CN117710830A - Method, device and equipment for acquiring fire image based on multi-source fire point data - Google Patents
Method, device and equipment for acquiring fire image based on multi-source fire point data Download PDFInfo
- Publication number
- CN117710830A CN117710830A CN202311781527.8A CN202311781527A CN117710830A CN 117710830 A CN117710830 A CN 117710830A CN 202311781527 A CN202311781527 A CN 202311781527A CN 117710830 A CN117710830 A CN 117710830A
- Authority
- CN
- China
- Prior art keywords
- fire
- satellite
- fire point
- point
- space
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000003068 static effect Effects 0.000 claims description 36
- 238000001914 filtration Methods 0.000 claims description 12
- 230000011218 segmentation Effects 0.000 claims description 11
- 230000005484 gravity Effects 0.000 claims description 7
- 238000012545 processing Methods 0.000 claims description 6
- 238000005520 cutting process Methods 0.000 claims description 5
- 238000012795 verification Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000003384 imaging method Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 241000208818 Helianthus Species 0.000 description 2
- 235000003222 Helianthus annuus Nutrition 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011084 recovery Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000000779 smoke Substances 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The application discloses a method, a device and equipment for acquiring fire images based on multi-source fire points, which relate to the technical field of satellite remote sensing, and the method comprises the following steps: the method comprises the steps of acquiring multi-source satellite fire point data, wherein the multi-source satellite fire point data comprises a plurality of fire point records comprising fire point longitude and latitude coordinates and fire point recording time, generating fire point pixel patches corresponding to each fire point record based on the fire point longitude and latitude coordinates of each fire point record, acquiring a space buffer zone by taking the union of the fire point pixel patches with space intersection among the plurality of fire point pixel patches, determining a fire occurrence time range based on the fire point recording time corresponding to the fire point pixel patches in the space buffer zone, forming a space-time buffer zone by the fire occurrence time range and the space buffer zone, and acquiring a target satellite fire image based on retrieval through the space-time buffer zone. The accuracy and efficiency of determining the real fire disaster are improved through mutual verification of fire point records acquired by different satellites, and satellite images of the real fire disaster are obtained.
Description
Technical Field
The application relates to the technical field of satellite remote sensing, in particular to a method, a device and equipment for acquiring fire images based on multi-source fire data.
Background
Satellite remote sensing is an important means for monitoring and early warning natural fires in forests, grasslands and the like, wherein fire identification is a key technology of the satellite remote sensing technology in natural fire monitoring. With the continuous development of artificial intelligence technology, more and more intelligent models and algorithms are applied to the fire identification of satellite images.
The artificial intelligence technology is utilized to process massive satellite images and identify fires, a large number of real fire images are needed to be used as samples for training, and the fire satellite samples are manufactured by screening and cutting out images corresponding to the truly generated fires from the massive satellite images. However, using known fire information, a great deal of manual operations are required to label the fire satellite images, and meanwhile, professional knowledge and experience are combined to judge and screen, so that the data quality is difficult to guarantee.
At present, the first method is to search natural fire information such as forests, grasslands and the like through a network, verify and confirm the authenticity of the fire and the information such as the scale, the occurrence time and the fire position of the fire from news reports or government information disclosure websites by utilizing information clues obtained through searching, and further search and acquire satellite images and mark the fire position according to corresponding fire information. The method can acquire the determined fire satellite image sample, but is completely dependent on manual operation, so that the efficiency is low.
The second method is to use satellite fire point data products generated by the traditional fire recognition algorithm as fire information, search corresponding satellite images, cut and mark fire images, and the traditional fire recognition algorithm is mostly realized by infrared or thermal infrared channels and threshold segmentation. At present, fire point data products generated by the algorithm are issued by satellites including wind cloud FY, a medium resolution imaging spectrometer MODIS, sunflower Himaware, a terrestrial satellite Landsat and the like. For fire point data products generated by the algorithm, the fire point data products are affected by factors such as imaging time, observation conditions, satellite resolution, a fire point extraction algorithm and the like, the corresponding fire scale, fire property, fire range and the like are inconsistent, a certain false alarm rate exists, meanwhile, a large number of fire points have no clear optical characteristics such as smoke and the like on satellite images, and the highlight points in the infrared wave band are not fire points, so that the authenticity of fire at the fire point can not be confirmed in practical application, and the obtained satellite images are difficult to guarantee the authenticity of the fire.
Therefore, how to efficiently and accurately acquire satellite images of a real fire disaster is a technical problem to be solved.
Disclosure of Invention
In view of the foregoing, a main object of the present application is to provide a method, apparatus and device for acquiring fire images based on multi-source fire data, so as to efficiently and accurately acquire satellite images of a real fire.
The first aspect of the application provides a method for acquiring fire images based on multi-source fire points, which comprises the following steps:
acquiring multi-source satellite fire point data, wherein the multi-source satellite fire point data comprises a plurality of fire point records, and the fire point records comprise: the longitude and latitude coordinates of the fire points and the time for recording the fire points are recorded as a static orbit satellite and a polar orbit satellite;
in the multi-source satellite fire point data, generating pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point recorded by each fire point respectively so as to obtain a plurality of fire point pixel patches;
obtaining a space buffer area by taking the union of the fire point pixel patches with the space intersection among the plurality of fire point pixel patches;
determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone;
And searching through the space-time buffer area to obtain a fire image of the target satellite, wherein the fire image of the target satellite is an image acquired by a static orbit satellite or a polar orbit satellite.
In some implementations of the first aspect of the present application, the method further includes:
cutting an image with preset resolution from the fire image of the target satellite by taking the center of gravity of the circumscribed rectangle of the space buffer as a center point to obtain a fire sample image;
taking a preset thermal emission band of the fire sample image, performing self-adaptive threshold segmentation processing on the fire sample image, marking pixels larger than a preset threshold in the fire sample image as fire pixels, and recording coordinates of the fire pixels;
determining a satellite for acquiring a fire image of a target satellite as the target satellite, and determining fire recording time corresponding to a fire pixel patch corresponding to the target satellite in a space buffer zone, wherein the target satellite is a static orbit satellite or a polar orbit satellite;
and generating a fire satellite image sample based on the fire sample image, the circumscribed rectangle of the space buffer, the coordinates of the fire pixel and the fire recording time corresponding to the fire pixel patch corresponding to the target satellite.
In some implementations of the first aspect of the present application, the method further includes:
In the multi-source satellite fire data, fire records acquired by stationary orbiting satellites are filtered based on fire records acquired by polar orbiting satellites.
In some implementations of the first aspect of the present application, filtering the fire record acquired by the stationary orbiting satellite based on the fire record acquired by the polar orbiting satellite includes:
taking the fire point recording time of each fire point record acquired by the polar orbit satellite as the interval center, and determining a plurality of fire point selection intervals;
and filtering out the fire record of which the fire record time does not fall into any fire record selection interval from the fire records acquired by the static orbit satellite.
In some implementations of the first aspect of the present application, generating a fire pixel patch corresponding to each fire record based on a fire longitude and latitude coordinate of each fire record, includes:
and for each fire point record, taking the longitude and latitude coordinates of the fire point as the coordinates of the center point of the pixel, and generating a fire point pixel patch based on the coordinates of the center point of the pixel and the resolution of the satellite for acquiring the fire point record.
In some implementations of the first aspect of the present application, determining a fire occurrence time range based on a fire recording time corresponding to each fire pixel patch in the spatial buffer includes:
Based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, taking a section formed by the earliest fire point recording time and the latest fire point recording time as a fire occurrence time range.
In some implementations of the first aspect of the present application, retrieving the target satellite fire image through a space-time buffer includes:
and searching on the geographic information searching platform according to the fire occurrence time range and the circumscribed rectangle of the space buffer area to obtain a target satellite image.
In some implementations of the first aspect of the present application, the fire satellite image sample is in HDF format.
The second aspect of the present application provides a device for acquiring fire images based on multi-source fire data, the device comprising:
the data acquisition module is used for acquiring multi-source satellite fire point data, the multi-source satellite fire point data comprises a plurality of fire point records, and the fire point records comprise: the longitude and latitude coordinates of the fire points and the time for recording the fire points are recorded as a static orbit satellite and a polar orbit satellite;
the pixel patch generation module is used for generating the corresponding fire point pixel patches of each fire point record based on the longitude and latitude coordinates of the fire point of each fire point record in the multi-source satellite fire point data so as to obtain a plurality of fire point pixel patches;
The space buffer area determining module is used for obtaining a space buffer area from a plurality of fire point pixel patches by taking the union of the fire point pixel patches with the space intersection;
the space-time buffer zone determining module is used for determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone;
the image retrieval module is used for retrieving a fire image of a target satellite through the space-time buffer zone, wherein the fire image of the target satellite is an image acquired by a static orbit satellite or a polar orbit satellite.
A third aspect of the present application provides a device for acquiring fire images based on multi-source fire data, comprising: the system comprises a memory and a processor, wherein the processor is used for executing a program stored in the memory and running the method for acquiring the fire image based on the multi-source fire point data provided by the first aspect of the application.
The technical scheme provided by the application has the following beneficial effects:
in the technical scheme provided by the application, firstly, multi-source satellite fire point data are acquired, the multi-source satellite fire point data are composed of a plurality of fire point records at least comprising fire point longitude and latitude coordinates and fire point recording time, and the fire point records in the multi-source satellite fire point data are derived from a static orbit satellite and a polar orbit satellite. And then, in the multi-source satellite fire point data, generating fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point of each fire point record respectively so as to obtain a plurality of fire point pixel patches. And in a plurality of fire point pixel patches, the spatial buffer area is obtained by taking the union of the fire point pixel patches with the spatial intersection, so that the mutual verification of the fire point records acquired by the static orbit satellite and the polar orbit satellite is realized, and the spatial buffer area obtained by taking the union can reflect the spatial range of the real fire disaster. And then determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone, wherein the space-time buffer zone reflects the occurrence time range and the space range of the real fire. Finally, based on the target satellite fire image obtained through the space-time buffer area search, the obtained satellite image is the satellite image of the real fire. Therefore, the accuracy of determining the real fire disaster is improved while the automatic determination of the real fire disaster is realized based on the mutual verification of the fire point records acquired by the satellites in different orbits, and further the satellite images of the real fire disaster are obtained with both efficiency and precision.
Drawings
Fig. 1 is a flow chart of a method for acquiring fire images based on multi-source fire data according to an embodiment of the present application;
FIG. 2 is a flowchart of another method for acquiring fire images based on multi-source fire data according to an embodiment of the present disclosure;
FIG. 3 is a schematic diagram of a spatial buffer according to an embodiment of the present disclosure;
FIG. 4 is a flowchart of another method for acquiring fire images based on multi-source fire data according to an embodiment of the present disclosure;
FIG. 5 is a schematic diagram of a clipping target satellite fire image according to an embodiment of the present disclosure;
FIG. 6 is a schematic diagram of a fire sample image according to an embodiment of the present disclosure;
FIG. 7 is a schematic diagram of adaptive thresholding of fire sample images according to an embodiment of the present disclosure;
fig. 8 is a schematic flow chart of generating a fire satellite image sample according to an embodiment of the present application;
fig. 9 is a schematic structural diagram of a device for acquiring fire images based on multi-source fire data according to an embodiment of the present application;
fig. 10 is a schematic structural diagram of a device for acquiring fire images based on multi-source fire data according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
The terms "first," "second," "third," "fourth" and the like in the description and in the claims of this application and in the above-described figures, if any, are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments described herein may be implemented in other sequences than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Besides the fact that the prior art in the background art cannot achieve efficient and accurate acquisition of satellite images of a real fire, the prior art has the following disadvantages:
for the first method, the number of fires that can be obtained is limited because of the large number of wild natural fires that are not reported or recorded in association; in addition, the fire information obtained by partial retrieval lacks of integrity, such as no occurrence time or exact position, and the report of the fire is only qualitatively described, so that quantitative parameters such as the scale, the grade and the like of the fire cannot be determined.
In the second method, because the occurrence of field fire is dynamically changed, when fire data products of a certain satellite are used for other satellite data retrieval, the problem that the transit time difference and the fire position are not completely overlapped exists, and even if the transit time is close, the coordinate positions of the fire points are not completely matched due to the difference of satellite resolution, and certain difficulties are brought to data retrieval and fire point marking; in addition, when the fire data acquired by the high-spatial-resolution satellite is used for retrieving the satellite image with low spatial resolution, the corresponding fire may not respond to the image with low spatial resolution, so that the fire data acquired by the high-spatial-resolution satellite cannot be used for producing the fire data set of the satellite image with low spatial resolution, and most of satellites with low spatial resolution do not have corresponding fire products, which all cause the problem that the fire data set of the satellite with low spatial resolution is difficult to produce.
In view of the problems mentioned in the background art, referring to fig. 1, an embodiment of the present application provides a method for acquiring a fire image based on multi-source fire data, which aims to efficiently and accurately acquire a satellite image of a real fire, and specifically includes the following steps:
S101: acquiring multi-source satellite fire point data, wherein the multi-source satellite fire point data comprises a plurality of fire point records, and the fire point records comprise: the longitude and latitude coordinates of the fire points and the time of recording the fire points are recorded as the collected by the static orbit satellite and the polar orbit satellite.
In the embodiment of the present application, the multi-source satellite fire data refers to fire vector data generated based on data of suspected fires monitored by a satellite remote sensing technology, and the multi-source satellite fire data may be obtained by the following manner: downloading the fire point data products of the polar orbit satellite and the static orbit satellite, acquiring the longitude and latitude, time, temperature, area and other information related to the fire point recorded by the satellite, and generating the fire point vector data with longitude and latitude coordinates and time attributes.
The fire point record refers to fire point vector data at least with longitude and latitude coordinates and time data, the longitude and latitude coordinates of a fire point refer to longitude and latitude coordinates of a suspected fire point acquired by an orbit satellite, and the fire point record time refers to time when the orbit satellite acquires fire point related information, and can also be understood as time when the longitude and latitude coordinates of the suspected fire point are acquired. It will be appreciated that an orbiting satellite may acquire multiple multi-point latitude and longitude coordinates at the same time.
The multiple fire records in the multi-source satellite fire data are acquired by at least one stationary orbiting satellite and at least one polar orbiting satellite. Wherein, the stationary orbit satellite refers to a satellite which runs in synchronous orbit around the equatorial plane of the earth, such as a wind cloud FY and a satellite like sunflower Himaware; polar orbit satellites refer to satellites that travel along the earth's dipolar directions, such as those of the medium resolution imaging spectrometer MODIS, the visible infrared imaging radiometer VIIRS, and the like.
Usually, the static orbit satellite repeatedly shoots at fixed time intervals, and has higher time resolution; the polar orbit satellite passes the border twice a day for a certain position on the ground, and the repeated observation period is longer than that of the static orbit satellite, but the polar orbit satellite has higher spatial resolution. Overall, the time resolution of the stationary orbiting satellites is better than the polar orbiting satellites, while the spatial resolution is lower than the polar orbiting satellites. Based on the above, the purpose of selecting the multi-source satellite fire point data is that the static orbit satellite can provide the fire point information with high time resolution, so that the real-time monitoring of fire is realized; the polar orbit satellite can provide fire information with high spatial resolution, so that accurate positioning of fire is realized. It will be appreciated that, because the stationary orbiting satellites have a higher time resolution, the number of fire records corresponding to the stationary orbiting satellites is greater in the multi-source satellite fire data.
In some implementations of the embodiments of the present application, before executing S102, in order to further implement mutual verification of fire records corresponding to different satellites in time, the following steps may be further executed: in the multi-source satellite fire data, fire records acquired by stationary orbiting satellites are filtered based on fire records acquired by polar orbiting satellites.
Specifically, because the spatial resolution of the polar orbit satellite is higher than that of the static orbit satellite, the fire points recorded by the polar orbit satellite are more accurate, and therefore, the fire point records collected by the static orbit satellite are further screened in a filtering mode by taking the fire point records collected by the polar orbit satellite as a reference.
In some implementations of the embodiments of the present application, taking a time of a fire point record of a polar orbit satellite as a time center point, selecting a fire point record of a stationary orbit satellite within a certain period of time, as shown in fig. 2, specifically includes the following steps:
s201: and determining a plurality of fire point selection sections by taking the fire point recording time of each fire point record acquired by the polar orbit satellite as the section center.
In the embodiment of the present application, the fire point data selection interval refers to a numerical interval determined based on a fire point recording time corresponding to a polar orbit satellite. For example, for each fire record acquired by a polar orbit satellite, a fire selection section is constituted by ten minutes before and after each fire record time.
S202: and filtering out the fire record of which the fire record time does not fall into any fire record selection interval from the fire records acquired by the static orbit satellite.
In the embodiment of the application, the fire records of the stationary orbit satellite, which are adjacent to the fire records acquired by the polar orbit satellite in the time of the fire records, can be screened out by filtering out the fire records which do not fall into any fire selecting section in each fire record acquired by the stationary orbit satellite. For example, it is ensured that only the fire record of the stationary orbiting satellite within ten minutes before and after the fire record of the polar orbiting satellite exists in the multi-source fire satellite data.
S102: and in the multi-source satellite fire point data, generating fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point recorded by each fire point respectively so as to obtain a plurality of fire point pixel patches.
In the embodiment of the application, the fire point pixel patch refers to a pixel patch representing one pixel generated based on the longitude and latitude coordinates of a fire point, and it should be noted that the longitude and latitude coordinates of the fire point may be coordinate values of one pixel 1*1, for example, a fire point position is marked in an image shot by a certain satellite, and the coordinate values of the pixel marked as the fire point position in the satellite image marked with the longitude and latitude coordinates in advance are the longitude and latitude coordinate values of the fire point. The fire spot pixel patches are used to reflect the location and extent of the fire spot on the satellite image. In some implementations of embodiments of the present application, specific ways of generating the fire pixel plaque may be found in the following steps: and for each fire point record, taking the longitude and latitude coordinates of the fire point as the coordinates of the center point of the pixel, and generating a fire point pixel patch based on the coordinates of the center point of the pixel and the resolution of the satellite for acquiring the fire point record. For example, when the spatial resolution of the satellite for acquiring the fire record a is 1000m and the longitude and latitude coordinates of the fire record a are actually 1*1 pixels, a vector patch of 1000m×1000m is generated by taking the longitude and latitude of the fire as the center, and the vector patch is used as the fire pixel patch.
S103: and obtaining a space buffer area by taking the union of the fire point pixel patches with the space intersection among the plurality of fire point pixel patches.
In the embodiment of the present application, based on the principle that a certain fire time is captured by a plurality of satellites at the same time, the fire event can be determined as a fire actually occurring, and mutual verification of each satellite fire record in space is realized through S103. The fire point pixel patches where there is a spatial intersection may refer to adjacent or overlapping fire point pixel patches, and the spatial buffer refers to a set of adjacent or overlapping fire point pixel patches. For example, three fire pixel patches A, B, C, where A is adjacent to B and B is adjacent to C, the spatial buffer is A, B, C the union of the three fire pixel patches. For another example, where A is adjacent to B and B overlaps with C, then the spatial buffer is also the union of A, B, C. Referring specifically to fig. 3, the left side of fig. 3 shows a plurality of fire pixel patches generated based on longitude and latitude coordinates of a fire record acquired by FY, MODIS, VIIRS, himawari and the corresponding satellite resolution, and by taking the union of adjacent and overlapping fire pixel patches, a space buffer area as shown on the right side of fig. 3 can be obtained. In addition, fig. 3 also illustrates that the present application generates the spot pixel patches based on the respective resolutions of the satellites, and the spot pixel patches are different in size corresponding to different satellites.
S104: and determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone.
In the embodiment of the present application, the fire occurrence time range refers to a time interval constituted based on the fire recording time of the fire pixel plaque in the space buffer, and the space-time buffer refers to a combination of the space buffer and the fire occurrence time range for reflecting the time range and the space range of the fire occurrence. Thereby providing space-time constraint for subsequent retrieval of the target satellite fire images and further improving the efficiency and accuracy of real fire determination.
In some implementations of embodiments of the present application, the time range of fire occurrence is determined specifically by: based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, taking a section formed by the earliest fire point recording time and the latest fire point recording time as a fire occurrence time range. Here, the fire occurrence time range refers to a section formed by the earliest and latest fire recording times in the space buffer, and the fire pixel patches can be time-aggregated by determining the fire occurrence time range at the earliest and latest times. For example, assuming that the earliest fire recording time in the spatial buffer is 2023, 4, 1, 8, 00 and the latest fire recording time is 2023, 4, 3, 10, 00, the fire occurrence time ranges from 2023, 4, 1, 8, 00 to 2023, 4, 3, 10, 00.
S105: and searching through the space-time buffer area to obtain a fire image of the target satellite, wherein the fire image of the target satellite is an image acquired by a static orbit satellite or a polar orbit satellite.
In the embodiments of the present application, the target satellite fire image refers to a satellite image acquired based on a stationary orbit satellite or a polar orbit satellite retrieved by a space-time buffer.
In some implementations of the embodiments of the present application, the retrieval of the target satellite fire image specifically includes the following steps: and searching on the geographic information searching platform according to the fire occurrence time range and the circumscribed rectangle of the space buffer area to obtain a target satellite image. The external rectangle of the space buffer zone refers to an external rectangle formed by a union of a plurality of fire pixel patches, and it should be noted that the existing geographic information retrieval platform has a corresponding function to realize the retrieval of satellite images through the external rectangle for reflecting the fire range and the fire occurrence time range.
In the flow shown in fig. 1, first, multi-source satellite fire data is acquired, the multi-source satellite fire data is composed of a plurality of fire records including at least a fire longitude and latitude coordinate and a fire record time, and the fire records in the multi-source satellite fire data are derived from a stationary orbit satellite and a polar orbit satellite. And then, in the multi-source satellite fire point data, generating fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point of each fire point record respectively so as to obtain a plurality of fire point pixel patches. And obtaining a space buffer area by taking the union of the fire point pixel patches with the space intersection among the plurality of fire point pixel patches. Therefore, mutual verification of fire point records acquired by the static orbit satellite and the polar orbit satellite is realized, and the space buffer area obtained by taking the union sets can reflect the space range of the real fire disaster. And then determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone, wherein the space-time buffer zone reflects the occurrence time range and the space range of the real fire. Finally, based on the target satellite fire image obtained through the space-time buffer area search, the obtained satellite image is the satellite image of the real fire. Therefore, the accuracy of determining the real fire disaster is improved while the automatic determination of the real fire disaster is realized based on the mutual verification of the fire point records acquired by the satellites in different orbits, and further the satellite images of the real fire disaster are obtained while the efficiency and the accuracy are considered.
In addition, the space-time buffer area is finally obtained based on mutual verification of the polar orbit satellite and the static orbit satellite in time and space, the fire images of the target satellite can be easily searched based on the space-time buffer area, and difficulties brought to image search due to different satellite resolutions and fire position misalignment are eliminated.
Referring to fig. 4, an embodiment of the present application provides a method for acquiring a fire image based on multi-source fire points, which aims to make a fire satellite image sample, and specifically includes the following steps:
s401: acquiring multi-source satellite fire point data, wherein the multi-source satellite fire point data comprises a plurality of fire point records, and the fire point records comprise: the longitude and latitude coordinates of the fire points and the time of recording the fire points are recorded as the collected by the static orbit satellite and the polar orbit satellite.
In the embodiment of the present application, the specific implementation manner of S401 is the same as S101, and the description thereof is omitted herein.
S402: and in the multi-source satellite fire point data, generating fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point recorded by each fire point respectively so as to obtain a plurality of fire point pixel patches.
In the embodiment of the present application, the specific implementation manner of S402 is the same as S102, and the description thereof is omitted herein.
S403: and obtaining a space buffer area by taking the union of the fire point pixel patches with the space intersection among the plurality of fire point pixel patches.
In the embodiment of the present application, the specific implementation manner of S403 is the same as S103, and the description thereof is omitted herein.
S404: and determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone.
In the embodiment of the present application, the specific implementation manner of S404 is the same as S104, and the description thereof is omitted herein.
S405: and searching through the space-time buffer area to obtain a fire image of the target satellite, wherein the fire image of the target satellite is an image acquired by a static orbit satellite or a polar orbit satellite.
In the embodiment of the present application, the specific implementation manner of S405 is the same as S105, and the description thereof is omitted herein.
S406: and cutting an image with preset resolution from the fire image of the target satellite by taking the center of gravity of the circumscribed rectangle of the space buffer as a center point so as to obtain a fire sample image.
In the embodiment of the present application, the fire sample image refers to a local image including a fire image, which is cut out from a target satellite fire image, and preferably, the cut-out image has a size of 100×100. The effect of cutting the fire image of the target satellite to obtain the fire sample image is to extract the real fire phenomenon from the fire image of the target satellite, so as to reduce irrelevant background information. Referring to fig. 5, fig. 5 shows a schematic diagram of obtaining a fire sample image by clipping a target satellite fire image based on a circumscribed rectangle of a space buffer, and in the case that the clipping image size is determined to be 100×100, the clipped fire sample image can be specifically shown in fig. 6.
The center of gravity of the circumscribed rectangle of the space buffer can be obtained by taking the average value of the coordinates of each corner point of the circumscribed rectangle of the space buffer. For example, the bounding rectangle of the spatial buffer includes the following corner coordinates: (x 1, y 1), (x 2, y 2), (xn, yn), the center of gravity is ((x1+x2+ & gt+xn)/n, (y1+y2+ & gt+yn)/n). Note that, the angular point coordinates are latitude and longitude coordinates.
S407: and taking a preset thermal emission band of the fire sample image, performing self-adaptive threshold segmentation processing on the fire sample image, marking pixels larger than a preset threshold in the fire sample image as fire pixels, and recording coordinates of the fire pixels.
In the embodiment of the application, the thermal emission band refers to an emission band of infrared radiation, and the earth surface temperature can be reflected in the satellite remote sensing technology; the self-adaptive threshold segmentation refers to an image segmentation method for automatically determining a proper threshold according to the gray level distribution of an image and segmenting the image into at least two areas; by taking a preset thermal emission wave band from the fire sample image and then performing self-adaptive threshold segmentation, the fire pixels and the non-fire pixels can be distinguished. Here, the fire point pixel refers to a pixel reflecting a real fire phenomenon in the fire sample image. The coordinates of the fire point pixels recorded here are longitude and latitude coordinates.
The effect achieved by the self-adaptive threshold segmentation can be shown as in fig. 7, and the fire point pixels and the non-fire point pixels can be visually distinguished in fig. 7, so that the fire point distribution in the fire sample image can be visually seen.
In some implementations of embodiments of the present application, a thermal emission band around 1200nm of the fire sample image may be taken, for example, a thermal emission band of 1150nm to 1250nm.
S408: determining a satellite for acquiring a fire image of a target satellite as the target satellite, and determining the fire recording time corresponding to the fire pixel patch corresponding to the target satellite in the space buffer zone, wherein the target satellite is a static orbit satellite or a polar orbit satellite.
In the embodiment of the application, the target satellite refers to a satellite for acquiring a fire image of the target satellite; determining the fire point recording time corresponding to the fire point pixel plaque corresponding to the target satellite in the space buffer zone refers to the time of the target satellite recording the fire point in the space range reflected by the space buffer zone. For example, the space buffer area respectively comprises at least one fire pixel plaque corresponding to each of the following satellites: FY, MODIS, VIIRS, himawari; if the target satellite is an FY satellite, determining the fire point recording time corresponding to the fire point pixel patch corresponding to the target satellite in the space buffer refers to determining the fire point recording time corresponding to the fire point pixel patch corresponding to the FY satellite in the space buffer. The corresponding relationship between the fire point pixel patch corresponding to the same satellite and the fire point recording time is a many-to-one relationship, the fire point pixel patch is generated based on the longitude and latitude coordinates of the fire point, and the satellite can record the longitude and latitude coordinates of a plurality of fire points at the same time.
S409: and generating a fire satellite image sample based on the fire sample image, the circumscribed rectangle of the space buffer, the coordinates of the fire pixel and the fire recording time corresponding to the fire pixel patch corresponding to the target satellite.
In embodiments of the present application, the circumscribed rectangle of the spatial buffer may be used to reflect the general area where the fire occurred; coordinates of the fire point pixels may be used to reflect a specific location of the occurrence of the fire within the range of occurrence of the fire; the time of fire record corresponding to the target satellite may be used to reflect the specific time of fire occurrence. The generated fire satellite image sample is a data set containing information such as fire sample images, fire ranges, fire point marks, fire time and the like, and a large number of fire satellite image samples can be generated by repeatedly executing the method shown in fig. 4, so that a fire satellite sample set is obtained and used for training an artificial intelligent model or algorithm, and the accuracy of the model or algorithm in the field of fire identification is improved.
Referring to fig. 8, fig. 8 shows a complete generation flow of a fire satellite image sample, where the flow can be divided into three parts, and the first part is determination of a real fire, specifically includes performing fire data preprocessing, fire pixel recovery and multi-source satellite fire pixel patch superposition analysis on multi-source satellite fire data, so as to obtain a space-time buffer zone, and the space-time buffer zone can determine occurrence of the real fire. The second part is the retrieval of the fire satellite image, and specifically comprises the retrieval of the image based on the space-time buffer zone, and then the fire satellite image is obtained by clipping, namely a fire satellite image sample. And the third part only carries out self-adaptive threshold segmentation of the thermal infrared band on the fire satellite image sample so as to mark the fire point pixels.
In some implementations of embodiments of the present application, the fire satellite image samples are in a hierarchical data format HDF.
In the flow shown in fig. 4, after the target satellite fire image is retrieved, first, the target satellite fire image is cut out with the center of gravity of the circumscribed rectangle of the space buffer as the center point to obtain a fire sample image, so as to reduce irrelevant background information. And then, taking a thermal emission wave band preset by the fire sample image, performing self-adaptive threshold segmentation, marking a fire point pixel on the fire sample image, and recording the coordinates of the fire point pixel. And finally, generating a fire satellite image sample based on the fire sample image, the circumscribed rectangle of the space buffer, the coordinates of the fire pixels and the fire recording time corresponding to the fire pixel patches corresponding to the target satellite. Therefore, in the technical scheme provided by the application, the target satellite fire image is obtained based on the space-time buffer area retrieval and is automatically cut, so that the fire pixels and the marks of the fire pixels can be automatically determined in a local small range by using a simple self-adaptive threshold segmentation method, the automatic production of the fire satellite image sample is realized, and the manpower consumption and the uncertainty of the fire sample are greatly reduced.
In addition, the fire satellite image sample generated by the method relates to the range, time, image, fire point and the like, and has complete fire information. In addition, the method and the device are based on mutual verification of the high-spatial-resolution polar orbit satellite and the low-spatial-resolution static orbit satellite in space, and can assist the static orbit satellite to find unresponsive fire points when a target satellite fire image related to the static orbit satellite is searched based on a space-time buffer zone in the follow-up process, so that the production of a fire satellite image sample with low spatial resolution is realized.
Referring to fig. 9, an embodiment of the present application provides a method for acquiring fire images based on multi-source fire data, where the method includes:
the data acquisition module 901 is configured to acquire multi-source satellite fire point data, where the multi-source satellite fire point data includes a plurality of fire point records, and the fire point records include: the longitude and latitude coordinates of the fire points and the time for recording the fire points are recorded as a static orbit satellite and a polar orbit satellite;
the pixel patch generation module 902 is configured to generate, in the multi-source satellite fire point data, fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of the fire point recorded by each fire point, so as to obtain a plurality of fire point pixel patches;
The space buffer area determining module 903 is configured to obtain a space buffer area from a union of the fire pixel patches with space intersection;
the space-time buffer area determining module 904 is configured to determine a fire occurrence time range based on a fire point recording time corresponding to each fire point pixel plaque in the space buffer area, and form a space-time buffer area from the fire occurrence time range and the space buffer area;
the image retrieval module 905 is configured to retrieve a fire image of a target satellite through the space-time buffer, where the fire image of the target satellite is an image acquired by a stationary orbit satellite or a polar orbit satellite.
In some implementations of embodiments of the present application, the apparatus further includes:
the image clipping module is used for clipping images with preset resolution in the fire images of the target satellite by taking the center of gravity of the circumscribed rectangle of the space buffer as a center point so as to obtain fire sample images;
the fire point marking module is used for taking a preset thermal emission band of the fire sample image, performing self-adaptive threshold segmentation processing on the fire sample image, marking pixels larger than a preset threshold in the fire sample image as fire point pixels, and recording coordinates of the fire point pixels;
The time determining module is used for determining that a satellite for acquiring a fire image of a target satellite is the target satellite, determining the fire recording time corresponding to a fire pixel patch corresponding to the target satellite in the space buffer zone, wherein the target satellite is a static orbit satellite or a polar orbit satellite;
the sample generation module is used for generating a fire satellite image sample based on the fire sample image, the circumscribed rectangle of the space buffer, the coordinates of the fire pixels and the fire recording time corresponding to the fire pixel patch corresponding to the target satellite.
In some implementations of embodiments of the present application, the apparatus further includes:
and the filtering module is used for filtering the fire record acquired by the static orbit satellite based on the fire record acquired by the polar orbit satellite in the multi-source satellite fire data.
In some implementations of embodiments of the present application, filtering the fire records acquired by the stationary orbiting satellites based on the fire records acquired by the polar orbiting satellites includes:
taking the fire point recording time of each fire point record acquired by the polar orbit satellite as the interval center, and determining a plurality of fire point selection intervals;
and filtering out the fire record of which the fire record time does not fall into any fire record selection interval from the fire records acquired by the static orbit satellite.
In some implementations of the embodiments of the present application, generating pixel patches corresponding to each fire record based on longitude and latitude coordinates of the fire record, respectively, includes:
and for each fire point record, taking the longitude and latitude coordinates of the fire point as the coordinates of the pixel center point, and generating pixel patches based on the coordinates of the pixel center point and the resolution of the satellite for acquiring the fire point record.
In some implementations of embodiments of the present application, determining a fire occurrence time range based on a fire recording time corresponding to each fire pixel patch in a spatial buffer includes:
based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, taking a section formed by the earliest fire point recording time and the latest fire point recording time as a fire occurrence time range.
In some implementations of embodiments of the present application, retrieving a target satellite fire image via a space-time buffer includes:
and searching on the geographic information searching platform according to the fire occurrence time range and the circumscribed rectangle of the space buffer area to obtain a target satellite image.
In some implementations of embodiments of the present application, the fire satellite image samples are in HDF format.
As shown in fig. 10, the embodiment of the present application further provides a fire image acquiring device based on multi-source fire points, including: memory 1001, processor 1002;
Wherein the memory 1001 is used for storing a program;
the processor 1002 is configured to execute a program in the memory to implement the steps of a method for acquiring a fire image based on multi-source fire data according to the embodiments of the present application.
Finally, it should also be noted that in the embodiments of the present application, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for acquiring a fire image based on multi-source fire points, the method comprising:
acquiring multi-source satellite fire data, the multi-source satellite fire data comprising a plurality of fire records, the fire records comprising: the fire point longitude and latitude coordinates and the fire point recording time are recorded as the collected by the static orbit satellite and the polar orbit satellite;
in the multi-source satellite fire point data, generating fire point pixel patches corresponding to each fire point record based on the longitude and latitude coordinates of each fire point record respectively so as to obtain a plurality of fire point pixel patches;
obtaining a space buffer area by taking the union of the fire point pixel patches with the space intersection among the plurality of fire point pixel patches;
Determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone;
and searching through the space-time buffer area to obtain a target satellite fire image, wherein the target satellite fire image is an image acquired by the static orbit satellite or the polar orbit satellite.
2. The method according to claim 1, wherein the method further comprises:
cutting an image with preset resolution from the target satellite fire image by taking the center of gravity of the circumscribed rectangle of the space buffer as a center point to obtain a fire sample image;
taking a preset thermal emission wave band of the fire sample image, performing self-adaptive threshold segmentation processing on the fire sample image, marking pixels larger than a preset threshold in the fire sample image as fire pixels, and recording coordinates of the fire pixels;
determining a satellite for acquiring the fire disaster image of the target satellite as a target satellite, and determining the fire point recording time corresponding to the fire point pixel patch corresponding to the target satellite in the space buffer zone, wherein the target satellite is the static orbit satellite or the polar orbit satellite;
And generating a fire satellite image sample based on the fire sample image, the circumscribed rectangle of the space buffer, the coordinates of the fire pixel and the fire recording time corresponding to the fire pixel patch corresponding to the target satellite.
3. The method according to claim 1, wherein the method further comprises:
and in the multi-source satellite fire point data, filtering fire point records acquired by the static orbit satellite based on the fire point records acquired by the polar orbit satellite.
4. The method of claim 3, wherein the filtering the fire records acquired by the stationary orbiting satellite based on the fire records acquired by the polar orbiting satellite comprises:
taking the fire point recording time of each fire point record acquired by the polar orbit satellite as a section center, and determining a plurality of fire point selection sections;
and filtering out the fire record of which the fire record time does not fall into any one of the fire record selection intervals in each fire record acquired by the static orbit satellite.
5. The method of claim 1, wherein generating a fire pixel patch corresponding to each of the fire records based on the fire latitude and longitude coordinates of each of the fire records, respectively, comprises:
And for each fire point record, taking the longitude and latitude coordinates of the fire point as the coordinates of a pixel center point, and generating a fire point pixel patch based on the coordinates of the pixel center point and the resolution of a satellite for acquiring the fire point record.
6. The method of claim 1, wherein said determining a fire occurrence time range based on said fire record time for each of said fire pixel patches in said spatial buffer comprises:
and taking a section formed by the earliest fire point recording time and the latest fire point recording time as a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel patch in the space buffer zone.
7. The method of claim 1, wherein said retrieving the target satellite fire image via the space-time buffer comprises:
and searching the target satellite image on a geographic information searching platform according to the fire occurrence time range and the circumscribed rectangle of the space buffer area.
8. The method of claim 2, wherein the fire satellite video samples are in HDF format.
9. A device for acquiring fire images based on multi-source fire points, the device comprising:
The system comprises a data acquisition module, a data processing module and a data processing module, wherein the data acquisition module is used for acquiring multi-source satellite fire point data, the multi-source satellite fire point data comprises a plurality of fire point records, and the fire point records comprise: the fire point longitude and latitude coordinates and the fire point recording time are recorded as the collected by the static orbit satellite and the polar orbit satellite;
the pixel patch generation module is used for generating a fire point pixel patch corresponding to each fire point record based on the longitude and latitude coordinates of each fire point record in the multi-source satellite fire point data so as to obtain a plurality of fire point pixel patches;
the space buffer area determining module is used for obtaining a space buffer area from the fire point pixel patches with space intersection by taking the union of the fire point pixel patches;
the space-time buffer zone determining module is used for determining a fire occurrence time range based on the fire point recording time corresponding to each fire point pixel plaque in the space buffer zone, and forming a space-time buffer zone by the fire occurrence time range and the space buffer zone;
and the image retrieval module is used for retrieving and obtaining a target satellite fire image through the space-time buffer zone, wherein the target satellite fire image is an image acquired by the static orbit satellite or the polar orbit satellite.
10. An apparatus for acquiring a fire image based on a multi-source fire, the apparatus comprising: comprising a memory and a processor for executing a program stored in the memory, running a fire determination method based on multi-source satellite fire data as claimed in any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311781527.8A CN117710830A (en) | 2023-12-22 | 2023-12-22 | Method, device and equipment for acquiring fire image based on multi-source fire point data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311781527.8A CN117710830A (en) | 2023-12-22 | 2023-12-22 | Method, device and equipment for acquiring fire image based on multi-source fire point data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117710830A true CN117710830A (en) | 2024-03-15 |
Family
ID=90158736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311781527.8A Pending CN117710830A (en) | 2023-12-22 | 2023-12-22 | Method, device and equipment for acquiring fire image based on multi-source fire point data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117710830A (en) |
-
2023
- 2023-12-22 CN CN202311781527.8A patent/CN117710830A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Over et al. | Processing coastal imagery with Agisoft Metashape Professional Edition, version 1.6—Structure from motion workflow documentation | |
| US10032267B2 (en) | Automating the assessment of damage to infrastructure assets | |
| Hess et al. | Geocoded digital videography for validation of land cover mapping in the Amazon basin | |
| Wick et al. | Description and validation of an automated objective technique for identification and characterization of the integrated water vapor signature of atmospheric rivers | |
| WO2019032692A1 (en) | Multi-resolution multi-spectral deep learning based change detection for satellite images | |
| JPWO2008016153A1 (en) | Disaster countermeasure support method | |
| CN111340012A (en) | Geological disaster interpretation method and device and terminal equipment | |
| KR102399089B1 (en) | A system and method for flood damage detection using satellite image information | |
| Liu | Preparing a landslide and shadow inventory map from high-spatial-resolution imagery facilitated by an expert system | |
| Khan et al. | Investigating the potential of a global precipitation forecast to inform landslide prediction | |
| Liu et al. | Single tree segmentation and diameter at breast height estimation with mobile LiDAR | |
| Karantanellis et al. | 3D hazard analysis and object-based characterization of landslide motion mechanism using UAV imagery | |
| Jain et al. | Modeling urban land cover growth dynamics based on land change modeler (LCM) using remote sensing: a case study of Gurgaon, India | |
| Symeonakis et al. | Multi-temporal land-cover classification and change analysis with conditional probability networks: The case of Lesvos Island (Greece) | |
| Bin et al. | Parcel-based change detection in land-use maps by adopting the holistic feature | |
| Jing et al. | Efficient point cloud corrections for mobile monitoring applications using road/rail-side infrastructure | |
| Wickramasinghe et al. | Satellite remote-sensing monitoring of a railway construction project | |
| Yin et al. | Object-based larch tree-crown delineation using high-resolution satellite imagery | |
| CN117710830A (en) | Method, device and equipment for acquiring fire image based on multi-source fire point data | |
| Ozisik et al. | Post-earthquake damage assessment using satellite and airborne data in the case of the 1999 Kocaeli earthquake, Turkey | |
| Galarreta | Urban Structural Damage Assessment Using Object-Oriented Analysis and Semantic Reasoning | |
| Kraus et al. | Breaking the curse of visual data exploration: improving analyses by building bridges between data world and real world | |
| Rulinda et al. | Visualizing and quantifying the movement of vegetative drought using remote-sensing data and GIS | |
| Vandaele et al. | Calibrated river-level estimation from river cameras using convolutional neural networks | |
| Ramlal et al. | A Rapid Post-Hurricane Building Damage Assessment Methodology using Satellite Imagery. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |