale the detection image to obtain a detection grayscale image; obtain the grayscale value of each pixel in the detection grayscale image and mark it as the detection grayscale value; Step S602, calculating the mean of all detected grayscale values, and marking it as the detected grayscale mean; Step S603: determine whether the detected grayscale mean is within the flame grayscale threshold range. If so, the dynamic point cloud area is determined to be an abnormal area, and a flame generation signal is issued.

[0032] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media containing computer-usable program code. The storage medium may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0033] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

Claims

1. A laser radar smoke and fire monitoring visualization analysis method for forest protection, characterized in that: The steps include: Continuously acquire forest point cloud images through laser radar, obtain a two-dimensional projection image of the forest point cloud image, and mark it as a two-dimensional point cloud image; Obtain dynamic point cloud areas based on continuous two-dimensional point cloud images; Obtain an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image and mark it as a detection image; Obtain a first number of historical flame graphs, and grayscale the historical flame graphs to obtain historical flame grayscale graphs; Obtaining a flame grayscale threshold range based on a historical flame grayscale image; Based on the flame grayscale threshold range and the detection map, it is determined whether the dynamic point cloud area is an abnormal area. If so, a flame generation signal is issued.

2. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 1 is characterized in that: Acquiring a dynamic point cloud region based on a continuous two-dimensional point cloud image includes the following sub-steps: Mark the two-dimensional point cloud image acquired in real time as the first point cloud image, and mark the two-dimensional point cloud image acquired in the previous frame as the second point cloud image; Establish a plane rectangular coordinate system, marked as the cloud point coordinate system, place the first point cloud image in the first quadrant, and place the second point cloud image at the same position of the first point cloud image; Mark each cloud point in the first point cloud image as the first cloud point; Mark each cloud point of the second point cloud image as a second cloud point; In the cloud point coordinate system, mark the second cloud point closest to the first cloud point as an adjacent cloud point; Get the distance between each first cloud point and its corresponding adjacent cloud point, and mark it as the adjacent distance.

3. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 2 is characterized in that: Acquiring a dynamic point cloud region based on a continuous two-dimensional point cloud image also includes the following sub-steps: Obtaining a second number of adjacent distances of historical flame occurrence portions, and marking them as historical flame distances; Get the range of historical flame distance, marked as flame movement range; Determine whether each adjacent distance is within the flame movement range. If so, mark the corresponding first cloud point as a point to be analyzed.

4. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 3 is characterized in that: Acquiring a dynamic point cloud region based on a continuous two-dimensional point cloud image also includes the following sub-steps: Get any point to be analyzed, mark it as the starting analysis point, establish an a×a analysis area with the starting analysis point as the center, and mark it as the starting division area; Determine whether the starting partition area contains a new point to be analyzed. If not, continue to search for the next point to be analyzed as the starting analysis point. If so, establish an a×a analysis area with the new point to be analyzed as the center, mark it as the search partition area, and again determine whether the search partition area contains a new point to be analyzed. Then repeat the establishment of a new search partition area and the judgment until the new search partition area does not contain a new point to be analyzed. Mark the continuously searched points to be analyzed as a coherent group analysis point, and use the coherent group analysis points of each search as a set of analysis data.

5. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 4 is characterized in that: Acquiring a dynamic point cloud region based on a continuous two-dimensional point cloud image also includes the following sub-steps: Under a set of analysis data, a coherent group of analysis points with a minimum value of the horizontal coordinate and a maximum value of the horizontal coordinate are obtained, and are marked as the first analysis point and the second analysis point respectively. A coherent group of analysis points with a minimum value of the vertical coordinate and a maximum value of the vertical coordinate are obtained, and are marked as the third analysis point and the fourth analysis point respectively. Then, a line segment passing through the first analysis point and parallel to the Y axis, a line segment passing through the second analysis point and parallel to the Y axis, a line segment passing through the third analysis point and parallel to the X axis, and a line segment passing through the fourth analysis point and parallel to the X axis are obtained, and are marked as the first line segment, the second line segment, the third line segment, and the fourth line segment respectively. The rectangular area formed by the first line segment, the second line segment, the third line segment, and the fourth line segment is marked as the dynamic point cloud area.

6. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 5 is characterized in that: Grayscale processing of the historical flame graph to obtain the historical flame grayscale graph includes the following sub-steps: Get the RGB value of each pixel in the historical flame graph and mark it as the historical flame RGB value; The grayscale conversion formula is used to convert all historical flame RGB values ​​in the historical flame graph into grayscale values ​​to obtain the historical flame grayscale graph.

7. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 6 is characterized in that: Obtaining the flame grayscale threshold range based on the historical flame grayscale image includes the following sub-steps: Mark the grayscale value of the pixel in the historical flame grayscale image as the flame grayscale value; Divide the range of all flame grayscale values ​​into b equal ranges, marked as flame grayscale division ranges; Count the frequency of each flame grayscale division range and mark it as the divided flame grayscale frequency; A histogram is drawn with the flame grayscale value as the X-axis, the flame grayscale frequency as the Y-axis, and the flame grayscale division range as the histogram interval, which is marked as the flame grayscale histogram.

8. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 7 is characterized in that: Obtaining the flame grayscale threshold range based on the historical flame grayscale image includes the following sub-steps: Get the sum of the divided grayscale frequencies and mark it as the historical total flame frequency; The first frequency threshold is calculated as: C1=0.1×D1 / b; where C1 is the first frequency threshold and D1 is the total historical flame frequency; In the flame grayscale histogram, starting from the leftmost divided flame grayscale frequency of the flame grayscale histogram and going rightward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the minimum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening minimum value; In the flame grayscale histogram, starting from the rightmost divided flame grayscale frequency of the flame grayscale histogram and going leftward, determine whether the divided flame grayscale frequency is less than the first frequency threshold. If so, delete the part of the flame grayscale histogram corresponding to the divided flame grayscale frequency, and stop when it is not. Obtain the maximum value of the horizontal coordinate of the flame grayscale histogram after deletion, and mark it as the screening maximum value; The range from the screening minimum value to the screening maximum value is labeled as the flame grayscale threshold range.

9. The laser radar smoke and fire monitoring visualization analysis method for forest protection according to claim 8, characterized in that: Based on the flame grayscale threshold range and the detection map, it is determined whether the dynamic point cloud area is an abnormal area. If so, the flame signal is generated, including the following steps: Grayscale the detection image to obtain a detection grayscale image; obtain the grayscale value of each pixel in the detection grayscale image and mark it as the detection grayscale value; Calculate the mean of all detected grayscale values ​​and mark it as the detected grayscale mean; The detection grayscale mean is obtained to determine whether it is within the flame grayscale threshold range. If so, the dynamic point cloud area is determined to be an abnormal area, and a flame generation signal is issued.

10. A laser radar smoke and fire monitoring visualization analysis system for forest protection, used to implement the laser radar smoke and fire monitoring visualization analysis method for forest protection according to any one of claims 1 to 9, characterized in that: It includes point cloud acquisition module, area acquisition module, image acquisition module, grayscale processing module, threshold acquisition module and abnormality judgment module; The point cloud image acquisition module is used to continuously acquire a forest point cloud image through a laser radar, and obtain a two-dimensional projection image of the forest point cloud image, which is marked as a two-dimensional point cloud image; The region acquisition module is used to acquire a dynamic point cloud region based on a continuous two-dimensional point cloud image; The image acquisition module is used to acquire an image of the dynamic point cloud area in the projection direction of the two-dimensional point cloud image, which is marked as a detection image; The grayscale processing module is used to obtain a first number of historical flame graphs, and grayscale the historical flame graphs to obtain historical flame grayscale graphs; The threshold acquisition module is used to obtain the flame grayscale threshold range based on the historical flame grayscale image; The abnormality judgment module is used to judge whether the dynamic point cloud area is an abnormal area based on the flame grayscale threshold range and the detection map, and if so, send a flame generation signal.

Images