CN105791686A - Multi-moving object capturing and tracking device and multi-moving object capturing and tracking method - Google Patents
Multi-moving object capturing and tracking device and multi-moving object capturing and tracking method Download PDFInfo
- Publication number
- CN105791686A CN105791686A CN201610121742.9A CN201610121742A CN105791686A CN 105791686 A CN105791686 A CN 105791686A CN 201610121742 A CN201610121742 A CN 201610121742A CN 105791686 A CN105791686 A CN 105791686A
- Authority
- CN
- China
- Prior art keywords
- tracked
- moving target
- acousto
- optic modulator
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/698—Control of cameras or camera modules for achieving an enlarged field of view, e.g. panoramic image capture
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/68—Control of cameras or camera modules for stable pick-up of the scene, e.g. compensating for camera body vibrations
- H04N23/681—Motion detection
Abstract
The invention provides a multi-moving object capturing and tracking device and a multi-moving object capturing and tracking method. The device comprises a panoramic image collector, a controller electrically connected with the panoramic image collector, and multiple branch lens photographing systems electrically connected with the controller. The panoramic image collector transmits collected panoramic environment images to the controller. The controller determines multiple to-be-tracked moving objects and the positions of the to-be-tracked moving objects according to the panoramic environment images and sends corresponding moving object tracking instructions to the branch lens photographing systems; the branch lens photographing systems reflect the light rays from the to-be-tracked moving objects and capture and track the multiple to-be-tracked moving objects by using a light ray reflection principle when lens directions are unchanged. According to the method and the device, the multiple moving objects are captured and tracked through the light ray reflection principle when the directions of the photographing lenses are kept unchanged, the problem that the adjusting speed is low when the lenses are adjusted mechanically, and the capturing and tracking speeds of the multiple moving objects are improved.
Description
Technical field
The present invention relates to field of optical equipment, particularly relate to multiple mobile object seizure and catch and tracking with tracking equipment and multiple mobile object.
Background technology
Multiple mobile object seizure and tracking are the study hotspots in computer vision field, and multiple mobile object catches and tracking refers to and utilizes computer to determine the movement locus that the position of multiple self-movement targets, size and each moving target interested, to have certain marked feature is complete in the video sequence.In recent years, along with being skyrocketed through and the development of image analysis technology of computer digital animation ability, the real-time capture of moving target and tracking technique are shown one's talent.Multiple moving targets are carried out real-time capture and tracking, has important practical value in fields such as video monitoring, video compression coding, robot navigation and location, intelligent human-machine interaction and virtual realities.
Prior art provides seizure and the tracking equipment of multiple mobile object, and this equipment includes controller, the monopod video camera being electrically connected with the controller.The operation principle of this equipment is: obtains the image of moving target by monopod video camera and transmits to controller, the controller image to receiving carries out moving object detection and tracking, determine the multiple moving targets needing to follow the tracks of, control instruction is sent to monopod video camera, after monopod video camera receives control instruction, the shooting direction of adjustable lens under the control of control instruction, is tracked shooting to needing the multiple moving targets followed the tracks of.
In prior art, utilizing monopod video camera that multiple moving targets are tracked shooting, owing to the position of multiple moving targets dynamically changes, therefore monopod video camera needs timely adjustable lens direction.But, monopod video camera adopts the mode adjustable lens direction of mechanical rotation, has the defect that governing speed is slow, causes that the seizure tracking velocity to multiple moving targets reduces.
Summary of the invention
The invention provides multiple mobile object seizure to catch and tracking with tracking equipment and multiple mobile object, seizure and the tracking of moving target can be carried out by light principle of reflection when keeping taking lens to point to constant, avoid the problem that mechanical adjustment lens direction governing speed is slow, improve the seizure tracking velocity of multiple moving target.
First aspect, embodiments provide a kind of multiple mobile object and catch and tracking equipment, including panoramic picture harvester, and the controller electrically connected with described panoramic picture harvester, the multiple points of mirror camera systems being electrically connected with described controller；Multiple described point mirror camera systems are respectively arranged at the appointment position in the pickup area of described panoramic picture harvester；Described panoramic picture harvester is used for gathering panorama ambient image, by described panorama ambient image transmission to described controller；Described controller is used for receiving described panorama ambient image, multiple moving target to be tracked is determined according to described panorama ambient image, generate and multiple described moving target to be tracked motion target tracking instructions one to one, multiple described motion target tracking instructions are transmitted separately to point mirror camera system of correspondence；Described point of mirror camera system is used for receiving described motion target tracking instruction, according to described motion target tracking instruction, the light carrying out self-corresponding described moving target to be tracked is reflected, utilize light principle of reflection to catch and follow the tracks of the moving target described to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target described to be tracked and transmit to described controller；Described controller is additionally operable to receive the image of the multiple described moving target to be tracked that each described point mirror camera system returns, and described image carries out sorting process, generates and each described moving target to be tracked movement locus video one to one.
In conjunction with first aspect, embodiments provide the first possible embodiment of first aspect, wherein, described controller includes the GPU (GraphicProcessingUnit being electrically connected to each other, graphic process unit) and CPU (CentralProcessingUnit, central processing unit)；Described CPU electrically connects with described panoramic picture harvester, for receiving the current frame image of described panorama ambient image, described current frame image is transferred to described GPU；Described GPU, for described current frame image is carried out moving object detection, obtains the characteristics of image of each moving target in described current frame image and sends to described CPU；Wherein, the characteristics of image of described moving target includes the image coordinate of described moving target, color characteristic and textural characteristics；Described CPU electrically connects with multiple described point mirror camera systems respectively, for receiving the characteristics of image of each described moving target, the characteristics of image of each described moving target is merged respectively, and mate with the characteristics of image prestored, the position of multiple moving target to be tracked and each described moving target to be tracked is determined according to described characteristic matching result；Position according to moving target to be tracked each described generates motion target tracking instruction one to one；Determine point mirror camera system that the position of moving target to be tracked with each described is corresponding respectively；Described in determining, mirror camera system is divided by each the described motion target tracking instruction generated corresponding transmission respectively.
In conjunction with the first possible embodiment of first aspect, embodiments providing the embodiment that first aspect the second is possible, wherein, described GPU for carrying out pretreatment to described current frame image, and described pretreatment includes decoding and/or denoising；Adopt mixed Gaussian to carry out background modeling, generate the background model of predetermined number；Described pretreated current frame image is carried out moving object detection by the background model according to described predetermined number, obtains the characteristics of image of each moving target in described current frame image and sends to described CPU.
In conjunction with the first possible embodiment of first aspect, embodiments provide the third possible embodiment of first aspect, wherein, described CPU is analyzed for the position according to moving target to be tracked each described, generate light deflection parameter corresponding to each described moving target to be tracked, focus adjustment parameter and acquisition parameters, generate motion target tracking instruction according to described light deflection parameter, focus adjustment parameter and acquisition parameters；Point mirror camera system that each described moving target to be tracked is corresponding respectively is determined according to position set in advance corresponding relation；Described in determining, mirror camera system is divided by each the described motion target tracking instruction generated corresponding transmission respectively.
In conjunction with first aspect, embodiments providing the 4th kind of possible embodiment of first aspect, wherein, described motion target tracking instruction carries light deflection parameter and acquisition parameters；Described point of mirror camera system includes acousto-optic modulator and high-definition camera；Described acousto-optic modulator electrically connects with described controller, receive described motion target tracking instruction, according to the light deflection parameter that described motion target tracking instruction is carried, the light carrying out self-corresponding described moving target to be tracked is reflected, described light is reflexed in described high-definition camera；Described high-definition camera electrically connects with described controller, receives described motion target tracking instruction, corresponding moving target described to be tracked is shot when camera lens points to constant according to the acquisition parameters that described motion target tracking instruction is carried.
In conjunction with the 4th kind of possible embodiment of first aspect, embodiments provide the 5th kind of possible embodiment of first aspect, wherein, described acousto-optic modulator includes longitudinal acousto-optic modulator and horizontal acousto-optic modulator, described longitudinal acousto-optic modulator, described horizontal acousto-optic modulator and described high-definition camera are arranged in order, and described longitudinal acousto-optic modulator and described horizontal acousto-optic modulator electrically connect with described controller respectively；Described longitudinal acousto-optic modulator receives described motion target tracking instruction, and the light of the light deflection parameter carried according to described motion target tracking instruction self-corresponding described moving target to be tracked in the future longitudinally reflexes to described horizontal acousto-optic modulator；Described horizontal acousto-optic modulator receives described motion target tracking instruction, is laterally reflexed in described high-definition camera by the light after described longitudinally reflection according to the light deflection parameter that described motion target tracking instruction is carried.
In conjunction with the 5th kind of possible embodiment of first aspect, embodiments provide the 6th kind of possible embodiment of first aspect, wherein, described point of mirror camera system also includes liquid crystal lens, and described longitudinal acousto-optic modulator, described horizontal acousto-optic modulator, described liquid crystal lens and described high-definition camera are arranged in order；Described liquid crystal lens electrically connects with described controller；Described motion target tracking instruction carries focus adjustment parameter；Described liquid crystal lens receives described motion target tracking instruction, carries out focus adjustment according to the focus adjustment parameter that described motion target tracking instruction is carried；Light after described laterally reflection enters described high-definition camera through described liquid crystal lens.
In conjunction with the 4th kind of possible embodiment of first aspect, embodiments provide the 7th kind of possible embodiment of first aspect, wherein, described acousto-optic modulator includes horizontal acousto-optic modulator and longitudinal acousto-optic modulator, described point of mirror camera system also includes liquid crystal lens, and described horizontal acousto-optic modulator, described longitudinal acousto-optic modulator, described liquid crystal lens and described high-definition camera are arranged in order；Described horizontal acousto-optic modulator, described longitudinal acousto-optic modulator and described liquid crystal lens electrically connect with described controller respectively.
In conjunction with the 6th kind of possible embodiment of first aspect, embodiments provide the 8th kind of possible embodiment of first aspect, wherein, described point of mirror camera system also includes liquid crystal lens installing rack, high-definition camera installing rack, horizontal acousto-optic modulator installing rack and longitudinal acousto-optic modulator installing rack；Described liquid crystal lens installing rack, described high-definition camera installing rack, described horizontal acousto-optic modulator installing rack and described longitudinal acousto-optic modulator installing rack are respectively arranged with regulating element；The installation site of described liquid crystal lens, described high-definition camera, described horizontal acousto-optic modulator and described longitudinal acousto-optic modulator is regulated respectively by described regulating element.
Second aspect, embodiments provides a kind of multiple mobile object and catches and tracking, including: gather panorama ambient image, determine multiple moving target to be tracked according to described panorama ambient image；Generate and multiple described moving target to be tracked motion target tracking instructions one to one；Multiple described motion target tracking instructions are transmitted separately to point mirror camera system of correspondence；Wherein, the appointment position in image acquisition region is respectively arranged with multiple described point mirror camera system；The light carrying out self-corresponding described moving target to be tracked is reflected by described point of mirror camera system according to described motion target tracking instruction, utilize light principle of reflection to catch and follow the tracks of the moving target described to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target described to be tracked；The image of moving target to be tracked each described is carried out sorting process, generates and each described moving target to be tracked movement locus video one to one.
In the embodiment of the present invention, the light from moving target to be tracked is reflected by a point mirror camera system, utilizes light principle of reflection to catch and follow the tracks of multiple moving target to be tracked when camera lens points to constant.Owing to when catching and follow the tracks of multiple moving target to be tracked, camera lens points to and need not change, therefore the Apparatus and method in the present embodiment can improve the seizure tracking velocity of multiple moving target, it is to avoid the problem that in prior art, mechanical adjustment lens direction governing speed is slow.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, the accompanying drawing used required in embodiment will be briefly described below, it is to be understood that, the following drawings illustrate only certain embodiments of the present invention, therefore the restriction to scope it is not construed as, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other relevant accompanying drawings according to these accompanying drawings.
Fig. 1 illustrates that the multiple mobile object that first embodiment of the invention provides catches and a kind of structural representation of the equipment of tracking；
Fig. 2 illustrates that the multiple mobile object that first embodiment of the invention provides catches and the another kind of structural representation of the equipment of tracking；
Fig. 3 illustrates a kind of structural representation dividing mirror camera system that first embodiment of the invention provides；
Fig. 4 illustrates the another kind of structural representation dividing mirror camera system that first embodiment of the invention provides；
Fig. 5 illustrates the schematic flow sheet of the image processing algorithm that first embodiment of the invention provides；
Fig. 6 illustrates that the multiple mobile object that second embodiment of the invention provides catches the schematic flow sheet with tracking.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearly, below in conjunction with accompanying drawing in the embodiment of the present invention, technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Generally can with various different configurations arrange and design with the assembly of the embodiment of the present invention that illustrate described in accompanying drawing herein.Therefore, below the detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit claimed the scope of the present invention, but is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, the every other embodiment that those skilled in the art obtain under the premise not making creative work, broadly fall into the scope of protection of the invention.
Consider prior art to utilize monopod video camera multiple moving targets are tracked shooting, monopod video camera adopts the mode adjustable lens direction of mechanical rotation, there is the defect that governing speed is slow, cause that the seizure tracking velocity to multiple moving targets reduces, the invention provides multiple mobile object seizure to catch and tracking with tracking equipment and multiple mobile object, be specifically described below in conjunction with embodiment.
Embodiment one
Fig. 1 illustrates that the multiple mobile object that first embodiment of the invention provides catches and a kind of structural representation of the equipment of tracking.As it is shown in figure 1, the multiple mobile object in the present embodiment catches and includes with tracking equipment 1:
Panoramic picture harvester 10, the controller 20 electrically connected with panoramic picture harvester 10, the multiple points of mirror camera systems 30 being electrically connected with controller 20；Multiple points of mirror camera systems 30 are respectively arranged at the appointment position in the pickup area of panoramic picture harvester 10；
Panoramic picture harvester 10 is used for gathering panorama ambient image, by panorama ambient image transmission to controller 20；
Controller 20 is used for receiving panorama ambient image, multiple moving target to be tracked is determined according to panorama ambient image, generate and multiple moving target to be tracked motion target tracking instructions one to one, multiple motion target tracking instructions are transmitted separately to point mirror camera system 30 of correspondence；
Mirror camera system 30 is divided to be used for receiving motion target tracking instruction, according to motion target tracking instruction, the light carrying out self-corresponding moving target to be tracked is reflected, utilize light principle of reflection to catch and follow the tracks of the moving target to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target to be tracked and transmit to controller 20；
Controller 20 is additionally operable to receive the image of the moving target multiple to be tracked that each point of mirror camera system 30 returns, and this image carries out sorting process, generates and each moving target to be tracked movement locus video one to one.
In the embodiment of the present invention, the light from moving target to be tracked is reflected by a point mirror camera system 30, utilizes light principle of reflection to catch and follow the tracks of multiple moving target to be tracked when camera lens points to constant.Owing to when catching and follow the tracks of multiple moving target to be tracked, camera lens points to and need not change, therefore the equipment in the present embodiment can improve the seizure tracking velocity of multiple moving target, it is to avoid the problem that in prior art, mechanical adjustment lens direction governing speed is slow.
In the present embodiment, the pickup area of panoramic picture harvester 10 is divided into some subregions, and every sub regions all demarcates coordinate, and each sub regions is all correspondingly arranged a point of mirror camera system 30.When moving target to be tracked is positioned at a certain subregion, carried out seizure and the tracking of moving target to be tracked by point mirror camera system 30 in this subregion.
In the present embodiment, controller 20 determines the position of multiple moving target to be tracked and each moving target to be tracked according to panorama ambient image；Position according to each moving target to be tracked generates motion target tracking instruction one to one；Determine point mirror camera system 30 corresponding respectively with the position of each moving target to be tracked；By each motion target tracking instruction generated corresponding transmission respectively to point mirror camera system 30 determined.
Table 1
After table 1 illustrates that controller 20 receives panorama ambient image, the corresponding a kind of form signal dividing mirror camera system is distinguished in the moving target multiple to be tracked, the position of each moving target to be tracked that obtain after panorama ambient image is analyzed with each moving target to be tracked motion target tracking instruction one to one and the position of each moving target to be tracked.As shown in table 1, controller is in order to carry out catching to 1,2, No. 3 targets and follow the tracks of, No. 1 motion target tracking instruction is sent to No. 3 points of mirror camera systems being positioned at 3 work song regions, No. 2 motion target tracking instructions are sent to No. 1 point of mirror camera system being positioned at 1 work song region, No. 3 motion target tracking instructions are sent to No. 2 points of mirror camera systems being positioned at 2 work song regions.By table 1 it can be seen that controller 20 can generate corresponding motion target tracking instruction for each moving target to be tracked, and sending to corresponding point mirror camera system 30, therefore the equipment in the present embodiment can catch and follow the tracks of multiple moving target to be tracked.
During owing to catching and following the tracks of multiple moving target to be tracked, point mirror camera system 30 corresponding with each moving target to be tracked all can be transmitted back to the image of corresponding moving target to be tracked to controller 20, these images mix, cannot be distinguished by each moving target to be tracked, therefore the image that each point of mirror camera system 30 is also transmitted back to by controller 20 carries out sorting process, the image of each moving target to be tracked is individually sorted out, generate and each moving target to be tracked movement locus video file one to one, thus recording the movement locus of each moving target to be tracked.
In the present embodiment, panoramic picture harvester 10 is arranged on predeterminated position.Panoramic picture harvester 10 preferably employs CMOS (ComplementaryMetalOxideSemiconductor, complementary metal oxide semiconductors (CMOS)) panoramic high-definition 1080P video camera, CMOS panoramic high-definition 1080P video camera with the speed of 25 frames per second by resolution be 1080P (1920 × 1080) video streaming image with the speed of 1000Mbps by network cable real-time Transmission to controller 20.The video flowing resolution adopted in prior art is usually VGA (videographicsarray, Video Graphics Array) (640 × 480), seldom can reach 720P (1280 × 720).Compared with prior art, the present embodiment can get, by CMOS panoramic high-definition 1080P video camera, the panorama ambient image that definition is high, thus improving the accuracy of successive image analyzing and processing, and, the raising of video resolution can make the details that controller 20 processes more, improves the detection success rate to small volume or distant moving target.
Fig. 2 illustrates that the multiple mobile object that first embodiment of the invention provides catches and the another kind of structural representation of the equipment of tracking.In Fig. 2, it is described for a point of mirror camera system 30.As in figure 2 it is shown, in the present embodiment, controller 20 includes GPU21 and the CPU22 being electrically connected to each other；CPU22 electrically connects with panoramic picture harvester 10, for receiving the current frame image of panorama ambient image, current frame image is transferred to GPU21；GPU21, for current frame image is carried out moving object detection, obtains the characteristics of image of each moving target in current frame image and sends to CPU22；Wherein, the characteristics of image of moving target includes the image coordinate of moving target, color characteristic and textural characteristics；CPU22 electrically connects with multiple points of mirror camera systems 30 respectively, for receiving the characteristics of image of each moving target, the characteristics of image of each moving target is merged respectively, and mate with the characteristics of image prestored, according to characteristic matching result, each moving target in current frame image is screened, it is determined that the position of multiple moving targets to be tracked and each moving target to be tracked；Position according to each moving target to be tracked generates motion target tracking instruction one to one；Determine point mirror camera system 30 corresponding respectively with the position of each moving target to be tracked；By each motion target tracking instruction generated corresponding transmission respectively to point mirror camera system 30 determined.Wherein, the position of moving target to be tracked includes image coordinate and the space coordinates of moving target to be tracked.
Current frame image is carried out moving object detection by above-mentioned GPU21, obtains the characteristics of image of each moving target in current frame image and sends to the detailed process of CPU22 and be: current frame image is carried out pretreatment by GPU21, and this pretreatment includes decoding and/or denoising；Adopt mixed Gaussian to carry out background modeling, generate the background model of predetermined number；Pretreated current frame image is carried out moving object detection by the background model according to predetermined number, obtains the characteristics of image of each moving target in current frame image and sends to CPU22.
Specifically, current frame image is decoded by GPU21, or current frame image carries out denoising, or current frame image is decoded and denoising.GPU21 adopts mixed Gaussian to carry out background modeling, generates 5-7 background model.Pretreated current frame image is carried out moving object detection according to 5-7 background model by GPU21, background image is rejected from current frame image, obtain each moving target in current frame image and the characteristics of image of each moving target in current frame image, and the characteristics of image of each moving target in current frame image is sent to CPU22.Compared with prior art, in the present embodiment, GPU21 adopts 5-7 model to set up Steady Background Light, and bigger calculating scale causes the raising of sport foreground and background segment accuracy, and GPU21 parallel computation characteristic ensures completing in real time of a large amount of calculating.
By GPU21, current frame image is carried out moving object detection, can detect from current frame image and obtain multiple moving target, and the multiple respective image coordinate of moving target, color characteristic and textural characteristics, wherein color characteristic can be represented by color histogram, and textural characteristics can be represented by local invariable rotary texture, histogram of gradients.Respective for multiple moving targets image coordinate, color characteristic and textural characteristics are sent to CPU22 by GPU21.
In the present embodiment, GPU21 is assigned to thousands of computing units and processing in thread carries out the processing procedure of high definition digital image is divided into minor matrix block simultaneously, drastically increases processing speed.GPU21 utilizes embedded moving target and feature detection algorithm the most of garbage in image to be filtered out, and retains effective information, and namely the characteristics of image of multiple moving targets in current frame image, is only transferred to CPU22 by effective information only small for data scale.
CPU22 receives the respective image coordinate of each moving target, color characteristic and textural characteristics, for each moving target, its respective image coordinate, color characteristic and textural characteristics are merged, and the feature after merging is mated with the characteristics of image prestored, if the match is successful, then determine that moving target corresponding to the feature after this fusion is moving target to be tracked, and by the characteristic storage after this fusion in internal memory, as the template of next frame matching primitives.CPU22 also determines that the position of moving target to be tracked, such as image coordinate, this image coordinate is converted to space coordinates, space coordinates according to moving target to be tracked generates motion target tracking instruction, and determine corresponding point mirror camera system 30 according to the space coordinates of moving target to be tracked, the motion target tracking instruction of generation is transmitted to point mirror camera system 30 determined.In the present embodiment, CPU22 can filter out multiple moving target to be tracked according to characteristic matching result, and send motion target tracking instruction respectively to multiple points of mirror camera systems 30, thus control multiple points of mirror camera systems 30 to follow the tracks of each moving target to be tracked respectively.
In order to make up the background modeling method inaccuracy to the relatively low situation of target speed, when in the present embodiment, CPU22 carries out moving target screening, the multiple characteristics of image of fusional movement target, including image coordinate, color characteristic and textural characteristics, further increase the accuracy of prospect and background separation, it is achieved the accurate detection to moving target interested in complex environment.In the present embodiment, CPU22 achieves moving target position and the matching algorithm of color, Texture Feature Fusion, it is achieved that the quick and precisely description to target trajectory.
Owing to point mirror camera system needs the space coordinates according to moving target to carry out seizure and the tracking of moving target, therefore the detailed process that CPU22 generates motion target tracking instruction one to one according to the space coordinates of each moving target to be tracked is: the CPU22 position according to each moving target to be tracked, the space coordinates of moving target as to be tracked in each is analyzed, generate the light deflection parameter that each moving target to be tracked is corresponding, focus adjustment parameter and acquisition parameters, according to light deflection parameter, focus adjustment parameter and acquisition parameters generate motion target tracking instruction.Additionally, CPU22 is always according to position set in advance corresponding relation, corresponding relation such as space coordinates and the position of point mirror camera system, determine point mirror camera system corresponding respectively with the space coordinates of each moving target to be tracked, by each motion target tracking instruction generated corresponding transmission respectively to point mirror camera system 30 determined.
Table 2
Table 2 illustrates the corresponding relation signal table between point mirror camera system that motion target tracking instruction that CPU22 generates, the space coordinates of each moving target to be tracked, each moving target to be tracked are corresponding.As shown in Table 2, each moving target to be tracked has the space coordinates of correspondence, motion target tracking instruction and point mirror camera system.In motion target tracking instruction, light deflection parameter includes light deflection angle, and focus adjustment parameter includes focus adjustment distance, and acquisition parameters includes shutter duration, aperture size etc..
In the present embodiment, controller 20 adopts GPU21 and CPU22 heterogeneous parallel processing mode that ambient image is processed, and can be mentioned the work efficiency of GPU21 and CPU22 by this heterogeneous parallel processing mode, thus improving image processing speed.
As in figure 2 it is shown, a point mirror camera system 30 includes acousto-optic modulator 31 and high-definition camera 32；Above-mentioned motion target tracking instruction carries light deflection parameter and acquisition parameters；Acousto-optic modulator 31 electrically connects with controller 20, receives motion target tracking instruction, according to the light deflection parameter that motion target tracking instruction is carried, the light carrying out self-corresponding moving target to be tracked is reflected, is reflexed to by light in high-definition camera 32；High-definition camera 32 electrically connects with controller 20, receives motion target tracking instruction, corresponding moving target to be tracked is shot when camera lens points to constant according to the acquisition parameters that motion target tracking instruction is carried.
Being described for a point of mirror camera system, in Fig. 2, the line segment between acousto-optic modulator 31 and high-definition camera 32 represents paths direction.In the present embodiment, acousto-optic modulator 31 carries out light line reflection according to light deflection parameter, and this light deflection parameter includes the data such as light deflection angle.After being reflected in high-definition camera 32 from the light of moving target to be tracked by acousto-optic modulator 31, high-definition camera 32 shoots according to acquisition parameters, such as parameters such as shutter duration, aperture size, thus reaching to catch and follow the tracks of the effect of moving target to be tracked.
Fig. 3 is the structural representation of point mirror camera system 30 in the present embodiment, as shown in Figure 3, in order to ensure that light is accurately deflected by acousto-optic modulator 31, acousto-optic modulator 31 includes longitudinal acousto-optic modulator 311 and horizontal acousto-optic modulator 312, longitudinal acousto-optic modulator 311, horizontal acousto-optic modulator 312 and high-definition camera 32 are arranged in order, and longitudinal acousto-optic modulator 311 and horizontal acousto-optic modulator 312 electrically connect with controller 20 respectively；Longitudinal acousto-optic modulator 311 receives motion target tracking instruction, and the light of the light deflection parameter self-corresponding moving target to be tracked in the future carried according to motion target tracking instruction longitudinally reflexes to horizontal acousto-optic modulator 312；Horizontal acousto-optic modulator 312 receives motion target tracking instruction, is laterally reflexed in high-definition camera 32 by the light after longitudinally reflection according to the light deflection parameter that motion target tracking instruction is carried.
In the present embodiment, light from moving target to be tracked passes through longitudinal acousto-optic modulator 311, utilize the feature of Donald Bragg diffraction produced by acoustooptical effect, realize the control to the light direction of propagation, undertaken light longitudinally deflecting into the angle specified, through horizontal acousto-optic modulator 312, light is carried out lateral deflection, converge in high-definition camera 32, by the covering power of high-definition camera 32, it is achieved the seizure of moving target to be tracked and tracking.
Consider that high-definition camera 32 also needs to when catching moving target to be tracked carry out focus adjustment, in order to avoid the problem that the mechanical adjustment focus adjustment time is slow, as shown in Figure 3, in the present embodiment, dividing mirror camera system 30 also to include liquid crystal lens 33, longitudinal acousto-optic modulator 311, horizontal acousto-optic modulator 312, liquid crystal lens 33 and high-definition camera 32 are arranged in order；Liquid crystal lens 33 electrically connects with controller 20；Motion target tracking instruction carries focus adjustment parameter；Liquid crystal lens 33 receives motion target tracking instruction, carries out focus adjustment according to the focus adjustment parameter that motion target tracking instruction is carried；Light after above-mentioned laterally reflection enters high-definition camera 32 through liquid crystal lens 33.
In Fig. 3, longitudinal acousto-optic modulator 311, horizontal acousto-optic modulator 312, line segment between liquid crystal lens 33 and high-definition camera 32 represent paths direction.By point mirror camera system 30 in Fig. 3, the light from moving target to be tracked first passes around longitudinal reflection, is then passed through horizontal reflection, eventually passes through liquid crystal lens 33 and enters high-definition camera 32.
In the present embodiment, electrically connected with controller 20 by liquid crystal lens 33 and carry out shooting focal length adjustment, it is possible to reach the fireballing effect of focus adjustment, it is to avoid the problem that in prior art, mechanical adjustment shooting focal length governing speed is slow.
In the present embodiment, acousto-optic modulator 31 can also replace by high-speed vibrating mirror.Point mirror camera system 30 adds liquid crystal lens 33, acousto-optic modulator 31 or high-speed vibrating mirror before high-definition camera 32, acousto-optic modulator 31 or the high-speed vibrating mirror deviation effect to light path can be passed through according to light path principle of reversibility, the ken that light path is passed back is imaged on high-definition camera 32, high speed focusing action through liquid crystal lens 33, it is clear that ground obtains the image of moving target to be tracked clearly.
When a point mirror camera system 30 includes multiple, the image that each point of mirror camera system 30 obtains is returned to controller 20, the response speed of the equipment such as acousto-optic modulator 31 and liquid crystal lens 33 is all up nanosecond order, then divide the speed that mirror camera system 30 obtains image can reach musec order, pass speed image back and also can reach musec order, by controller 20 and coordinating of dividing mirror camera system 30, it is possible to achieve Microsecond grade moving target catches and follows the tracks of.If replacing acousto-optic modulator 31 by high-speed vibrating mirror, response speed reaches millisecond to musec order, then can realize Millisecond moving target and catch and follow the tracks of.
In practical application, the image of the different visual angles of same moving target to be tracked can be caught by multiple points of mirror camera systems 30, thus building the panoramic information of moving target to be tracked, it is possible to obtain moving target to be tracked at large range of concrete moving line.
As shown in Figure 4, point mirror camera system 30 in the present embodiment also has another kind of version, acousto-optic modulator 31 includes horizontal acousto-optic modulator 312 and longitudinal acousto-optic modulator 311, dividing mirror camera system 30 also to include liquid crystal lens 33, horizontal acousto-optic modulator 312, longitudinal acousto-optic modulator 311, liquid crystal lens 33 and high-definition camera 32 are arranged in order；Horizontal acousto-optic modulator 312, longitudinal acousto-optic modulator 311 and liquid crystal lens 33 electrically connect with controller 20 respectively.
In Fig. 4, longitudinal acousto-optic modulator 311, horizontal acousto-optic modulator 312, line segment between liquid crystal lens 33 and high-definition camera 32 represent paths direction.By point mirror camera system 30 in Fig. 4, the light from moving target to be tracked first passes around horizontal reflection, is then passed through longitudinal reflection, eventually passes through liquid crystal lens 33 and enters high-definition camera 32.
For the ease of regulating the position of each device in point mirror camera system 30, a point mirror camera system 30 also includes liquid crystal lens installing rack, high-definition camera installing rack, horizontal acousto-optic modulator installing rack and longitudinal acousto-optic modulator installing rack；Liquid crystal lens installing rack, high-definition camera installing rack, horizontal acousto-optic modulator installing rack and longitudinal acousto-optic modulator installing rack are respectively arranged with regulating element；Installation site by this regulating element adjustable liquid crystal display lens 33 respectively, high-definition camera 32, horizontal acousto-optic modulator 312 and longitudinal acousto-optic modulator 311.
Caught and the equipment of tracking by the multiple mobile object in the present embodiment, it is effectively increased the tracking efficiency of moving target, deformation can be eliminated, block, the rough sledding negative effect to tracking effect such as intersection, thus set up multiple moving target complete, long-time in motion trace data.
Fig. 5 illustrates the schematic flow sheet of the image processing algorithm that the present embodiment provides, and this image processing algorithm is performed jointly by CPU and GPU, as it is shown in figure 5, this image processing algorithm includes:
Step 502, carries out pretreatment to current frame image, and this pretreatment includes decoding and/or denoising；
Step 504, adopts mixed Gaussian to carry out background modeling, generates 5-7 background model；
Step 506, carries out moving object detection according to 5-7 background model to pretreated current frame image, obtains the characteristics of image of each moving target in current frame image；Wherein, the characteristics of image of moving target includes the image coordinate of moving target, color characteristic and textural characteristics；
Step 508, the characteristics of image of each moving target in current frame image is merged respectively, and mate with the characteristics of image prestored, according to characteristic matching result, the multiple moving targets in current frame image are screened, it is determined that the image coordinate of multiple moving targets to be tracked and each moving target to be tracked；
Step S510, this image coordinate is converted to space coordinates, space coordinates according to each moving target to be tracked generates motion target tracking instruction one to one, determine point mirror camera system corresponding respectively with the space coordinates of each moving target to be tracked, by each motion target tracking instruction generated corresponding transmission respectively to point mirror camera system determined；
Step S512, receives the image of the moving target multiple to be tracked that each point of mirror camera system returns, and this image carries out sorting process, generates and each moving target to be tracked movement locus video one to one.
Image processing algorithm shown in Fig. 5 is performed jointly by CPU and GPU, ensure that the accuracy of image procossing, rapidity and robustness, improve image processing speed, it is achieved that the fusion of moving target position, color and textural characteristics, it is achieved that describe the movement locus of moving target fast and accurately.
By the method in Fig. 5, adopt Fuzzy Set Theory fusional movement position and velocity estimation and color, the mating of Gradient Features, increase hunting zone in the historical data, it is effectively improved motion target tracking efficiency, can eliminate deform, block, the rough sledding negative effect to tracking effect such as intersection, set up moving target interior motion trace data complete, long-time.
Embodiment two
Fig. 6 illustrates that the multiple mobile object that second embodiment of the invention provides catches the schematic flow sheet with tracking, and as shown in Figure 6, the method includes:
Step 602, gathers panorama ambient image, determines multiple moving target to be tracked according to panorama ambient image；
Step 604, generates and multiple moving target to be tracked motion target tracking instructions one to one；
Multiple motion target tracking instructions are transmitted separately to point mirror camera system of correspondence by step 606；Wherein, the appointment position in image acquisition region is respectively arranged with multiple points of mirror camera systems；
Step 608, mirror camera system is divided according to motion target tracking instruction, the light carrying out self-corresponding moving target to be tracked to be reflected, utilize light principle of reflection to catch and follow the tracks of the moving target to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target to be tracked；
Step 610, carries out sorting process to the image of each moving target to be tracked, generates and each moving target to be tracked movement locus video one to one.
In the embodiment of the present invention, the light from moving target to be tracked is reflected by a point mirror camera system, utilizes light principle of reflection to catch and follow the tracks of multiple moving target to be tracked when camera lens points to constant.Owing to when catching and follow the tracks of multiple moving target to be tracked, camera lens points to and need not change, therefore the method in the present embodiment can improve the seizure tracking velocity of multiple moving target, it is to avoid the problem that in prior art, mechanical adjustment lens direction governing speed is slow.
In embodiment provided by the present invention, it should be understood that disclosed apparatus and method, it is possible to realize by another way.Device embodiment described above is merely schematic, such as, the division of described unit, it is only a kind of logic function to divide, actual can have other dividing mode when realizing, again such as, multiple unit or assembly can in conjunction with or be desirably integrated into another system, or some features can ignore, or do not perform.Another point, shown or discussed coupling each other or direct-coupling or communication connection can be through INDIRECT COUPLING or the communication connection of some communication interfaces, device or unit, it is possible to be electrical, machinery or other form.
The described unit illustrated as separating component can be or may not be physically separate, and the parts shown as unit can be or may not be physical location, namely may be located at a place, or can also be distributed on multiple NE.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the present embodiment scheme.
It addition, each functional unit in embodiment provided by the invention can be integrated in a processing unit, it is also possible to be that unit is individually physically present, it is also possible to two or more unit are integrated in a unit.
If described function is using the form realization of SFU software functional unit and as independent production marketing or use, it is possible to be stored in a computer read/write memory medium.Based on such understanding, part or the part of this technical scheme that prior art is contributed by technical scheme substantially in other words can embody with the form of software product, this computer software product is stored in a storage medium, including some instructions with so that a computer equipment (can be personal computer, server, or the network equipment etc.) perform all or part of step of method described in each embodiment of the present invention.And aforesaid storage medium includes: USB flash disk, portable hard drive, read only memory (ROM, Read-OnlyMemory), the various media that can store program code such as random access memory (RAM, RandomAccessMemory), magnetic disc or CD.
It should also be noted that similar label and letter below figure represent similar terms, therefore, once a certain Xiang Yi accompanying drawing is defined, then it need not be carried out definition further and explain in accompanying drawing subsequently, in addition, term " first ", " second ", " the 3rd " etc. are only used for distinguishing description, and it is not intended that indicate or hint relative importance.
Last it is noted that embodiment described above, it is only the specific embodiment of the present invention, in order to technical scheme to be described, it is not intended to limit, protection scope of the present invention is not limited thereto, although the present invention being described in detail with reference to previous embodiment, it will be understood by those within the art that: any those familiar with the art is in the technical scope that the invention discloses, technical scheme described in previous embodiment still can be modified by it maybe can readily occur in change, or wherein portion of techniques feature is carried out equivalent replacement；And these amendments, change or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of embodiment of the present invention technical scheme.All should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should described be as the criterion with scope of the claims.
Claims (10)
1. a multiple mobile object catches and the equipment of tracking, it is characterised in that including:
Panoramic picture harvester, the controller electrically connected with described panoramic picture harvester, the multiple points of mirror camera systems being electrically connected with described controller；Multiple described point mirror camera systems are respectively arranged at the appointment position in the pickup area of described panoramic picture harvester；
Described panoramic picture harvester is used for gathering panorama ambient image, by described panorama ambient image transmission to described controller；
Described controller is used for receiving described panorama ambient image, multiple moving target to be tracked is determined according to described panorama ambient image, generate and multiple described moving target to be tracked motion target tracking instructions one to one, multiple described motion target tracking instructions are transmitted separately to point mirror camera system of correspondence；
Described point of mirror camera system is used for receiving described motion target tracking instruction, according to described motion target tracking instruction, the light carrying out self-corresponding described moving target to be tracked is reflected, utilize light principle of reflection to catch and follow the tracks of the moving target described to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target described to be tracked and transmit to described controller；
Described controller is additionally operable to receive the image of the multiple described moving target to be tracked that each described point mirror camera system returns, and described image carries out sorting process, generates and each described moving target to be tracked movement locus video one to one.
2. equipment according to claim 1, it is characterised in that described controller includes GPU and the CPU being electrically connected to each other；
Described CPU electrically connects with described panoramic picture harvester, for receiving the current frame image of described panorama ambient image, described current frame image is transferred to described GPU；
Described GPU, for described current frame image is carried out moving object detection, obtains the characteristics of image of each moving target in described current frame image and sends to described CPU；Wherein, the characteristics of image of described moving target includes the image coordinate of described moving target, color characteristic and textural characteristics；
Described CPU electrically connects with multiple described point mirror camera systems respectively, for receiving the characteristics of image of each described moving target, the characteristics of image of each described moving target is merged respectively, and mate with the characteristics of image prestored, the position of multiple moving target to be tracked and each described moving target to be tracked is determined according to described characteristic matching result；Position according to moving target to be tracked each described generates motion target tracking instruction one to one；Determine point mirror camera system that the position of moving target to be tracked with each described is corresponding respectively；Described in determining, mirror camera system is divided by each the described motion target tracking instruction generated corresponding transmission respectively.
3. equipment according to claim 2, it is characterised in that described GPU for carrying out pretreatment to described current frame image, and described pretreatment includes decoding and/or denoising；Adopt mixed Gaussian to carry out background modeling, generate the background model of predetermined number；Described pretreated current frame image is carried out moving object detection by the background model according to described predetermined number, obtains the characteristics of image of each moving target in described current frame image and sends to described CPU.
4. equipment according to claim 2, it is characterized in that, described CPU is analyzed for the position according to moving target to be tracked each described, generate light deflection parameter corresponding to each described moving target to be tracked, focus adjustment parameter and acquisition parameters, generate motion target tracking instruction according to described light deflection parameter, focus adjustment parameter and acquisition parameters；Point mirror camera system that each described moving target to be tracked is corresponding respectively is determined according to position set in advance corresponding relation；Described in determining, mirror camera system is divided by each the described motion target tracking instruction generated corresponding transmission respectively.
5. equipment according to claim 1, it is characterised in that described motion target tracking instruction carries light deflection parameter and acquisition parameters；Described point of mirror camera system includes acousto-optic modulator and high-definition camera；
Described acousto-optic modulator electrically connects with described controller, receive described motion target tracking instruction, according to the light deflection parameter that described motion target tracking instruction is carried, the light carrying out self-corresponding described moving target to be tracked is reflected, described light is reflexed in described high-definition camera；
Described high-definition camera electrically connects with described controller, receives described motion target tracking instruction, corresponding moving target described to be tracked is shot when camera lens points to constant according to the acquisition parameters that described motion target tracking instruction is carried.
6. equipment according to claim 5, it is characterized in that, described acousto-optic modulator includes longitudinal acousto-optic modulator and horizontal acousto-optic modulator, described longitudinal acousto-optic modulator, described horizontal acousto-optic modulator and described high-definition camera are arranged in order, and described longitudinal acousto-optic modulator and described horizontal acousto-optic modulator electrically connect with described controller respectively；
Described longitudinal acousto-optic modulator receives described motion target tracking instruction, and the light of the light deflection parameter carried according to described motion target tracking instruction self-corresponding described moving target to be tracked in the future longitudinally reflexes to described horizontal acousto-optic modulator；
Described horizontal acousto-optic modulator receives described motion target tracking instruction, is laterally reflexed in described high-definition camera by the light after described longitudinally reflection according to the light deflection parameter that described motion target tracking instruction is carried.
7. equipment according to claim 6, it is characterised in that described point of mirror camera system also includes liquid crystal lens, and described longitudinal acousto-optic modulator, described horizontal acousto-optic modulator, described liquid crystal lens and described high-definition camera are arranged in order；Described liquid crystal lens electrically connects with described controller；Described motion target tracking instruction carries focus adjustment parameter；
Described liquid crystal lens receives described motion target tracking instruction, carries out focus adjustment according to the focus adjustment parameter that described motion target tracking instruction is carried；Light after described laterally reflection enters described high-definition camera through described liquid crystal lens.
8. equipment according to claim 5, it is characterized in that, described acousto-optic modulator includes horizontal acousto-optic modulator and longitudinal acousto-optic modulator, described point of mirror camera system also includes liquid crystal lens, and described horizontal acousto-optic modulator, described longitudinal acousto-optic modulator, described liquid crystal lens and described high-definition camera are arranged in order；Described horizontal acousto-optic modulator, described longitudinal acousto-optic modulator and described liquid crystal lens electrically connect with described controller respectively.
9. equipment according to claim 7, it is characterised in that described point of mirror camera system also includes liquid crystal lens installing rack, high-definition camera installing rack, horizontal acousto-optic modulator installing rack and longitudinal acousto-optic modulator installing rack；
Described liquid crystal lens installing rack, described high-definition camera installing rack, described horizontal acousto-optic modulator installing rack and described longitudinal acousto-optic modulator installing rack are respectively arranged with regulating element；
The installation site of described liquid crystal lens, described high-definition camera, described horizontal acousto-optic modulator and described longitudinal acousto-optic modulator is regulated respectively by described regulating element.
10. a multiple mobile object catches and tracking, it is characterised in that including:
Gather panorama ambient image, determine multiple moving target to be tracked according to described panorama ambient image；
Generate and multiple described moving target to be tracked motion target tracking instructions one to one；
Multiple described motion target tracking instructions are transmitted separately to point mirror camera system of correspondence；Wherein, the appointment position in image acquisition region is respectively arranged with multiple described point mirror camera system；
The light carrying out self-corresponding described moving target to be tracked is reflected by described point of mirror camera system according to described motion target tracking instruction, utilize light principle of reflection to catch and follow the tracks of the moving target described to be tracked of correspondence when camera lens points to constant, obtain the image of corresponding moving target described to be tracked；
The image of moving target to be tracked each described is carried out sorting process, generates and each described moving target to be tracked movement locus video one to one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121742.9A CN105791686B (en) | 2016-03-03 | 2016-03-03 | Multiple mobile object captures and tracking equipment and multiple mobile object capture and tracking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121742.9A CN105791686B (en) | 2016-03-03 | 2016-03-03 | Multiple mobile object captures and tracking equipment and multiple mobile object capture and tracking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105791686A true CN105791686A (en) | 2016-07-20 |
| CN105791686B CN105791686B (en) | 2019-03-15 |
Family
ID=56387902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610121742.9A Active CN105791686B (en) | 2016-03-03 | 2016-03-03 | Multiple mobile object captures and tracking equipment and multiple mobile object capture and tracking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105791686B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106791294A (en) * | 2016-11-25 | 2017-05-31 | 益海芯电子技术江苏有限公司 | Motion target tracking method |
| CN109040664A (en) * | 2018-06-01 | 2018-12-18 | 深圳市商汤科技有限公司 | video stream processing method and device, electronic equipment and storage medium |
| TWI690207B (en) * | 2018-05-29 | 2020-04-01 | 圓展科技股份有限公司 | Method for object tracking |
| CN111199203A (en) * | 2019-12-30 | 2020-05-26 | 广州幻境科技有限公司 | Motion capture method and system based on handheld device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1658670A (en) * | 2004-02-20 | 2005-08-24 | 上海银晨智能识别科技有限公司 | Intelligent tracking monitoring system with multi-camera |
| CN1976445A (en) * | 2006-12-15 | 2007-06-06 | 上海鼎业电子科技有限公司 | Multi-camera monitoring system based on three-dimensional video frequency dynamic tracking and tracking method thereof |
| CN103020990A (en) * | 2012-12-06 | 2013-04-03 | 华中科技大学 | Moving object detecting method based on graphics processing unit (GPU) |
| US20130088567A1 (en) * | 2011-10-07 | 2013-04-11 | L-3 Communications Cincinnati Electronics Corporation | Panoramic Imaging Systems Comprising Rotatable Mirrors for Image Stabilization |
| US20150085096A1 (en) * | 2013-09-23 | 2015-03-26 | Samsung Electronics Co., Ltd. | Tracking optics for a mobile device |
-
2016
- 2016-03-03 CN CN201610121742.9A patent/CN105791686B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1658670A (en) * | 2004-02-20 | 2005-08-24 | 上海银晨智能识别科技有限公司 | Intelligent tracking monitoring system with multi-camera |
| CN1976445A (en) * | 2006-12-15 | 2007-06-06 | 上海鼎业电子科技有限公司 | Multi-camera monitoring system based on three-dimensional video frequency dynamic tracking and tracking method thereof |
| US20130088567A1 (en) * | 2011-10-07 | 2013-04-11 | L-3 Communications Cincinnati Electronics Corporation | Panoramic Imaging Systems Comprising Rotatable Mirrors for Image Stabilization |
| CN103020990A (en) * | 2012-12-06 | 2013-04-03 | 华中科技大学 | Moving object detecting method based on graphics processing unit (GPU) |
| US20150085096A1 (en) * | 2013-09-23 | 2015-03-26 | Samsung Electronics Co., Ltd. | Tracking optics for a mobile device |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106791294A (en) * | 2016-11-25 | 2017-05-31 | 益海芯电子技术江苏有限公司 | Motion target tracking method |
| TWI690207B (en) * | 2018-05-29 | 2020-04-01 | 圓展科技股份有限公司 | Method for object tracking |
| CN109040664A (en) * | 2018-06-01 | 2018-12-18 | 深圳市商汤科技有限公司 | video stream processing method and device, electronic equipment and storage medium |
| CN111199203A (en) * | 2019-12-30 | 2020-05-26 | 广州幻境科技有限公司 | Motion capture method and system based on handheld device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105791686B (en) | 2019-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wu et al. | Squeezesegv2: Improved model structure and unsupervised domain adaptation for road-object segmentation from a lidar point cloud | |
| EP3709216B1 (en) | Methods and apparatuses for object detection in a scene represented by depth data of a range detection sensor and image data of a camera | |
| US11733386B2 (en) | Methods and apparatuses for object detection in a scene based on lidar data and radar data of the scene | |
| CN105592269B (en) | Moving target captures and tracking equipment and moving target capture and tracking | |
| CN109800689B (en) | Target tracking method based on space-time feature fusion learning | |
| CN105513087A (en) | Laser aiming and tracking equipment and method for controlling same | |
| US20130215228A1 (en) | Method and apparatus for robustly collecting facial, ocular, and iris images using a single sensor | |
| CN105744163A (en) | Video camera and video recording method performing tracking focusing based on depth information | |
| CN112785628B (en) | Track prediction method and system based on panoramic view angle detection tracking | |
| CN105791686A (en) | Multi-moving object capturing and tracking device and multi-moving object capturing and tracking method | |
| WO2020207172A1 (en) | Method and system for optical monitoring of unmanned aerial vehicles based on three-dimensional light field technology | |
| Bose et al. | Visual odometry for pixel processor arrays | |
| Xu et al. | Dynamic obstacle detection based on panoramic vision in the moving state of agricultural machineries | |
| CN103617631B (en) | A kind of tracking based on Spot detection | |
| Huang et al. | Event-based simultaneous localization and mapping: A comprehensive survey | |
| Baisware et al. | Review on recent advances in human action recognition in video data | |
| CN205490869U (en) | Motion target is caught and tracking device | |
| CN108184062B (en) | High-speed tracking system and method based on multi-level heterogeneous parallel processing | |
| CN205490868U (en) | Target of doing more physical exercises is caught and tracking device | |
| Faruqui et al. | Trackez: an iot-based 3d-object tracking from 2d pixel matrix using mez and fsl algorithm | |
| CN205486308U (en) | Laser aims and tracking device | |
| Song et al. | A weld feature points detection method based on improved YOLO for welding robots in strong noise environment | |
| CN116828305A (en) | Method for automatically tracking target object by holder based on YOLOv5 algorithm | |
| WO2023105800A1 (en) | Object detection device, object detection method, and object detection system | |
| Ni | Application of motion tracking technology in movies, television production and photography using big data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100088 1, 111, 4 Building 44, North Third Ring Road, Haidian District, Beijing. Applicant after: Sima great (Beijing) Intelligent Systems Ltd Address before: 100080 Beijing, Haidian District, Shanghai 26 Road, 10 floor, 1019 room. Applicant before: Beijing Zhiguang Boyuan Technology Co., Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |