CN115908475B - Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function - Google Patents
Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function Download PDFInfo
- Publication number
- CN115908475B CN115908475B CN202310220865.8A CN202310220865A CN115908475B CN 115908475 B CN115908475 B CN 115908475B CN 202310220865 A CN202310220865 A CN 202310220865A CN 115908475 B CN115908475 B CN 115908475B
- Authority
- CN
- China
- Prior art keywords
- feature point
- image
- point set
- photoelectric
- airborne
- 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
Links
Images
Classifications
-
- 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
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a realization method and a system for an airborne photoelectric reconnaissance pod image pre-tracking function, wherein the method comprises the following steps: the airborne photoelectric reconnaissance pod transmits the original photoelectric image to an airborne image tracking processing module for pre-tracking processing; the airborne image tracking processing module downloads the photoelectric image and the total feature point set obtained after the pre-tracking processing to command and control software in the ground command and control cabin; the command software primarily screens feature points according to the point selection and the total feature point set to serve as a concerned feature point set; the onboard image tracking processing module calculates the geometric center pixel coordinates of the focus feature point set in the current original photoelectric image; the command software can continuously receive the external re-input clicking position, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focused feature point set. The invention can solve the problem of difficult target capture and tracking caused by poor radio link signals or large delay when a satellite communication link is used in the airborne photoelectric reconnaissance pod.
Description
Technical Field
The invention relates to the technical field of unmanned aerial vehicles, in particular to a method and a system for realizing an airborne photoelectric reconnaissance pod image pre-tracking function.
Background
The airborne photoelectric reconnaissance pod is widely applied to a wide area reconnaissance monitoring configuration of the unmanned aerial vehicle, is one of the most basic and important task loads of the unmanned aerial vehicle, and has functions and performances which reflect the application scene and tactical value of the unmanned aerial vehicle when the unmanned aerial vehicle executes the wide area reconnaissance monitoring task. The airborne photoelectric reconnaissance pod is mainly used for carrying out real-time imaging reconnaissance on a target area in a large range, all weather and full spectrum, and realizing tactical functions of searching, identifying, tracking, ranging, positioning and the like on an interested target in the area.
Because a certain radio link delay (usually about 0.5 s) exists when an operator in the ground command control cabin controls the airborne photoelectric reconnaissance pod, the condition that the operator is difficult to stably capture and track for targets with small size, unobvious characteristics, complex background and continuous movement in a long-distance and small-view use scene is caused. In a remote mountain area or a highland combat environment, the condition that signals are poor or the condition that the signal cannot meet the viewing condition can be caused by a radio link of the unmanned aerial vehicle, at the moment, the unmanned aerial vehicle is controlled by adopting a satellite communication link, and compared with the radio link, the satellite communication link has larger delay (usually about 1 s), so that the control difficulty of stably tracking a combat target when an on-board photoelectric reconnaissance nacelle is used by ground operators is also greatly improved.
Disclosure of Invention
In order to solve the problems, the invention provides a realization method and a system of an image pre-tracking function of an airborne photoelectric reconnaissance pod, which are based on the image pre-tracking function of feature point set selection and multiple screening, and can solve the problem of difficult target capturing and tracking caused by poor radio link signals or large delay when a satellite communication link is used by the airborne photoelectric reconnaissance pod.
The technical scheme adopted by the invention is as follows:
the implementation method of the airborne photoelectric reconnaissance pod image pre-tracking function comprises the following steps:
s1, transmitting an original photoelectric image to an airborne image tracking processing module by an airborne photoelectric reconnaissance pod for pre-tracking processing, wherein the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set;
s2, the onboard image tracking processing module downloads the photoelectric image subjected to the pre-tracking processing and the total feature point set into command software in a ground command control cabin through a radio link or a satellite communication link;
s3, the finger control software displays the pre-tracked photoelectric image in real time and receives the click position input from the outside, simultaneously, all feature points in a preset radius range which take the click position as the center of a circle are screened out for the first time according to the total feature point set, and all feature points in the focus feature point set are projected and displayed by adopting a pre-tracking frame; uploading the attention feature point number to the airborne image tracking processing module through a radio link or a satellite communication link;
s4, after receiving the attention feature point numbers uploaded on the ground, the airborne image tracking processing module queries feature points which are matched and consistent with the attention feature point numbers in the total feature point set of the current original photoelectric image, and eliminates the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; calculating the geometric center pixel coordinates of the focus feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focus feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focus feature point set by adopting a pixel tracking mode;
s5, after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the concerned feature point set, the finger control software can continuously receive the external re-input selected position, if the pixel coordinates of the selected position are in the current concerned feature point set, the concerned feature point set is gradually reduced, and finally the concerned feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, a new focus feature point set is redefined, and the steps S3 and S4 are repeated, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
Further, in step S1, the extracting feature points to form a total feature point set includes: all edge pixel points are obtained through calculation based on an edge detection algorithm, numbering is carried out, and a total feature point set is formed.
Further, the edge detection algorithm comprises a Canny edge detection algorithm, and the method for extracting the characteristic points in the photoelectric image based on the Canny edge detection algorithm comprises the following steps:
s101, smoothing an original photoelectric image by adopting a Gaussian filter built in the airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range by adopting a double-threshold detection method, reserving edge pixel points with gradient values larger than the maximum threshold value and defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
Further, the information of the total feature point set includes the number and pixel coordinates of each feature point.
Further, in step S5, the step-down the feature point set of interest includes step-down the screening radius, and repeating steps S3 and S4 until the screening radius is reduced to a threshold value.
An implementation system of an airborne photoelectric reconnaissance pod image pre-tracking function, comprising:
the system comprises an airborne image tracking processing module, an airborne photoelectric detection pod and an airborne image processing module, wherein the airborne image tracking processing module is configured to perform pre-tracking processing according to original photoelectric image transmission transmitted by the airborne photoelectric detection pod, and the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set; the photoelectric image after the pre-tracking treatment and the total feature point set are downloaded to command and control software in a ground command and control cabin through a radio link or a satellite communication link; after receiving the attention feature point numbers uploaded by the command software, inquiring feature points which are matched and consistent with the attention feature point numbers in the total feature point set of the current original photoelectric image, and eliminating the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; and calculating the geometric center pixel coordinates of the focus feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focus feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focus feature point set by adopting a pixel tracking mode.
The control software is configured to display the photoelectric image pre-tracked by the airborne image tracking processing module in real time and receive the click position input from the outside, and simultaneously screen out all feature points in a preset radius range with the click position as the center of a circle as a concerned feature point set for the first time according to the total feature point set, and perform projection display on all feature points in the concerned feature point set by adopting a pre-tracking frame; uploading the attention feature point numbers to the airborne image tracking processing module through a radio link or a satellite communication link; after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the attention feature point set, the command software can continuously receive the external re-input click position, if the pixel coordinates of the click position are in the current attention feature point set, the attention feature point set is gradually reduced, and finally the attention feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, and a new focus feature point set is redefined, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
Further, the extracting feature points by the onboard image tracking processing module to form a total feature point set includes: all edge pixel points are obtained through calculation based on an edge detection algorithm, numbering is carried out, and a total feature point set is formed.
Further, the edge detection algorithm comprises a Canny edge detection algorithm, and the method for extracting the characteristic points in the photoelectric image based on the Canny edge detection algorithm comprises the following steps:
s101, smoothing an original photoelectric image by adopting a Gaussian filter built in the airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range by adopting a double-threshold detection method, reserving edge pixel points with gradient values larger than the maximum threshold value and defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
Further, the information of the total feature point set includes the number and pixel coordinates of each feature point.
Further, the step-down the feature point set of interest includes step-down a screening radius, causing the onboard photo-electric reconnaissance pod to automatically track the geometric center of the new feature point set of interest until the screening radius is reduced to a threshold value.
The invention has the beneficial effects that:
according to the invention, the on-board image tracking processing module is added on the unmanned aerial vehicle to pre-track the photoelectric image transmitted by the on-board photoelectric reconnaissance pod, and upload the feature point numbers in the focused feature point set generated during pointing by the pointing control software, the on-board image tracking processing module finds the feature point consistent with the focused feature point numbers in the current original photoelectric image according to the number matching principle, calculates the geometric center pixel coordinate of the focused feature point set in the current original photoelectric image according to the pixel coordinate of each feature point, so that the on-board photoelectric reconnaissance pod primarily stably tracks the target nearby area of interest, and finally gradually reduces the screening radius through multiple screening to concentrate the focused feature point on the target to be tracked, thereby finally realizing the stable automatic tracking of the target of interest.
By adopting the image pre-tracking function based on feature point set selection and multiple screening, the problem that a manual tracking target is difficult to capture due to large delay time of an airborne link can be effectively avoided, the feature point set of interest is generated in real time by clicking a downloaded photoelectric image in a ground command control cabin, and then feature point numbers in the feature point set of interest are uploaded to an aircraft for tracking processing and gradually narrowing a tracking range, so that ground operators can more quickly and conveniently accurately track an interested target in various complex environments when controlling the airborne photoelectric reconnaissance hanging cabin.
Drawings
Fig. 1 is a flowchart of an implementation method of an airborne photoelectric reconnaissance pod image pre-tracking function according to embodiment 1 of the present invention.
Detailed Description
Specific embodiments of the present invention will now be described in order to provide a clearer understanding of the technical features, objects and effects of the present invention. It should be understood that the particular embodiments described herein are illustrative only and are not intended to limit the invention, i.e., the embodiments described are merely some, but not all, of the embodiments of the invention. All other embodiments, which can be made by a person skilled in the art without making any inventive effort, are intended to be within the scope of the present invention.
Example 1
As shown in fig. 1, the embodiment provides a method for implementing an airborne photoelectric reconnaissance pod image pre-tracking function, which includes the following steps:
s1, transmitting an original photoelectric image to an airborne image tracking processing module by an airborne photoelectric reconnaissance pod for pre-tracking processing, wherein the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set;
s2, the airborne image tracking processing module downloads the photoelectric image and the total feature point set which are subjected to pre-tracking processing into command control software in the ground command control cabin through a radio link or a satellite communication link;
s3, the command software displays the photoelectric image after the pre-tracking processing in real time and receives the click position input from the outside, simultaneously, according to the total feature point set, all feature points in the preset radius range with the click position as the center of a circle are screened out for the first time to be used as the attention feature point set, and the pre-tracking frame is adopted to perform projection display on all feature points in the attention feature point set; the focused feature point numbers are uploaded to an onboard image tracking processing module through a radio link or a satellite communication link;
s4, after receiving the attention feature point numbers uploaded on the ground, the airborne image tracking processing module inquires feature points, which are matched and consistent with the attention feature point numbers, in the total feature point set of the current original photoelectric image, and eliminates the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; calculating the geometric center pixel coordinates of the focused feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focused feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focused feature point set by adopting a pixel tracking mode;
s5, after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the focused feature point set, the finger control software can continuously receive the external re-input click position, if the pixel coordinates of the click position are in the current focused feature point set, the focused feature point set is gradually reduced, and finally the focused feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, a new focus feature point set is redefined, and the steps S3 and S4 are repeated, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
Preferably, in step S1, the method for extracting the feature points to form the total feature point set may be that all edge pixel points are calculated based on an edge detection algorithm, and numbering recording is performed to form the total feature point set.
More preferably, the edge detection algorithm may be a Canny edge detection algorithm, and the method for extracting feature points in the photoelectric image based on the Canny edge detection algorithm includes the following steps:
s101, considering that edge detection of an image is easily affected by noise, smoothing an original photoelectric image by adopting a Gaussian filter built in an airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range (based on engineering experience), reserving edge pixel points with gradient values larger than the maximum threshold value, defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
Preferably, the information of the total feature point set includes the number and pixel coordinates of each feature point.
Preferably, in step S5, gradually narrowing the feature point set of interest includes gradually decreasing the screening radius, and repeating steps S3 and S4 until the screening radius decreases to the threshold value.
More preferably, the radius of the primary screening circular region may be 200 pixels; in the screening again, the screening radius can be reduced by 50 pixels each time, and the steps S3 and S4 are repeated until the screening radius is reduced to 50 pixels.
Example 2
The embodiment provides a realization system of an airborne photoelectric reconnaissance pod image pre-tracking function, which comprises an airborne image tracking processing module and command software, wherein the airborne image tracking processing module is installed in an airborne photoelectric reconnaissance pod, and the command software is installed in a ground command control pod, and is specifically described as follows.
The airborne image tracking processing module is configured to perform pre-tracking processing according to the original photoelectric image transmission transmitted by the airborne photoelectric reconnaissance pod, wherein the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set; the photoelectric image and the total characteristic point set after the pre-tracking treatment are downloaded to command and control software in the ground command and control cabin through a radio link or a satellite communication link; after receiving the attention feature point numbers uploaded by the command software, inquiring feature points which are matched and consistent with the attention feature point numbers in the total feature point set of the current original photoelectric image, and eliminating the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; and calculating the geometric center pixel coordinates of the focused feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focused feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focused feature point set by adopting a pixel tracking mode.
The control software is configured to display the photoelectric image pre-tracked by the onboard image tracking processing module in real time and receive the click position input from the outside, and simultaneously screen all the characteristic points in the preset radius range with the click position as the center of a circle according to the total characteristic point set for the first time as the attention characteristic point set, and perform projection display on all the characteristic points in the attention characteristic point set by adopting a pre-tracking frame; the focused feature point numbers are uploaded to an airborne image tracking processing module through a radio link or a satellite communication link; after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the focused feature point set, the finger control software can continuously receive the externally input click position again, if the pixel coordinates of the click position are in the current focused feature point set, the focused feature point set is gradually reduced, and finally the focused feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, and a new focus feature point set is redefined, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
Preferably, the method for extracting the feature points by the onboard image tracking processing module to form the total feature point set can be based on all edge pixel points obtained by calculation of an edge detection algorithm, numbering recording is carried out, and the total feature point set is formed.
More preferably, the edge detection algorithm may be a Canny edge detection algorithm, and the method for extracting feature points in the photoelectric image based on the Canny edge detection algorithm includes the following steps:
s101, considering that edge detection of an image is easily affected by noise, smoothing an original photoelectric image by adopting a Gaussian filter built in an airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range (based on engineering experience), reserving edge pixel points with gradient values larger than the maximum threshold value, defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
Preferably, the information of the total feature point set includes the number and pixel coordinates of each feature point.
Preferably, progressively shrinking the feature point set of interest includes progressively reducing the screening radius such that the onboard photo-electric reconnaissance pod automatically tracks the geometric center of the new feature point set of interest until the screening radius is reduced to a threshold value.
More preferably, the radius of the primary screening circular region may be 200 pixels; in the screening again, the screening radius can be reduced by 50 pixels each time, and the steps S3 and S4 are repeated until the screening radius is reduced to 50 pixels.
It should be noted that, for the sake of simplicity of description, the foregoing method embodiments are expressed as a series of combinations of actions, but it should be understood by those skilled in the art that the present application is not limited by the order of actions described, as some steps may be performed in other order or simultaneously according to the present application. Further, those skilled in the art will also appreciate that the embodiments described in the specification are all preferred embodiments, and that the acts and modules referred to are not necessarily required in the present application.
Claims (10)
1. The implementation method of the airborne photoelectric reconnaissance pod image pre-tracking function is characterized by comprising the following steps of:
s1, transmitting an original photoelectric image to an airborne image tracking processing module by an airborne photoelectric reconnaissance pod for pre-tracking processing, wherein the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set;
s2, the onboard image tracking processing module downloads the photoelectric image subjected to the pre-tracking processing and the total feature point set into command software in a ground command control cabin through a radio link or a satellite communication link;
s3, the finger control software displays the pre-tracked photoelectric image in real time and receives the click position input from the outside, simultaneously, all feature points in a preset radius range which take the click position as the center of a circle are screened out for the first time according to the total feature point set, and all feature points in the focus feature point set are projected and displayed by adopting a pre-tracking frame; the focused feature point numbers are uploaded to the airborne image tracking processing module through a radio link or a satellite communication link;
s4, after receiving the attention feature point numbers uploaded on the ground, the airborne image tracking processing module queries feature points which are matched and consistent with the attention feature point numbers in the total feature point set of the current original photoelectric image, and eliminates the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; calculating the geometric center pixel coordinates of the focus feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focus feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focus feature point set by adopting a pixel tracking mode;
s5, after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the concerned feature point set, the finger control software can continuously receive the external re-input selected position, if the pixel coordinates of the selected position are in the current concerned feature point set, the concerned feature point set is gradually reduced, and finally the concerned feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, a new focus feature point set is redefined, and the steps S3 and S4 are repeated, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
2. The method for implementing the airborne photoelectric reconnaissance pod image pre-tracking function according to claim 1, wherein in step S1, the extracting feature points to form a total feature point set includes: all edge pixel points are obtained through calculation based on an edge detection algorithm, numbering is carried out, and a total feature point set is formed.
3. The method for implementing the airborne photoelectric reconnaissance pod image pre-tracking function according to claim 2, wherein the edge detection algorithm comprises a Canny edge detection algorithm, and the method for extracting feature points in the photoelectric image based on the Canny edge detection algorithm comprises the following steps:
s101, smoothing an original photoelectric image by adopting a Gaussian filter built in the airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range by adopting a double-threshold detection method, reserving edge pixel points with gradient values larger than the maximum threshold value and defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
4. A method of implementing an on-board photo-electric reconnaissance pod image pre-tracking function according to any of claims 1-3, characterized in that the information of the total feature point set comprises the number and pixel coordinates of each feature point.
5. A method of implementing an on-board photo-electric reconnaissance pod image pre-tracking function according to any of claims 1-3, wherein in step S5, said gradually narrowing the feature point set of interest comprises gradually decreasing the screening radius, and repeating steps S3 and S4 until the screening radius decreases to a threshold value.
6. The implementation system of the airborne photoelectric reconnaissance pod image pre-tracking function is characterized by comprising:
the system comprises an airborne image tracking processing module, an airborne photoelectric detection pod and an airborne image processing module, wherein the airborne image tracking processing module is configured to perform pre-tracking processing according to original photoelectric image transmission transmitted by the airborne photoelectric detection pod, and the pre-tracking processing comprises the steps of converting the original photoelectric image into a gray image and extracting characteristic points to form a total characteristic point set; the photoelectric image after the pre-tracking treatment and the total feature point set are downloaded to command and control software in a ground command and control cabin through a radio link or a satellite communication link; after receiving the attention feature point numbers uploaded by the command software, inquiring feature points which are matched and consistent with the attention feature point numbers in the total feature point set of the current original photoelectric image, and eliminating the attention feature point numbers which cannot be found in the total feature point set due to the change of the original photoelectric image; calculating the geometric center pixel coordinates of the focused feature point set in the current original photoelectric image according to all the queried feature point pixel coordinates consistent with the focused feature point numbers, and enabling the airborne photoelectric reconnaissance pod to automatically track the geometric center of the focused feature point set by adopting a pixel tracking mode;
the control software is configured to display the photoelectric image pre-tracked by the airborne image tracking processing module in real time and receive the click position input from the outside, and simultaneously screen out all feature points in a preset radius range with the click position as the center of a circle as a concerned feature point set for the first time according to the total feature point set, and perform projection display on all feature points in the concerned feature point set by adopting a pre-tracking frame; uploading the attention feature point numbers to the airborne image tracking processing module through a radio link or a satellite communication link; after the airborne photoelectric reconnaissance pod stably tracks the geometric center of the attention feature point set, the command software can continuously receive the external re-input click position, if the pixel coordinates of the click position are in the current attention feature point set, the attention feature point set is gradually reduced, and finally the attention feature point is concentrated on a target to be tracked; if the selected position is not in the current focus feature point set, the screening radius is not reduced, and a new focus feature point set is redefined, so that the airborne photoelectric reconnaissance pod automatically tracks the geometric center of the new focus feature point set.
7. The system for implementing the on-board photo-scout pod image pre-tracking function of claim 6, wherein the extracting feature points to form a total feature point set comprises: all edge pixel points are obtained through calculation based on an edge detection algorithm, numbering is carried out, and a total feature point set is formed.
8. The system for implementing the pre-tracking function of the airborne photoelectric reconnaissance pod image according to claim 7, wherein the edge detection algorithm comprises a Canny edge detection algorithm, and the method for extracting the feature points in the photoelectric image based on the Canny edge detection algorithm comprises the following steps:
s101, smoothing an original photoelectric image by adopting a Gaussian filter built in the airborne image tracking processing module, and eliminating image noise;
s102, calculating the gradient size and the gradient direction of each pixel point in the photoelectric image after noise reduction by using a Sobel operator;
s103, eliminating adverse effects caused by edge detection by using a non-maximum suppression method; traversing all pixel points in the photoelectric image, judging whether the current pixel point is the maximum value with the gradient in the same direction in surrounding pixel points, and extracting the pixel points meeting the judging conditions as edge pixel points;
s104, determining a minimum threshold value and a maximum threshold value according to a characteristic rule and a preset range by adopting a double-threshold detection method, reserving edge pixel points with gradient values larger than the maximum threshold value and defining the edge pixel points as boundaries, discarding edge pixel points with gradient values smaller than the minimum threshold value, and reserving pixel points with gradient values between the minimum threshold value and the maximum threshold value but connected with the boundaries;
s105, numbering and recording all edge pixel points obtained in the step S104, and forming a total feature point set.
9. The system for implementing the pre-tracking function of the image of the airborne photoelectric reconnaissance pod according to any one of claims 6 to 8, wherein the information of the total feature point set includes the number and the pixel coordinates of each feature point.
10. The system for implementing an on-board photo-electric reconnaissance pod image pre-tracking function according to any of claims 6-8, wherein said step-down the feature point set of interest comprises step-down a screening radius, causing the on-board photo-electric reconnaissance pod to automatically track the geometric center of the new feature point set of interest until the screening radius is reduced to a threshold value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310220865.8A CN115908475B (en) | 2023-03-09 | 2023-03-09 | Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310220865.8A CN115908475B (en) | 2023-03-09 | 2023-03-09 | Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115908475A CN115908475A (en) | 2023-04-04 |
CN115908475B true CN115908475B (en) | 2023-05-19 |
Family
ID=85728499
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310220865.8A Active CN115908475B (en) | 2023-03-09 | 2023-03-09 | Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115908475B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117011296B (en) * | 2023-10-07 | 2024-06-25 | 长春通视光电技术股份有限公司 | Method, equipment and storage medium for quickly detecting tracking precision of photoelectric pod |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101364046B1 (en) * | 2012-11-05 | 2014-02-19 | 재단법인대구경북과학기술원 | Method and apparatus for object tracking in video sequences |
CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
CN106292710A (en) * | 2016-10-20 | 2017-01-04 | 西北工业大学 | Four rotor wing unmanned aerial vehicle control methods based on Kinect sensor |
CN110400330A (en) * | 2019-08-13 | 2019-11-01 | 湖南海迅自动化技术有限公司 | Photoelectric nacelle image tracking method and tracking system based on fusion IMU |
WO2021189507A1 (en) * | 2020-03-24 | 2021-09-30 | 南京新一代人工智能研究院有限公司 | Rotor unmanned aerial vehicle system for vehicle detection and tracking, and detection and tracking method |
CN113744307A (en) * | 2021-08-06 | 2021-12-03 | 上海有个机器人有限公司 | Image feature point tracking method and system based on threshold dynamic adjustment |
CN115439424A (en) * | 2022-08-23 | 2022-12-06 | 成都飞机工业(集团)有限责任公司 | Intelligent detection method for aerial video image of unmanned aerial vehicle |
CN115601471A (en) * | 2022-12-16 | 2023-01-13 | 四川腾盾科技有限公司(Cn) | Drawing method based on large unmanned aerial vehicle photoelectric reconnaissance swept area |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20140108047A (en) * | 2013-02-28 | 2014-09-05 | 삼성전자주식회사 | Method for tracking a moving object and a controlling apparatus capable of tracking a moving object |
JP6129981B2 (en) * | 2013-10-01 | 2017-05-17 | 株式会社日立製作所 | Moving object position estimation apparatus and moving object position estimation method |
-
2023
- 2023-03-09 CN CN202310220865.8A patent/CN115908475B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101364046B1 (en) * | 2012-11-05 | 2014-02-19 | 재단법인대구경북과학기술원 | Method and apparatus for object tracking in video sequences |
CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
CN106292710A (en) * | 2016-10-20 | 2017-01-04 | 西北工业大学 | Four rotor wing unmanned aerial vehicle control methods based on Kinect sensor |
CN110400330A (en) * | 2019-08-13 | 2019-11-01 | 湖南海迅自动化技术有限公司 | Photoelectric nacelle image tracking method and tracking system based on fusion IMU |
WO2021189507A1 (en) * | 2020-03-24 | 2021-09-30 | 南京新一代人工智能研究院有限公司 | Rotor unmanned aerial vehicle system for vehicle detection and tracking, and detection and tracking method |
CN113744307A (en) * | 2021-08-06 | 2021-12-03 | 上海有个机器人有限公司 | Image feature point tracking method and system based on threshold dynamic adjustment |
CN115439424A (en) * | 2022-08-23 | 2022-12-06 | 成都飞机工业(集团)有限责任公司 | Intelligent detection method for aerial video image of unmanned aerial vehicle |
CN115601471A (en) * | 2022-12-16 | 2023-01-13 | 四川腾盾科技有限公司(Cn) | Drawing method based on large unmanned aerial vehicle photoelectric reconnaissance swept area |
Non-Patent Citations (3)
Title |
---|
Development of UAV-Based Target Tracking and Recognition Systems;Shuaijun Wang等;IEEE Transactions on Intelligent Transportation Systems;第21卷(第8期);第3409-3422页 * |
基于CamShift融合局部特征匹配的无人机目标跟踪研究;刘亚伟;李小民;杨森;;电子技术应用(09);第13-16页 * |
空对地可见光图像制导跟踪算法研究;丁洋;中国优秀硕士学位论文全文数据库 信息科技辑(第8期);第I138-1185页 * |
Also Published As
Publication number | Publication date |
---|---|
CN115908475A (en) | 2023-04-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8102423B2 (en) | Method and system for performing adaptive image acquisition | |
KR102400452B1 (en) | Context-aware object detection in aerial photographs/videos using travel path metadata | |
CN110622211B (en) | System and method for reducing low frequency non-uniformities in images | |
CN115908475B (en) | Implementation method and system for airborne photoelectric reconnaissance pod image pre-tracking function | |
EP2780862B1 (en) | Mobile and one-touch tasking and visualization of sensor data | |
CN108108697B (en) | Real-time unmanned aerial vehicle video target detection and tracking method | |
EP2867873B1 (en) | Surveillance process and apparatus | |
CN109218598B (en) | Camera switching method and device and unmanned aerial vehicle | |
Kejriwal et al. | A hybrid filtering approach of digital video stabilization for UAV using kalman and low pass filter | |
CA3005733C (en) | Predicting radio frequency interference events | |
Kumari et al. | Artificial intelligent based smart system for safe mining during foggy weather | |
CN115760910A (en) | Target tracking method and device of gun and ball linkage equipment, terminal and storage medium | |
US10200574B2 (en) | Method and system for generating a video frame | |
KR102168438B1 (en) | Method and server for detecting abnormal activity using drones | |
CN116935328B (en) | Image recognition method and device, storage medium and electronic equipment | |
KR101697211B1 (en) | Image recognition apparatus, method and system for providing parking space information based on interaction and image recognition | |
JP2022138333A (en) | Image processing device, image processing method, and image processing program | |
Hayrapetyan et al. | Border Surveillance Using UAVs with Thermal Camera | |
US12028619B2 (en) | Image processing device and image processing method | |
JP7212294B2 (en) | Wireless transmission system, wireless transmission device, wireless transmission method, and program | |
Lian et al. | Research on Image Processing and Application of Air-to-Ground Platforms | |
EP4300131A1 (en) | Automatic optimization of obstacle detection | |
US20240020801A1 (en) | Variable distortion correction for panoramic images | |
CN117270580A (en) | Servo control method, system and equipment for tracking unmanned aerial vehicle photoelectric pod target | |
JP2010148085A (en) | Apparatus and method for collecting and processing data of reception sensitivity by positions |
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 |