A system for continuously tracking and identifying moving targets
By setting up information acquisition, processing and analysis modules in the photoelectric pod camera, combined with deep convolutional neural network and signal intensity analysis, the problem of moving targets being separated from the shooting range is solved, and the targets are continuously tracked and identified, ensuring the stability of the shooting effect.
Patent Information
- Application Number
- CN202210371584.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-04-11
- Publication Date
- 2025-08-29
- Estimated Expiration
- 2042-04-11
AI Technical Summary
In the prior art, during the movement of the photoelectric pod camera, the moving target is prone to leave the camera's shooting range or is not in the center of the picture, resulting in poor shooting effect.
By setting up an information acquisition module, an information processing module, an information analysis module and an object tracking module in the monitoring center, the deep convolutional neural network model and signal intensity analysis can monitor the position of the moving target in real time, and adjust the shooting range of the information acquisition module to keep the target in the positioning area.
The movement target is achieved within the shooting range and in the center of the picture, ensuring the stability and clarity of the shooting effect.
Smart Images

Figure CN114693738B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of visual tracking technology, and in particular to a system for continuously tracking and identifying moving targets. Background Art
[0002] A stationary camera without a motion feedback device can only observe the movement of the target within its field of view. However, the camera mounted on the optoelectronic pod can obtain a clear image of the target and track and shoot the moving target under the movement of the image stabilization mechanism. The effectiveness of achieving real-time target tracking depends largely on the motion control algorithm of the image stabilization mechanism.
[0003] When an object is in motion, it often leaves the camera's shooting range or is not in the center of the picture, resulting in poor shooting results. How to ensure that the object is within the shooting range and in the center of the picture is a problem we need to solve. We now provide a system for continuously tracking and identifying moving targets. Summary of the Invention
[0004] The purpose of the present invention is to provide a system for continuously tracking and identifying moving targets.
[0005] The object of the present invention can be achieved by the following technical solutions: A system for continuously tracking and identifying a moving target, comprising a monitoring center, wherein the monitoring center is communicatively and / or electrically connected to an information acquisition module, an information processing module, an information analysis module, and an object tracking module;
[0006] The information acquisition module is used to obtain dynamic images and signal strength of moving targets;
[0007] The information processing module is used to process the received dynamic image signal strength of the moving target, perform feature recognition on the moving target, and output the recognition result;
[0008] The information analysis module is used to analyze whether the moving target is located in the positioning area during the movement process, and determine whether the moving target is out of the tracking range;
[0009] The object tracking module is used to correct the shooting range of the information acquisition module when the moving object leaves the positioning area.
[0010] Furthermore, the process of the information acquisition module acquiring the dynamic image and signal strength of the moving target includes:
[0011] Construct a visual capture area according to the image shooting range of the information acquisition module, and establish a two-dimensional coordinate system with the center point of the visual capture area as the origin;
[0012] A positioning area is set with the origin as the center; two mutually perpendicular diameters are made in the positioning area, and the four intersection points generated by the intersection of the two diameters and the circular area are marked as positioning points; a motion image of the moving target is obtained through the information acquisition module; a signal transmitter is installed on the moving target to obtain the signal strength of the signal transmitter received by the four positioning points.
[0013] Furthermore, the positioning points include positioning point A, positioning point B, positioning point C and positioning point D; and positioning point A and positioning point B are located at two ends of the same diameter, and positioning point C and positioning point D are located at two ends of the same diameter.
[0014] Furthermore, the process of the information processing module processing the received dynamic image signal strength of the moving target includes:
[0015] Establish a deep convolutional neural network model, and input the dynamic image of the moving target obtained by the information acquisition module into the deep convolutional neural network model; extract the recognition features of the moving target in the obtained dynamic image of the moving target through the deep convolutional neural network model;
[0016] Establish a two-dimensional coordinate system of time with respect to signal strength, and establish a signal strength variation curve with respect to distance based on the signal transmitter installed on the moving target according to the four positioning points;
[0017] Establishing a signal attenuation variation curve of signal strength with respect to distance based on a signal transmitter installed on a moving target;
[0018] According to the signal strength received by the positioning point, the distance between the corresponding positioning point and the signal transmitting end is obtained on the signal attenuation change curve.
[0019] Furthermore, the information analysis module analyzes whether the moving target is out of the tracking range during the movement process, including:
[0020] The obtained dynamic image of the moving target is rasterized to obtain a video frame of the moving target; the video frame of the moving target is numbered as i, i=1, 2, ..., n, where n is an integer; the distances from the four positioning points in the video frames numbered i=1 and i=n to the signal transmitting end of the moving target are obtained respectively; and the distance differences between the four positioning points and the signal transmitting end are obtained;
[0021] Set the distance threshold L. When the distance difference between positioning points A and B and the signal transmitter is CAB, and the distance difference between positioning points C and D and the signal transmitter is CCD, and the distance between at least one of the four positioning points and the signal transmitter is less than L, the target is being tracked. When the distance between all positioning points and the signal transmitter is not less than L, the target is out of tracking range.
[0022] When the distance difference between positioning point A and positioning point B to the signal transmitter is CAB and is not within the distance deviation threshold range, or the distance difference between positioning point C and positioning point D to the signal transmitter is CCD and is not within the distance deviation threshold range, it indicates that the moving target has left the positioning area.
[0023] Furthermore, the process of the object tracking module correcting the shooting range of the information acquisition module includes:
[0024] When the moving target deviates from the positioning area, the distance between the positioning point and the signal transmitter is determined based on the signal strength obtained from the signal transmitter;
[0025] Adjust the shooting direction of the information acquisition module so that the distance from the signal transmitter to the four positioning points is consistent. Then adjust the distance between the information acquisition module and the moving target so that the distance between the moving target and the information acquisition module is always within a reasonable range and the moving target is located within the positioning area.
[0026] Compared with the prior art, the beneficial effect of the present invention is: by obtaining the signal strength between the information acquisition module and the moving target, the distance between the information acquisition module and the moving target is determined, and the distance between the information acquisition module and the moving target is monitored, so that the moving target can always be in the positioning area, and at the same time the distance between the information acquisition module and the moving target is maintained in an appropriate range. When the moving target leaves the positioning area or the distance between the moving target and the information acquisition module is too far, the information acquisition module is adjusted, so that the distance between the information acquisition module and the moving target remains stable, and at the same time the moving target is always within the positioning area. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 This is a schematic diagram of the present invention. DETAILED DESCRIPTION
[0028] like Figure 1 As shown, a system for continuously tracking and identifying a moving target includes a monitoring center, wherein the monitoring center is communicatively and / or electrically connected to an information acquisition module, an information processing module, an information analysis module, and an object tracking module;
[0029] The information acquisition module is used to obtain dynamic images and signal strength of moving targets. The specific process includes:
[0030] Construct a visual capture area according to the image shooting range of the information acquisition module, and establish a two-dimensional coordinate system with the center point of the visual capture area as the origin;
[0031] A positioning area is set with the origin as the center; it should be further explained that, in the specific implementation process, the positioning area is a circular area with the origin as the center and a radius of R;
[0032] Draw two mutually perpendicular diameters within the positioning area, and mark the four intersection points of the two diameters and the circular area as positioning points, which are respectively recorded as positioning point A, positioning point B, positioning point C, and positioning point D. It should be further explained that in the specific implementation process, positioning point A and positioning point B are located at the ends of the same diameter, and positioning point C and positioning point D are located at the ends of the same diameter;
[0033] Acquire the motion image of the moving target through the information acquisition module;
[0034] Install a signal transmitter on the moving target and obtain the signal strength of the signal transmitter received by the four positioning points;
[0035] The dynamic image of the moving target acquired by the information acquisition module and the signal strength received by the four positioning points are sent to the information processing module.
[0036] The information processing module is used to process the received dynamic image of the moving target and the signal strength received by the four positioning points. The specific processing process includes:
[0037] Establish a deep convolutional neural network model, and input the dynamic image of the moving target obtained by the information acquisition module into the deep convolutional neural network model;
[0038] Establishing a moving target feature library within the deep convolutional neural network model, extracting identification features of the moving target from the obtained dynamic image of the moving target through the deep convolutional neural network model, and outputting the extracted identification features of the moving target;
[0039] According to the extracted identification features of the moving target, the moving target is identified and marked within the visual capture area;
[0040] Establish a two-dimensional coordinate system of time with respect to signal strength, and establish a signal strength change curve of signal strength with respect to distance based on the signal transmitter installed on the moving target according to the four positioning points, and map the signal strength change curve into the two-dimensional coordinate system to obtain a signal strength change graph;
[0041] A signal attenuation curve of signal strength with respect to distance is established based on the signal transmitter installed on the moving target. It should be further explained that, in a specific implementation process, the signal strength of the signal transmitted by the signal transmitter received by the positioning point will attenuate as the distance between the positioning point and the signal transmitter increases. Therefore, the distance between the positioning point and the signal transmitter can be inferred based on the signal strength of the signal transmitter received by the positioning point.
[0042] According to the signal strength received by the positioning point, the distance between the corresponding positioning point and the signal transmitter is obtained on the signal attenuation change curve, and the distances between positioning point A, positioning point B, positioning point C, and positioning point D and the signal transmitter are marked as JA, JB, JC, and JD respectively;
[0043] The obtained data is sent to the information analysis module.
[0044] The information analysis module is used to analyze whether the moving target is located in the positioning area during the movement process. The specific analysis process includes:
[0045] Set a time period threshold T, and every T time period, rasterize the dynamic images of the moving target obtained within the T time period to obtain the video frames of the moving target, and mark the video frames of the moving target respectively;
[0046] The video frames of the moving target are labeled as i, i=1, 2, ..., n, where n is an integer; the video frames with label i=1 and label i=n are labeled, and the distances from the four positioning points in the video frames with label i=1 and label i=n to the signal transmitting end of the moving target are obtained respectively;
[0047] Set the distance deviation threshold interval (0, a) to obtain the distance differences between the four positioning points and the signal transmitter. That is, the distance difference between positioning points A and B and the signal transmitter is CAB; the distance difference between positioning points C and D and the signal transmitter is CCD. Where CAB = |JA - JB|, CCD = |JC - JB|.
[0048] It should be further explained that, in the specific implementation process, in the initial state, the distance difference between positioning point A and positioning point B and the signal transmitting end CAB = 0; the distance difference between positioning point C and positioning point D and the signal transmitting end CCD = 0, that is, the moving target is at the center point of the positioning area;
[0049] Set the distance threshold L. When the distance difference between positioning points A and B and the signal transmitter is CAB, and the distance difference between positioning points C and D and the signal transmitter is CCD, and the distance between at least one of the four positioning points and the signal transmitter is less than L, the target is being tracked. When the distance between all positioning points and the signal transmitter is not less than L, the target is out of tracking range.
[0050] If the distance difference between positioning point A and positioning point B and the signal transmitter is CAB and is not within the distance deviation threshold range, or the distance difference between positioning point C and positioning point D and the signal transmitter is CCD and is not within the distance deviation threshold range, it means that the moving target has left the positioning area;
[0051] When the moving target leaves the positioning area, a tracking correction instruction is generated and sent to the object tracking module.
[0052] The object tracking module is used to correct the shooting range of the information acquisition module when the moving object leaves the positioning area. The specific process includes:
[0053] When the moving target deviates from the positioning area, the distance between the positioning point and the signal transmitter is determined based on the signal strength obtained from the signal transmitter;
[0054] Adjust the shooting direction of the information acquisition module so that the distance from the signal transmitter to the four positioning points is consistent. Then adjust the distance between the information acquisition module and the moving target so that the distance between the moving target and the information acquisition module is always within a reasonable range and the moving target is located within the positioning area.
[0055] The above embodiments are only used to illustrate the technical method of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.
Claims
1. A system for continuously tracking and identifying moving targets, including a monitoring center, characterized in that: The monitoring center is communicatively and / or electrically connected to an information acquisition module, an information processing module, an information analysis module, and an object tracking module; The information acquisition module is used to obtain dynamic images and signal strength of moving targets; The information processing module is used to process the received dynamic image signal strength of the moving target, perform feature recognition on the moving target, and output the recognition result; The information analysis module is used to analyze whether the moving target is located in the positioning area during the movement process, and determine whether the moving target is out of the tracking range; The object tracking module is used to correct the shooting range of the information acquisition module when the moving object leaves the positioning area; The process of the information acquisition module acquiring the dynamic image and signal strength of the moving target includes: Construct a visual capture area according to the image shooting range of the information acquisition module, and establish a two-dimensional coordinate system with the center point of the visual capture area as the origin; Set up a positioning area with the origin as the center; draw two mutually perpendicular diameters within the positioning area, and mark the four intersection points of the two diameters and the circular area as positioning points; obtain the moving image of the moving target through the information acquisition module; install a signal transmitter on the moving target and obtain the signal strength of the signal transmitter received at the four positioning points; The positioning points include positioning point A, positioning point B, positioning point C and positioning point D; positioning point A and positioning point B are located at the two ends of the same diameter, and positioning point C and positioning point D are located at the two ends of the same diameter; The information analysis module analyzes whether the moving target is out of the tracking range during the movement process, including: The obtained dynamic image of the moving target is rasterized to obtain a video frame of the moving target; the video frame of the moving target is numbered as i, i=1, 2, ..., n, where n is an integer; the distances from the four positioning points in the video frames numbered i=1 and i=n to the signal transmitting end of the moving target are obtained respectively; and the distance differences between the four positioning points and the signal transmitting end are obtained; Set the distance threshold L. When the distance difference between positioning points A and B and the signal transmitter is CAB, and the distance difference between positioning points C and D and the signal transmitter is CCD, and the distance between at least one of the four positioning points and the signal transmitter is less than L, the target is being tracked. When the distance between all positioning points and the signal transmitter is not less than L, the target is out of tracking range. When the distance difference between positioning point A and positioning point B to the signal transmitter is CAB and is not within the distance deviation threshold range, or the distance difference between positioning point C and positioning point D to the signal transmitter is CCD and is not within the distance deviation threshold range, it indicates that the moving target has left the positioning area.
2. A system for continuously tracking and identifying moving targets according to claim 1, characterized in that: The process of the information processing module processing the received dynamic image signal strength of the moving target includes: Establish a deep convolutional neural network model, and input the dynamic image of the moving target obtained by the information acquisition module into the deep convolutional neural network model; extract the recognition features of the moving target in the obtained dynamic image of the moving target through the deep convolutional neural network model; Establish a two-dimensional coordinate system of time with respect to signal strength, and establish a signal strength variation curve with respect to distance based on the signal transmitter installed on the moving target according to the four positioning points; Establishing a signal attenuation variation curve of signal strength with respect to distance based on a signal transmitter installed on a moving target; According to the signal strength received by the positioning point, the distance between the corresponding positioning point and the signal transmitting end is obtained on the signal attenuation change curve.
3. A system for continuously tracking and identifying moving targets according to claim 1, characterized in that: The process of the object tracking module correcting the shooting range of the information acquisition module includes: When the moving target deviates from the positioning area, the distance between the positioning point and the signal transmitter is determined based on the signal strength obtained from the signal transmitter; Adjust the shooting direction of the information acquisition module so that the distance from the signal transmitter to the four positioning points is consistent. Then adjust the distance between the information acquisition module and the moving target so that the distance between the moving target and the information acquisition module is always within a reasonable range and the moving target is located within the positioning area.
Citation Information
Patent Citations
Indoor positioning method and indoor positioning system
CN106842123A
Moving target tracking method
CN113873152A