CN110765944A - Target identification method, device, equipment and medium based on multi-source remote sensing image - Google Patents
Target identification method, device, equipment and medium based on multi-source remote sensing image Download PDFInfo
- Publication number
- CN110765944A CN110765944A CN201911013119.1A CN201911013119A CN110765944A CN 110765944 A CN110765944 A CN 110765944A CN 201911013119 A CN201911013119 A CN 201911013119A CN 110765944 A CN110765944 A CN 110765944A
- Authority
- CN
- China
- Prior art keywords
- target
- remote sensing
- suspected
- visible light
- image
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/194—Terrestrial scenes using hyperspectral data, i.e. more or other wavelengths than RGB
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The application discloses a target identification method, a device, equipment and a medium based on multi-source remote sensing images, wherein the method comprises the following steps: constructing a feature database of the target; collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment mounted on a remote sensing platform; respectively carrying out comparative analysis on the multispectral image and spectral characteristics and spatial characteristics in a characteristic database, and selecting a suspected target from the targets to be identified according to an analysis result; controlling an optical axis of a visible light camera installed on a remote sensing platform to point to a suspected target for fixed-point shooting, and acquiring a visible light image; and comparing and analyzing the visible light image and the morphological characteristics in the characteristic database, and identifying the target according to the analysis result. The method comprehensively utilizes the multispectral remote sensing technology and the visible light remote sensing technology, and realizes the identification of the target through the large sample data comparison based on the combined analysis of the spectral characteristics and the spatial characteristics of the object in the aerial photographing area by the multisource remote sensing data, thereby improving the identification efficiency and the accuracy.
Description
Technical Field
The invention relates to the technical field of remote sensing, in particular to a target identification method, a target identification device, target identification equipment and a target identification medium based on a multi-source remote sensing image.
Background
The remote sensing technology is performed in the air by using various satellites, airplanes, airships, balloons, and the like as sensor vehicles. At present, the remote sensing technology is an effective way for target identification, and has the advantages of mature technology, large imaging scale, high resolution, suitability for large-area topographic mapping and small-area detailed investigation, no need of complex processing equipment and the like, but the object type is complex, the characteristics are various, and the efficiency and precision requirements of target identification are difficult to meet only by a single remote sensing means.
Therefore, how to improve the recognition efficiency of the target and ensure the recognition accuracy thereof is a technical problem to be urgently solved by the technical personnel in the field.
Disclosure of Invention
In view of the above, the present invention provides a method, an apparatus, a device and a medium for identifying a target based on a multi-source remote sensing image, which can improve the identification efficiency and the identification accuracy of the target. The specific scheme is as follows:
a target identification method based on multi-source remote sensing images comprises the following steps:
constructing a feature database of the target;
collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment mounted on a remote sensing platform;
respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to an analysis result;
controlling an optical axis of a visible light camera installed on a remote sensing platform to point to the suspected target for fixed-point shooting, and acquiring a visible light image of the suspected target;
and comparing and analyzing the acquired visible light image and the morphological characteristics in the constructed characteristic database, and identifying the target according to the analysis result.
Preferably, in the object identification method based on multi-source remote sensing imagery provided in the embodiment of the present invention, before acquiring, by a spectrum measurement device installed on a remote sensing platform, a multispectral image of all objects to be identified in an aerial photography area, the method further includes:
and determining the characteristic wave band of the target according to the built spectral characteristics in the characteristic database.
Preferably, in the method for identifying a target based on a multi-source remote sensing image provided in the embodiment of the present invention, the comparing and analyzing the collected multispectral image with the spectral feature and the spatial feature in the constructed feature database, and selecting a suspected target from the targets to be identified according to an analysis result includes:
carrying out spectrum contrast analysis on the collected multispectral image and the built spectral features in the feature database, and screening out a first suspected target from the target to be identified through a target screening algorithm;
and in the screened area where the first suspected target is located, carrying out similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database, extracting the first suspected target with similarity higher than a first set threshold value, and defining the first suspected target as a second suspected target.
Preferably, in the method for identifying a target based on a multi-source remote sensing image provided in an embodiment of the present invention, the step of comparing and analyzing the acquired visible light image with morphological features in the constructed feature database, and identifying the target according to an analysis result specifically includes:
and performing similarity analysis on the acquired visible light image of the second suspected target and the constructed morphological characteristics in the characteristic database, and determining the second suspected target with the similarity higher than a second set threshold as the target.
Preferably, in the object identification method based on the multi-source remote sensing image provided in the embodiment of the present invention, the method further includes:
and obtaining evidence of the identified target.
Preferably, in the object identification method based on the multi-source remote sensing image provided in the embodiment of the present invention, the obtaining evidence of the identified object specifically includes:
constructing a multi-source load coordinate conversion model in the motion process of the remote sensing platform;
calculating the geographic coordinate position of the identified target according to the constructed multi-source load coordinate conversion model;
and archiving the calculated geographic position and the acquired visible light image of the target to finish the evidence obtaining of the target.
The embodiment of the invention also provides a target identification device based on the multi-source remote sensing image, which comprises the following components:
the database construction module is used for constructing a characteristic database of the target;
the spectral measurement equipment is arranged on the remote sensing platform and is used for collecting multispectral images of all targets to be identified in the aerial photography area;
the suspected target extraction module is used for respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to an analysis result;
the visible light camera is arranged on the remote sensing platform and used for shooting the suspected target at a fixed point to obtain a visible light image of the suspected target;
and the sensitive target identification module is used for comparing and analyzing the acquired visible light image and the morphological characteristics in the constructed characteristic database and identifying the target according to the analysis result.
Preferably, in the object recognition device based on the multi-source remote sensing image provided in the embodiment of the present invention, the suspected object extraction module specifically includes:
the first suspected target screening unit is used for carrying out spectrum comparison analysis on the collected multispectral image and the built spectral features in the feature database, and screening out a first suspected target from the target to be identified through a target screening algorithm;
and the second suspected target extracting unit is used for carrying out similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database in the screened area where the first suspected target is located, extracting the first suspected target with the similarity higher than a first set threshold value, and defining the first suspected target as a second suspected target.
The embodiment of the invention also provides target identification equipment based on the multi-source remote sensing image, which comprises a processor and a memory, wherein the target identification method based on the multi-source remote sensing image provided by the embodiment of the invention is realized when the processor executes a computer program stored in the memory.
The embodiment of the invention also provides a computer-readable storage medium for storing a computer program, wherein the computer program is executed by a processor to implement the above target identification method based on multi-source remote sensing images provided by the embodiment of the invention.
According to the technical scheme, the target identification method, the device, the equipment and the medium based on the multi-source remote sensing image, which are provided by the invention, comprise the following steps: constructing a feature database of the target; collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment mounted on a remote sensing platform; respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to the analysis result; controlling an optical axis of a visible light camera installed on a remote sensing platform to point to a suspected target for fixed-point shooting, and acquiring a visible light image of the suspected target; and comparing and analyzing the acquired visible light image with morphological characteristics in the constructed characteristic database, and identifying the target according to the analysis result.
The invention comprehensively utilizes the multispectral remote sensing technology and the visible light remote sensing technology, fully excavates effective information in the multispectral image and the visible light image, jointly analyzes the spectral characteristics and the spatial characteristics of an object in an aerial photographing area based on multisource remote sensing data, compares multidimensional information, and realizes the identification of a certain target according to the analysis result of large sample data, thereby improving the identification efficiency of the target and ensuring the identification accuracy of the target.
Drawings
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 description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a target identification method based on a multi-source remote sensing image according to an embodiment of the present invention;
fig. 2 is a specific flowchart of a target identification method based on a multi-source remote sensing image according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a target identification device based on a multi-source remote sensing image according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention provides a target identification method based on multi-source remote sensing images, which comprises the following steps as shown in figure 1:
s101, constructing a feature database of a target;
in practical applications, for a target sample, data may be collected through a plurality of fields, a plurality of measurement modes, a plurality of shooting conditions, and the like, for example, a spectral measurement device (e.g., a multispectral camera) collects a multispectral image of the target sample, reflection spectral information and spatial information (e.g., a shape, a contour, and the like) of the multispectral image may be obtained from the multispectral image, a visible light camera collects a high-resolution visible light image of the target sample, and detailed information of morphological characteristics of the high-resolution visible light image may be interpreted from the high-resolution visible light image; therefore, the spectral characteristics and the spatial characteristics of the reflection information of a specific target sample are analyzed, the reflection spectrum and the morphological characteristics of the specific target sample are grasped, and the obtained characteristic big data are collated to construct a more comprehensive target characteristic database.
S102, collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment installed on a remote sensing platform;
it should be noted that the spectral measurement device used in step S101 may be located at any position in order to complete the database, and the spectral measurement device is installed on the remote sensing platform when step S102 is executed. When the remote sensing platform takes off stably and vertically to the ground by one key, and the height reaches a certain set threshold value, the overlapping degree triggering mode of the spectrum measuring equipment is triggered. The remote sensing platform can be manually operated through a software login interface, can automatically plan a route operation (the area on a map can be selected by a frame by clicking the area on the map), and can also be intelligently operated according to a conventional route. Specific parameter settings on the software login interface comprise course overlapping degree (with default overlapping degree, the overlapping degree can be changed) of the spectrum measurement equipment, flying speed, flying height and the like of the remote sensing platform, residual time length, flying area and the like can be displayed, and one-key return voyage is supported. In operation, the firmware versions of the remote sensing platform and the load need to be detected, and the battery electric quantity, compass state, magnetic field interference degree, residual storage capacity of the SD card of the camera, and the like of the remote sensing platform, the remote controller and the station software are monitored in real time. The specific navigation page can use a satellite map as a base map, displays the flight position, the flight track and the planned route in a real-time overlapping mode, and also displays marks of exposure points of the spectral measurement equipment.
S103, respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to the analysis result;
specifically, multispectral data reading is carried out on the collected multispectral images, suspected targets are selected, and the areas where the suspected targets are located are subjected to next shooting and proofreading analysis, so that the complex background can be effectively suppressed, the scene does not need to be subjected to global calculation, the calculation amount is effectively reduced, and the working efficiency is improved.
S104, controlling an optical axis of a visible light camera installed on the remote sensing platform to point to the suspected target for fixed-point shooting, and obtaining a visible light image of the suspected target;
in practical applications, the visible light camera is initialized to the default focal length. And synchronously controlling the visible light camera to enter a normal monitoring state during takeoff, namely when the specified flying height is reached, the visible light camera is vertically opposite to the ground and zooms to an initial default focal length. The real-time video display frame of the visible light camera is interchanged with one key of the position, the size and the like of a navigation page, and the collected visible light images are high-resolution and can interpret the detail information of the suspected target morphological characteristics.
S105, comparing and analyzing the acquired visible light image with morphological characteristics in the constructed characteristic database, and identifying a target according to an analysis result;
in practical application, the high-resolution visible light image is compared with the high-spatial-resolution morphological characteristics of the target in the characteristic database for analysis, the identified suspected target is the final identification result of the target, and the identification method comprises but is not limited to manual identification, automatic algorithm identification, intelligent identification and the like.
In the target identification method based on the multi-source remote sensing image provided by the embodiment of the invention, the multispectral remote sensing technology and the visible light remote sensing technology are comprehensively utilized, effective information in the multispectral image and the visible light image is fully mined, the spectral characteristic and the spatial characteristic of an object in an aerial photographing area are jointly analyzed based on the multi-source remote sensing data, the multi-dimensional information is compared, and identification of a certain target is realized according to a large sample data analysis result, so that the identification efficiency of the target is improved, and the identification accuracy of the target is ensured.
It should be noted that the invention jointly applies multisource remote sensing data to perform target identification based on multisource remote sensing images, wherein the multisource remote sensing data includes but is not limited to multispectral remote sensing data and visible light remote sensing data.
Further, in specific implementation, in the method for identifying an object based on a multi-source remote sensing image provided in the embodiment of the present invention, as shown in fig. 2, the method may further include:
and S106, obtaining evidence of the identified target.
The step S106 may specifically include: firstly, constructing a multi-source load coordinate conversion model in the motion process of a remote sensing platform; then, calculating the geographic coordinate position of the identified target according to the constructed multi-source load coordinate conversion model; and finally, archiving the calculated geographic position and the acquired visible light image of the target to finish the evidence obtaining of the target. In practical application, the calculation of the geographic coordinate position can be obtained through information such as a central point GPS, optical system parameters, pixel positions and the like.
Further, in specific implementation, in the above object identification method based on multi-source remote sensing images provided by the embodiment of the present invention, before the step S102 is executed to collect multispectral images of all objects to be identified in an aerial photographing area through a spectrum measurement device installed on a remote sensing platform, as shown in fig. 2, the method may further include the following steps:
s201, determining a characteristic wave band of the target according to the spectral characteristics in the constructed characteristic database.
According to the characteristic wave band of the target, when the step S102 is executed, the multispectral images of all targets to be identified in the aerial photography area only need to be collected corresponding to the characteristic wave band, so that the identification efficiency can be further improved.
Further, in a specific implementation, in the above target identification method based on a multi-source remote sensing image provided in the embodiment of the present invention, as shown in fig. 2, step S103 is to compare and analyze the collected multi-spectral image with a spectral feature and a spatial feature in the constructed feature database, and select a suspected target from targets to be identified according to an analysis result, which may specifically include the following steps:
s202, carrying out spectrum contrast analysis on the collected multispectral image and the spectral characteristics in the constructed characteristic database, and screening out a first suspected target from the targets to be identified through a target screening algorithm;
in practical application, the target screening algorithm is constructed by the spectral information of the characteristic wave band, and is an algorithm based on spectral dimension and spatial dimension characteristic analysis of a multispectral image.
S203, in the area where the screened first suspected target is located, carrying out similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database, extracting the first suspected target with similarity higher than a first set threshold value, and defining the first suspected target as a second suspected target.
And when the single-channel multispectral image of the first suspected target is received, calling a similarity analysis algorithm to identify the form of the first suspected target, eliminating obvious false extraction, and determining the second suspected target. At this time, step S104 may directly execute controlling the optical axis of the visible light camera installed on the remote sensing platform to point to the second suspected target for fixed-point shooting, so as to obtain a visible light image of the second suspected target.
Further, in a specific implementation, in the above method for identifying a target based on a multi-source remote sensing image provided in the embodiment of the present invention, step S105 may compare and analyze the acquired visible light image with morphological features in the constructed feature database, and identify the target according to an analysis result, which specifically includes the following steps:
and S204, performing similarity analysis on the acquired visible light image of the second suspected target and morphological characteristics in the constructed characteristic database, and determining the second suspected target with the similarity higher than a second set threshold as the target.
Therefore, reasonable feature similarity threshold setting is carried out according to the analysis result of the large sample data, and the accuracy of the identification result can be further ensured. When the second suspected target is obtained, an alarm can be synchronously given at the corresponding positions of all the second suspected targets on the base map of the navigation page. And single-point proofreading or one-key proofreading can be carried out according to the alarm position.
In the specific operation process, when the identification method is manual identification, any alarm icon can be clicked, the unmanned aerial vehicle is triggered to hover, the exposure of the spectrum measuring equipment is stopped, meanwhile, the pitch angle and the azimuth angle of the optical axis of the visible light camera pointing to the alarm position are automatically calculated, the optical axis of the visible light camera is controlled to point to the alarm position, the focal distance of the visible light camera is lengthened to a certain specific value, fixed-point video stream shooting is carried out, and the visible light video popup frame is displayed in real time (the visible light camera real-time video display frame is automatically hidden in a conventional state); clicking a 'confirmation' button, shooting a static image by a visible light camera, archiving the static image and corresponding position information together, and prompting 'please continue after evidence obtaining is finished'; clicking the ignore button, enabling the video popup frame at the corresponding position to disappear (the visible light camera real-time video display frame automatically appears in a conventional state), prompting 'please continue', and manually judging the alarm icon point by point; and automatically detecting whether all the second suspected targets under the current view are processed or not, and if yes, popping a frame to prompt that whether the processing is finished or whether the flying is continued or not. Clicking 'yes', continuously flying by the unmanned aerial vehicle, enabling the spectrum measuring equipment to enter an overlapping rate triggering mode, and enabling the visible light camera to enter a conventional monitoring state; clicking 'no', hovering the unmanned aerial vehicle, and waiting for a further instruction; when the planning task is completed, the user can return to the working mode selection interface.
When the identification method is automatic identification, if a second suspected target exists, the unmanned aerial vehicle is automatically triggered to hover, the exposure of the spectral measurement equipment is stopped, and the judgment sequence of the second suspected target is calculated; then, automatically calculating the pitch angles and azimuth angles of the optical axis of the visible light camera pointing to all alarm positions; then, controlling the focal length of the visible light camera to be lengthened to a certain specific value, sequentially controlling the optical axis of the visible light camera to point to all 'alarm' positions, and shooting a high-resolution visible light image; and after receiving the visible light image, calling a background form recognition algorithm to give the similarity, and inlaying and displaying a similarity value at a corresponding position. At this time, the single point collation mode may be selected: clicking any alarm icon, amplifying and displaying a corresponding visible light image suspension frame (automatically hiding a visible light camera real-time video display frame in a conventional state), namely clicking a confirmation button, archiving a calculation result and a visible light image, prompting that evidence obtaining is finished and the user continues, clicking a neglect button, and prompting that the user continues; the alarm icons can be checked point by point; and automatically detecting whether all second suspected targets under the current view are processed or not, if so, popping a frame to prompt that whether the processing is finished or not and whether the flying is continued or not, namely clicking yes, continuing to execute the rest route, restoring the interface to the conventional monitoring state, clicking no, hovering the unmanned aerial vehicle, restoring the interface to the conventional monitoring state, and waiting for a next instruction. One-touch collation mode may also be selected: clicking 'one-key confirmation', archiving a suspected target calculation result and a visible light image under the current view, recovering the interface to a conventional monitoring state, automatically continuing the journey, and prompting 'evidence obtaining is finished, and automatically continuing the journey' by the interface; clicking 'one-key recovery', recovering the interface to a conventional monitoring state, automatically continuing the journey, and prompting 'proofreading is finished and automatically continuing the journey' by the interface; and after the planning task is finished, automatically returning and returning to the working mode selection interface.
When the identification method is intelligent identification, if a second suspected target exists, the unmanned aerial vehicle is automatically triggered to hover, the exposure of the spectral measurement equipment is stopped, and the judgment sequence of the second suspected target is calculated; then, automatically calculating the pitch angles and azimuth angles of the optical axis of the visible light camera pointing to all alarm positions; then, controlling the focal length of the visible light camera to be lengthened to a certain fixed value (long focal length state), sequentially controlling the optical axis of the visible light camera to point to all 'alarm' positions, and shooting a high-resolution visible light image; after receiving the visible light image, calling a background form recognition algorithm, giving out similarity, setting a second set threshold (with a default value and capable of being changed) of the similarity, and automatically archiving a calculation result and the visible light image when the similarity of the second suspected target is greater than the second set threshold; the similarity is displayed in a mosaic mode at the corresponding position; after the identification of the second suspected target under the current view is completed, automatic cruising is carried out, the overlapping rate triggering mode of the spectrum measuring equipment is started, the cradle head is stable and vertical to the ground, and the visible light camera enters a conventional monitoring state; and after the planning task is finished, automatically returning and returning to the working mode selection interface.
Based on the same invention concept, the embodiment of the invention also provides a target recognition device based on the multi-source remote sensing image, and as the principle of solving the problem of the target recognition device based on the multi-source remote sensing image is similar to that of the target recognition method based on the multi-source remote sensing image, the implementation of the target recognition device based on the multi-source remote sensing image can refer to the implementation of the target recognition method based on the multi-source remote sensing image, and repeated parts are not repeated.
In specific implementation, the target identification device based on the multi-source remote sensing image provided by the embodiment of the present invention, as shown in fig. 3, specifically includes:
a database construction module 11, configured to construct a feature database of a target;
the spectral measurement equipment 12 is arranged on the remote sensing platform and used for collecting multispectral images of all targets to be identified in the aerial photography area;
a suspected target extraction module 13, configured to compare and analyze the acquired multispectral image with spectral features and spatial features in the constructed feature database, respectively, and select a suspected target from the targets to be identified according to an analysis result;
the visible light camera 14 is installed on the remote sensing platform and used for shooting the suspected target at a fixed point to obtain a visible light image of the suspected target;
and the sensitive target identification module 15 is used for comparing and analyzing the acquired visible light image and morphological characteristics in the constructed characteristic database, and identifying a target according to an analysis result.
In the target recognition device based on the multi-source remote sensing image provided by the embodiment of the invention, effective information in a multispectral image and a visible light image can be fully mined, the spectral characteristic and the spatial characteristic of a target are jointly analyzed, the multidimensional information is compared, reasonable judgment is carried out according to a large sample data analysis result, the accuracy of a target recognition result based on the multi-source remote sensing image is ensured, and the target recognition efficiency is improved.
Further, in a specific implementation, in the object recognition device based on the multi-source remote sensing image provided in the embodiment of the present invention, as shown in fig. 3, the suspected object extracting module 13 may specifically include:
the first suspected target screening unit 131 is configured to perform spectrum comparison analysis on the acquired multispectral image and the spectral features in the constructed feature database, and screen out a first suspected target from the targets to be identified through a target screening algorithm;
the second suspected target extracting unit 132 is configured to perform similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database in the region where the screened first suspected target is located, extract the first suspected target with the similarity higher than the first set threshold, and define the first suspected target as the second suspected target.
Further, in a specific implementation, in the object recognition apparatus based on a multi-source remote sensing image provided in an embodiment of the present invention, the apparatus may further include: and the sensitive target forensics module is used for forensics of the identified target. Specifically, the sensitive target forensics module is specifically used for constructing a multi-source load coordinate conversion model in the motion process of the remote sensing platform; calculating the geographic coordinate position of the identified target according to the constructed multi-source load coordinate conversion model; and archiving the calculated position of the geographic landmark and the acquired visible light image of the target to finish the evidence obtaining of the target.
For more specific working processes of the modules and the components, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not described here again.
Correspondingly, the embodiment of the invention also discloses target identification equipment based on the multi-source remote sensing image, which comprises a processor and a memory; when the processor executes the computer program stored in the memory, the target identification method based on the multi-source remote sensing image disclosed by the embodiment is realized.
For more specific processes of the above method, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
Further, the present invention also discloses a computer readable storage medium for storing a computer program; when being executed by a processor, the computer program realizes the object identification method based on the multi-source remote sensing image.
For more specific processes of the above method, reference may be made to corresponding contents disclosed in the foregoing embodiments, and details are not repeated here.
The embodiments are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same or similar parts among the embodiments are referred to each other. The device, the equipment and the storage medium disclosed by the embodiment correspond to the method disclosed by the embodiment, so that the description is relatively simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The embodiment of the invention provides a target identification method, a device, equipment and a medium based on a multi-source remote sensing image, which comprises the following steps: constructing a feature database of the target; collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment mounted on a remote sensing platform; respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to the analysis result; controlling an optical axis of a visible light camera installed on a remote sensing platform to point to a suspected target for fixed-point shooting, and acquiring a visible light image of the suspected target; and comparing and analyzing the acquired visible light image with morphological characteristics in the constructed characteristic database, and identifying the target according to the analysis result. The invention comprehensively utilizes the multispectral remote sensing technology and the visible light remote sensing technology, fully excavates effective information in the multispectral image and the visible light image, jointly analyzes the spectral characteristics and the spatial characteristics of an object in an aerial photographing area based on multisource remote sensing data, compares multidimensional information, and realizes the identification of a certain target according to the analysis result of large sample data, thereby improving the identification efficiency of the target and ensuring the identification accuracy of the target.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be 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. Also, 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 an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The method, the device, the equipment and the medium for identifying the target based on the multi-source remote sensing image provided by the invention are introduced in detail, a specific example is applied in the text to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (10)
1. A target identification method based on multi-source remote sensing images is characterized by comprising the following steps:
constructing a feature database of the target;
collecting multispectral images of all targets to be identified in an aerial photography area through spectral measurement equipment mounted on a remote sensing platform;
respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to an analysis result;
controlling an optical axis of a visible light camera installed on a remote sensing platform to point to the suspected target for fixed-point shooting, and acquiring a visible light image of the suspected target;
and comparing and analyzing the acquired visible light image and the morphological characteristics in the constructed characteristic database, and identifying the target according to the analysis result.
2. The method for identifying the target based on the multisource remote sensing image according to claim 1, wherein before collecting multispectral images of all targets to be identified in the aerial photographing area through a spectral measurement device installed on a remote sensing platform, the method further comprises the following steps:
and determining the characteristic wave band of the target according to the built spectral characteristics in the characteristic database.
3. The method for identifying a target based on a multi-source remote sensing image according to claim 2, wherein the step of comparing and analyzing the collected multi-spectral image with spectral features and spatial features in the constructed feature database respectively, and selecting a suspected target from the targets to be identified according to an analysis result specifically comprises the steps of:
carrying out spectrum contrast analysis on the collected multispectral image and the built spectral features in the feature database, and screening out a first suspected target from the target to be identified through a target screening algorithm;
and in the screened area where the first suspected target is located, carrying out similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database, extracting the first suspected target with similarity higher than a first set threshold value, and defining the first suspected target as a second suspected target.
4. The method for identifying the target based on the multi-source remote sensing image according to claim 3, wherein the step of comparing and analyzing the acquired visible light image and the morphological features in the constructed feature database and identifying the target according to the analysis result specifically comprises the steps of:
and performing similarity analysis on the acquired visible light image of the second suspected target and the constructed morphological characteristics in the characteristic database, and determining the second suspected target with the similarity higher than a second set threshold as the target.
5. The method for identifying the target based on the multi-source remote sensing image according to claim 1, further comprising:
and obtaining evidence of the identified target.
6. The method for identifying the target based on the multi-source remote sensing image according to claim 5, wherein the evidence obtaining of the identified target specifically comprises:
constructing a multi-source load coordinate conversion model in the motion process of the remote sensing platform;
calculating the geographic coordinate position of the identified target according to the constructed multi-source load coordinate conversion model;
and archiving the calculated geographic position and the acquired visible light image of the target to finish the evidence obtaining of the target.
7. A target recognition device based on multi-source remote sensing images is characterized by comprising:
the database construction module is used for constructing a characteristic database of the target;
the spectral measurement equipment is arranged on the remote sensing platform and is used for collecting multispectral images of all targets to be identified in the aerial photography area;
the suspected target extraction module is used for respectively carrying out comparative analysis on the collected multispectral image and the spectral characteristics and the spatial characteristics in the constructed characteristic database, and selecting a suspected target from the targets to be identified according to an analysis result;
the visible light camera is arranged on the remote sensing platform and used for shooting the suspected target at a fixed point to obtain a visible light image of the suspected target;
and the sensitive target identification module is used for comparing and analyzing the acquired visible light image and the morphological characteristics in the constructed characteristic database and identifying the target according to the analysis result.
8. The object recognition device based on the multi-source remote sensing image according to claim 7, wherein the suspected object extraction module specifically comprises:
the first suspected target screening unit is used for carrying out spectrum comparison analysis on the collected multispectral image and the built spectral features in the feature database, and screening out a first suspected target from the target to be identified through a target screening algorithm;
and the second suspected target extracting unit is used for carrying out similarity analysis on the spatial features of the multispectral image of the first suspected target and the spatial features in the feature database in the screened area where the first suspected target is located, extracting the first suspected target with the similarity higher than a first set threshold value, and defining the first suspected target as a second suspected target.
9. An object recognition device based on multi-source remote sensing images, which is characterized by comprising a processor and a memory, wherein the processor executes a computer program stored in the memory to realize the object recognition method based on multi-source remote sensing images according to any one of claims 1 to 6.
10. A computer-readable storage medium storing a computer program, wherein the computer program is executed by a processor to implement the method for object recognition based on multi-source remote sensing images according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911013119.1A CN110765944A (en) | 2019-10-23 | 2019-10-23 | Target identification method, device, equipment and medium based on multi-source remote sensing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911013119.1A CN110765944A (en) | 2019-10-23 | 2019-10-23 | Target identification method, device, equipment and medium based on multi-source remote sensing image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110765944A true CN110765944A (en) | 2020-02-07 |
Family
ID=69333250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911013119.1A Pending CN110765944A (en) | 2019-10-23 | 2019-10-23 | Target identification method, device, equipment and medium based on multi-source remote sensing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110765944A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111597377A (en) * | 2020-04-08 | 2020-08-28 | 广东省国土资源测绘院 | Deep learning technology-based field investigation method and system |
| CN111924101A (en) * | 2020-08-31 | 2020-11-13 | 金陵科技学院 | Unmanned aerial vehicle double-tripod-head camera and working method thereof |
| CN115393813A (en) * | 2022-08-18 | 2022-11-25 | 中国人民公安大学 | Road identification method, device and equipment based on remote sensing image and storage medium |
| CN116630825A (en) * | 2023-06-09 | 2023-08-22 | 北京佳格天地科技有限公司 | Satellite remote sensing data and monitoring video fusion method and system |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102789578A (en) * | 2012-07-17 | 2012-11-21 | 北京市遥感信息研究所 | Infrared remote sensing image change detection method based on multi-source target characteristic support |
| CN103090868A (en) * | 2012-12-31 | 2013-05-08 | 华中科技大学 | Method for preparing typical ground object reference picture by using multispectral image classification |
| CN105303184A (en) * | 2015-11-25 | 2016-02-03 | 中国矿业大学(北京) | Method for accurately identifying ground features in satellite remote-sensing image |
| CN106651880A (en) * | 2016-12-27 | 2017-05-10 | 首都师范大学 | Method for detecting marine moving target of thermal infrared remote sensing image based on multi-feature fusion |
| US20170358190A1 (en) * | 2014-12-17 | 2017-12-14 | Honeywell International Inc. | Detection system and method featuring multispectral imaging device |
| CN107609534A (en) * | 2017-09-28 | 2018-01-19 | 北京市遥感信息研究所 | An automatic testing method of mooring a boat is stayed in a kind of remote sensing based on harbour spectral information |
| CN109800380A (en) * | 2019-01-30 | 2019-05-24 | 上海卫星工程研究所 | The tight imaging geometry model building method that satellite-borne microwave remote sensing instrument detects over the ground |
| CN109918531A (en) * | 2019-03-06 | 2019-06-21 | 长光禹辰信息技术与装备(青岛)有限公司 | A kind of the seeking method, apparatus and computer readable storage medium of mother drug plants |
| CN110188716A (en) * | 2019-06-04 | 2019-08-30 | 中国煤炭地质总局勘查研究总院 | A kind of mineralization information extracting method and system |
-
2019
- 2019-10-23 CN CN201911013119.1A patent/CN110765944A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102789578A (en) * | 2012-07-17 | 2012-11-21 | 北京市遥感信息研究所 | Infrared remote sensing image change detection method based on multi-source target characteristic support |
| CN103090868A (en) * | 2012-12-31 | 2013-05-08 | 华中科技大学 | Method for preparing typical ground object reference picture by using multispectral image classification |
| US20170358190A1 (en) * | 2014-12-17 | 2017-12-14 | Honeywell International Inc. | Detection system and method featuring multispectral imaging device |
| CN105303184A (en) * | 2015-11-25 | 2016-02-03 | 中国矿业大学(北京) | Method for accurately identifying ground features in satellite remote-sensing image |
| CN106651880A (en) * | 2016-12-27 | 2017-05-10 | 首都师范大学 | Method for detecting marine moving target of thermal infrared remote sensing image based on multi-feature fusion |
| CN107609534A (en) * | 2017-09-28 | 2018-01-19 | 北京市遥感信息研究所 | An automatic testing method of mooring a boat is stayed in a kind of remote sensing based on harbour spectral information |
| CN109800380A (en) * | 2019-01-30 | 2019-05-24 | 上海卫星工程研究所 | The tight imaging geometry model building method that satellite-borne microwave remote sensing instrument detects over the ground |
| CN109918531A (en) * | 2019-03-06 | 2019-06-21 | 长光禹辰信息技术与装备(青岛)有限公司 | A kind of the seeking method, apparatus and computer readable storage medium of mother drug plants |
| CN110188716A (en) * | 2019-06-04 | 2019-08-30 | 中国煤炭地质总局勘查研究总院 | A kind of mineralization information extracting method and system |
Non-Patent Citations (2)
| Title |
|---|
| LIU CHUANLI 等: "Research on plant information identification, extraction and change monitoring based on multi-source remote sensing data", 《2010 SECOND IITA INTERNATIONAL CONFERENCE ON GEOSCIENCE AND REMOTE SENSING》 * |
| 戴玉超 等: "深度残差网络的多光谱遥感图像显著目标检测", 《测绘学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111597377A (en) * | 2020-04-08 | 2020-08-28 | 广东省国土资源测绘院 | Deep learning technology-based field investigation method and system |
| CN111924101A (en) * | 2020-08-31 | 2020-11-13 | 金陵科技学院 | Unmanned aerial vehicle double-tripod-head camera and working method thereof |
| CN111924101B (en) * | 2020-08-31 | 2024-04-09 | 金陵科技学院 | Unmanned aerial vehicle double-cradle head camera and working method thereof |
| CN115393813A (en) * | 2022-08-18 | 2022-11-25 | 中国人民公安大学 | Road identification method, device and equipment based on remote sensing image and storage medium |
| CN115393813B (en) * | 2022-08-18 | 2023-05-02 | 中国人民公安大学 | Road identification method, device, equipment and storage medium based on remote sensing image |
| CN116630825A (en) * | 2023-06-09 | 2023-08-22 | 北京佳格天地科技有限公司 | Satellite remote sensing data and monitoring video fusion method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110765944A (en) | Target identification method, device, equipment and medium based on multi-source remote sensing image | |
| US7630990B2 (en) | Endmember spectrum database construction method, endmember spectrum database construction apparatus and endmember spectrum database construction program | |
| US8855427B2 (en) | Systems and methods for efficiently and accurately detecting changes in spatial feature data | |
| US8832593B2 (en) | Systems and methods for efficient spatial feature analysis | |
| JP5535025B2 (en) | Outdoor feature detection system, program for outdoor feature detection system, and recording medium for program for outdoor feature detection system | |
| CN113196331A (en) | Application service providing device and method using satellite image | |
| CN108445489A (en) | The method for determining tropical agriculture loss based on satellite remote sensing and unmanned aerial vehicle remote sensing | |
| JP2017102672A (en) | Geographic position information specification system and geographic position information specification method | |
| CN114252884A (en) | Method and device for positioning and monitoring roadside radar, computer equipment and storage medium | |
| CN113741510A (en) | Routing inspection path planning method and device and storage medium | |
| JP2008287379A (en) | Road sign data input system | |
| CN111556157A (en) | Crop distribution monitoring method, equipment, storage medium and device | |
| JP6828448B2 (en) | Information processing equipment, information processing systems, information processing methods, and information processing programs | |
| CN111881984A (en) | Target detection method and device based on deep learning | |
| JP4788447B2 (en) | Image processing program and image processing apparatus | |
| US20150317828A1 (en) | Method and system for geo-referencing at least one sensor image | |
| JP2022018783A (en) | Inspection method | |
| CN115112100B (en) | Remote sensing control system and method | |
| CN116052023A (en) | Three-dimensional point cloud-based electric power inspection ground object classification method and storage medium | |
| JP6060377B2 (en) | Image cluster analyzer | |
| CN113096129B (en) | Method and device for detecting cloud cover in hyperspectral satellite image | |
| JP2005056186A (en) | Traffic condition observation system | |
| JP6524794B2 (en) | Object identification apparatus, object identification system, object identification method and object identification program | |
| EP3816938A1 (en) | Region clipping method and recording medium storing region clipping program | |
| JP6509546B2 (en) | Image search system and image search method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200207 |
|
| RJ01 | Rejection of invention patent application after publication |