Methods, systems, devices, storage media, and electronic equipment for natural disaster monitoring

Abstract

This application relates to the field of disaster monitoring technology and provides a natural disaster monitoring method, device, storage medium, and electronic device. The natural disaster monitoring method provided in this application responds to the acquisition of a natural disaster event stream obtained by a dynamic visual sensor, and reconstructs a natural disaster image based on the natural disaster event stream; identifies the natural disaster image through edge detection algorithm and optical flow estimation algorithm to obtain the occurrence location and spread characteristics of the natural disaster; identifies the natural disaster image through spatial wavelet transform algorithm and spatial neighborhood filtering algorithm to obtain the disaster level of the natural disaster; when the disaster level meets the early warning conditions, generates early warning information based on the occurrence location, spread characteristics, disaster type, and disaster level, and outputs the early warning information.

Description

Technical Field

[0001] This application belongs to the field of disaster monitoring technology, and more specifically, relates to a natural disaster monitoring method, system, device, storage medium and electronic equipment. Background Technology

[0002] Wildfires are fires that occur in forests, grasslands, or other natural vegetation areas, often caused by factors such as cigarette butts or campfires.

[0003] Currently, the main methods for monitoring wildfires include satellite remote sensing, drone patrols, and ground monitoring. However, satellite remote sensing has limited resolution and cannot accurately locate small-scale fire sources. Drone patrols are greatly affected by weather conditions; strong winds and severe weather can affect drone flight and monitoring. Furthermore, their limited battery life necessitates frequent battery replacements or recharging.

[0004] It is evident that the data transmission and processing of the aforementioned traditional methods are limited by transmission bandwidth and power consumption, resulting in latency and other issues, making real-time monitoring impossible and leading to poor accuracy in disaster monitoring. Summary of the Invention

[0005] The purpose of this application is to provide a natural disaster monitoring method, system, device, storage medium, and electronic device, aiming to solve the technical problem in the prior art where data transmission and processing during disaster monitoring are limited by transmission bandwidth and power consumption, resulting in poor accuracy of disaster monitoring.

[0006] To achieve the above objectives, according to the first aspect of this application, a natural disaster monitoring method is provided, the method comprising:

[0007] In response to acquiring a natural disaster event stream obtained by a dynamic visual sensor, a natural disaster image is reconstructed based on the natural disaster event stream;

[0008] The natural disaster images are identified using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disasters.

[0009] The natural disaster images are identified using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster;

[0010] When the disaster level meets the early warning conditions, an early warning message is generated and output based on the location of occurrence, spread characteristics, disaster type, and disaster level.

[0011] Optionally, in one possible implementation of the first aspect, the natural disaster image is identified using an edge detection algorithm and an optical flow estimation algorithm to obtain the location and spread characteristics of the natural disaster, including:

[0012] The gradient magnitude, direction, and motion vector of each pixel in the natural disaster image are calculated using the edge detection algorithm and the optical flow estimation algorithm.

[0013] Based on the gradient magnitude, direction, and motion vector of each pixel, an oriented gradient histogram and an optical flow histogram are constructed.

[0014] The location and spread characteristics of the natural disaster are determined based on the directional gradient histogram and the optical flow histogram.

[0015] Optionally, in one possible implementation of the first aspect, determining the location and spread characteristics of the natural disaster based on the directional gradient histogram and the optical flow histogram includes:

[0016] In the case where the natural disaster is a wildfire, based on the directional gradient histogram and the optical flow histogram, at least one of the following information for the flames and smoke in the wildfire is determined: motion pattern, outline, and motion characteristics.

[0017] Based on at least one of the information about the flames and the smoke, and the natural wind force at which the wildfire occurred, the spread characteristics of the wildfire disaster are determined;

[0018] Based on the monitoring points and spatiotemporal distribution information of the dynamic visual sensor, the area where the spread characteristics occurred is located, and the location of the wildfire is obtained.

[0019] Optionally, in one possible implementation of the first aspect, the natural disaster image is identified using a spatial wavelet transform algorithm and a spatial neighborhood filtering algorithm to obtain the disaster level of the natural disaster, including:

[0020] In the case where the natural disaster is a wildfire, the fuzzy edge features of smoke and flames in the wildfire are extracted using a spatial wavelet transform algorithm.

[0021] Based on the blurred edge features of the smoke and the flames, the energy flicker features of the flame region in the natural disaster image are identified;

[0022] The spatial neighborhood filtering algorithm is used to identify the low-noise dynamic event stream corresponding to the natural disaster image. The energy flashing feature and the low-noise dynamic event stream are input into a Bayesian classifier for matching and classification to obtain the disaster level of the natural disaster.

[0023] Alternatively, in one possible implementation of the first aspect, the dynamic visual sensor acquires the natural disaster event stream in the following manner:

[0024] Light intensity is detected in multiple pixel regions within the monitoring range to obtain the light energy change value of each pixel region;

[0025] When the light energy change value of any pixel region exceeds a predetermined change threshold, a natural disaster event is output, wherein the natural disaster event includes: change time, pixel position and light intensity information;

[0026] The natural disaster event stream is generated based on the multiple natural disaster events output.

[0027] According to a second aspect of this application, a natural disaster monitoring system is provided, comprising:

[0028] Dynamic visual sensors are used to acquire data on natural disaster events.

[0029] A data processor, connected to the dynamic vision sensor, is used to reconstruct natural disaster images based on the natural disaster event stream; identify the natural disaster images using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disaster; identify the natural disaster images using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster; and, when the disaster level meets the warning conditions, generate and output warning information based on the location, spread characteristics, disaster type, and disaster level.

[0030] Optionally, in one possible implementation of the first aspect, the pixel unit hardware circuit of the dynamic vision sensor includes: a logarithmic photoreceptor, a differential amplifier circuit, and a comparator; wherein:

[0031] The logarithmic photoreceptor is used to detect light intensity in multiple pixel regions within the monitoring range and convert the light intensity signal of each pixel region into a corresponding voltage signal.

[0032] The differential amplifier circuit is used to transmit the voltage value obtained by differentially amplifying the voltage signal to the first terminal of the comparator, wherein the second terminal of the comparator is a voltage threshold.

[0033] A comparator is used to determine whether to output a natural disaster event based on the relationship between the voltage value and the voltage threshold.

[0034] The second aspect and any implementation thereof correspond to the first aspect and any implementation thereof, respectively. The technical effects of the second aspect and any implementation thereof can be found in the technical effects of the first aspect and any implementation thereof, as described above, and will not be repeated here.

[0035] According to a third aspect of this application, a natural disaster monitoring device is provided, the device comprising:

[0036] A determination unit is used to reconstruct a natural disaster image based on a natural disaster event stream acquired by a dynamic visual sensor in response to the acquisition of the natural disaster event stream.

[0037] The identification unit is used to identify the natural disaster image through edge detection algorithm and optical flow estimation algorithm to obtain the location and spread characteristics of the natural disaster; and to identify the natural disaster image through spatial wavelet transform algorithm and spatial neighborhood filtering algorithm to obtain the disaster level of the natural disaster.

[0038] The early warning unit is used to generate and output early warning information based on the location of occurrence, spread characteristics, disaster type and disaster level when the disaster level meets the early warning conditions.

[0039] Fourthly, embodiments of this application provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any one of the above.

[0040] Fifthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in any of the above.

[0041] In a sixth aspect, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in any one of the first aspects.

[0042] It is understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here.

[0043] The natural disaster monitoring system provided in this application uses a dynamic visual sensor to acquire natural disaster event streams, and a data processor to reconstruct natural disaster images based on the natural disaster event streams; it identifies natural disaster images using edge detection algorithms and optical flow estimation algorithms to obtain the location and spread characteristics of natural disasters; it identifies natural disaster images using spatial wavelet transform algorithms and spatial neighborhood filtering algorithms to obtain the disaster level of natural disasters; and when the disaster level meets the early warning conditions, it generates and outputs early warning information based on the location, spread characteristics, disaster type, and disaster level.

[0044] By working in tandem with dynamic visual sensors and a data processor, real-time monitoring, accurate identification, and timely early warning of wildfire disasters can be achieved. Specifically, the dynamic visual sensors, with their high sensitivity and rapid response, can reflect the dynamic development of disasters in real time. The data processor, based on edge detection, optical flow estimation, spatial wavelet transform, and spatial neighborhood filtering algorithms, learns the feature patterns of disaster images. The monitoring and identification process is not limited by transmission bandwidth and power consumption, enabling real-time monitoring of natural disaster phenomena and improving monitoring accuracy. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic structural diagram of a natural disaster monitoring system provided in an embodiment of this application;

[0047] Figure 2 This is a schematic flowchart illustrating an optional natural disaster monitoring method provided in an embodiment of this application;

[0048] Figure 3 This is a schematic flowchart illustrating an optional natural disaster monitoring method provided in an embodiment of this application;

[0049] Figure 4 This is a schematic flowchart illustrating an optional natural disaster monitoring method provided in an embodiment of this application;

[0050] Figure 5 This is a schematic flowchart illustrating an optional natural disaster monitoring method provided in an embodiment of this application;

[0051] Figure 6 This is a schematic diagram of the structure of a natural disaster monitoring device provided in an embodiment of this application;

[0052] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0053] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0054] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0055] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0056] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0057] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0058] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0059] Wildfires are fires that occur in forests, grasslands, or other natural vegetation areas, often caused by factors such as cigarette butts or campfires.

[0060] Currently, the main methods for monitoring wildfires include satellite remote sensing, drone patrols, and ground monitoring. Satellite remote sensing uses sensors onboard satellites to monitor the Earth's surface over a wide area. Satellite remote sensing technology can detect thermal anomalies and smoke associated with wildfires, identify high-temperature areas through infrared imaging, and detect the presence and spread of fires. Drone patrols rely on drones equipped with high-resolution cameras and thermal imagers to conduct detailed observations and monitoring of wildfire disasters at low altitudes and close range. Drones can transmit image and video data in real time, providing timely disaster information. Ground monitoring uses cameras, weather stations, and other sensors installed on the ground to monitor the environmental conditions of designated areas in real time and issue timely alerts when anomalies are detected.

[0061] However, satellite remote sensing technology has limited resolution and cannot accurately locate small-scale fire sources. Drone patrols are greatly affected by weather conditions; strong winds and severe weather can affect drone flight and monitoring. Furthermore, their limited flight time necessitates frequent battery replacements or recharging. The data transmission and processing of these traditional methods are limited by bandwidth and power consumption, resulting in latency and other issues that prevent real-time monitoring, leading to poor accuracy in identifying wildfires.

[0062] First, some terms used in the embodiments of this application will be explained to facilitate understanding by those skilled in the art.

[0063] The Sobe operator is an edge detection operator used in image processing. It detects edges by calculating the gradient of the image's grayscale values. It is a discrete difference operator consisting of two 3×3 convolution kernels, used to detect edges in the horizontal and vertical directions, respectively. By performing convolution operations on the image with these two kernels, the gradient magnitude and direction of the image can be obtained, resulting in an oriented gradient histogram to represent the local features of the image.

[0064] Optical flow refers to the change in pixel intensity over time and space, which can reflect the motion of an object.

[0065] Optical flow histogram is a method for statistical analysis of optical flow fields. It generates a histogram by calculating the distribution of optical flow vectors in different directions and amplitudes.

[0066] Wavelet transform is a mathematical tool for signal analysis that can simultaneously provide information about a signal in both the time and frequency domains. Spatial wavelet transform can be used in image processing to decompose an image into sub-images of different resolutions through multi-scale analysis, thereby capturing details and textures within the image.

[0067] Time wavelet transform is used for time series analysis. By decomposing the signal at multiple scales, it can extract instantaneous features and trend information from the signal.

[0068] A Bayesian classifier is a statistical classification method based on Bayes' theorem. Bayes' theorem describes the formula for calculating posterior probability, which can be used to statistically classify processed data.

[0069] The above is a brief introduction to the terms used in the embodiments of this application, and will not be repeated below.

[0070] According to an embodiment of this application, a natural disaster monitoring system is provided. Please refer to... Figure 1 As shown, Figure 1 A schematic structural diagram of a natural disaster monitoring system provided in this application is shown, including:

[0071] Dynamic vision sensor 101 (can be one or more, Figure 1 Taking a dynamic visual sensor as an example, it is used to collect data on natural disaster events.

[0072] The data processor 102, connected to the dynamic vision sensor 101, is used to reconstruct natural disaster images based on natural disaster event streams; identify natural disaster images using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of natural disasters; identify natural disaster images using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of natural disasters; and, when the disaster level meets the warning conditions, generate and output warning information based on the location, spread characteristics, disaster type, and disaster level.

[0073] Optionally, this application example may be applicable, but is not limited to, online monitoring of natural disasters. Taking wildfires as an example, dynamic visual sensors, due to their high sensitivity and rapid response, can reflect the dynamic development of wildfires in real time. By using dynamic visual sensors installed at monitoring points to continuously monitor the field of view, various changes occurring during a wildfire can be captured, forming a natural disaster event stream. This event stream contains rich information, such as the appearance of flames and smoke, changes in brightness, and movement.

[0074] After receiving natural disaster event streams from dynamic visual sensors, the data processor processes these event streams using specific algorithms (illustrated in the example below) to reconstruct natural disaster images. As an example, and not a limitation, the reconstruction process is similar to combining discrete event points to restore the true appearance of a wildfire, allowing for a direct observation of the wildfire disaster situation.

[0075] For example, a data processor can use edge detection algorithms (Sobe I operator) to detect the edges of objects in natural disaster images. If the identified natural disaster image is a wildfire, it can clearly outline the contours of flames and smoke. By analyzing this edge information, the data processor can determine the location of the wildfire. The data processor can use optical flow estimation algorithms to calculate the motion vector of each pixel in the natural disaster image, thereby determining the spread characteristics of the wildfire, such as its direction and speed, to predict its development trend and help relevant departments take timely and effective countermeasures. The data processor can use spatial wavelet transform algorithms to extract multi-scale features from natural disaster images. In the case of wildfire images, it can capture the features of flames and smoke of different sizes. These features can reflect the scale and severity of the wildfire. The data processor can use spatial neighborhood filtering algorithms to remove noise from natural disaster images, making the images clearer. Simultaneously, it can also extract useful features, such as the texture features of flames and smoke. By analyzing these features, the data processor can determine the severity level of the wildfire.

[0076] Subsequently, when the data processor determines that the wildfire's hazard level meets the warning criteria, it can generate early warning information based on information such as the wildfire's location, spread characteristics, hazard type, and hazard level. This information can include the wildfire's specific location, spread rate, expected impact area, and hazard level. Finally, the data processor will output these early warning messages so that relevant departments can take timely countermeasures, such as organizing personnel evacuation and dispatching fire brigades to extinguish the fire.

[0077] The natural disaster monitoring system provided in this application, through the collaborative work of dynamic visual sensors and data processors, can achieve real-time monitoring, accurate identification, and timely early warning of wildfire disasters.

[0078] In one possible implementation, the pixel unit hardware circuit of the dynamic vision sensor includes: a logarithmic photosensor, a differential amplifier circuit, and a comparator; wherein:

[0079] A logarithmic photoreceptor is used to detect light intensity in multiple pixel areas within a monitoring range, converting the light intensity signal of each pixel area into a corresponding voltage signal.

[0080] A differential amplifier circuit is used to transmit the voltage values ​​obtained by differentially amplifying the voltage signal to the first terminal of a comparator, wherein the second terminal of the comparator is the voltage threshold.

[0081] A comparator is used to determine whether to output a natural disaster event based on the relationship between the voltage value and a voltage threshold.

[0082] Optionally, in this application example, the logarithmic photoreceptor plays a crucial front-end sensing role in the entire pixel unit hardware circuit of the dynamic vision sensor, enabling detailed light intensity detection of multiple pixel areas within the monitoring range. Different pixel areas may receive light of varying intensities, whether it be weak ambient light or strong reflected light from flames or smoke. When light shines on the logarithmic photoreceptor, it converts the light intensity signal of each pixel area into a corresponding voltage signal through a specific physical or electronic process. This conversion process allows the light intensity information to be processed in the form of an electrical signal, providing the raw data source for subsequent circuits.

[0083] The differential amplifier circuit receives the voltage signals output from the logarithmic photosensor and amplifies these signals differentially. Differential amplification effectively amplifies the difference between the two input signals while suppressing common-mode signals, such as environmental noise. The amplified voltage values ​​are then transmitted to the first terminal of the comparator. This amplified voltage value more accurately reflects changes in light intensity, enhancing signal stability and discernibility, and providing a more reliable basis for subsequent comparison with a voltage threshold.

[0084] The comparator performs the decision-making function in the entire pixel unit hardware circuit. The second terminal of the comparator is a pre-set voltage threshold, which should be understood as being set based on the needs and actual conditions of wildfire disaster monitoring. When the comparator receives a voltage value from the differential amplifier circuit, it compares this voltage value with the voltage threshold. If the voltage value exceeds the threshold, the comparator determines that the current light intensity may correspond to a natural disaster event, and outputs a natural disaster event. This output can be a specific electrical or digital signal used to trigger subsequent data processing and early warning systems. Conversely, if the voltage value does not exceed the threshold, the comparator does not output a natural disaster event, indicating that the current light intensity is within the normal range and no natural disaster such as a wildfire has occurred.

[0085] This application provides an example of a natural disaster monitoring method. Please refer to... Figure 2 As shown, Figure 2 A schematic flowchart of a natural disaster monitoring method provided in this application is shown. It is provided as an example and not as a limitation. This method can be applied to a natural disaster monitoring system.

[0086] S201, in response to acquiring the natural disaster event stream obtained by the dynamic visual sensor, a natural disaster image is reconstructed based on the natural disaster event stream.

[0087] S202 identifies natural disaster images using edge detection and optical flow estimation algorithms, obtaining the location and spread characteristics of natural disasters.

[0088] S203 uses spatial wavelet transform and spatial neighborhood filtering algorithms to identify natural disaster images and obtain the disaster level of the natural disaster.

[0089] S204: When the disaster level meets the early warning conditions, generate and output early warning information based on the location of occurrence, spread characteristics, disaster type and disaster level.

[0090] In this application example, a dynamic visual sensor is used to continuously monitor the natural environment and capture various changes related to natural disasters. For example, in a wildfire disaster, the sensor will detect changes in light intensity caused by flames and smoke. These changes are recorded in the form of an event stream to obtain a natural disaster event stream.

[0091] In one example, optionally, when the data processor acquires the natural disaster event stream collected by the dynamic visual sensor, it transforms or restores the natural disaster event stream into an intuitive natural disaster image to more clearly see the specific form and extent of the disaster.

[0092] First, the data processor identifies natural disaster images using edge detection and optical flow estimation algorithms. Specifically, edge detection algorithms can be used to detect the edges of objects in the images. For example, in wildfire disasters, the outlines of flames and smoke can be clearly delineated. By analyzing this edge information, the location of the natural disaster can be determined. For instance, taking wildfire disasters as an example, the edges of flames are usually sharp, while the edges of smoke are relatively blurred. By identifying important features in the natural disaster image and reducing unnecessary information interference, the location of the wildfire can be accurately pinpointed.

[0093] Furthermore, the data processor employs an optical flow estimation algorithm to calculate the motion vector of each pixel in the image. In natural disaster monitoring, this can help understand the spread characteristics of natural disasters. For example, in wildfire disasters, the optical flow estimation algorithm can determine the direction and speed of movement of flames and smoke, thereby predicting the spread trend of wildfires.

[0094] Subsequently, the data processor can identify natural disaster images using spatial wavelet transform and spatial neighborhood filtering algorithms. Specifically, the spatial wavelet transform algorithm can extract multi-scale features from natural disaster images. For natural disaster images, it can capture disaster features of different sizes, such as small-scale flames and large-scale smoke, to reflect the scale and severity of the natural disaster, thereby helping to determine the disaster level.

[0095] It should be noted that the spatial wavelet transform algorithm can analyze images at different scales, providing more comprehensive disaster information. The spatial neighborhood filtering algorithm can remove noise from images, making them clearer. In natural disaster monitoring, noise may originate from environmental factors, the sensors themselves, etc. Spatial neighborhood filtering can improve image quality and make the extracted disaster features more accurate. Simultaneously, spatial neighborhood filtering can also extract useful features, such as the texture features of flames and smoke. These features can also be used to determine the severity of natural disasters.

[0096] Furthermore, after determining the disaster level of a natural disaster through the above steps, it is determined whether the disaster level meets the early warning conditions. Early warning conditions can be set according to different types and severity of natural disasters. For example, for wildfires, an early warning message can be triggered when the fire area reaches a certain size, the spread is rapid, or the smoke concentration is too high.

[0097] If the disaster level meets the early warning criteria, an early warning message is generated based on information such as the location, spread characteristics, type, and level of the natural disaster. This information may include the specific location of the disaster, its direction and speed of spread, the expected impact area, and the disaster level. The early warning message can be in the form of text, images, or sound, so that relevant personnel can quickly and accurately understand the disaster situation.

[0098] Finally, by issuing early warning information, relevant departments and personnel can take timely countermeasures. Early warning information can be sent to command centers, fire departments, emergency rescue teams, etc., and can also be disseminated to the public through radio, television, and mobile phone text messages to raise public awareness and response capabilities.

[0099] The natural disaster monitoring method provided in this application, by comprehensively utilizing multiple algorithms and technologies, can achieve real-time monitoring, accurate identification, and timely early warning of natural disasters, providing strong support for protecting people's lives and property and the ecological environment.

[0100] In one possible implementation, the dynamic visual sensor acquires the natural disaster event stream in the following way:

[0101] Light intensity is detected in multiple pixel areas within the monitoring range to obtain the light energy change value of each pixel area;

[0102] When the light energy change value of any pixel region exceeds a predetermined change threshold, a natural disaster event is output. The natural disaster event includes: change time, pixel location, and light intensity information.

[0103] Generate a natural disaster event stream based on the output of multiple natural disaster events.

[0104] Dynamic vision sensors can simultaneously monitor lighting conditions at multiple different locations. First, they detect light intensity in multiple pixel areas within the monitoring range. Each pixel area can independently sense the light intensity and convert it into a corresponding electrical or digital signal.

[0105] By continuously monitoring light intensity, dynamic vision sensors can calculate the change in light energy for each pixel region, reflecting how the light intensity of that pixel region changes over a period of time. For example, if a pixel region changes from a darker state to a brighter state, its change in light energy will increase accordingly.

[0106] The dynamic vision sensor compares the change in light energy in each pixel area with a predetermined threshold. If the change in light energy in a pixel area exceeds the predetermined threshold, it indicates that a change related to a natural disaster may have occurred in that area. This threshold is set based on the needs and actual conditions of natural disaster monitoring.

[0107] When the light energy change in any pixel region exceeds a predetermined threshold, the dynamic vision sensor outputs a natural disaster event. This event contains important information, such as the time of change, pixel location, and light intensity. Over time, the dynamic vision sensor continuously detects light energy changes in pixel regions and outputs multiple natural disaster events. These events, combined in chronological order, form a natural disaster event stream.

[0108] It should be understood that natural disaster event streams contain a wealth of information that can provide a foundation for subsequent data analysis and processing. By analyzing event streams, it is possible to reconstruct natural disaster images, identify the location and spread characteristics of natural disasters, and determine disaster severity levels.

[0109] In summary, dynamic vision sensors detect light intensity in multiple pixel regions, output natural disaster events exceeding a change threshold, and generate natural disaster event streams, providing an important data source for natural disaster monitoring.

[0110] In one possible implementation, please refer to Figure 3 As shown, Figure 3 This paper illustrates a schematic flowchart of a natural disaster monitoring method provided in this application. The method identifies natural disaster images using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disaster, including:

[0111] S301 calculates the gradient magnitude, direction, and motion vector of each pixel in a natural disaster image using edge detection and optical flow estimation algorithms.

[0112] S302, based on the gradient magnitude, direction and motion vector of each pixel, constructs the directional gradient histogram and optical flow histogram.

[0113] S303. Based on the directional gradient histogram and optical flow histogram, determine the location and spread characteristics of natural disasters.

[0114] In this example, the data processor determines the location of edges by analyzing the brightness changes of pixels in an image using an edge detection algorithm. For each pixel, the edge detection algorithm calculates its gradient magnitude and direction in the horizontal and vertical directions. The gradient magnitude represents the drasticness of the brightness change of the pixel, while the gradient direction indicates the direction of the brightness change.

[0115] For example, in images of wildfires, the edges of flames and smoke often exhibit large gradient magnitudes because these areas show significant brightness variations. By calculating the gradient magnitude and direction of each pixel, the initial location of the natural disaster's edge can be determined.

[0116] In this example, the data processor calculates the motion vector of each pixel in a natural disaster image using an optical flow estimation algorithm. Optical flow reflects the motion of objects in the image; by comparing the positional changes of pixels in two or more consecutive frames, the motion direction and velocity of each pixel can be determined.

[0117] In wildfire disasters, flames and smoke typically exhibit different movement patterns. Flames may flicker or dance rapidly, while smoke may drift with the wind. By calculating the motion vector of each pixel using optical flow estimation algorithms, we can gain a deeper understanding of the dynamic changes in natural disasters.

[0118] Next, the data processor constructs a Histogram of Oriented Gradients (HOG) based on the gradient magnitude and direction of each pixel. HOG divides the image into multiple small cells, and for each cell, it statistically analyzes the gradient direction distribution of the pixels within that cell. By combining the histograms of gradient orientations of these cells, the directional gradient features of the entire image can be obtained. It should be understood that in natural disaster monitoring, HOG can capture the shape and texture features of natural disasters. For example, wildfire flames typically have specific shapes and directions, and HOG can better describe these features.

[0119] Similarly, the data processor constructs an optical flow histogram based on the motion vector of each pixel. The optical flow histogram divides the image into multiple regions, and for each region, it statistically analyzes the distribution of motion directions of the pixels within it. By combining the optical flow direction histograms of these regions, the optical flow characteristics of the entire image can be obtained. In natural disaster monitoring, optical flow histograms can reflect the movement patterns of natural disasters. For example, the spread direction and speed of wildfires can be described using optical flow histograms.

[0120] Finally, the data processor can determine the location of natural disasters by analyzing the histogram of directional gradients and the optical flow histogram. For example, in wildfires, the edges of flames and smoke typically show distinct features in the histogram of directional gradients, while the optical flow histogram reflects the direction of movement of flames and smoke. By combining information from these two histograms, the location of wildfires can be accurately determined.

[0121] Meanwhile, histograms of directional gradients and optical flow histograms can also provide information on the spread characteristics of natural disasters. For example, by observing the distribution of motion vectors in the optical flow histogram, the direction and speed of wildfire spread can be determined. Histograms of directional gradients can reflect the diffusion of flames and smoke. By analyzing these characteristics, we can better predict the development trend of natural disasters and provide a basis for taking timely and effective response measures.

[0122] In summary, by using edge detection algorithms and optical flow estimation algorithms to calculate the gradient magnitude, direction, and motion vector of pixels, we can construct directional gradient histograms and optical flow histograms. Based on these histograms, we can determine the location and spread characteristics of natural disasters, thereby effectively monitoring natural disasters.

[0123] In one possible implementation, please refer to Figure 4 As shown, Figure 4 A schematic flowchart of a natural disaster monitoring method provided in this application is shown. Based on the directional gradient histogram and optical flow histogram, the method determines the location and spread characteristics of natural disasters, including:

[0124] S401, In the case of a wildfire disaster, based on the directional gradient histogram and the optical flow histogram, determine at least one of the following information for flames and smoke in the wildfire disaster: motion pattern, outline, and motion characteristics.

[0125] S402, Determine the spread characteristics of a wildfire disaster based on at least one of the information on flames and smoke, as well as the natural wind force at which the wildfire disaster occurs.

[0126] S403, based on the monitoring points and spatiotemporal distribution information of the dynamic visual sensor, locates the area where the spread characteristics occur, and obtains the location of the wildfire.

[0127] In wildfire disasters, histograms of oriented gradients (OARs) can provide information about the shape and texture of flames and smoke. For example, the edges of flames are usually sharp, and their OARs will show a specific gradient direction distribution. Smoke, on the other hand, is relatively diffuse, and its OARs will also exhibit different characteristics.

[0128] It should be understood that optical flow histograms can reflect the motion patterns of flames and smoke. Flames typically flicker or dance rapidly, resulting in a more concentrated distribution of motion vectors in their optical flow histograms with significant directional changes. Smoke, on the other hand, drifts with the wind, moving relatively slowly and in a more regular pattern, leading to a different distribution of motion vectors in its optical flow histograms.

[0129] By analyzing these two histograms, information such as the movement patterns, outlines, and motion characteristics of the flames and smoke can be determined. The movement patterns help determine the development stage and size of the wildfire, the outline information helps determine the extent of the wildfire, and the motion characteristics provide a basis for predicting the direction of wildfire spread.

[0130] Furthermore, since natural wind is a crucial factor influencing the spread of wildfires, the magnitude and direction of wind must be considered when determining the spread characteristics of a wildfire disaster. Consequently, by combining information such as the movement patterns, outlines, and characteristics of flames and smoke with the conditions of natural wind, the spread direction, speed, and extent of the wildfire can be determined more accurately. For example, if the wind is strong and its direction is the same as the direction of flame movement, the wildfire may spread faster. Conversely, if the wind is weak or its direction is opposite to the direction of flame movement, the wildfire may spread slower.

[0131] Furthermore, since the monitoring point locations and spatiotemporal distribution information of dynamic visual sensors can provide important information for locating wildfires, in this application example, the approximate location of a wildfire can be determined by analyzing image data collected by sensors at different times and locations. Combining the motion and spread characteristics of flames and smoke further narrows down the range of the wildfire's location, improving the accuracy of the location. For example, if multiple sensors detect flame and smoke characteristics in the same area, and these characteristics conform to the wildfire's spread pattern, that area can be identified as the location of the wildfire.

[0132] In summary, by analyzing the directional gradient histogram and optical flow histogram to determine the information of flames and smoke, combining it with natural wind force to determine the spread characteristics of wildfire disasters, and then using the monitoring points and spatiotemporal distribution information of dynamic visual sensors to locate the location of wildfires, we can effectively monitor wildfire disasters.

[0133] In one possible implementation, please refer to Figure 5 As shown, Figure 5 This application provides a schematic flowchart of a natural disaster monitoring method that identifies natural disaster images using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level, including:

[0134] S501, in the case of a wildfire disaster, uses a spatial wavelet transform algorithm to extract the fuzzy edge features of smoke and flames in the wildfire disaster.

[0135] S502, based on the blurred edge features of smoke and flames, identifies the energy flickering features of flame areas in natural disaster images.

[0136] S503 uses a spatial neighborhood filtering algorithm to identify the low-noise dynamic event stream corresponding to the natural disaster image. It inputs the energy flashing features and the low-noise dynamic event stream into a Bayesian classifier for matching and classification to obtain the disaster level of the natural disaster.

[0137] Because smoke and flames in wildfires possess different characteristics—for example, smoke is typically diffuse and irregular in shape with relatively blurred edges, while flames, although their edges may be sharp in some areas, also exhibit a degree of blurriness overall—spatial wavelet transform algorithms can be used to perform multi-scale analysis on wildfire images, thereby accurately extracting the blurred edge features of both smoke and flames.

[0138] Based on the extracted blurred edge features of smoke and flame, energy flashing characteristics are obtained by analyzing the flame region. Flames typically have a dynamic, pulsating shape, and their energy flashing characteristics can reflect the activity and intensity of the flame. For example, the stronger the energy flashing of the flame, the larger the fire is usually, and the higher the disaster level may be.

[0139] Subsequently, spatial neighborhood filtering algorithms can be used to process natural disaster images and identify low-noise dynamic event streams. This algorithm can effectively remove noise from images, improve image quality, and retain useful dynamic information.

[0140] In wildfire disaster monitoring, low-noise dynamic event streams can provide information about the movement and changes of flames and smoke. Energy flicker features are input into a Bayesian classifier for matching and classification. A Bayesian classifier is a probability-based classification method that can classify new data based on known features and category information.

[0141] In wildfire disaster monitoring, Bayesian classifiers can classify wildfire disasters into different levels based on the characteristics of flames and smoke, as well as historical data. For example, wildfire disasters can be classified into different levels such as small fires, medium fires, and large fires.

[0142] By combining the spatial wavelet transform algorithm, the spatial neighborhood filtering algorithm, and the classification function of the Bayesian classifier through the above embodiments of this application, the disaster level of wildfire disasters can be effectively determined, providing an important basis for the monitoring and response to wildfire disasters.

[0143] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0144] Corresponding to the natural disaster monitoring method in the above embodiments, Figure 6 This is a schematic diagram of the structure of a natural disaster monitoring device provided in an embodiment of this application. The device can be implemented as part or all of a computer device using software, hardware, or a combination of both. This computer device can be... Figure 7 The electronic device shown.

[0145] Reference Figure 6 The natural disaster monitoring device includes:

[0146] The determining unit 601 is used to reconstruct a natural disaster image based on the natural disaster event stream obtained by the dynamic visual sensor in response to acquiring the natural disaster event stream.

[0147] The identification unit 602 is used to identify natural disaster images through edge detection algorithms and optical flow estimation algorithms to obtain the location and spread characteristics of natural disasters; and to identify natural disaster images through spatial wavelet transform algorithms and spatial neighborhood filtering algorithms to obtain the disaster level of natural disasters.

[0148] The early warning unit 603 is used to generate and output early warning information based on the location of occurrence, spread characteristics, disaster type and disaster level when the disaster level meets the early warning conditions.

[0149] It should be noted that the natural disaster monitoring device provided in the above embodiments is only an example of the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0150] The functional units and modules in the above embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of the embodiments of this application.

[0151] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0152] This application also provides an electronic device, which includes one or more processors and a memory;

[0153] The memory is coupled to one or more processors. The memory is used to store computer program code, which includes computer instructions. One or more processors invoke the computer instructions to cause the electronic device to perform the natural disaster monitoring method described above.

[0154] Electronic devices can be mobile phones, smart screens, tablets, wearable electronic devices, in-vehicle electronic devices, augmented reality (AR) devices, virtual reality (VR) devices, laptops, ultra-mobile personal computers (UMPCs), netbooks, personal digital assistants (PDAs), projectors, or communication devices such as servers, storage devices, and base stations, or smart cars, etc. This application does not limit the specific type of electronic device.

[0155] This application also provides a computer-readable storage medium storing computer instructions; when the computer-readable storage medium is used on an electronic device, the electronic device performs the aforementioned natural disaster monitoring method.

[0156] The aforementioned computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the aforementioned computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The aforementioned computer-readable storage medium can be any available medium that a computer can access or can include one or more data storage devices such as servers or data centers that can be integrated with media. The aforementioned available medium can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media, or semiconductor media (e.g., solid-state drives (SSDs)).

[0157] This application also provides a computer program product containing computer instructions, which, when run on an electronic device, enables the electronic device to execute the aforementioned natural disaster monitoring method.

[0158] The computer storage medium and computer program product provided in the embodiments of this application are used to execute the methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects corresponding to the methods provided above, and will not be repeated here.

[0159] In the above embodiments, implementation can also be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented, in whole or in part, as a computer program product. The aforementioned computer program product includes one or more computer instructions. When these computer instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The aforementioned computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The aforementioned computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the aforementioned computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line, DSL) or wireless (e.g., infrared, wireless, microwave, etc.) means. The aforementioned computer-readable storage medium can be any available medium accessible to a computer, or a data storage device such as a server or data center that integrates one or more available media. The aforementioned available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., Digital Versatile Discs (DVDs)), or semiconductor media (e.g., Solid State Drives (SSDs)).

[0160] Figure 7 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 700 can be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, etc.

[0161] The memory 701 can be used to store computer software programs 702 and modules. The processor 703 executes various functions and data processing of the mobile phone by running the software programs and modules stored in the memory 701. The memory 701 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory 701 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0162] The processor 703 may include one or more processors such as a central processing unit (CPU), an application processor (AP), and a baseband processor. The processor can serve as the nerve center and command center of the wireless router. The processor 703 can generate operation control signals based on instruction opcodes and timing signals to control instruction fetching and execution. The memory 701 can be used to store executable program code, including instructions. The processor 703 executes various functional applications and data processing of the network device by running the instructions stored in the memory. The memory 701 may include a program storage area and a data storage area, such as storing data for audio signals to be played. For example, the memory may be Double Data Rate Synchronous Dynamic Random Access Memory (DDR) or Flash memory.

[0163] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0164] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments claimed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0165] In the embodiments provided in this application, it should be understood that the disclosed apparatus / network devices and methods can be implemented in other ways. For example, the apparatus / network device embodiments described above are merely illustrative. For instance, the division of modules or units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0166] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0167] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for monitoring natural disasters, characterized in that, include: In response to acquiring a natural disaster event stream obtained by a dynamic visual sensor, a natural disaster image is reconstructed based on the natural disaster event stream; The natural disaster images are identified using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disasters. The natural disaster images are identified using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster; When the disaster level meets the early warning conditions, an early warning message is generated and output based on the location of occurrence, spread characteristics, disaster type, and disaster level.

2. The method according to claim 1, characterized in that, The natural disaster images are identified using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disasters, including: The gradient magnitude, direction, and motion vector of each pixel in the natural disaster image are calculated using the edge detection algorithm and the optical flow estimation algorithm. Based on the gradient magnitude, direction, and motion vector of each pixel, an oriented gradient histogram and an optical flow histogram are constructed. The location and spread characteristics of the natural disaster are determined based on the directional gradient histogram and the optical flow histogram.

3. The method according to claim 2, characterized in that, The step of determining the location and spread characteristics of the natural disaster based on the directional gradient histogram and the optical flow histogram includes: In the case where the natural disaster is a wildfire, based on the directional gradient histogram and the optical flow histogram, at least one of the following information for the flames and smoke in the wildfire is determined: motion pattern, outline, and motion characteristics. Based on at least one of the information about the flames and the smoke, and the natural wind force at which the wildfire occurred, the spread characteristics of the wildfire disaster are determined; Based on the monitoring points and spatiotemporal distribution information of the dynamic visual sensor, the area where the spread characteristics occurred is located, and the location of the wildfire is obtained.

4. The method according to claim 1, characterized in that, The natural disaster images are identified using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster, including: In the case where the natural disaster is a wildfire, the fuzzy edge features of smoke and flames in the wildfire are extracted using a spatial wavelet transform algorithm. Based on the blurred edge features of the smoke and the flames, the energy flicker features of the flame region in the natural disaster image are identified; The spatial neighborhood filtering algorithm is used to identify the low-noise dynamic event stream corresponding to the natural disaster image. The energy flashing feature and the low-noise dynamic event stream are input into a Bayesian classifier for matching and classification to obtain the disaster level of the natural disaster.

5. The method according to any one of claims 1 to 4, characterized in that, The dynamic visual sensor acquires the natural disaster event stream in the following manner: Light intensity is detected in multiple pixel regions within the monitoring range to obtain the light energy change value of each pixel region; When the light energy change value of any pixel region exceeds a predetermined change threshold, a natural disaster event is output, wherein the natural disaster event includes: change time, pixel position and light intensity information; The natural disaster event stream is generated based on the multiple natural disaster events output.

6. A natural disaster monitoring system, characterized in that, include: Dynamic visual sensors are used to acquire data on natural disaster events. A data processor, connected to the dynamic vision sensor, is used to reconstruct natural disaster images based on the natural disaster event stream; identify the natural disaster images using edge detection and optical flow estimation algorithms to obtain the location and spread characteristics of the natural disaster; and identify the natural disaster images using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster. And when the disaster level meets the warning conditions, generate warning information based on the location of occurrence, spread characteristics, disaster type and disaster level, and output the warning information.

7. The system according to claim 6, characterized in that, The pixel unit hardware circuit of the dynamic vision sensor includes: a logarithmic photoreceptor, a differential amplifier circuit, and a comparator; wherein: The logarithmic photoreceptor is used to detect light intensity in multiple pixel regions within the monitoring range and convert the light intensity signal of each pixel region into a corresponding voltage signal. The differential amplifier circuit is used to transmit the voltage value obtained by differentially amplifying the voltage signal to the first terminal of the comparator, wherein the second terminal of the comparator is a voltage threshold. A comparator is used to determine whether to output a natural disaster event based on the relationship between the voltage value and the voltage threshold.

8. A natural disaster monitoring device, characterized in that, include: A determination unit is used to reconstruct a natural disaster image based on a natural disaster event stream acquired by a dynamic visual sensor in response to the acquisition of the natural disaster event stream. The identification unit is used to identify the natural disaster image through edge detection algorithm and optical flow estimation algorithm to obtain the location and spread characteristics of the natural disaster; The natural disaster images are identified using spatial wavelet transform and spatial neighborhood filtering algorithms to obtain the disaster level of the natural disaster; The early warning unit is used to generate and output early warning information based on the location of occurrence, spread characteristics, disaster type and disaster level when the disaster level meets the early warning conditions.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images