CN216916282U - Infrared and visible light cross-modal data fusion search robot - Google Patents
Infrared and visible light cross-modal data fusion search robot Download PDFInfo
- Publication number
- CN216916282U CN216916282U CN202123175266.4U CN202123175266U CN216916282U CN 216916282 U CN216916282 U CN 216916282U CN 202123175266 U CN202123175266 U CN 202123175266U CN 216916282 U CN216916282 U CN 216916282U
- Authority
- CN
- China
- Prior art keywords
- visible light
- pan
- tilt
- infrared
- zoom
- 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
Landscapes
- Closed-Circuit Television Systems (AREA)
Abstract
The utility model relates to the technical field of security monitoring, in particular to a searching robot for infrared and visible light cross-modal data fusion. The infrared visible light cross-modal data fusion search robot comprises a support frame, a three-dimensional pan-tilt zoom pan-tilt camera and a GPU edge computer platform, wherein an unmanned aerial vehicle is installed at the top end of the support frame, a mounting plate is installed below the support frame, the three-dimensional pan-tilt zoom pan-tilt camera is installed at the bottom end of the mounting plate, the three-dimensional pan-tilt zoom pan-tilt camera comprises a thermal imaging machine core, a zooming visible light machine core, an augmentation stabilizing pan-tilt, a digital-image integrated system and an electrical interface, the GPU edge computer platform is fixedly installed at the upper end of the mounting plate, and the GPU edge computer platform comprises a video acquisition module, a moving target detection module, a target tracking module and a control module. The utility model has the advantages of low cost, wide coverage, high intellectualization, rapid deployment and wider applicability.
Description
Technical Field
The utility model relates to the technical field of security monitoring, in particular to a searching robot for infrared and visible light cross-modal data fusion.
Background
The security protection measure of the important target point location is a prominent practical problem in national defense construction. In actual security and protection requirements, an effective coverage area of security and protection is generally required to be more than 1 kilometer, and all-weather uninterrupted monitoring of an open space and three-dimensional interval is required to be completed simultaneously. The video monitoring system is a group of equipment consisting of acquisition, transmission, storage, control and display parts, and the equipment can record, playback and process the production and the life of various industries in an image or video mode.
Traditional security protection adopts modes such as sentry patrol, fixed point video monitoring more, has following shortcoming:
1. a plurality of cameras need to be installed in a distributed mode, the cost of the system is too high, and operation and maintenance are difficult;
2. the ability to adjust the monitoring angle and focus in real time is lacking;
3. lack of interactivity and cannot guarantee real-time analysis and prompt sending of potential risks.
Therefore, there is a need to provide a new searching robot for infrared-visible light cross-modal data fusion to solve the above technical problems.
SUMMERY OF THE UTILITY MODEL
In order to solve the technical problems, the utility model provides the infrared and visible light cross-modal data fusion search robot which is low in cost, wide in coverage range, high in intelligence, rapid in deployment and wider in applicability.
The utility model provides an infrared and visible light cross-modal data fusion searching robot, which comprises: the top end of the support frame is fixedly provided with the unmanned aerial vehicle, and the support frame is positioned below the unmanned aerial vehicle and is fixedly provided with the mounting plate;
three-dimensional cloud platform zooms cloud platform camera, three-dimensional cloud platform zooms cloud platform camera and installs in the mounting panel bottom, and three-dimensional cloud platform zooms cloud platform camera and includes thermal imaging core, zooms visible light core, increases steady cloud platform, a digital map system and electrical interface, thermal imaging core, zooms visible light core, digital map integrated system and electrical interface set up in three-dimensional cloud platform zooms cloud platform camera organism and through increasing steady cloud platform and mounting panel fixed connection:
the system comprises a GPU edge computer platform, wherein the GPU edge computer platform is fixedly arranged at the upper end of a mounting plate and is electrically connected with an unmanned aerial vehicle and a three-dimensional pan-tilt-zoom pan-tilt camera through cables, the GPU edge computer platform comprises a video acquisition module, a moving target detection module, a target tracking module and a control module, the video acquisition module is used for acquiring video data, the moving target detection module utilizes a proper motion detection algorithm to extract a moving target, the target tracking module tracks the target by utilizing a target tracking algorithm to determine the position of the target, and the control module judges whether to send a signal to a control center according to the position of the target to control the three-dimensional pan-tilt-zoom pan-tilt camera and the unmanned aerial vehicle to move.
Preferably, the thermal imaging cartridge has a specification of 320x240 thermal imaging blade.
Preferably, the zooming visible light movement adopts low illumination SENSOR, the effective pixel is 400 thousands, and 10 times optical zooming is integrated in the zooming visible light movement.
Preferably, the stability augmentation cloud platform is a high-precision professional two-axis brushless stability augmentation zooming dual-optical pod, and the stability augmentation cloud platform adopts a high-precision encoder FOC control scheme and is internally provided with a damping ball.
Preferably, the stability augmentation cloud platform is a high-precision professional two-axis brushless stability augmentation zooming dual-optical pod, and the stability augmentation cloud platform adopts a high-precision encoder FOC control scheme and is internally provided with a damping ball.
Preferably, the digital-image integrated system supports network picture-in-picture output, local TF storage, two-way synchronous video recording, ground station multiple picture-in-picture mode switching, optical zooming and thermal imaging digital zooming.
Preferably, the electrical interface includes, but is not limited to, a power interface, a communication interface, a TF card interface, and an antenna interface.
Compared with the related technology, the infrared and visible light cross-modal data fusion searching robot provided by the utility model has the following beneficial effects:
1. the utility model provides a search robot for infrared-visible light cross-modal data fusion, which comprises a three-dimensional pan-tilt zoom pan-tilt camera, a search module, a display module and a search module, wherein the three-dimensional pan-tilt zoom pan-tilt camera is used for performing significant area analysis through visible light and infrared dual-mode images to search suspicious areas; then, acquiring a high-definition image of a suspicious region through the rotation of the stability-increasing cradle head and the zooming of the camera; the monitoring angle and the focusing capability can be adjusted in real time, and the level of video monitoring is greatly improved. Not only the effective monitoring range is increased; meanwhile, the influence of factors such as low image resolution on the recognition result is reduced, whether the target area has potential danger or not is analyzed through technologies such as target detection based on the high-definition image of the local area, and a prompt is sent;
2. and acquiring data by using a data acquisition module by using a GPU edge computer platform, and tracking the advancing of the pedestrian and making a pan-tilt camera follow up according to a target tracking algorithm after a suspicious region is detected. And then, the pan-tilt camera carries out intelligent zooming, and suspicious target judgment is realized by acquiring a high-definition image of a suspicious region after the pan-tilt zooms.
Drawings
Fig. 1 is a schematic structural diagram of a search robot with cross-modal infrared and visible light data fusion according to a preferred embodiment of the present invention;
fig. 2 is a schematic structural diagram of another view angle of the infrared-visible light cross-modal data fusion search robot provided by the present invention;
fig. 3 is a schematic structural diagram of the three-dimensional pan/tilt/zoom camera shown in fig. 1;
FIG. 4 is a block diagram of the GPU-edge computer platform shown in FIG. 1;
fig. 5 is a flowchart of the work of the infrared-visible light cross-modal data fusion search robot provided by the present invention.
Reference numbers in the figures: 1. a support frame; 2. an unmanned aerial vehicle; 3. mounting a plate; 4. a three-dimensional pan-tilt zoom pan-tilt camera; 41. a thermal imaging engine core; 42. zooming the visible light movement; 43. a stability augmentation holder; 44. a figure-counting integrated system; 45. an electrical interface; 5. a GPU-edge computer platform; 51. a video acquisition module; 52. a moving object detection module; 53. a target tracking module; 54. and a control module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the utility model and are not intended to limit the utility model.
Specific implementations of the present invention are described in detail below with reference to specific embodiments.
Referring to fig. 1 to fig. 5, an infrared-visible light cross-modal data fusion search robot according to an embodiment of the present invention includes: the system comprises a support frame 1, an unmanned aerial vehicle 2, a mounting plate 3, a three-dimensional pan-tilt-zoom camera 4 and a GPU edge computer platform 5.
The three-dimensional pan-tilt-zoom pan-tilt camera 4 is installed at the bottom end of the installation plate 3, the three-dimensional pan-tilt-zoom pan-tilt camera 4 comprises a thermal imaging core 41, a zooming visible light core 42, a stability-increasing pan-tilt 43, a digital-image integrated system 44 and an electrical interface 45, and the thermal imaging core 41, the zooming visible light core 42, the digital-image integrated system 44 and the electrical interface 45 are integrally arranged in the three-dimensional pan-tilt-zoom pan-tilt camera 4 and are fixedly connected with the installation plate 3 through the stability-increasing pan-tilt 43;
the GPU edge computer platform 5 is fixedly installed at the upper end of the installing plate 3 and is electrically connected with the unmanned aerial vehicle 2 and the three-dimensional pan-tilt-zoom-pan-tilt camera 4 through cables, the GPU edge computer platform 5 comprises a video acquisition module 51, a moving target detection module 52, a target tracking module 53 and a control module 54, the video acquisition module 51 is used for acquiring video data, the moving target detection module 52 extracts moving targets by using a proper motion detection algorithm, the target tracking module 53 tracks the targets by using a target tracking algorithm and determines the positions of the targets, and the control module 54 judges whether to send signals to a control center according to the positions of the targets and controls the three-dimensional pan-tilt-zoom-pan-tilt camera 4 and the unmanned aerial vehicle 2 to move.
It should be noted that: when the system is used, the thermal imaging movement 41 and the zooming visible light movement 42 of the three-dimensional pan-tilt-zoom camera 4 are used for infrared and visible light double-mode significant area analysis to search suspicious areas; then, a high-definition image of a suspicious region is obtained through the rotation of the stability-increasing cradle head 43 and the zooming of the camera; based on the high-definition image of the local area, analyzing whether a potential danger exists in the target area through technologies such as target detection and the like, sending a prompt, then starting suspicious area intelligent detection when the video acquisition module 51 of the GPU edge computer platform 5 is used for acquiring the attitude information and video stream information of the camera, carrying out suspicious area detection on the visual field area of the three-dimensional pan/tilt/zoom camera 4, when the suspicious area is detected, tracking the progress of a pedestrian by using a moving target detection module 52 and a target tracking module 53 according to a target tracking algorithm, controlling the unmanned aerial vehicle 2 to drive the three-dimensional pan/tilt/zoom camera 4 to follow up through a control module 54, then carrying out intelligent zooming by the three-dimensional pan/tilt/zoom camera 4, and acquiring a suspicious area high-definition image to realize suspicious target judgment after the three-dimensional pan/tilt/zoom camera 4, the whole robot has the advantages of low cost, wide coverage range, high intellectualization, rapid deployment and wider applicability.
The specification of the thermal imaging engine core 41 is 320 × 240 thermal imaging barrier, which is convenient for protecting the thermal imaging engine core 41.
The zooming visible light movement 42 adopts low-illumination SENSOR, the effective pixels are 400 thousands, 10 times of optical zooming is integrated in the zooming visible light movement, flexible zooming adjustment can be carried out according to the visible light environments with different illumination intensities, and clearer images can be shot.
In an embodiment of the present invention, please refer to fig. 1 and fig. 3, the stability-increasing cradle 43 is a high-precision professional two-axis brushless stability-increasing zooming dual-optical pod, and the stability-increasing cradle 43 adopts a high-precision encoder FOC control scheme and is internally provided with a damping ball.
It should be noted that: the high-precision professional two-axis brushless stability-increasing zooming dual-optical pod is a professional video camera pan-tilt commonly used in the prior art, and is internally provided with a damping ball, so that a three-dimensional pan-tilt zoom pan-tilt camera 4 is more stable in the running and advancing process, the shooting is convenient, the monitoring angle and the focusing capability can be adjusted in real time, the level of video monitoring is greatly improved, and the effective monitoring range is enlarged; meanwhile, the influence of factors such as low image resolution on the recognition result is reduced.
In an embodiment of the present invention, referring to fig. 1 and 5, the digital image integrated system 44 supports network pip output, local TF storage, two-way simultaneous video recording, ground station pip mode switching, optical zoom, thermal imaging digital zoom.
It should be noted that: thus, after the three-dimensional pan/tilt/zoom camera 4 shoots, the video acquisition, the local TF storage of image data, and the network picture-in-picture output and switching can be performed in real time through the integrated image counting system 44, which is convenient for ground monitoring.
In an embodiment of the present invention, referring to fig. 1 and 2, the electrical interface 45 includes, but is not limited to, a power interface, a communication interface, a TF card interface, and an antenna interface.
It should be noted that: the power interface of the electrical interface 45 is used for switching on a power supply, the video and control can be carried out through the antenna interface, the two-way communication with the ground station in a wireless mode is achieved, the TF card interface is inserted into the TF memory card for local data storage, the communication interface comprises IBUS, SBUS and USB interfaces, the posture of the cradle head can be controlled through the serial port command, the IBUS and the SBUS to move at a certain angular rate in the course and the pitching direction, and the USB interface can be used for debugging the stability-increasing parameters of the cradle head.
The working principle of the infrared visible light cross-modal data fusion searching robot provided by the utility model is as follows:
when the system is used, the thermal imaging movement 41 and the zooming visible light movement 42 of the three-dimensional pan-tilt-zoom camera 4 are used for infrared and visible light double-mode significant area analysis to search suspicious areas; then, a high-definition image of a suspicious region is obtained through the rotation of the stability-increasing cradle head 43 and the zooming of the camera; based on the high-definition image of the local area, analyzing whether a potential danger exists in the target area through technologies such as target detection and the like, sending a prompt, then starting suspicious area intelligent detection when the video acquisition module 51 of the GPU edge computer platform 5 is used for acquiring the attitude information and video stream information of the camera, carrying out suspicious area detection on the visual field area of the three-dimensional pan/tilt/zoom camera 4, when the suspicious area is detected, tracking the progress of a pedestrian by using a moving target detection module 52 and a target tracking module 53 according to a target tracking algorithm, controlling the unmanned aerial vehicle 2 to drive the three-dimensional pan/tilt/zoom camera 4 to follow up through a control module 54, then carrying out intelligent zooming by the three-dimensional pan/tilt/zoom camera 4, and acquiring a suspicious area high-definition image to realize suspicious target judgment after the three-dimensional pan/tilt/zoom camera 4, finally, video acquisition, image data storage of local TF, network picture-in-picture output and switching can be performed in real time through the integrated digital-picture system 44, so that ground monitoring is facilitated, and security and protection processing can be performed in time.
The circuits and controls involved in the present invention are prior art and will not be described in detail herein.
The above description is only an embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (6)
1. An infrared-visible cross-modal data-fused search robot, comprising:
the support frame (1), the top end of the support frame (1) is fixedly provided with an unmanned aerial vehicle (2), and the support frame (1) is positioned below the unmanned aerial vehicle (2) and is fixedly provided with a mounting plate (3);
it is characterized by also comprising:
three-dimensional cloud platform zooms cloud platform camera (4), three-dimensional cloud platform zooms cloud platform camera (4) and installs in mounting panel (3) bottom, and three-dimensional cloud platform zooms cloud platform camera (4) including thermal imaging core (41), zoom visible light core (42), increase steady cloud platform (43), digital picture body system (44) and electrical interface (45), thermal imaging core (41), zoom visible light core (42), digital picture body system (44) and electrical interface (45) integration set up in three-dimensional cloud platform zooms cloud platform camera (4) organism and through increasing steady cloud platform (43) and mounting panel (3) fixed connection:
a GPU edge computer platform (5), wherein the GPU edge computer platform (5) is fixedly arranged at the upper end of the mounting plate (3) and is electrically connected with the unmanned aerial vehicle (2) and the three-dimensional pan-tilt-zoom-pan-tilt camera (4) through cables, and the GPU edge computer platform (5) comprises a video acquisition module (51), a moving object detection module (52), an object tracking module (53) and a control module (54), the video acquisition module (51) is used for obtaining video data, the moving object detection module (52) extracts moving objects by applying a proper motion detection algorithm, the target tracking module (53) tracks the target by using a target tracking algorithm to determine the position of the target, and the control module (54) judges whether to send a signal to a control center according to the position of the target, and controls the three-dimensional pan-tilt-zoom pan-tilt camera (4) and the unmanned aerial vehicle (2) to move.
2. The infrared-visible cross-modal data fusion search robot of claim 1, wherein the thermal imaging engine core (41) is of a specification of 320x240 thermal imaging blade.
3. The infrared-visible light cross-modal data fusion search robot as recited in claim 1, wherein the zoom visible light engine (42) employs a low-light SENSOR, an effective pixel of 400 thousands, and a 10-fold optical zoom is integrated therein.
4. The infrared visible light cross-modal data fusion search robot as recited in claim 1, characterized in that the stability augmentation cloud deck (43) is a high-precision professional two-axis brushless stability augmentation zooming dual-light pod, and the stability augmentation cloud deck (43) adopts a high-precision encoder FOC control scheme and is internally provided with a damping ball.
5. The infrared visible light cross-modal data fusion search robot of claim 1, wherein the map-integrated system (44) supports network picture-in-picture output, local TF storage, two-way synchronous video, ground station multiple picture-in-picture mode switching, optical zoom, thermal imaging digital zoom.
6. The infrared-visible light cross-modal data fusion search robot of claim 1, wherein the electrical interface (45) includes, but is not limited to, a power interface, a communication interface, a TF card interface, and an antenna interface.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202123175266.4U CN216916282U (en) | 2021-12-16 | 2021-12-16 | Infrared and visible light cross-modal data fusion search robot |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202123175266.4U CN216916282U (en) | 2021-12-16 | 2021-12-16 | Infrared and visible light cross-modal data fusion search robot |
Publications (1)
Publication Number | Publication Date |
---|---|
CN216916282U true CN216916282U (en) | 2022-07-08 |
Family
ID=82252420
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202123175266.4U Active CN216916282U (en) | 2021-12-16 | 2021-12-16 | Infrared and visible light cross-modal data fusion search robot |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN216916282U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115361499A (en) * | 2022-08-16 | 2022-11-18 | 南京甄视智能科技有限公司 | Dual-computer cooperative frontier defense target identification and tracking system and method |
-
2021
- 2021-12-16 CN CN202123175266.4U patent/CN216916282U/en active Active
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115361499A (en) * | 2022-08-16 | 2022-11-18 | 南京甄视智能科技有限公司 | Dual-computer cooperative frontier defense target identification and tracking system and method |
CN115361499B (en) * | 2022-08-16 | 2024-03-12 | 小视科技(江苏)股份有限公司 | Dual-machine cooperative border defense target recognition and tracking system and method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105744163B (en) | A kind of video camera and image capture method based on depth information tracking focusing | |
JP4188394B2 (en) | Surveillance camera device and surveillance camera system | |
CN104639916A (en) | Large-scene multi-target tracking shooting video monitoring system and monitoring method thereof | |
CN1937766A (en) | Surveillance camera apparatus and surveillance camera system | |
CN104125372B (en) | Target photoelectric search and detection method | |
CN112699839B (en) | Automatic video target locking and tracking method under dynamic background | |
CN110580054B (en) | Control system and method of photoelectric pod based on autonomous visual tracking | |
CN206260046U (en) | A kind of thermal source based on thermal infrared imager and swarm into tracks of device | |
CN210526874U (en) | Airborne three-light photoelectric pod system | |
CN101511005A (en) | Multi-item panorama infrared locking and tracking camera shot system | |
CN216916282U (en) | Infrared and visible light cross-modal data fusion search robot | |
CN109917364A (en) | A kind of warning system of integrated radar guiding and photoelectric tracking function | |
CN111432187B (en) | Method, system and device for tracking transaction target | |
WO2015185161A1 (en) | System and method for remote monitoring at least one observation area | |
CN105245845A (en) | Method for controlling camera to follow and shoot automatically based on gathering trend in match field | |
CN105208349A (en) | Method for controlling automatic following shot of cameras according to number of people on game site | |
CN103108164A (en) | Compound eye type panorama continuous tracking and monitoring system | |
CN112040140A (en) | Wide-view-field high-resolution hybrid imaging device based on light field | |
CN114554093B (en) | Image acquisition system and target tracking method | |
CN104113733A (en) | Television search detection method for small slow low-altitude target | |
CN103634509A (en) | Automatic tracking recording method and system | |
CN113743286A (en) | Target monitoring system and method for multi-source signal fusion | |
KR101118926B1 (en) | System for observation moving objects | |
CN109785562A (en) | A kind of vertical photoelectricity ground based threats warning system and suspicious object recognition methods | |
CN112764052A (en) | Air defense missile flight monitoring system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |