CN108156419A - More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms - Google Patents
More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms Download PDFInfo
- Publication number
- CN108156419A CN108156419A CN201711402320.XA CN201711402320A CN108156419A CN 108156419 A CN108156419 A CN 108156419A CN 201711402320 A CN201711402320 A CN 201711402320A CN 108156419 A CN108156419 A CN 108156419A
- Authority
- CN
- China
- Prior art keywords
- video
- picture
- camera
- linkage
- focal length
- 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.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/90—Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/14—Picture signal circuitry for video frequency region
- H04N5/144—Movement detection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/265—Mixing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/222—Studio circuitry; Studio devices; Studio equipment
- H04N5/262—Studio circuits, e.g. for mixing, switching-over, change of character of image, other special effects ; Cameras specially adapted for the electronic generation of special effects
- H04N5/268—Signal distribution or switching
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N5/00—Details of television systems
- H04N5/76—Television signal recording
- H04N5/91—Television signal processing therefor
- H04N5/93—Regeneration of the television signal or of selected parts thereof
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Studio Devices (AREA)
Abstract
The present invention discloses a kind of more focal length lens ultrahigh resolutions linkage imaging camera machine system of multiple target tracking combined based on multiple features combining with Camshift algorithms, the system is mainly by video acquisition device, Video Composition handles output system, video management software sharing and two-path video output interface, video acquisition device includes one for acquiring the general picture collecting unit of the scene general picture of low resolution, ten details collecting units for acquisition detail section ultrahigh resolution picture jointly, the video of acquisition flows through Video Composition processing output to video management software, it is used to implement scene general picture picture and the collaboration linkage of ultrahigh resolution picture.The invention has the advantages that:The ultrahigh resolution splicing picture obtained using the scene general picture of single-lens acquisition with multiple and different focal length lens is linked, and realizes and scene is monitored without dead angle all standing, effectively improve monitoring efficiency.
Description
Technical field
The invention belongs to electronic information subject fields, are related to Image Information Processing technology, more particularly to a kind of based on mostly special
More focal length lens ultrahigh resolutions of multiple target tracking linkage imaging camera machine system that sign joint is combined with Camshift algorithms.
Background technology
At present, it is supervised in the video of the extensive areas such as government square, large parking lot, airport Waiting Lounge, station hall
In prosecutor case, traditional scheme is mainly coordinated using more gunlocks and clipping the ball.But the gunlock of telephoto lens is installed, field range has
Limit, installs the gunlocks of short-focus lens, the lack of resolution of collected distant object, and clipping the ball range of observation is limited, it is impossible to same
When take into account multiple scenes, can there are blind areas during use;And for a user, the picture of multiple cameras is monitored and
Playback, it is very inconvenient, monitoring efficiency is reduced, increases monitoring difficulty.
With the development of Video Supervision Technique, high-resolution sensitive chip technology is more and more ripe, and market has 12,000,000
The CMOS of pixel, but separate unit video camera is still difficult the application demand for meeting extensive area.
Invention content
The technical problem to be solved by the present invention is to:In order to overcome existing for above-mentioned traditional extensive area video surveillance program
Deficiency proposes a kind of more focal length lens super-resolutions of the multiple target tracking combined based on multiple features combining with Camshift algorithms
Rate linkage imaging camera machine system, by using the scene general picture of single-lens unit acquisition and the lens unit of multiple and different focal lengths
The ultrahigh resolution splicing picture linkage of acquisition, realizes and scene is monitored without dead angle all standing.
The technical scheme is that:A kind of more focal length lens ultrahigh resolution linkage imaging camera machine systems are provided, it should
More focal length lens ultrahigh resolution linkage imaging camera machine systems mainly include a video based on Multiobjective Intelligent tracking and adopt
Acquisition means, Video Composition processing output system, video management software and two-path video output interface;
The video acquisition device includes a general picture collecting unit, for acquiring the scene general picture of low resolution;Ten
A details collecting unit, for acquiring detail section ultrahigh resolution picture jointly;And two blocks of auxiliary functional circuit plates, one piece
It converts for providing supply voltage and provides power supply into front-end camera and interchanger, while provide gigabit networking switch function
And front-end camera is connected, another piece then carries out the Multiobjective Intelligent tracking combined based on multiple features combining with Camshift algorithms
System writing and inputting, and the Multiobjective Intelligent of front-end camera is controlled to track;
Described Video Composition processing output system, including one the multi-channel video synthesis system based on FPGA, for connecing
The video flowing for entering, head end video harvester being forwarded to obtain, while multi-path video stream is converted into HD video storage all the way simultaneously
Output;
The video management software is used to implement scene general picture picture and the collaboration linkage preview of ultrahigh resolution picture,
Playback and scaling;
The two-path video output interface, video is general picture video all the way, is all the way 4K high definition details videos.
The general picture collecting unit uses 1080P web cameras, and coding mode uses h264 standards;Video camera front end
Short-focus lens is connected, transverse field angle is more than 90 degree;Ten details collecting units are web camera, are used
12000000 pixel CMOS sensor, video acquisition resolution ratio are 4000 × 3000, and coding mode uses h264 standards.
Ten details collecting units are web camera, wherein 3 video camera front end connection mid-focal length camera lenses,
Separate unit video camera transverse field angle is more than 30 degree, and 3 video cameras are placed in approximately the same plane according to left, center, right, collectively form transverse direction
90 degree of field angle, the field range of each two video camera have certain overlapping, are placed in intermediate video camera, visual field direction with it is general
Looks collecting unit is consistent;Other 7 video camera front ends connect long focal length lens, and separate unit video camera transverse field angle is more than 13 degree, 7
Platform video camera is divided to two rows to be placed on two parallel planes, and the visual field direction of video camera is in approximately the same plane, according to different angle
It puts, collectively forms laterally 90 degree of field angle, the field range of each two video camera has certain overlapping;Place mid-focal length camera shooting
The plane of machine and the plane for placing long-focus video camera, shape is at a certain angle in the longitudinal direction, makes middle focus video camera and long-focus
Video camera has certain overlapping region in longitudinal field range.
The video acquisition device further includes housing and stent, and the housing is used to protect whole device, all camera shootings
Machine and auxiliary functional circuit plate are all installed inside housings;The stent is mounted on housing bottom, passes through fixed screw, Ke Yijin
Row regulation of longitudinal angle is adjusted.
The Video Composition processing output system includes two independent gigabit network interfaces, and a multichannel based on FPGA regards
Frequency acquisition display system, for the video flowing for accessing, forwarding head end video harvester to obtain, one is converted to by multi-path video stream
Road HD video is stored and is exported;Head end video harvester is accessed by front end network interface, multi-path video stream access is obtained and is based on
The multi-channel video synthesis system of FPGA will be converted to HD video all the way and store, simultaneously after multi-path video stream merging treatment
The video of general picture all the way of general picture collecting unit is stored, finally according to rear end video management software requirement to two-path video into
Row output.
The video management software, be used to implement scene general picture picture and ultrahigh resolution picture collaboration linkage preview and
It plays back and video is zoomed in and out, is called;The video management software major function includes Internet video storage device pipe
Manage the spatial registration function of function, general picture collecting unit and details collecting unit, equipment collaboration linkage preview function, equipment collaboration
Link playback function and log management function, subscriber management function and system configuration features.
The Internet video storage device management function, user can pass through onvif protocol search and access Internet video
Storage device;Can also Configuration network video storaging equipment network address, storage parameter, system configuration;
The general picture collecting unit and the spatial registration function of details collecting unit, by scene general picture picture and carefully
Section shows that picture samples respectively, and spatial registration information is calculated;Spatial registration information is used for scene general picture unit and details list
The matching linkage of member;
The equipment collaboration linkage preview function, user need the details of preview in scene general picture picture and selecting, lead to
After crossing the calculating of spatial registration information, in ultrahigh resolution picture display window, linkage is carried out to the details of user's selection
Dosing thorough browse;
Described equipment collaboration linkage playback function, user can be and right by onvif protocol search and inquiry video file
The video file inquired is played back;In replayed section, user is in the scene checked of scene general picture picture and selecting needs
Details after being calculated by spatial registration information, in ultrahigh resolution picture display window, carries out the details of user's selection
Linkage matching amplification playback.
The log management function can retrieve video management software local operation and alarm log;
The subscriber management function can add and delete user account, and user right is configured;
The system configuration features can carry out language, NTP service functions, cpu performance configuration.
The general picture collecting unit and the details collecting unit realize spatial registration by following steps:
(1) single-lens unit acquisition picture coordinate system is established, using spherical model, for mapping real space coordinate to sense
The correspondence of the plane coordinates of optical device;
(2) video pictures of scene general picture picture and details collecting unit are sampled respectively, the mapping for calculating sampled point is closed
System;Computational methods are, the pixel coordinate of 6 points is up-sampled, and be converted to space coordinate in scene general picture picture, obtain 6 to
Measure v1, v2 ... v6;The pixel coordinate of 6 points is up-sampled, and be converted to space coordinate in the video pictures of details collecting unit, obtained
To v1 ', v2 ' ... v6 ', using following computational methods, spin matrix MatH2D is obtained by this 6 groups of spatial coordinates calculations:
MatD=[v1, v2, v3, v4, v5, v6],
MatH=[v1', v2', v3', v4', v5', v6'],
MatH2D=MatD × PINV (MatH), wherein PINV (Matrix) are the generalized inverse of matrix Matrix;
(3) when matching linkage preview or playback, its space is calculated by the coordinate (x, y) on scene general picture picture first and is sat
V is marked, and space coordinate v ' of this under ultrahigh resolution picture coordinate system is obtained by v × MatH2D=v ', is finally counted by v '
Calculate the pixel coordinate (x ', y ') of single U10X.
Multiple target tracking algorithm realization in the Multiobjective Intelligent tracking system mainly includes the following steps that:
(1) multiple mobile object is gone out based on Background difference and frame difference method fusion detection;Multiple mobile object detection is main
It is divided into three steps:First step initial background model;Second step utilizes frame difference method context update, then carries out
Binaryzation;Then image is carried out binaryzation by third step using Background difference background difference;
(2) image of binaryzation is subjected to further denoising and Segmentation of Multi-target;
(3) moving foreground object tracking is carried out using improved Camshift algorithms;
Wherein, step (1), step (2) are based on necessary to institute's video image to be handled, are to carry out multiple target tracking
Preceding pretreatment.
The present invention is had the advantages that using above-mentioned technical proposal:Use single-lens unit to the system innovation
The ultrahigh resolution splicing picture linkage that the scene general picture of acquisition is obtained with the lens unit of multiple and different focal lengths, it is not only real simultaneously
Now the picture of multiple cameras is monitored and played back, also realizes and scene is monitored, and without dead angle all standing away from device
In the range of 130m, the picture element density for acquiring image is more than 125 pixels/rice, improves the efficiency and precision of monitoring.
Description of the drawings
Fig. 1 is more focal length lens ultrahigh resolution linkage imaging system connection diagrams;
Fig. 2 is fixed position schematic diagram in video acquisition unit housing;
Fig. 3 is Multiobjective Intelligent track algorithm overall flow figure described in embodiment;
Fig. 4 is the whole hardware chart of Video Composition processing system;
Fig. 5 is rear end video management software interface;
Fig. 6 is the picture coordinate system model of single-lens unit acquisition.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, with reference to diagram to embodiment of the present invention
It is described in further detail.
Fig. 1 show a kind of more focal length lens of the multiple target tracking combined based on multiple features combining with Camshift algorithms
Ultrahigh resolution linkage imaging camera machine system, including a set of head end video harvester, a set of back-end network video storage dress
It puts, a set of Internet video Video Composition processing output system and a set of rear end video management software, specific embodiment are as follows:
1st, head end video harvester, including a general picture collecting unit, for acquiring the scene general picture of low resolution;Ten
A details collecting unit, for acquiring ultrahigh resolution picture jointly;Two blocks of auxiliary functional circuit plates, one piece is used to provide power supply
Voltage is converted and gigabit networking switch function;Another piece is used to input Multiobjective Intelligent tracking module, housing and stent.
Its middle casing is made of heat-conducting metal, and inner horizontal places three pieces of metal plates, for fixing video camera.Wherein put
It is placed among housing and the tablet of bottom is parallel with Housing Base, be positioned over the tablet of upper part of the housing and the upper shape in Housing Base longitudinal direction
At a certain angle, corner dimension is determined by longitudinal field range of mid-focal length video camera and long-focus video camera.Stent is installed
In housing bottom, by fixed screw, regulation of longitudinal angle adjusting can be carried out.
General picture collecting unit uses web camera of the resolution ratio for 1080P.Video camera front end connects short-focus lens, and
It is fixed on pedestal.Pedestal is fixed on the center of intermetallic metal tablet in housing.
Ten details collecting units are web camera, using 12,000,000 pixel CMOS sensor, video acquisition point
Resolution is 4000 × 3000.Wherein 3 video camera front end connection mid-focal length camera lenses, put according to left, center, right, are fixed in housing
On the metal plate on top, 3 video cameras collectively form laterally 90 degree of field angle, and the field range of each two video camera has one
Fixed overlapping;Other 7 video camera front ends connect long focal length lens, wherein 3 are put according to left, center, right, are fixed on bottom in housing
Metal plate on, in addition 4 are distributed in general picture collecting unit both sides, are fixed in housing on intermediate tablet, 7 video cameras
Laterally 90 degree of field angle is collectively formed, the field range of each two video camera has certain overlapping.Head end video harvester packet
The fixed position of all lens units contained is as shown in Figure 2.
Two blocks of auxiliary functional circuit plates of housing back fixation.One piece of major function is converted including supply voltage, will be made
12VDC is converted to 24VAC supply voltages, for all collecting units and power supply for exchange;There is provided gigabit networking interchanger simultaneously
Function, for connecting the network interface of all collecting units;Auxiliary functional circuit plate externally provides power supply interface and gigabit networking
Interface.Another piece of input for being mainly used for carrying out Multiobjective Intelligent tracking module, multiple target tracking modular algorithm realize main packet
It includes:
(1) multiple mobile object is gone out based on Background difference and frame difference method fusion detection;Multiple mobile object detection is main
It is divided into three steps:First step initial background model;Second step utilizes frame difference method context update, then carries out
Binaryzation;Then image is carried out binaryzation by third step using Background difference background difference;
(2) image of binaryzation is subjected to further denoising and Segmentation of Multi-target;
(3) moving foreground object tracking is carried out using improved Camshift algorithms;
Wherein, step (1), step (2) are based on necessary to institute's video image to be handled, are to carry out multiple target tracking
Preceding pretreatment.Based on the multiple target tracking algorithm that above step carries out, computational efficiency is higher, high treating effect.Multiple target intelligence
Energy tracking module algorithm overall flow figure is as shown in Figure 3.
2nd, Internet video synthesis processing output system, for accessing, forwarding the video flowing of head end video harvester acquisition,
Multichannel details video flowing is converted to HD video all the way to store and export, while another way general picture video is stored and logical simultaneously
Cross interface output.It is characterized in that comprising two independent gigabit network interfaces, head end video harvester is accessed by front end network interface,
Multi-channel video synthesis system of the multi-path video stream access based on FPGA is obtained, system will be converted to one after multi-channel video merging treatment
Road HD video is stored and is exported according to rear end video management software requirement.It can simultaneously be realized to multichannel by the system
The acquisition and display of vision signal, can also be to certain full frame amplification display of signal progress all the way.Specifically design includes two to the system
A major part:Hardware platform is built, the denoising algorithm after Video Composition.
1) hardware platform is built;
For inputting multi-path high-definition video, the hardware configuration of system is as shown in Figure 4.Video decoding chip mainly is used,
The devices such as video coding chip, FPGA and DDR2SRAM.Video decoding chip selects SAA7113H as video A/D converter
Part;Video coding chip selects SAA7121 as digital-to-analogue conversion device;Chip external memory SRAM using IS61LV51216 and
FPGA uses the DM6467 chips of Texas Instruments.Design inside wherein FPGA mainly includes key control unit, data
Acquisition and extraction unit, frame synthesis processing unit and frame deposit control unit.System described in the present embodiment receives 16 routing high definitions
The analog video signal of CMOS camera acquisition, the digital video signal of pal mode is converted to by video decoding chip, is passed through
The Data Format Transforms such as FPGA extracts it and frame synthesizes, then shown by the digital-to-analogue conversion feeding VGA of video coding chip
Show device, can show 1,4,9,16 road divided frames, and pass through button Switch Video picture in turn.
2) synthetic video denoising algorithm;
Common Denoising Algorithm has mean filter and medium filtering, and the modified hydrothermal process on the basis of them, intermediate value filter
Wave filters out noise while the original feature of protection signal as much as possible, thus is widely used.The original substantially of medium filtering
Reason is the value of pixel each in digital picture to be set as the intermediate value of all pixels point value in the point neighborhood window, because making an uproar
The characteristics of sound, the intermediate value in window is generally not noise spot, so as to achieve the purpose that inhibit noise.
The present invention is using a kind of improved " Max/Median medium filterings " algorithm:
If am,nThe pixel value for being image at (m, n), Z is the filter window of one (2N+1) × (2N+1):
It is obtained by above formula (1) with am,nCentered on 4 child windows intermediate value, then again by the following formula in 4
Value is maximized, the filter value as window:
Y (m, n)=max [Z1(m,n),Z2(m,n),Z3(m,n),Z4(m,n)]
(2) it based on above-mentioned formula (2), can be improved:
Z is first obtained by above-mentioned improvement formula (3)1, Z2, am,nAnd Z3,Z4,am,nIntermediate value, then obtained again by following formula (4)
The value of filtering:
Y (m, n)=median [Y1(m,n),Y2(m,n),am,n] (4)
Algorithm as shown in formula (2) remains the geometrical property of signal, and two dimension is done with reference to all one-dimensional characteristics of image
Estimation, retains the one-dimensional of two dimensional image, but it be not substantially it is symmetrical, to noise than comparatively dense region performance not
It is fine.Algorithm as shown in formula (4) simultaneously, solves its shortcoming keep former algorithm the advantages of.The present invention
Innovatory algorithm shown in formula (4) is realized on FPGA, the logical construction that algorithm for design is realized is completed to after synthesis
Video all the way denoising.
3rd, rear end video management software is mainly used for realizing that scene general picture picture and the collaboration linkage of ultrahigh resolution picture are pre-
It lookes at and playback function and the spatial registration of general picture collecting unit and details collecting unit.As shown in Figure 5.Other functions are main
Include Internet video storage device management, log management, user management and system configuration.
Internet video storage device management function refers to that user passes through onvif agreements, search and access Internet video storage
Equipment and the network address of Configuration network video storaging equipment, storage parameter, system configuration.
After software access Internet video storage device, by spatial registration function, scene general picture picture and details are shown
Picture samples respectively, and spatial registration information is calculated.Spatial registration information is used for of scene general picture unit and detailed elements
With linkage.
Equipment collaboration linkage preview function refers to that user needs the details of preview in scene general picture picture and selecting, passes through
Spatial registration information calculates, and after obtaining the serial number of details collecting unit and the corresponding coordinate of picture, is shown in ultrahigh resolution picture
Show in window, linkage matching amplification preview is carried out to the details of user's selection.
Equipment collaboration linkage playback function refers to that user can be and right by onvif protocol search and inquiry video file
The video file inquired is played back.In replayed section, user is in the scene checked of scene general picture picture and selecting needs
Details, after being calculated by spatial registration information, after obtaining the serial number of details collecting unit and the corresponding coordinate of picture, in superelevation point
In resolution picture display window, linkage matching amplification playback is carried out to the details of user's selection.
Log management function is used to retrieve software local operation and alarm log;Subscriber management function is used to add and delete
User account and configuration user right;System configuration features are for carrying out language configuration, NTP service functions are configured, cpu performance
Configuration.
Rear end video management software is to realize general picture collecting unit and details collecting unit by spatial registration method
Spatial registration solves the problems, such as the matching linkage between lens unit.The step of spatial registration method, mainly includes:
(1) single-lens unit acquisition picture coordinate system is established, as shown in fig. 6, using spherical model, for mapping practical sky
Between coordinate to the plane coordinates of sensor devices correspondence.
(2) video pictures of scene general picture picture and details collecting unit are sampled respectively, the mapping for calculating sampled point is closed
System.Computational methods are, the pixel coordinate of 6 points is up-sampled, and be converted to space coordinate in scene general picture picture, obtain 6 to
Measure v1, v2 ... v6;The pixel coordinate of 6 points is up-sampled, and be converted to space coordinate in the video pictures of details collecting unit, obtained
To v1 ', v2 ' ... v6 ', using following computational methods, spin matrix MatH2D is obtained by this 6 groups of spatial coordinates calculations:
MatD=[v1, v2, v3, v4, v5, v6],
MatH=[v1', v2', v3', v4', v5', v6'],
MatH2D=MatD × PINV (MatH), wherein PINV (Matrix) are the generalized inverse of matrix Matrix.
(3) when matching linkage preview or playback, its space is calculated by the coordinate (x, y) on scene general picture picture first and is sat
V is marked, and space coordinate v ' of this under ultrahigh resolution picture coordinate system is obtained by v × MatH2D=v ', is finally counted by v '
Calculate the pixel coordinate (x ', y ') of single U10X.
4th, with more focal length lens ultrahigh resolutions linkage imaging system of Multiobjective Intelligent tracking, it can realize that scene is general
The linkage surveillance of looks and ultrahigh resolution picture, and equipment transverse field angle is more than 90 degree, so can be in a wide range of scene
It is realized in monitoring without dead angle all standing, and in the range of away from device 130m, the picture element density for acquiring image is more than 125 pixels/rice.
Above to the more of the multiple target tracking provided by the present invention combined based on multiple features combining with Camshift algorithms
Focal length lens ultrahigh resolution linkage imaging camera machine system is described in detail, and is applied in present embodiment specific
A example is expounded the principle of the present invention and embodiment, and the explanation of above example is only intended to help to understand the present invention
Method and its core concept;Meanwhile for those of ordinary skill in the art, thought according to the present invention is being embodied
There will be changes in mode and application range, in conclusion the content of the present specification should not be construed as limiting the invention.
Claims (10)
1. a kind of more focal length lens ultrahigh resolutions linkage imaging camera machine system, more focal length lens ultrahigh resolutions linkages into
As camera chain mainly includes a video acquisition device tracked based on Multiobjective Intelligent, Video Composition processing output system
System, video management software and two-path video output interface, it is characterised in that:
The video acquisition device includes a general picture collecting unit, for acquiring the scene general picture of low resolution;It is ten thin
Collecting unit is saved, for acquiring detail section ultrahigh resolution picture jointly;And two blocks of auxiliary functional circuit plates, one piece is used for
Supply voltage is provided to convert and provide power supply into front-end camera and interchanger, while gigabit networking switch function and company are provided
Front-end camera is connect, another piece then carries out the Multiobjective Intelligent tracking system combined based on multiple features combining with Camshift algorithms
Write and input, the Multiobjective Intelligent of front-end camera is controlled to track;
Described Video Composition processing output system including one the multi-channel video synthesis system based on FPGA, for accessing, turns
It sends out the video flowing that head end video harvester obtains, while multi-path video stream is converted into HD video all the way and stores and exports;
The video management software is used to implement scene general picture picture and the collaboration linkage preview of ultrahigh resolution picture, playback
And scaling;
The two-path video output interface, video is general picture video all the way, is all the way 4K high definition details videos.
2. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 1, it is characterised in that:Institute
The general picture collecting unit stated uses 1080P web cameras, and coding mode uses h264 standards;Video camera front end connects short focus
Camera lens, transverse field angle are more than 90 degree;Ten details collecting units are web camera, using 12,000,000 pixels
CMOS sensor, video acquisition resolution ratio are 4000 × 3000, and coding mode uses h264 standards.
3. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 2, it is characterised in that:
Ten details collecting units are web camera, wherein 3 video camera front end connection mid-focal length camera lenses, separate unit
Video camera transverse field angle is more than 30 degree, and 3 video cameras are placed in approximately the same plane according to left, center, right, collectively forms laterally 90 degree
Field angle, the field range of each two video camera has certain overlapping, is placed in intermediate video camera, visual field direction is adopted with general picture
It is consistent to collect unit;Other 7 video camera front ends connect long focal length lens, and separate unit video camera transverse field angle is more than 13 degree, and 7 are taken the photograph
Camera is divided to two rows to be placed on two parallel planes, and the visual field direction of video camera is put according to different angle in approximately the same plane,
Laterally 90 degree of field angle is collectively formed, the field range of each two video camera has certain overlapping;Place mid-focal length video camera
Plane and the plane for placing long-focus video camera, shape is at a certain angle in the longitudinal direction, images middle focus video camera and long-focus
Machine has certain overlapping region in longitudinal field range.
4. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 1, it is characterised in that:Institute
The video acquisition device stated further includes housing and stent, and the housing is used to protect whole device, all video cameras and auxiliary work(
Energy circuit board is all installed inside housings;The stent is mounted on housing bottom, by fixed screw, can carry out regulation of longitudinal angle
It adjusts.
5. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 1, it is characterised in that:Institute
The Video Composition processing output system stated includes two independent gigabit network interfaces, and a multi-channel video capturing based on FPGA is shown
Multi-path video stream for the video flowing for accessing, forwarding head end video harvester to obtain, is converted to HD video all the way by system
It stores and exports;Head end video harvester is accessed by front end network interface, obtains multichannel of the multi-path video stream access based on FPGA
Image synthesizing system will be converted to HD video all the way and store, while general picture is acquired after multi-path video stream merging treatment
The video of general picture all the way of unit is stored, and finally two-path video is exported according to rear end video management software requirement.
6. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 1, it is characterised in that:Institute
The video management software stated is used to implement scene general picture picture and ultrahigh resolution picture collaboration linkage preview and playback and right
Video is zoomed in and out, is called;The video management software major function includes Internet video storage device management function, general picture
The spatial registration function of collecting unit and details collecting unit, equipment collaboration linkage preview function, equipment collaboration linkage playback work(
Energy and log management function, subscriber management function and system configuration features.
7. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 6, it is characterised in that:
The Internet video storage device management function, user can be stored by onvif protocol search and access Internet video
Equipment;Can also Configuration network video storaging equipment network address, storage parameter, system configuration;
The general picture collecting unit and the spatial registration function of details collecting unit, by being shown to scene general picture picture and details
Show that picture samples respectively, spatial registration information is calculated;Spatial registration information is for scene general picture unit and detailed elements
Matching linkage;
The equipment collaboration linkage preview function, user need the details of preview in scene general picture picture and selecting, pass through sky
Between registration information calculate after, in ultrahigh resolution picture display window, to user selection details carry out linkage matching put
Thorough browse;
Described equipment collaboration linkage playback function, user can be by onvif protocol search and inquiry video file, and to inquiry
To video file played back;In replayed section, user in the scene details checked of scene general picture picture and selecting needs,
After being calculated by spatial registration information, in ultrahigh resolution picture display window, link to the details of user's selection
Matching amplification playback.
8. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 6, it is characterised in that:
The log management function can retrieve video management software local operation and alarm log;
The subscriber management function can add and delete user account, and user right is configured;
The system configuration features can carry out language, NTP service functions, cpu performance configuration.
9. more focal length lens ultrahigh resolution linkage imaging camera machine systems according to claim 7, it is characterised in that:Institute
The general picture collecting unit and the details collecting unit stated realize spatial registration by following steps:
(1) single-lens unit acquisition picture coordinate system is established, using spherical model, for mapping real space coordinate to photoreceptor
The correspondence of the plane coordinates of part;
(2) video pictures of scene general picture picture and details collecting unit are sampled respectively, calculates the mapping relations of sampled point;Meter
Calculation method is to up-sample the pixel coordinate of 6 points, and be converted to space coordinate in scene general picture picture, obtain 6 vector vs 1,
v2…v6;The pixel coordinate of 6 points is up-sampled, and be converted to space coordinate in the video pictures of details collecting unit, obtained
V1 ', v2 ' ... using following computational methods, spin matrix MatH2D is obtained by this 6 groups of spatial coordinates calculations by v6 ':
MatD=[v1, v2, v3, v4, v5, v6],
MatH=[v1', v2', v3', v4', v5', v6'],
MatH2D=MatD × PINV (MatH), wherein PINV (Matrix) are the generalized inverse of matrix Matrix;
(3) when matching linkage preview or playback, its space coordinate v is calculated by the coordinate (x, y) on scene general picture picture first,
And space coordinate v ' of this under ultrahigh resolution picture coordinate system is obtained by v × MatH2D=v ', it is finally calculated by v ' single
The pixel coordinate (x ', y ') of a U10X.
It is 10. special according to the more focal length lens ultrahigh resolutions linkage imaging camera machine systems of claim 1-8 any one of them
Sign is:Multiple target tracking algorithm realization in the Multiobjective Intelligent tracking system mainly includes the following steps that:
(1) multiple mobile object is gone out based on Background difference and frame difference method fusion detection;Multiple mobile object detection is broadly divided into three steps:The
One step initial background model;Second step utilizes frame difference method context update, then carries out binaryzation;Third step utilizes Background difference
Then image is carried out binaryzation by background difference;
(2) image of binaryzation is subjected to further denoising and Segmentation of Multi-target;
(3) moving foreground object tracking is carried out using improved Camshift algorithms;
Wherein, step (1), step (2) are based on necessary to institute's video image to be handled, are before carrying out multiple target tracking
Pretreatment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711402320.XA CN108156419A (en) | 2017-12-22 | 2017-12-22 | More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711402320.XA CN108156419A (en) | 2017-12-22 | 2017-12-22 | More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108156419A true CN108156419A (en) | 2018-06-12 |
Family
ID=62464171
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711402320.XA Pending CN108156419A (en) | 2017-12-22 | 2017-12-22 | More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108156419A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109165629A (en) * | 2018-09-13 | 2019-01-08 | 百度在线网络技术(北京)有限公司 | It is multifocal away from visual barrier cognitive method, device, equipment and storage medium |
CN111914715A (en) * | 2020-07-24 | 2020-11-10 | 廊坊和易生活网络科技股份有限公司 | Intelligent vehicle target real-time detection and positioning method based on bionic vision |
CN112135046A (en) * | 2020-09-23 | 2020-12-25 | 维沃移动通信有限公司 | Video shooting method, video shooting device and electronic equipment |
CN112383747A (en) * | 2020-11-02 | 2021-02-19 | 长沙超创电子科技有限公司 | Long-distance full-range ultrahigh-resolution intelligent perception compound eye monitoring camera |
US10984588B2 (en) | 2018-09-07 | 2021-04-20 | Baidu Online Network Technology (Beijing) Co., Ltd | Obstacle distribution simulation method and device based on multiple models, and storage medium |
US11047673B2 (en) | 2018-09-11 | 2021-06-29 | Baidu Online Network Technology (Beijing) Co., Ltd | Method, device, apparatus and storage medium for detecting a height of an obstacle |
US11205289B2 (en) | 2018-09-07 | 2021-12-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, device and terminal for data augmentation |
US11307302B2 (en) | 2018-09-07 | 2022-04-19 | Baidu Online Network Technology (Beijing) Co., Ltd | Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium |
US11718318B2 (en) | 2019-02-22 | 2023-08-08 | Apollo Intelligent Driving (Beijing) Technology Co., Ltd. | Method and apparatus for planning speed of autonomous vehicle, and storage medium |
US11780463B2 (en) | 2019-02-19 | 2023-10-10 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120169842A1 (en) * | 2010-12-16 | 2012-07-05 | Chuang Daniel B | Imaging systems and methods for immersive surveillance |
CN105744239A (en) * | 2016-05-11 | 2016-07-06 | 湖南源信光电科技有限公司 | Multi-focal-length lens ultrahigh resolution linkage imaging device |
CN105933678A (en) * | 2016-07-01 | 2016-09-07 | 湖南源信光电科技有限公司 | Multi-focal length lens linkage imaging device based on multi-target intelligent tracking |
-
2017
- 2017-12-22 CN CN201711402320.XA patent/CN108156419A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120169842A1 (en) * | 2010-12-16 | 2012-07-05 | Chuang Daniel B | Imaging systems and methods for immersive surveillance |
CN105744239A (en) * | 2016-05-11 | 2016-07-06 | 湖南源信光电科技有限公司 | Multi-focal-length lens ultrahigh resolution linkage imaging device |
CN105933678A (en) * | 2016-07-01 | 2016-09-07 | 湖南源信光电科技有限公司 | Multi-focal length lens linkage imaging device based on multi-target intelligent tracking |
Non-Patent Citations (2)
Title |
---|
侯红娜 等: "基于多特征自适应融合的Camshift行人跟踪算法", 《电子质量》 * |
刘阳: "基于多特征融合的CamShift跟踪算法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10984588B2 (en) | 2018-09-07 | 2021-04-20 | Baidu Online Network Technology (Beijing) Co., Ltd | Obstacle distribution simulation method and device based on multiple models, and storage medium |
US11205289B2 (en) | 2018-09-07 | 2021-12-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, device and terminal for data augmentation |
US11307302B2 (en) | 2018-09-07 | 2022-04-19 | Baidu Online Network Technology (Beijing) Co., Ltd | Method and device for estimating an absolute velocity of an obstacle, and non-volatile computer-readable storage medium |
US11519715B2 (en) | 2018-09-11 | 2022-12-06 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, device, apparatus and storage medium for detecting a height of an obstacle |
US11047673B2 (en) | 2018-09-11 | 2021-06-29 | Baidu Online Network Technology (Beijing) Co., Ltd | Method, device, apparatus and storage medium for detecting a height of an obstacle |
CN109165629A (en) * | 2018-09-13 | 2019-01-08 | 百度在线网络技术(北京)有限公司 | It is multifocal away from visual barrier cognitive method, device, equipment and storage medium |
CN109165629B (en) * | 2018-09-13 | 2019-08-23 | 百度在线网络技术(北京)有限公司 | It is multifocal away from visual barrier cognitive method, device, equipment and storage medium |
US11126875B2 (en) | 2018-09-13 | 2021-09-21 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium |
US11780463B2 (en) | 2019-02-19 | 2023-10-10 | Baidu Online Network Technology (Beijing) Co., Ltd. | Method, apparatus and server for real-time learning of travelling strategy of driverless vehicle |
US11718318B2 (en) | 2019-02-22 | 2023-08-08 | Apollo Intelligent Driving (Beijing) Technology Co., Ltd. | Method and apparatus for planning speed of autonomous vehicle, and storage medium |
CN111914715A (en) * | 2020-07-24 | 2020-11-10 | 廊坊和易生活网络科技股份有限公司 | Intelligent vehicle target real-time detection and positioning method based on bionic vision |
CN111914715B (en) * | 2020-07-24 | 2021-07-16 | 廊坊和易生活网络科技股份有限公司 | Intelligent vehicle target real-time detection and positioning method based on bionic vision |
CN112135046B (en) * | 2020-09-23 | 2022-06-28 | 维沃移动通信有限公司 | Video shooting method, video shooting device and electronic equipment |
CN112135046A (en) * | 2020-09-23 | 2020-12-25 | 维沃移动通信有限公司 | Video shooting method, video shooting device and electronic equipment |
CN112383747B (en) * | 2020-11-02 | 2021-06-22 | 长沙超创电子科技有限公司 | Long-distance full-range ultrahigh-resolution intelligent perception compound eye monitoring camera |
CN112383747A (en) * | 2020-11-02 | 2021-02-19 | 长沙超创电子科技有限公司 | Long-distance full-range ultrahigh-resolution intelligent perception compound eye monitoring camera |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108156419A (en) | More focal length lens linkage imaging camera machine system based on multiple features combining and Camshift algorithms | |
CN105933678B (en) | More focal length lens linkage imaging device based on Multiobjective Intelligent tracking | |
CN108171723A (en) | Based on more focal length lens of Vibe and BP neural network algorithm linkage imaging camera machine system | |
CN105744239B (en) | More focal length lens ultrahigh resolutions linkage imaging devices | |
US20130107072A1 (en) | Multi-resolution ip camera | |
CN104301677B (en) | The method and device monitored towards the panoramic video of large scene | |
Lei et al. | Depth map super-resolution considering view synthesis quality | |
Pozo et al. | An integrated 6DoF video camera and system design | |
JP4546565B2 (en) | Digital image processing | |
CN107659774A (en) | A kind of video imaging system and method for processing video frequency based on multiple dimensioned camera array | |
CN104270570A (en) | Binocular video camera and image processing method thereof | |
CN104580858B (en) | A kind of unmanned plane is imaged load with multi-source | |
CN102857739A (en) | Distributed panorama monitoring system and method thereof | |
WO2008031089A2 (en) | System and method for high performance image processing | |
Yong et al. | Panoramic background image generation for PTZ cameras | |
CN108076297A (en) | Camera chain based on the target tracking algorism that Kalman filter is combined with Camshift algorithms | |
Soyak et al. | Low-complexity tracking-aware H. 264 video compression for transportation surveillance | |
CN106060470A (en) | Video monitoring method and system | |
CN107018386A (en) | A kind of video flowing multiresolution observation system | |
CN105959575A (en) | Zooming panoramic camera | |
EP3466051A1 (en) | Three-dimensional noise reduction | |
CN109636763A (en) | A kind of intelligence compound eye monitoring system | |
Quevedo et al. | Video enhancement using spatial and temporal super-resolution from a multi-camera system | |
CN113436130B (en) | Intelligent sensing system and device for unstructured light field | |
JP2005175970A (en) | Imaging system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180612 |
|
RJ01 | Rejection of invention patent application after publication |