CN115294439B - Method, system, equipment and storage medium for detecting air weak and small moving target - Google Patents

Method, system, equipment and storage medium for detecting air weak and small moving target Download PDF

Info

Publication number
CN115294439B
CN115294439B CN202210923739.4A CN202210923739A CN115294439B CN 115294439 B CN115294439 B CN 115294439B CN 202210923739 A CN202210923739 A CN 202210923739A CN 115294439 B CN115294439 B CN 115294439B
Authority
CN
China
Prior art keywords
image
moving target
weak
band
wave
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.)
Active
Application number
CN202210923739.4A
Other languages
Chinese (zh)
Other versions
CN115294439A (en
Inventor
贺广均
陈千千
马天舒
刘世烁
冯鹏铭
符晗
常江
金世超
邹同元
张鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Institute of Satellite Information Engineering
Original Assignee
Beijing Institute of Satellite Information Engineering
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Institute of Satellite Information Engineering filed Critical Beijing Institute of Satellite Information Engineering
Priority to CN202210923739.4A priority Critical patent/CN115294439B/en
Publication of CN115294439A publication Critical patent/CN115294439A/en
Application granted granted Critical
Publication of CN115294439B publication Critical patent/CN115294439B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/215—Motion-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10032—Satellite or aerial image; Remote sensing
    • G06T2207/10036—Multispectral image; Hyperspectral image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
    • Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to a method, a system, equipment and a storage medium for detecting an aerial weak and small moving target. The method realizes the detection of the small and weak moving target in the air in the multispectral images with different spatial resolutions in a wide imaging mode, avoids the influence of the spatial resolution on the detection precision in the traditional method and the limitation in practical application, makes up the defects of the means and the method in the prior art, and improves the detection and identification precision of the small and weak moving target in the air.

Description

Method, system, equipment and storage medium for detecting air weak and small moving target
Technical Field
The invention relates to a method, a system, equipment and a storage medium for detecting an aerial weak and small moving target.
Background
The detection and identification of the air weak moving targets are crucial to space safety and airspace management, the monitoring of the air weak moving targets is mainly based on a foundation radar or an air target automatic identification system at present, no single technology can monitor all the air weak moving targets in the global range under the current complex aviation environment, the foundation radar can only detect 30% of moving airplane targets, and the air target automatic identification system cannot cover all ocean areas and can only monitor cooperative targets. The satellite remote sensing is used as a supplementary means for detecting and identifying the small and weak moving target in the air, and has the advantages of wide monitoring range, strong detection capability of a non-cooperative target and the like.
The satellite remote sensing image is a real-time record of the electromagnetic spectrum characteristics of the ground features, and researchers can realize the detection and identification of specific targets in the image according to the target characteristics (the spectral characteristics, the spatial characteristics, the time displacement characteristics and the like of the targets) recorded on the remote sensing image. At present, most of aerial target detection and identification methods based on satellite remote sensing images are based on high-spatial-resolution remote sensing data. Due to the limitation of the satellite imaging mode, the width of the satellite imaging in the high-resolution imaging mode is often small, and the method has great limitation in aerial target monitoring application. Meanwhile, the current main imaging-based aerial target detection methods focus on static target detection in a specific scene, such as airplane target detection parked at an airport. In recent years, methods such as deep learning have significantly improved the accuracy of detecting and identifying objects in images, but these methods require that the objects have a certain size in the images, and are not suitable for detecting and identifying weak and small objects. Therefore, in a satellite wide-width imaging mode, detection of a weak and small aerial weak and small moving target in an image is an urgent problem to be solved.
Disclosure of Invention
In view of the technical problems, the invention provides a method for detecting a small and medium moving target in the air of a satellite remote sensing image, which is used for detecting the small and medium moving target in the air of multispectral satellite remote sensing images with different resolutions in a wide imaging mode.
The technical solution for realizing the purpose of the invention is as follows: a method for detecting an aerial weak and small moving target comprises the following steps:
the method comprises the following steps of S1, reading satellite remote sensing data, and selecting a plurality of wave bands with the maximum spectral reflection intensity of a space moving target;
s2, determining a candidate region of the air weak and small moving target by using the selected multiple wave bands;
s3, carrying out image slicing according to the target candidate area, and determining an aerial weak and small moving target by combining a plurality of selected wave bands;
and S4, converting coordinates and projection of the detected aerial weak and small moving target image, and outputting a detection result.
According to an aspect of the present invention, in step S1, the method specifically includes:
s11, reading satellite remote sensing image data, wherein the satellite remote sensing image data is a satellite remote sensing image with imaging parallax among at least three different wave bands;
s12, reading an original multispectral data source, acquiring the total number of wave bands, sequentially acquiring pixel matrixes of different wave bands, and storing the pixel matrixes and projection information together into a single-wave-band image;
and S13, analyzing the brightness characteristics of the images of different wave bands in a wide imaging mode, selecting a plurality of wave bands with the maximum spectral reflection intensity of the space moving target, and synthesizing a plurality of single-wave-band images to obtain a synthesized image.
According to an aspect of the present invention, in step S2, the method specifically includes:
s21, extracting a reflection abnormal area of the moving target in a middle wave band of a plurality of wave bands based on wave band operation and logic operation;
and S22, determining a candidate area of the air weak and small moving target by using the reflection abnormal area according to a preset judgment condition.
According to an aspect of the present invention, three bands with the maximum spectral reflection intensity of the space moving target are selected, which are respectively a B1 band, a B2 band and a B3 band, and in step S21, the method specifically includes:
step S211, calculating difference images of the B2 band and the B1 band and the B3 band, which are denoted as B21 and B23, with the B2 band as the middle band, and the calculation formula is:
B21[i,j]＝B2[i,j]-B1[i,j]，
B23[i,j]＝B2[i,j]-B3[i,j]，
b21[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B1 at the position (i, j), B23[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B3 at the position (i, j), i and j are values from 1 to m and n respectively, and m and n are the height and width of the image respectively;
step S212, judging whether the pixel values of the pixel difference images B21 and B31 and the wave band B2 image are all in a preset range, and calculating to obtain a binary image B2gray [ i, j ], wherein the calculation formula is as follows:
Figure BDA0003778754300000031
wherein T1, T2 and T3 are preset threshold values;
and step S213 and step S212, the pixel with the pixel value of 1 in the binary image is the abnormal reflection area of the moving object.
According to an aspect of the present invention, in the step S22, the method specifically includes:
step S221, mapping the acquired two-dimensional binary image and the reflection abnormal area to a three-dimensional true color image to obtain a three-dimensional binary image B123gray, wherein the formula is as follows:
Figure BDA0003778754300000032
b123[ i, j, k ] represents the pixel value of a position (i, j) in the kth image channel, the value of k is 0, 1 and 2, and the region with the pixel value of 1 is a candidate region of the air weak and small moving target;
step S222, obtaining a connected region of the three-dimensional image by using a region connection technology of image processing, and when the area of the connected region is smaller than a certain threshold value, making the pixel value of the B123gray be 0, wherein the reserved connected region is the region where the candidate aerial weak and small moving target is located.
According to an aspect of the present invention, in step S3, the method specifically includes:
s31, obtaining a region centroid according to the target candidate region, and cutting the region centroid as a center into image slices;
and S32, respectively extracting reflection abnormal areas of B1 wave band, B2 wave band and B3 wave band based on image slices, extracting linear combination characteristics and proportional displacement characteristics among different wave bands of a candidate area of the air weak small moving target, and determining the air weak small moving target by combining preset judgment conditions.
According to an aspect of the present invention, in step S31, the method specifically includes:
and according to the three-dimensional binary image, taking a two-dimensional binary image of the first channel, acquiring a connected region of the image and a centroid of the connected region according to the two-dimensional image, and cutting the two-dimensional binary image into image slices by taking the centroid as a center.
According to an aspect of the present invention, there is provided an aerial weak moving object detection system, including:
the reading unit is used for reading satellite remote sensing data;
the determining unit is used for determining a candidate region of the aerial weak and small moving target by using the selected multiple wave bands;
an image processing unit for performing image slicing according to the target candidate region;
the determining unit is further used for determining the air weak and small moving target by combining the selected multiple wave bands;
the image processing unit is also used for carrying out coordinate and projection conversion on the detected aerial weak and small moving target image and outputting a detection result.
According to an aspect of the present invention, there is provided an electronic apparatus including: one or more processors, one or more memories, and one or more computer programs; wherein, a processor is connected with a memory, the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute a method for detecting a moving object with weak air and small air as described in any one of the above technical solutions.
According to an aspect of the present invention, there is provided a computer readable storage medium for storing computer instructions, which when executed by a processor, implement a method for detecting an airborne weak moving object according to any one of the above technical solutions.
According to the concept of the invention, a method for detecting an aerial weak and small moving target is provided, which comprises the steps of firstly reading a satellite remote sensing image with imaging parallax among at least three different wave bands, obtaining a multispectral data source, wherein each wave band has certain time deviation when imaging the target, the time deviation can cause the displacement parallax of the moving target in the multispectral image, determining the aerial weak and small moving target by using the displacement parallax, finally carrying out coordinate and projection conversion on the detected aerial weak and small moving target image, outputting a detection result, realizing the detection of the aerial weak and small moving target in the multispectral image with different spatial resolutions in a wide-width imaging mode, avoiding the influence of the spatial resolution on the detection precision and the limitation in practical application in the traditional method, making up the defects of the prior art means and methods, and improving the detection and identification precision of the aerial weak and small moving target.
The method can be expanded and applied to scenes such as target identification, target tracking and the like, and has the characteristics of high accuracy, good stability and strong practicability.
Drawings
FIG. 1 is a schematic flow chart illustrating a method for detecting an airborne weak moving object according to an embodiment of the present invention;
FIG. 2 is a diagram schematically illustrating the detection result of an airborne weak moving object according to an embodiment of the present invention;
FIG. 3 is a schematic representation of a partial enlarged view of the detection results of weak and small moving objects in the air in capital international airports and surrounding areas according to one embodiment of the invention;
FIG. 4 is a flow diagram schematically illustrating detection of a small and weak moving object based on a candidate region image slice, according to an embodiment of the present invention;
FIG. 5 is a flow diagram schematically illustrating a method for detecting a weak moving object in the air according to another embodiment of the present invention;
FIG. 6 schematically shows a flow chart of step S1 according to an embodiment of the invention;
FIG. 7 schematically shows a partial flow diagram of step S2 according to an embodiment of the present invention;
fig. 8 schematically shows a schematic diagram of an aerial weak moving object detection system framework according to an embodiment of the invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the embodiments will be briefly described below. It is obvious that the drawings in the following description are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
The present invention is described in detail below with reference to the drawings and the specific embodiments, which are not repeated herein, but the embodiments of the present invention are not limited to the following embodiments.
As shown in fig. 1 to 8, the method for detecting a weak moving object in the air of the present invention includes the following steps:
the method comprises the following steps of S1, reading satellite remote sensing data, and selecting a plurality of wave bands with the maximum spectral reflection intensity of a space moving target;
s2, determining a candidate region of the air weak and small moving target by using the selected multiple wave bands;
s3, carrying out image slicing according to the target candidate area, and determining an aerial weak and small moving target by combining a plurality of selected wave bands;
and S4, converting coordinates and projection of the detected aerial weak and small moving target image, and outputting a detection result.
In the embodiment, a satellite remote sensing image with imaging parallax among at least three different wave bands is read, a certain time deviation exists when each wave band in an acquired multispectral data source images a target, the time deviation can cause the displacement parallax of a moving target in the multispectral image, the displacement parallax is utilized to determine a small moving target in the air, finally, the coordinate and projection conversion is carried out on the detected small moving target in the air, a detection result is output, the detection of the small moving target in the air in the multispectral image with different spatial resolutions in a wide-amplitude imaging mode can be realized, the influence of the spatial resolution on the detection precision and the limitation in practical application in the traditional method are avoided, the defects of the prior art means and the method are overcome, and the detection and identification precision of the small moving target in the air is improved.
The method can be expanded and applied to scenes such as target identification, target tracking and the like, and has the characteristics of high accuracy, good stability and strong practicability.
As shown in fig. 6, in an embodiment of the present invention, preferably, in step S1, the method specifically includes:
s11, reading satellite remote sensing image data, wherein the satellite remote sensing image data is a satellite remote sensing image with imaging parallax among at least three different wave bands;
s12, reading an original multispectral data source, acquiring the total number of wave bands, sequentially acquiring pixel matrixes of different wave bands, and storing the pixel matrixes and projection information together into a single-wave-band image;
and S13, analyzing the brightness characteristics of the images of different wave bands in a wide imaging mode, selecting a plurality of wave bands with the maximum spectral reflection intensity of the space moving target, and synthesizing a plurality of single-wave-band images to obtain a synthesized image.
In this embodiment, the acquired multispectral data is full-spectral-band spectral imager data of a high-resolution five-number satellite, the total number of the bands is 12, the spatial resolution in the band of 0.45 to 2.35 μm is 20m, the spatial resolution in the band of 3.5 to 12.5 μm is 40 m, the imaging width is 60km, 3 candidate bands are selected from the 12 spectral bands, the bands are selected according to the conditions that the signal-to-noise ratio is high and the brightness characteristics of the small and weak moving targets in the air are obvious, the selected bands are respectively marked as B1, B2 and B3, and in a true color image formed by the candidate 3 bands, the small and weak moving targets in the air present a red-green-blue or blue-green-red combination which is approximately linearly distributed.
As shown in fig. 7, in an embodiment of the present invention, preferably, in the step S2, the method specifically includes:
s21, extracting a reflection abnormal area of the moving target in a middle wave band of a plurality of wave bands based on wave band operation and logic operation;
and S22, determining a candidate area of the air weak and small moving target by using the reflection abnormal area according to a preset judgment condition.
In this embodiment, the band operation calculates the pixel difference between the bands, and the logical operation is to determine whether the pixel values of the pixel difference image and the mid-band image are both within a predetermined range.
In an embodiment of the present invention, preferably, three bands with the maximum spectral reflection intensity of the spatially moving object are selected, which are respectively a B1 band, a B2 band, and a B3 band, and in step S21, the method specifically includes:
step S211, calculating difference images of the B2 band and the B1 band and the B3 band, which are denoted as B21 and B23, with the B2 band as the middle band, and the calculation formula is:
B21[i,j]＝B2[i,j]-B1[i,j]，
B23[i,j]＝B2[i,j]-B3[i,j]，
b21[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B1 at the position (i, j), B23[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B3 at the position (i, j), i and j are values from 1 to m and n respectively, and m and n are the height and width of the image respectively;
step S212, judging whether the pixel values of the pixel difference images B21 and B31 and the wave band B2 image are all in a preset range, and calculating to obtain a binary image B2gray [ i, j ], wherein the calculation formula is as follows:
Figure BDA0003778754300000091
wherein T1, T2 and T3 are preset threshold values;
and step S213 and step S212, the pixel with the pixel value of 1 in the binary image obtained in step S is the abnormal reflection area of the moving object.
In the embodiment, three wave bands with the maximum spectral reflection intensity of the space moving target are selected, the middle wave band is the B2 wave band, T1, T2 and T3, and the optimal solutions of the three threshold values are respectively 150, 150 and 900 through tests and statistical analysis.
Further, if four bands with the maximum spectral reflection intensity of the spatially moving object are selected, i.e., B1, B2, B3, and B4, the intermediate band may be either B2 or B3.
In an embodiment of the present invention, preferably, in the step S22, specifically including:
step S221, mapping the obtained two-dimensional binary image and the reflection abnormal area to a three-dimensional true color image to obtain a three-dimensional binary image B123gray, wherein the formula is as follows:
Figure BDA0003778754300000092
b123gray [ i, j, k ] represents a pixel value of a position (i, j) in a kth image channel, k is 0, 1 and 2, and an area with the pixel value of 1 is a candidate area of an aerial weak and small moving target;
step S222, obtaining a connected region of the three-dimensional image by using a region connection technology of image processing, and when the area of the connected region is smaller than a certain threshold value, making the pixel value of the B123gray be 0, wherein the reserved connected region is the region where the candidate aerial weak and small moving target is located.
In the embodiment, a connected domain is set to be 18 according to the pixel value of the three-dimensional true color image, the obtained image connected domain is further judged, when the number of the pixels of the connected domain is larger than 900, the connected domain is considered to be not in accordance with the small area characteristic of the moving target, the connected domain is removed from the candidate region of the moving target, a new two-dimensional image is obtained again, the pixel value is 1, the candidate region of the moving target is obtained, the connected region of the three-dimensional image is obtained, when the area of the connected region is smaller than a certain threshold value, the pixel value of the B123gray is made to be 0, and the reserved connected region is the region where the candidate weak moving target in the air is located.
In an embodiment of the present invention, preferably, in the step S3, specifically including:
s31, obtaining a region centroid according to the target candidate region, and cutting the region centroid as a center into image slices;
and S32, respectively extracting reflection abnormal areas of B1 wave band, B2 wave band and B3 wave band based on image slices, extracting linear combination characteristics and proportional displacement characteristics among different wave bands of a candidate area of the air weak small moving target, and determining the air weak small moving target by combining preset judgment conditions.
In this embodiment, based on the target candidate area image slice, reflection abnormal areas of B1, B2, and B3 wave bands are respectively extracted, linear combination features and proportional displacement features between different wave bands of the air weak moving target are extracted, and the process of determining the air weak moving target is as shown in fig. 4, and specifically includes:
s321, acquiring B2 wave band data of an image slice, extracting a reflection abnormal region of a B2 wave band to obtain a binary image B2Gray, solving a connected region of the B2Gray, and recording a mass center;
step S322, acquiring B1 wave band data of the image slice, and extracting a reflection abnormal area of the B1 wave band, wherein the process is as follows:
step S3221, calculating pixel difference images of a B1 wave band, a B2 wave band and a B3 wave band, and recording the pixel difference images as B12 and B13;
step S3222, determining whether pixel values of B12, B13, and B1 bands are within a certain range, defining four ranges r11, r12, r13, and r14, and obtaining a binary image as follows:
Figure BDA0003778754300000111
wherein r11, r12, r13 and r14 are preset thresholds, and the optimal values are 40, 90, 500 and 900 after multiple tests.
Step S3223, 8 connected regions of the B1gray and the centroids of the connected regions are obtained.
Step S323, B3 wave band data of the image slice is obtained, and a reflection abnormal area of the B3 wave band is extracted, wherein the process is as follows:
step S3231, calculating pixel difference images of a B3 wave band, a B1 wave band and a B2 wave band, and recording the image difference images as B31 and B32;
step S3232, judging whether the pixel values of the B31, B32 and B3 wave bands are in a certain range, defining four ranges r31, r32, r33 and r34, and obtaining a binary image as follows:
Figure BDA0003778754300000112
wherein r31, r32, r33 and r34 are preset thresholds, and the optimal values of 20, 60, 400 and 1100 are obtained through multiple tests.
Step S3233, 8 connected regions of the B3gray and the mass centers of the connected regions are obtained.
Step S324, the centroids of the current image slice in B1, B2, and B3 bands can be obtained through step S321, step S322, and step S323, and are respectively recorded as CenB1, cenB2, and CenB3;
if no centroid exists in any wave band in the image slice, the image slice is considered to have no detection of the aerial weak moving object;
if CenB1, cenB2 and CenB3 are not empty, calculating the distance from the mass center CenB2 of the B2 wave band to the mass centers CenB1 and CenB3 of the B1 wave band and the angle formed by the mass centers of 3 wave bands.
Wherein, the distance calculation formula is as follows:
Figure BDA0003778754300000121
/>
wherein, cenB2x and CenB2y are column number and row number of the centroid of B2 wave band, x and y are column number and row number of the centroid of B1 wave band or B3 wave band respectively;
the calculation formula of the angle is as follows:
Figure BDA0003778754300000122
wherein B represents a determinant of a 2 × 2 matrix formed by the centroid of the band B2 and the centroid of the band B1 or the centroid of the band B3, and C represents a vector product of the centroid of B2 and the centroid of the band B1 or the centroid of the band B3.
Step S325, determining whether the centroid, the distance, and the angle satisfy the linear combination characteristic and the proportional displacement characteristic, which are defined as follows:
3＜Distance≤100,90≤angle≤210；
when the above conditions are met, the presence of an aerial weak moving object in the image slice is judged, and the centroids of 3 wave bands (two centroids closest to the centroid of the B2 wave band) are returned.
In an embodiment of the present invention, preferably, in the step S31, specifically, the method includes:
and according to the three-dimensional binary image, taking a two-dimensional binary image of the first channel, acquiring a connected region of the image and a centroid of the connected region according to the two-dimensional image, and cutting the two-dimensional binary image into image slices by taking the centroid as a center.
In this embodiment, according to the three-dimensional binary image obtained in step S221, an 8-connected region in which the pixel value is 1 in the binary image is obtained, and the centroid of each connected region is calculated; because the small and weak moving target in the air is sparse in the multispectral, the original image is used as data input, so that the calculated amount is greatly increased, the calculation amount is reduced by extracting the moving target image slice, the accuracy is improved, and the multispectral true color image with the original size is generally cut into the image slice with the size of 21 multiplied by 21 by taking the center of mass as the center.
As shown in fig. 8, according to an aspect of the present invention, there is provided an air weak moving object detection system, including:
the reading unit is used for reading satellite remote sensing data;
the determining unit is used for determining a candidate region of the aerial weak and small moving target by using the selected multiple wave bands;
an image processing unit for performing image slicing according to the target candidate region;
the determining unit is further used for determining the small and weak moving target in the air by combining the selected multiple wave bands;
the image processing unit is also used for carrying out coordinate and projection conversion on the detected aerial weak and small moving target image and outputting a detection result.
According to an aspect of the present invention, there is provided an electronic apparatus including: one or more processors, one or more memories, and one or more computer programs; wherein, a processor is connected with a memory, the one or more computer programs are stored in the memory, and when the electronic device runs, the processor executes the one or more computer programs stored in the memory, so as to make the electronic device execute a method for detecting a moving object with weak air and small air as described in any one of the above technical solutions.
According to an aspect of the present invention, there is provided a computer readable storage medium for storing computer instructions, which when executed by a processor, implement a method for detecting an airborne weak moving object according to any one of the above technical solutions.
In summary, the invention provides a method, a system, a device and a storage medium for detecting an aerial weak and small moving target, which includes firstly reading a satellite remote sensing image with imaging parallax between at least three different bands, obtaining a certain time deviation when each band in a multispectral data source images the target, wherein the time deviation can cause the displacement parallax of the moving target in the multispectral image, determining the aerial weak and small moving target by using the displacement parallax, finally performing coordinate and projection conversion on the detected aerial weak and small moving target image, outputting a detection result, and realizing the detection of the aerial weak and small moving target in the multispectral image with different spatial resolutions in a wide-width imaging mode.
The method can be expanded and applied to scenes such as target identification, target tracking and the like, and has the characteristics of high accuracy, good stability and strong practicability.
Furthermore, it should be noted that the present invention may be provided as a method, apparatus or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied in the medium.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, an embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal 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 terminal. Without further limitation, an element defined by the phrases "comprising one of \ 8230; \8230;" does not exclude the presence of additional like elements in a process, method, article, or terminal device that comprises the element.
Finally, it should be noted that while the above describes a preferred embodiment of the invention, it will be appreciated by those skilled in the art that, once the basic inventive concepts have been learned, numerous changes and modifications may be made without departing from the principles of the invention, which shall be deemed to be within the scope of the invention. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Claims (5)

1. A method for detecting an aerial weak moving target comprises the following steps:
s1, reading satellite remote sensing data, and selecting a plurality of wave bands with the maximum spectral reflection intensity of a space moving target;
s2, reading an original multispectral data source, and determining a candidate area of the air weak and small moving target by using a plurality of selected wave bands;
s3, carrying out image slicing according to the target candidate area, and determining an aerial weak and small moving target by combining a plurality of selected wave bands;
s4, converting coordinates and projection of the detected aerial weak and small moving target image, and outputting a detection result;
in the step S1, the method specifically includes:
s11, reading satellite remote sensing image data, wherein the satellite remote sensing image data is a satellite remote sensing image with imaging parallax among at least three different wave bands;
s12, reading an original multispectral data source, acquiring the total number of wave bands, sequentially acquiring pixel matrixes of different wave bands, and storing the pixel matrixes and projection information together into a single-wave-band image;
s13, analyzing the brightness characteristics of images of different wave bands in a wide-width imaging mode, selecting a plurality of wave bands with the maximum spectral reflection intensity of the space moving target, and synthesizing a plurality of single-wave-band images to obtain a synthesized image;
in the step S2, the method specifically includes:
s21, extracting a reflection abnormal area of the moving target in a middle wave band of a plurality of wave bands based on wave band operation and logic operation;
s22, determining a candidate area of the air weak and small moving target by using the reflection abnormal area according to a preset judgment condition;
selecting three wave bands with the maximum spectral reflection intensity of the space moving target, wherein the three wave bands are respectively a B1 wave band, a B2 wave band and a B3 wave band, and in the step S21, the method specifically comprises the following steps:
step S211, calculating difference images of the B2 band and the B1 band and the B3 band, which are denoted as B21 and B23, with the B2 band as the middle band, and the calculation formula is:
B21[i,j]＝B2[i,j]-B1[i,j]，
B23[i,j]＝B2[i,j]-B3[i,j]，
b21[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B1 at the position (i, j), B23[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B3 at the position (i, j), i and j are values from 1 to m and n respectively, and m and n are the height and the width of the image respectively;
step S212, determining whether the pixel values of the pixel difference images B21 and B31 and the band B2 image are all within a predetermined range, and calculating to obtain a binary image B2gray [ i, j ], where the calculation formula is:
Figure FDA0004066097850000021
wherein T1, T2 and T3 are preset threshold values;
the pixel with the pixel value of 1 in the binary image obtained in the step S213 and the step S212 is a reflection abnormal area of the moving object;
in step S22, the method specifically includes:
step S221, mapping the acquired two-dimensional binary image and the reflection abnormal area to a three-dimensional true color image to obtain a three-dimensional binary image B123gray, wherein the formula is as follows:
Figure FDA0004066097850000022
wherein, B123gray [ i, j, k ] represents the pixel value of the position (i, j) in the kth image channel, k is 0, 1 and 2, and the region with pixel value of 1 is the candidate region of the air weak and small moving target;
step S222, acquiring a connected region of the three-dimensional image by using a region connection technology of image processing, and when the area of the connected region is smaller than a certain threshold value, making the pixel value of the B123gray be 0, wherein the reserved connected region is the region where the candidate aerial weak and small moving target is located;
in the step S3, the method specifically includes:
s31, obtaining a region centroid according to the target candidate region, and cutting the region centroid into image slices by taking the centroid as a center;
and S32, respectively extracting reflection abnormal areas of B1 wave band, B2 wave band and B3 wave band based on image slices, extracting linear combination characteristics and proportional displacement characteristics among different wave bands of a candidate area of the air weak small moving target, and determining the air weak small moving target by combining preset judgment conditions.
2. The method according to claim 1, wherein in step S31, specifically comprising:
and according to the three-dimensional binary image, taking a two-dimensional binary image of the first channel, acquiring a connected region of the image and a centroid of the connected region according to the two-dimensional image, and cutting the two-dimensional binary image into image slices by taking the centroid as a center.
3. An airborne weak moving object detection system, comprising:
the reading unit is used for reading satellite remote sensing data;
the determining unit is used for determining a candidate region of the aerial weak and small moving target by using the selected multiple wave bands;
an image processing unit for performing image slicing according to the target candidate region;
the determining unit is further used for determining the air weak and small moving target by combining the selected multiple wave bands;
the image processing unit is also used for carrying out coordinate and projection conversion on the detected aerial weak and small moving target image and outputting a detection result;
reading satellite remote sensing data, selecting a plurality of wave bands with the maximum spectral reflection intensity of a space moving target, and specifically comprising the following steps:
reading satellite remote sensing image data, wherein the satellite remote sensing image data is a satellite remote sensing image with imaging parallax between at least three different wave bands;
reading an original multispectral data source, acquiring the total number of wave bands, sequentially acquiring pixel matrixes of different wave bands, and storing the pixel matrixes and projection information together into a single-wave-band image;
under a wide-width imaging mode, analyzing the brightness characteristics of images of different wave bands, selecting a plurality of wave bands with the maximum spectral reflection intensity of a space moving target, and synthesizing a plurality of single-wave-band images to obtain a synthesized image;
reading an original multispectral data source, and determining a candidate region of an aerial weak and small moving target by using a plurality of selected wave bands, wherein the method specifically comprises the following steps:
extracting a reflection abnormal area of a moving target in a middle wave band in a plurality of wave bands based on wave band operation and logic operation;
determining a candidate area of the air weak and small moving target by using the reflection abnormal area according to a preset judgment condition;
selecting three wave bands with the maximum spectral reflection intensity of the space moving target, wherein the three wave bands are respectively a B1 wave band, a B2 wave band and a B3 wave band, and the method specifically comprises the following steps:
the B2 wave band is a middle wave band, difference value images of the B2 wave band, the B1 wave band and the B3 wave band are calculated and marked as B21 and B23, and the calculation formula is as follows:
B21[i,j]＝B2[i,j]-B1[i,j]，
B23[i,j]＝B2[i,j]-B3[i,j]，
b21[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B1 at the position (i, j), B23[ i, j ] refers to the pixel value of the difference image of the waveband B2 and the waveband B3 at the position (i, j), i and j are values from 1 to m and n respectively, and m and n are the height and width of the image respectively;
judging whether the pixel values of the pixel difference images B21 and B31 and the band B2 image are all in a preset range, and calculating to obtain a binary image B2gray [ i, j ], wherein the calculation formula is as follows:
Figure FDA0004066097850000051
wherein T1, T2 and T3 are preset threshold values;
the pixel with the pixel value of 1 in the obtained binary image is a reflection abnormal area of the moving target;
determining a candidate region of the air weak and small moving target by using the reflection abnormal region according to a preset judgment condition, wherein the method specifically comprises the following steps:
mapping the obtained two-dimensional binary image and the reflection abnormal area to a three-dimensional true color image to obtain a three-dimensional binary image B123gray, wherein the formula is as follows:
Figure FDA0004066097850000052
b123gray [ i, j, k ] represents a pixel value of a position (i, j) in a kth image channel, k is 0, 1 and 2, and an area with the pixel value of 1 is a candidate area of an aerial weak and small moving target;
obtaining a connected region of a three-dimensional image by using a region connection technology of image processing, and when the area of the connected region is smaller than a certain threshold value, making the pixel value of the B123gray be 0, and the reserved connected region be the region where the candidate aerial weak moving target is located;
performing image slicing according to the target candidate region, and determining the aerial weak and small moving target by combining the selected multiple wave bands, wherein the image slicing specifically comprises the following steps:
obtaining a region centroid according to the target candidate region, and cutting the region centroid into image slices by taking the centroid as a center;
based on image slices, respectively extracting reflection abnormal areas of a B1 wave band, a B2 wave band and a B3 wave band, extracting linear combination characteristics and proportional displacement characteristics among different wave bands of an air weak small moving target candidate area, and determining the air weak small moving target by combining preset judgment conditions.
4. An electronic device, comprising: one or more processors, one or more memories, and one or more computer programs; wherein a processor is connected to the memory, the one or more computer programs being stored in the memory, the processor executing the one or more computer programs stored in the memory when the electronic device is running, so as to cause the electronic device to perform the method for detecting a moving object with weak air as claimed in any one of claims 1-2.
5. A computer-readable storage medium storing computer instructions which, when executed by a processor, implement a method of airborne weak moving object detection as claimed in any one of claims 1-2.
CN202210923739.4A 2022-08-02 2022-08-02 Method, system, equipment and storage medium for detecting air weak and small moving target Active CN115294439B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210923739.4A CN115294439B (en) 2022-08-02 2022-08-02 Method, system, equipment and storage medium for detecting air weak and small moving target

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210923739.4A CN115294439B (en) 2022-08-02 2022-08-02 Method, system, equipment and storage medium for detecting air weak and small moving target

Publications (2)

Publication Number Publication Date
CN115294439A CN115294439A (en) 2022-11-04
CN115294439B true CN115294439B (en) 2023-04-07

Family

ID=83825445

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210923739.4A Active CN115294439B (en) 2022-08-02 2022-08-02 Method, system, equipment and storage medium for detecting air weak and small moving target

Country Status (1)

Country Link
CN (1) CN115294439B (en)

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103776540B (en) * 2013-12-30 2015-07-08 华中科技大学 Multiband common-optical-path spectrum combined remote sensing measurement system and method thereof
CN107256560B (en) * 2017-05-16 2020-02-14 北京环境特性研究所 Infrared weak and small target detection method and system thereof
CN111027497B (en) * 2019-12-17 2023-03-31 西安电子科技大学 Weak and small target rapid detection method based on high-resolution optical remote sensing image
CN112966635B (en) * 2021-03-19 2022-05-31 北京理工大学 Low-resolution time sequence remote sensing image-oriented moving ship detection method and device
CN114764801A (en) * 2022-05-19 2022-07-19 中国科学院微电子研究所 Weak and small ship target fusion detection method and device based on multi-vision significant features

Also Published As

Publication number Publication date
CN115294439A (en) 2022-11-04

Similar Documents

Publication Publication Date Title
US10839211B2 (en) Systems, methods and computer program products for multi-resolution multi-spectral deep learning based change detection for satellite images
Zhang et al. An unsupervised evaluation method for remotely sensed imagery segmentation
US20120328161A1 (en) Method and multi-scale attention system for spatiotemporal change determination and object detection
CN112712535B (en) Mask-RCNN landslide segmentation method based on simulation difficult sample
CN111814771B (en) Image processing method and device
WO2013088175A1 (en) Image processing method
CN111027497B (en) Weak and small target rapid detection method based on high-resolution optical remote sensing image
CN111401387B (en) Abnormal sample construction method, device, computer equipment and storage medium
CN111079596A (en) System and method for identifying typical marine artificial target of high-resolution remote sensing image
JP2018072938A (en) Number-of-targets estimation device, number-of-targets estimation method, and program
EP2447884B1 (en) Method for detecting and recognising an object in an image, and an apparatus and a computer program therefor
CN116091497B (en) Remote sensing change detection method, device, electronic equipment and storage medium
Zhao et al. SAR ship detection based on end-to-end morphological feature pyramid network
Che et al. 2-and 3-D urban change detection with quad-PolSAR data
CN109934170B (en) Mine resource statistical method based on computer vision
CN109409285B (en) Remote sensing video target detection method based on overlapped slices
CN114155278A (en) Target tracking and related model training method, related device, equipment and medium
CN107369163B (en) Rapid SAR image target detection method based on optimal entropy dual-threshold segmentation
CN115294439B (en) Method, system, equipment and storage medium for detecting air weak and small moving target
US7283664B2 (en) Interactive computer aided analysis of images from sensor combinations
Albrecht et al. Visual maritime attention using multiple low-level features and naive bayes classification
CN112651351B (en) Data processing method and device
CN114639015A (en) High-resolution image landslide detection method combining spectrum, vegetation index and textural features
CN111079797B (en) Image classification method and device and storage medium
CN113627292A (en) Remote sensing image identification method and device based on converged network

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
GR01 Patent grant
GR01 Patent grant