CN113793260B - Method and device for semi-automatically correcting target tracking frame and electronic equipment - Google Patents
Method and device for semi-automatically correcting target tracking frame and electronic equipment Download PDFInfo
- Publication number
- CN113793260B CN113793260B CN202110873739.3A CN202110873739A CN113793260B CN 113793260 B CN113793260 B CN 113793260B CN 202110873739 A CN202110873739 A CN 202110873739A CN 113793260 B CN113793260 B CN 113793260B
- Authority
- CN
- China
- Prior art keywords
- target
- tracking
- image
- frame
- area
- 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
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 230000011218 segmentation Effects 0.000 claims description 48
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 12
- 230000003321 amplification Effects 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 7
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 238000011022 operating instruction Methods 0.000 claims description 4
- 238000007619 statistical method Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 abstract description 5
- 230000000694 effects Effects 0.000 description 14
- 238000001514 detection method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000009825 accumulation Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 3
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method, a device and electronic equipment for semi-automatically correcting a target tracking frame, wherein the method comprises the following steps: receiving an operation instruction which is sent by an operator through a sending and controlling device and used for determining a target position, amplifying an image corresponding to a target area, and overlapping and displaying the image on an original image in a picture-in-picture mode; when receiving an operation instruction for correcting the target position sent by an operator through a sending and controlling device, correcting the target area according to the operation instruction for correcting the target position; preprocessing the image corresponding to the corrected target area, segmenting the target in the target area, and finding out the optimal target; and tracking the optimal target, initializing the tracker if an operation instruction for adjusting the position of the target sent by an operator through the transmission and control equipment is received in the tracking process, and otherwise, normally tracking. The invention can further improve the tracking precision of locking to tracking and can more effectively and accurately track the target.
Description
Technical Field
The invention relates to the field of image tracking algorithms, in particular to a method and a device for semi-automatically correcting a target tracking frame and electronic equipment.
Background
Object detection techniques are commonly used in object tracking systems. The target detection is generally divided into two stages, the first stage is that the operator locks the target, and the second stage is that the target tracking frame is corrected. One of the difficulties is to accurately detect the target and determine the position of the target tracking center and the size of the tracking frame.
The image tracking system adopts two modes of television and infrared to input video images at present, and ground targets generally comprise targets such as vehicles, buildings and the like. In the image tracking system, after the image is transmitted to the transmitting and controlling receiving end through an optical fiber or a wireless data link, an operator identifies the specific position of the target in the image by observing the image.
When the target is locked in the initial stage, due to the fact that the ground scene is complex, the distance between the tracker and the target is long, the number of target imaging pixels is small, energy of rain, fog, snow and sand dust on target radiation is seriously attenuated, and the intercepting distance of the tracker on the target is greatly reduced; the reflection of sunlight and its own heat radiation from the ground background make it difficult for the tracker to find objects in the background. Therefore, there may be a deviation in pixels between the tracking point position and the target center point position during locking, and the initial deviation is easy to accumulate continuously in the tracking process, and the error increases to the end of tracking, thereby affecting the hit probability. Therefore, we need to solve the problem one: the method solves the problem of locking and identifying the weak and small target with low signal-to-noise ratio in the complex background in real time.
Meanwhile, as the distance between the tracker and the target approaches, the affine perspective relationship in the scene changes due to different approach angles, the three-dimensional relative position relationship between the background around the target and the target reacts and strongly changes in the relative position relationship on the two-dimensional plane image, a series of complex conditions that the target is fused into the background, the foreground blocks the target, and the shadow under the target foot appears occur, and the tracking midpoint deviates from the target center. Therefore, we need to solve the second problem: the target needs to be corrected by manual assistance, otherwise, the background is easy to track, and the target is lost.
In order to solve the problem of target locking in the prior art, an adopted method is to manually set an initial tracking point, manually adjust the width and the height of a tracking frame, or intercept a certain area image around the tracking point and divide a target to obtain the width and the height of the tracking frame, so that the target is easily locked to have deviation. The cause of the deviation is: (1) and manually adjusting the deviation of the tracking frame. Because the image on the moving platform generates certain jitter, if the speed of the center point is determined manually and is slower than that of the platform, the locking position is deviated. (2) And segmenting the target to obtain the deviation of the tracking frame. Because the ground background is relatively complex, the segmentation result is easily interfered by background information, so that segmentation errors are caused, and errors easily occur when the target center is locked. Therefore, the prior art cannot guarantee the accuracy of the initial locking of the target under the condition of image shaking of the motion platform. Because the initial locking deviation can be accumulated continuously in the tracking process, the subsequent tracking process can be influenced, and therefore, the problem that the tracking frame is deviated due to the fact that the initial locking error accumulation in the approaching process cannot be solved in the prior art, and the problems that the target is merged into the background, the target is shielded and the like to influence the tracking caused by the change of the affine perspective relation cannot be solved.
Disclosure of Invention
In order to solve the technical problems in the prior art, the invention provides a method and a device for semi-automatically correcting a target tracking frame and electronic equipment.
The invention is realized in the following way:
in a first aspect, the present invention provides a method for semi-automatically correcting a target tracking frame, including the following steps:
receiving an operation instruction for determining a target position sent by an operator through a sending and controlling device, determining a target area according to the operation instruction for determining the target position, amplifying an image corresponding to the target area, and overlapping and displaying the image on an original image in a picture-in-picture mode;
when receiving an operation instruction for correcting the target position sent by an operator through a sending and controlling device, correcting the target area according to the operation instruction for correcting the target position;
preprocessing the image corresponding to the corrected target area, segmenting targets in the target area according to the preprocessed image, counting the number of the targets in the target area, and finding out an optimal target;
and tracking the optimal target, initializing the tracker if an operation instruction for adjusting the position of the target sent by an operator through the transmission and control equipment is received in the tracking process, and otherwise, normally tracking.
Further, the preprocessing the image corresponding to the corrected target area specifically includes:
and performing down-sampling on the image, performing Gaussian smoothing, and then performing image significance enhancement.
Further, the segmenting the target in the target region according to the preprocessed image specifically includes:
and calculating a segmentation threshold value by using a mean variance statistical method, wherein the segmentation threshold value is the mean value plus the variance coefficient, and segmenting the preprocessed image by using the threshold value according to the segmentation threshold value to segment the target in the target region.
Further, the counting the number of targets in the target area and finding out the optimal target specifically includes:
and counting the number of targets in the target area by using a connected area algorithm, establishing a pipeline for tracking each target, and finding a target closest to the position of the locked target from a plurality of targets which continuously exist as an optimal target.
Further, the counting the number of targets in the target area by using the connected area algorithm, and the tracking of the pipeline established by each target specifically comprises:
traversing a binarization search area image corresponding to a target area by using a connected area algorithm, marking each independent target in the binarization search area image, and counting width, height, central point and gray information of each target in the marking process;
and (4) associating the single-frame target with the pipeline target by calculating the similarity of the position, the size and the gray level of the single-frame target and the target in the pipeline.
Further, the method also comprises the following steps:
and carrying out secondary segmentation on the segmented image of the optimal target to calculate the target size, and tracking the secondarily segmented target.
Further, comparing the result of the second division with the result of the first division, if the pixel deviation of the central point is less than half of the width and the height of the frame of the first division and the width of the frame of the second division is more than 50% of the frame of the first division, the second division is valid, otherwise, the second division is invalid, and the result of the first division is still used.
In a second aspect, the present invention further provides an apparatus for semi-automatically correcting a target tracking frame, the apparatus comprising:
the target area amplification display module is used for receiving an operation instruction which is sent by an operator through the control device and used for determining a target position, determining a target area according to the operation instruction for determining the target position, amplifying an image corresponding to the target area and displaying the image on an original image in a picture-in-picture mode in an overlapping mode;
the target area correction module is used for correcting the target area according to an operation instruction for correcting the target position, which is sent by an operator through the sending and controlling equipment when the operation instruction for correcting the target position is received;
the tracking target determining module is used for preprocessing the image corresponding to the corrected target area, segmenting a target in the target area according to the preprocessed image, counting the number of the targets in the target area and finding out an optimal target;
and the target tracking and initializing module is used for tracking the optimal target, initializing the tracker if an operation instruction for adjusting the target position sent by an operator through the control device is received in the tracking process, and otherwise, normally tracking.
In a third aspect, the present invention further provides an electronic device for semi-automatically correcting a target tracking frame, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of any one of the above methods when executing the computer program.
In a fourth aspect, the present invention also provides a computer-readable storage medium, in which a computer program is stored, which computer program, when being executed by a processor, carries out the steps of the method as set forth in any of the above.
Compared with the prior art, the invention has the following beneficial effects:
(1) by adopting a suspected target amplification method, when the target imaging does not meet the locking condition, an operator can be assisted to pre-judge the target in advance, and when the target meets the locking condition, the target can be quickly locked from the background, so that the probability of locking the target is improved;
(2) the tracking frame deviation problem caused by accumulation of initial locking errors or appearance of shadows in the approaching process can be solved by adopting a manual correction tracking frame strategy;
(3) by adopting a quadratic segmentation algorithm, the problem that the primary segmentation is inaccurate under the complex background condition of the target can be solved;
(4) the method of modifying the strategy by responding to the tracking frame at any moment can meet the requirement of quickly changing the tracking target;
(5) by adopting the method for semi-automatically correcting the target tracking frame, the operation process of the whole target detection tracking system can be more convenient, the tracking precision of locking to tracking is further improved, and the target can be more effectively and accurately tracked.
Drawings
Fig. 1 is an overall flowchart of a method for semi-automatically correcting a target tracking frame according to an embodiment of the present invention;
FIG. 2 is a detailed flowchart of a method for semi-automatically correcting a target tracking frame according to an embodiment of the present invention;
FIG. 3 is an effect diagram of enlarging the image of the tracking target area by 1-10 times according to the embodiment of the present invention;
fig. 4 is an original image and a down-sampled image provided by an embodiment of the present invention;
FIG. 5 is a graph of Gaussian smoothing effect provided by an embodiment of the present invention;
fig. 6 is a graph of the effect of initializing significance enhancement according to an embodiment of the present invention;
FIG. 7 is a graph of the significance enhancement effect provided by the embodiment of the present invention;
FIG. 8 is a diagram illustrating the effect of threshold segmentation according to an embodiment of the present invention;
FIG. 9 is a diagram illustrating the effect of marking connected regions according to an embodiment of the present invention;
FIG. 10 is a schematic view of a pipeline provided by an embodiment of the present invention;
FIG. 11 is a schematic diagram of a segmentation process according to an embodiment of the present invention;
FIG. 12 is a schematic diagram of a secondary segmentation provided in accordance with an embodiment of the present invention;
fig. 13 is a block diagram of an apparatus for semi-automatically correcting a target tracking frame 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.
As shown in fig. 1 and fig. 2, an embodiment of the present invention provides a method for semi-automatically correcting a target tracking frame, including the following steps:
s101, receiving an operation instruction for determining a target position sent by an operator through a sending and controlling device, determining a target area according to the operation instruction for determining the target position, amplifying an image corresponding to the target area, and overlapping and displaying the image on an original image in a picture-in-picture mode;
and in the suspected target detection stage, the target is locked by an operator. Firstly, an operator sends an operation instruction for determining a target position through a sending device, the operation instruction for determining the target position can be an operation instruction for clicking the target position through a screen of the sending device by the operator, or can be an operation instruction for clicking the target position through a cursor, after the operation instruction for determining the target position is received, a target area is determined according to a preset shape and size by taking a point coordinate of the target position as a center, an image corresponding to the target area is intercepted, the intercepted image is amplified according to a preset amplification factor and is superposed and displayed on an original image, the amplification factor can be automatically adjusted, the image of the tracking target area is automatically selected by 1-10 times, the amplification effect of the image of the tracking target area is 1-10 times as shown in figure 3, the method can amplify the image of the target area, and the image of the central area of the image is amplified and displayed before the tracking target is locked, therefore, the number of suspected target pixels is increased, the small window displays an enlarged area, and an operator can further recognize the target conveniently.
S102, when an operation instruction for correcting the target position sent by an operator through a sending and controlling device is received, correcting the target area according to the operation instruction for correcting the target position;
an operator observes whether the target center point and the tracking frame center coincide or not through the amplified image, if the target center point and the tracking frame center coincide, the operator can click the target center point position again through the transmitting and controlling equipment, namely, an operating instruction for correcting the target position is sent out, and when the operating instruction for correcting the target position is received, the target tracking frame position is corrected, namely, a target area is corrected.
S103, preprocessing the image corresponding to the corrected target area, segmenting the target in the target area according to the preprocessed image, counting the number of the targets in the target area, and finding out the optimal target;
in order to avoid influence of details inside the target on segmentation, preprocessing the image corresponding to the corrected target area, specifically including: and performing down-sampling on the image, performing Gaussian smoothing, and then performing image significance enhancement. Wherein:
the target area is sampled specifically as follows: selecting down-sampling multiples according to the target size, calculating down-sampling images by using a bilinear interpolation method, wherein the down-sampling multiples range is [2, 4 ]; the original image and the down-sampled image are shown in fig. 4;
the gaussian smoothing is specifically: smoothing the down-sampled image by using a Gaussian operator to remove some noise influences around the target, wherein the Gaussian operator is preferably [0.5, 1, 0.5 ]; the Gaussian smoothing effect graph is shown in FIG. 5;
the significant enhancement is specifically as follows: traversing each pixel point on the image area to be enhanced, calculating the pixel mean value of each pixel point area, such as the pixel mean value of 8 neighborhoods [10, 20], and calculating the square of the difference between the pixel point and the mean value to be used as a significance enhancement initialization image; the significance enhancement initialization effect graph is shown in fig. 6;
counting a gray scale range [ min, max ] of the saliency enhanced initialized image, linearly mapping the gray scale range [ min, max ] to [0, 255], wherein the linear mapping formula is that the gray scale of the enhanced image is (gray scale-min of the saliency enhanced initialized image)/(max-min); the significant enhancement effect is shown in fig. 7.
Preferably, the segmenting the target in the target region according to the preprocessed image specifically includes:
calculating a segmentation threshold value by using a mean variance statistical method, wherein the segmentation threshold value is a mean value plus a variance coefficient and is within a coefficient range of [0.5-2 ]; performing threshold segmentation on the preprocessed image according to a segmentation threshold to segment a target in a target region; the threshold segmentation effect map is shown in fig. 8.
Preferably, the counting the number of the targets in the target area and finding out the optimal target specifically includes:
and (3) counting the number of targets in the target area by using a connected area algorithm, establishing a pipeline for tracking each target, and after accumulating [3,10] frames, finding a target closest to the position of the locked target from a plurality of targets which continuously exist as an optimal target.
Specifically, the counting of the number of targets in the target area by using the connected region algorithm, and the tracking of the pipeline established by each target specifically includes:
traversing a binarization search area image corresponding to a target area by using a connected area algorithm, marking each independent target in the binarization search area image, and counting related information such as width and height, a central point, gray level and the like of each target in the marking process; the individual regions are marked with different numbers and the connected region marking effect is shown in fig. 9.
Target establishment pipeline tracking: video transmission is a data stream formed by a series of images, and after the number of targets in a connected region on a single-frame image and the related information of the target in each connected region are counted, front and back frame data association, namely a target pipeline, is established for each target. The target pipeline is a data storage structure body, variables of the structure body comprise the number of existing frames of the target, and the position, the size, the gray level and other related information of the target in each frame of image, the storage space of the structure body of the target pipeline can be automatically set according to the requirement, and the length of the storage space is set to be 15 frames. A schematic of the piping is shown in fig. 10.
And (4) associating the single-frame target with the pipeline target by calculating the similarity of the position, the size and the gray level of the single-frame target and the target in the pipeline.
The optimal target selection method specifically comprises the following steps: when a plurality of targets exist, the optimal target is selected through a distance nearest method, the distance between each target and the position of the locking target is calculated according to the coordinate point of the position of the input locking target, and the target which is nearest to the position of the locking target is the target to be locked.
Because there is no prior information of the target and the size of the target is unknown, when the target is segmented at a time, in order to adapt to the segmentation conditions of different target sizes, a larger segmentation area is used to segment the target, when the target and the background are adhered, the target and the background are easily segmented together, so that the calculation of the tracking frame of the target is inaccurate when the target is segmented at a time, and a one-time segmentation effect graph is shown in fig. 11.
Preferably, after finding the optimal target, the method further comprises: and carrying out secondary segmentation on the segmented image of the optimal target to calculate the size of the target, and tracking the secondarily segmented target.
The secondary segmentation is adopted, primary segmentation treatment is carried out on the basis of primary segmentation, targets adhered to the background can be effectively extracted, and the actual size of the targets is accurately segmented. The second division is to reduce the target division range on the basis of the first division, and recalculate the gray level difference between the target and the background in the division frame by using the first division frame as the division range, so that the actual target can be divided more accurately, and the second division effect graph is shown in fig. 12.
Further preferably, the second-time division result is compared with the first-time division result, if the central-point pixel deviation is less than half of the width and the height of the first-time division frame, and the width and the height of the second-time division frame are greater than 50% of the width and the height of the first-time division frame, the second-time division is effective, otherwise, the second-time division is invalid, and the first-time division result is still adopted to prevent the second-time division from reaching an error target.
And S104, tracking the optimal target, initializing the tracker if an operation instruction for adjusting the position of the target, which is sent by an operator through the transmission and control equipment, is received in the tracking process, and otherwise, normally tracking.
And after the target detection is successful, tracking is carried out, in the tracking process, if an operator finds that the central point of the tracking frame and the central point of the target deviate, the position of the central point is adjusted through a control device, after the central point is corrected, the tracker is automatically reinitialized, a new target tracking point is tracked again, and semi-automatic correction tracking is finished.
Based on the same inventive concept, embodiments of the present invention further provide a device for semi-automatically correcting a target tracking frame, and since the principle of the device for solving the technical problem is similar to that of the method for semi-automatically correcting a target tracking frame, the implementation of the device may refer to the method embodiments, and repeated details are omitted.
As shown in fig. 13, an apparatus for semi-automatically modifying a target tracking frame according to an embodiment of the present invention is configured to implement the foregoing method embodiment, where the apparatus includes:
a target area enlargement display module 201, configured to receive an operation instruction for determining a target position sent by an operator through a control device, determine a target area according to the operation instruction for determining the target position, enlarge an image corresponding to the target area, and display the image in a picture-in-picture manner in an original image in an overlaid manner;
the target area correction module 202 is configured to, when receiving an operation instruction for correcting a target position, which is sent by an operator through a transmission and control device, correct the target area according to the operation instruction for correcting the target position;
the tracking target determining module 203 is configured to pre-process the image corresponding to the corrected target area, segment a target in the target area according to the pre-processed image, count the number of targets in the target area, and find out an optimal target;
and the target tracking and initializing module 204 is used for tracking the optimal target, initializing the tracker if an operation instruction for adjusting the target position sent by an operator through the control device is received in the tracking process, and otherwise, normally tracking.
The embodiment of the present invention further provides an electronic device for semi-automatically correcting a target tracking frame, which includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, where the processor implements the steps of the above method embodiments when executing the computer program.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the computer program implements the steps of the above method embodiment.
In conclusion, the invention has the following beneficial effects:
(1) by adopting a suspected target amplification method, when the target imaging does not meet the locking condition, an operator can be assisted to pre-judge the target in advance, and when the target meets the locking condition, the target can be quickly locked from the background, so that the probability of locking the target is improved;
(2) the tracking frame deviation problem caused by accumulation of initial locking errors or shadow in the approaching process can be solved by adopting a manual correction tracking frame strategy;
(3) by adopting a quadratic segmentation algorithm, the problem that the primary segmentation is inaccurate under the complex background condition of the target can be solved;
(4) the method for modifying the strategy by responding to the tracking frame at any moment can meet the requirement of quickly changing the tracking target;
(5) by adopting the method for semi-automatically correcting the target tracking frame, the operation process of the whole target detection tracking system can be more convenient, the tracking precision of locking to tracking is further improved, and the target can be more effectively and accurately tracked.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (8)
1. A method for semi-automatically correcting a target tracking frame is characterized by comprising the following steps:
receiving an operation instruction for determining a target position sent by an operator through a sending and controlling device, determining a target area according to the operation instruction for determining the target position, amplifying an image corresponding to the target area, and overlapping and displaying the image on an original image in a picture-in-picture mode;
when an operating instruction for correcting the target position sent by an operator through a sending and controlling device is received, correcting the target area according to the operating instruction for correcting the target position;
preprocessing the image corresponding to the corrected target area, segmenting the target in the target area according to the preprocessed image, counting the number of the targets in the target area, finding out the optimal target, performing secondary segmentation on the segmented image of the optimal target to calculate the size of the target, tracking the secondarily segmented target, wherein the tracking the secondarily segmented target comprises the following steps: comparing the secondary segmentation result with the primary segmentation result, if the pixel deviation of the central point is less than half of the width and the height of the primary segmentation frame, and the width and the height of the secondary segmentation frame are greater than 50% of the width and the height of the primary segmentation frame, indicating that the secondary segmentation is effective, otherwise, indicating that the secondary segmentation is ineffective, and still adopting the primary segmentation result; and tracking the optimal target, initializing the tracker if an operation instruction for adjusting the target position sent by an operator through the transmitting and controlling equipment is received in the tracking process, and otherwise, normally tracking.
2. The method of semi-automatically correcting a target tracking frame according to claim 1, wherein the preprocessing the image corresponding to the corrected target area specifically comprises:
and performing down-sampling on the image, performing Gaussian smoothing, and then performing image significance enhancement.
3. The method of semi-automatically modifying a target tracking frame according to claim 1, wherein the segmenting the target in the target region according to the preprocessed image specifically comprises:
and calculating a segmentation threshold value by using a mean variance statistical method, wherein the segmentation threshold value is mean value + variance coefficient, and performing threshold segmentation on the preprocessed image according to the segmentation threshold value to segment the target in the target region.
4. The method for semi-automatically correcting the target tracking frame according to claim 1, wherein the counting of the number of targets in the target area and the finding of the optimal target specifically comprise:
and counting the number of targets in the target area by using a connected region algorithm, establishing a pipeline for tracking each target, and finding the target closest to the position of the locked target from a plurality of targets which continuously exist as an optimal target.
5. The method for semi-automatically modifying the target tracking frame according to claim 4, wherein the step of counting the number of targets in the target area by using the connected area algorithm specifically comprises the following steps of:
traversing a binarization search area image corresponding to a target area by using a connected area algorithm, marking each independent target in the binarization search area image, and counting width, height, central point and gray information of each target in the marking process;
and (4) associating the single-frame target with the pipeline target by calculating the similarity of the position, the size and the gray level of the single-frame target and the target in the pipeline.
6. An apparatus for semi-automatically correcting a target tracking frame, the apparatus comprising:
the target area amplification display module is used for receiving an operation instruction which is sent by an operator through the control device and used for determining a target position, determining a target area according to the operation instruction for determining the target position, amplifying an image corresponding to the target area and displaying the image on an original image in a picture-in-picture mode in an overlapping mode;
the target area correction module is used for correcting the target area according to an operation instruction for correcting the target position, which is sent by an operator through the sending and controlling equipment when the operation instruction for correcting the target position is received;
a tracking target determining module, configured to pre-process the image corresponding to the corrected target area, segment the target in the target area according to the pre-processed image, count the number of targets in the target area, find out an optimal target, perform secondary segmentation on the segmented image of the optimal target to calculate the size of the target, track the secondarily segmented target, and track the secondarily segmented target, where the tracking of the secondarily segmented target includes: comparing the secondary segmentation result with the primary segmentation result, if the pixel deviation of the central point is less than half of the width and the height of the primary segmentation frame, and the width and the height of the secondary segmentation frame are greater than 50% of the width and the height of the primary segmentation frame, indicating that the secondary segmentation is effective, otherwise, indicating that the secondary segmentation is ineffective, and still adopting the primary segmentation result;
and the target tracking and initializing module is used for tracking the optimal target, initializing the tracker if an operation instruction for adjusting the target position sent by an operator through the control device is received in the tracking process, and otherwise, normally tracking.
7. An electronic device for semi-automatically correcting a target tracking frame, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method according to any one of claims 1 to 5 when executing the computer program.
8. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873739.3A CN113793260B (en) | 2021-07-30 | 2021-07-30 | Method and device for semi-automatically correcting target tracking frame and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110873739.3A CN113793260B (en) | 2021-07-30 | 2021-07-30 | Method and device for semi-automatically correcting target tracking frame and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113793260A CN113793260A (en) | 2021-12-14 |
| CN113793260B true CN113793260B (en) | 2022-07-22 |
Family
ID=79181445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110873739.3A Active CN113793260B (en) | 2021-07-30 | 2021-07-30 | Method and device for semi-automatically correcting target tracking frame and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113793260B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117714851A (en) * | 2022-05-25 | 2024-03-15 | 荣耀终端有限公司 | Video recording method, device and storage medium |
| CN117765031B (en) * | 2024-02-21 | 2024-05-03 | 四川盎芯科技有限公司 | Image multi-target pre-tracking method and system for edge intelligent equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116896A (en) * | 2013-03-07 | 2013-05-22 | 中国科学院光电技术研究所 | Automatic detection tracking method based on visual saliency model |
| CN108830219A (en) * | 2018-06-15 | 2018-11-16 | 北京小米移动软件有限公司 | Method for tracking target, device and storage medium based on human-computer interaction |
| CN110111283A (en) * | 2019-05-14 | 2019-08-09 | 武汉高德红外股份有限公司 | The reminding method and system of infrared suspected target under a kind of complex background |
| CN110291775A (en) * | 2018-05-29 | 2019-09-27 | 深圳市大疆创新科技有限公司 | A kind of track up method, equipment and storage medium |
| CN110414535A (en) * | 2019-07-02 | 2019-11-05 | 绵阳慧视光电技术有限责任公司 | A kind of manual initial block modification method and system based on background differentiation |
| CN112286430A (en) * | 2020-10-29 | 2021-01-29 | 维沃移动通信有限公司 | Image processing method, apparatus, device and medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10055666B2 (en) * | 2014-05-08 | 2018-08-21 | Aselsan Elektronik Sanayi Ve Ticaret Anonim Sirketi | Real-time, semi-automatic method for target tracking window initialization in thermal imagery |
-
2021
- 2021-07-30 CN CN202110873739.3A patent/CN113793260B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103116896A (en) * | 2013-03-07 | 2013-05-22 | 中国科学院光电技术研究所 | Automatic detection tracking method based on visual saliency model |
| CN110291775A (en) * | 2018-05-29 | 2019-09-27 | 深圳市大疆创新科技有限公司 | A kind of track up method, equipment and storage medium |
| CN108830219A (en) * | 2018-06-15 | 2018-11-16 | 北京小米移动软件有限公司 | Method for tracking target, device and storage medium based on human-computer interaction |
| CN110111283A (en) * | 2019-05-14 | 2019-08-09 | 武汉高德红外股份有限公司 | The reminding method and system of infrared suspected target under a kind of complex background |
| CN110414535A (en) * | 2019-07-02 | 2019-11-05 | 绵阳慧视光电技术有限责任公司 | A kind of manual initial block modification method and system based on background differentiation |
| CN112286430A (en) * | 2020-10-29 | 2021-01-29 | 维沃移动通信有限公司 | Image processing method, apparatus, device and medium |
Non-Patent Citations (2)
| Title |
|---|
| TubeTK: Adopting Tubes to Track Multi-Object in a One-Step Training Model;Bo Pang等;《arXiv:2006.05683v1》;20200610;第1-11页 * |
| 基于二次分割的彩色图像目标自动计数算法;杨吉宏 等;《计算机工程与设计》;20101231;第780-782、787页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113793260A (en) | 2021-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113793260B (en) | Method and device for semi-automatically correcting target tracking frame and electronic equipment | |
| CN108986037B (en) | Monocular vision odometer positioning method and positioning system based on semi-direct method | |
| EP3296953B1 (en) | Method and device for processing depth images | |
| JP6095018B2 (en) | Detection and tracking of moving objects | |
| EP2399239B1 (en) | Estimation of panoramic camera orientation relative to a vehicle coordinate frame | |
| CN110796010B (en) | Video image stabilizing method combining optical flow method and Kalman filtering | |
| CN110807809B (en) | Light-weight monocular vision positioning method based on point-line characteristics and depth filter | |
| CN112880687B (en) | Indoor positioning method, device, equipment and computer readable storage medium | |
| US11004212B1 (en) | Object tracking method and system using iterative template matching | |
| KR100996897B1 (en) | correction method of Radial Distortion Based on a Line-Fitting | |
| US20100014709A1 (en) | Super-resolving moving vehicles in an unregistered set of video frames | |
| CN111144213B (en) | Object detection method and related equipment | |
| CN111247526B (en) | Method and system for tracking position and direction of target object moving on two-dimensional plane | |
| CN112509003B (en) | Method and system for solving target tracking frame drift | |
| KR20210141668A (en) | Detection, 3D reconstruction and tracking of multiple orthopedic objects moving relative to each other | |
| US11880993B2 (en) | Image processing device, driving assistance system, image processing method, and program | |
| KR20230091906A (en) | Offset detection method based on binocular vision and symmetry | |
| CN116740126A (en) | Target tracking method, high-speed camera, and storage medium | |
| CN114821280A (en) | Sliding window-based SLAM local real-time relocation method | |
| CN113781523A (en) | Football detection tracking method and device, electronic equipment and storage medium | |
| CN113240638A (en) | Target detection method, device and medium based on deep learning | |
| CN111524167B (en) | Positioning correction and optimization method for mobile radioactive source | |
| CN109784229B (en) | Composite identification method for ground building data fusion | |
| KR102190297B1 (en) | Method for tracking air-targets using correlation filter and system thereof | |
| CN114037977B (en) | Road vanishing point detection method, device, equipment and storage medium |
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 |