CN106162144A - A kind of visual pattern processing equipment, system and intelligent machine for overnight sight - Google Patents
A kind of visual pattern processing equipment, system and intelligent machine for overnight sight Download PDFInfo
- Publication number
- CN106162144A CN106162144A CN201610580656.4A CN201610580656A CN106162144A CN 106162144 A CN106162144 A CN 106162144A CN 201610580656 A CN201610580656 A CN 201610580656A CN 106162144 A CN106162144 A CN 106162144A
- Authority
- CN
- China
- Prior art keywords
- information
- image
- monitoring
- infrared light
- intelligent machine
- 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
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/204—Image signal generators using stereoscopic image cameras
- H04N13/239—Image signal generators using stereoscopic image cameras using two 2D image sensors having a relative position equal to or related to the interocular distance
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/204—Image signal generators using stereoscopic image cameras
- H04N13/246—Calibration of cameras
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N13/00—Stereoscopic video systems; Multi-view video systems; Details thereof
- H04N13/20—Image signal generators
- H04N13/275—Image signal generators from 3D object models, e.g. computer-generated stereoscopic image signals
-
- 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
Abstract
The invention provides a kind of visual pattern processing equipment, system and intelligent machine for overnight sight, including: infrared light-emitting device, for the object emission infrared light in multidimensional image harvester vision monitoring space;Multidimensional image harvester, for gathering the infrared light of object reflection, is focused imaging processing to infrared light, obtains the video image of correspondence;Front-end image processing means, for setting up the threedimensional model that vision monitoring space is corresponding, and the spatial positional information that obtains object with this and monitoring information;Its light filling based on infrared light-emitting device, obtain the video image in vision monitoring space in real time, and build three dimensional local information model accordingly, obtain the 3 D monitoring information of object, carrying out monitoring under object security protection identification, safety monitoring, subdued light conditions measurement etc. accordingly, its accuracy of identification is higher, simultaneously, path planning can also be carried out so that automatic obstacle-avoiding walkaround inspection implemented by intelligent machine according to above-mentioned 3 D monitoring information.
Description
Technical field
The present invention relates to technical field of image processing, in particular to a kind of at the visual pattern of overnight sight
Reason equipment, system and intelligent machine.
Background technology
Along with expanding economy, the mankind after the primary demand meeting clothing, food, shelter, row, the person and wealth security protection
Demand has the most just risen to a new height.Video monitoring becomes the technology peace of everybody first-selection because of its intuitive and seriality
Anti-mode.
At present, infrared night video camera then clearly extensively should by a large amount of because meeting the hidden and shooting picture under night-environment
In video monitoring.Concrete, at monitoring place installation infrared night viewing camera, and gather it by infrared night video camera
Video image in coverage, and the video image of collection is stored, when monitoring suspicion object, from the figure gathered
As information identifies and positions current suspicion object, and this suspicion object is tracked, to realize this suspicion object
Video monitoring.
Inventor finds under study for action, and the monitoring being currently based on infrared night video camera also rests on the video of planar
Monitor stages, accuracy of identification is relatively low, and identification and tracking to personage, vehicle etc. have some limitations.
Summary of the invention
In view of this, the purpose of the embodiment of the present invention is that providing a kind of visual pattern for overnight sight to process sets
Standby, system and intelligent machine, it is possible to by under overnight sight according to carrying out Three-dimension object recognition and implementing avoidance so that accuracy of identification
Higher and avoidance is effective.
First aspect, embodiments provides a kind of visual pattern processing equipment for overnight sight, including: red
Outer luminescent device, multidimensional image harvester and front-end image processing means;
Described infrared light-emitting device, for the vision monitoring spatial emission infrared light of described multidimensional image harvester;
Described multidimensional image harvester, the infrared light of the object reflection in its vision monitoring space of Real-time Collection,
Described infrared light is focused imaging processing, obtains dynamic continuous print video image, described video image is sent to front end
Image processing apparatus;
Described front-end image processing means is connected with described multidimensional image harvester, for according to described video image
Carry out three-dimensional modeling, obtain the threedimensional model that the vision monitoring space of described multidimensional image harvester is corresponding, and according to described
Threedimensional model obtains the monitoring information of object;The following information of described monitoring information at least object includes: kind of information, size are believed
Breath and spatial positional information.
In conjunction with first aspect, embodiments provide the first possible embodiment of first aspect, wherein, use
In the visual pattern processing equipment of overnight sight, also include high accuracy assistant images harvester;
Described high accuracy assistant images harvester, for gather in its vision monitoring space object reflection infrared
Light, is focused imaging processing to described infrared light, obtains dynamic continuous print high accuracy video image, by described high accuracy video
Image sends to front-end image processing means;
Described front-end image processing means is additionally operable to, and according to described high accuracy video image, described threedimensional model is carried out school
Quasi-process, the threedimensional model after being calibrated, and from the threedimensional model after described calibration, obtain the monitoring information of described object.
In conjunction with the first possible embodiment of first aspect, embodiments provide the second of first aspect
Possible embodiment, wherein, described multidimensional image harvester includes: monocular camera lens or binocular head;Described binocular
Head includes: first group gathers camera lens and second group of collection camera lens;
Described monocular camera lens, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, to described red
Outer light is focused imaging processing, obtains dynamic continuous print video image, is sent with the first predeterminated frequency by described video image
To front-end image processing means;
Described first group gathers camera lens and described second group of collection camera lens, is respectively used to gather in corresponding vision monitoring space
Dynamic continuous print view data, by all of view data all with described first predeterminated frequency send to front-end image process dress
Put.
In conjunction with the embodiment that the second of first aspect is possible, embodiments provide the third of first aspect
Possible embodiment, wherein, described front-end image processing means includes: control device, special visual processing apparatus and storage
Device;
Described control device, for control algorithm in described special visual processing apparatus processing procedure and with described algorithm
Corresponding result;And, control the data stream of described memorizer storage and corresponding data flow circuit;
Described consecutive image data, under the control of described control device, are entered by described special visual processing apparatus
Row three-dimensional modeling processes, and obtains threedimensional model, and obtains the prison of object corresponding to described view data from described threedimensional model
Control information, and the barrier integrated information of correspondence is obtained according to described monitoring information;
Described memorizer, under the control of described control device, stores what described multidimensional image harvester sent
Dynamically continuous print view data, in high precision the high accuracy video image of assistant images harvester transmission, threedimensional model, and/or,
Threedimensional model after calibration, and, described monitoring information and described barrier integrated information.
In conjunction with the third possible embodiment of first aspect, embodiments provide the 4th kind of first aspect
Possible embodiment, wherein, described special visual processing apparatus includes:
Receiver module, for receiving two groups of video images that described binocular head sends, extracts video described in two groups respectively
Image feature data in image;
Identification module, trains described image feature data for the method using the degree of depth to learn, obtains described video image
In the kind of information of object;
Stereo matching module, mates under different points of view two groups of image feature datas are carried out solid, and to
The two groups of image feature datas being fitted on same position are corrected processing, and obtain two groups of image feature datas in same position
Parallax;
Distance size computing module, for according to described parallax and the attitude information of described binocular head, calculates present bit
Put and the distance of object in the vision monitoring space of described binocular head;
Three-dimensional modeling module, in the vision monitoring space according to current location and described binocular head object away from
From, build threedimensional model, and obtain the spatial positional information of object in described video image according to described threedimensional model.
In conjunction with the 4th kind of possible embodiment of first aspect, embodiments provide the 5th kind of first aspect
Possible embodiment, wherein, described front-end image processing means also includes: speech recognition synthesizer;
Described speech recognition synthesizer, for receiving the voice messaging that user sends, knows described voice messaging
Other places are managed, and obtain the destination address of correspondence;And, according to described destination address and described barrier integrated information, synthesis and institute
State the suggestion voice that destination address is corresponding, export described suggestion voice to point out.
In conjunction with the 5th kind of possible embodiment of first aspect, embodiments provide the 6th kind of first aspect
Possible embodiment, wherein, described front-end image processing means also includes: path planning subsystem;
Described path planning subsystem, for according to described barrier integrated information original guidance path being adjusted or
Person regenerates guidance path
In conjunction with the embodiment that the second of first aspect is possible, embodiments provide the 7th kind of first aspect
Possible embodiment, wherein, first group of collection camera lens and second group of collection camera lens all include a monocular camera lens or multiple list
Ocular head.
In conjunction with the first possible embodiment of first aspect, embodiments provide the 8th kind of first aspect
Possible embodiment, wherein, multidimensional image harvester includes monocular camera lens;
Monocular camera lens, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, infrared light is carried out
Focal imaging processes, and obtains dynamic continuous print video image, is sent to front-end image with the first predeterminated frequency by video image
Reason device.
The embodiment of the present invention provide a kind of visual pattern processing equipment for overnight sight, with prior art in based on
The monitoring of infrared night video camera also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, and it is based on red
The light filling of outer luminescent device, dynamic continuous print video image in obtaining its vision monitoring space in real time, and by front-end image
Reason device carries out real-time signal analysis to this video image, builds three dimensional local information model, obtains the 3 D monitoring letter of object
Breath, according to carrying out object security protection identification accordingly, to carry out monitoring under safety monitoring, subdued light conditions measurement etc., its accuracy of identification is relatively
Height, at the same time it can also be carry out path planning according to above-mentioned 3 D monitoring information so that intelligent machine is implemented automatic obstacle-avoiding and maked an inspection tour inspection
Look into.
Second aspect, the embodiment of the present invention additionally provides a kind of intelligent machine, including: intelligent machine body and above-mentioned first
The visual pattern processing equipment for overnight sight described in any one of aspect;Described visual pattern processing equipment embeds described intelligence
In energy machine body;
Described intelligent machine, monitors the 3 D monitoring information of object in its vision monitoring space in real time, and by described
3 D monitoring information stores, in order to follow-up carry out security protection identification;And/or, receive the voice messaging that user sends, to described
Voice messaging is identified processing, and obtains the destination address of correspondence;And, the barrier in detection vision monitoring space, front is combined
Conjunction information, according to described barrier integrated information, plans the walking path of described destination address, and controls according to the row after planning
Walk path walking, to implement avoidance.
A kind of intelligent machine that the embodiment of the present invention provides, monitoring based on infrared night video camera is also with prior art
Resting on the video monitoring stage of planar, accuracy of identification is relatively low to be compared, its light filling based on infrared light-emitting device, obtains in real time
Dynamic continuous print video image in taking its vision monitoring space, and by front-end image processing means, this video image is carried out reality
Time information analysis, build three dimensional local information model, obtain the 3 D monitoring information of object, according to carrying out object security protection knowledge accordingly
Not, to carry out monitoring under safety monitoring, subdued light conditions measurement etc., its accuracy of identification is higher, at the same time it can also be according to above-mentioned three-dimensional
Monitoring information carries out path planning, to implement automatic obstacle-avoiding walkaround inspection.
The third aspect, the embodiment of the present invention additionally provides a kind of visual pattern processing system for overnight sight, including:
Including: the intelligent machine described in above-mentioned any one of second aspect;Also include remote control terminal;
Described remote control terminal, for receiving the 3 D monitoring information of described intelligent machine transmission and storing;With/
Or, according to the spatial positional information of the object in the threedimensional model that described intelligent machine is corresponding, calculate described intelligent machine relative
In the distance of described object, and according to the track route of the result described intelligent machine of planning calculated, and control described intelligent machine
Device travels according to the track route after planning, to implement avoidance;And/or, according to the track route after planning to described intelligent machine
The track route of device is monitored.
The embodiment of the present invention provide a kind of visual pattern processing system for overnight sight, with prior art in based on
The monitoring of infrared night video camera also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, above-mentioned intelligence
Machine can light filling based on infrared light-emitting device, obtain in real time in its vision monitoring space dynamically continuous print video image, and lead to
Cross front-end image processing means and this video image is carried out real-time signal analysis, build three dimensional local information model, obtain object
3 D monitoring information, according to carrying out object security protection identification accordingly, to carry out monitoring under safety monitoring, subdued light conditions measurement etc.,
Its accuracy of identification is higher, at the same time it can also be carry out path planning according to above-mentioned 3 D monitoring information, to implement automatic obstacle-avoiding tour
Check.Further, system can also control above-mentioned intelligent machine by remote control terminal or be monitored above-mentioned intelligent machine,
Object preferably can be identified and control intelligent machine and implement avoidance.
For making the above-mentioned purpose of the present invention, feature and advantage to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme of the embodiment of the present invention, below by embodiment required use attached
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, and it is right to be therefore not construed as
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to this
A little accompanying drawings obtain other relevant accompanying drawings.
Fig. 1 shows the structure of a kind of visual pattern processing equipment for overnight sight that the embodiment of the present invention provided
Schematic diagram;
Fig. 2 shows another kind that the embodiment of the present invention the provided knot for the visual pattern processing equipment of overnight sight
Structure schematic diagram;
Fig. 3 shows another kind that the embodiment of the present invention the provided knot for the visual pattern processing equipment of overnight sight
Structure schematic diagram;
It is a kind of special in the visual pattern processing equipment of overnight sight that Fig. 4 shows that the embodiment of the present invention is provided
The overall structure schematic diagram of visual processing apparatus;
Fig. 5 shows a kind of front end in the visual pattern processing equipment of overnight sight that the embodiment of the present invention is provided
The overall structure schematic diagram of image processing apparatus;
Fig. 6 shows another kind that the embodiment of the present invention the provided visual pattern processing equipment special secondary school for overnight sight
With the overall structure schematic diagram of visual processing apparatus;
Fig. 7 shows the structure of a kind of visual pattern processing system for overnight sight that the embodiment of the present invention provided
Schematic diagram.
Major Symbol illustrates:
100, visual pattern processing equipment;200, intelligent machine;300, remote control terminal;10, infrared light-emitting device;20、
Multidimensional image harvester;30, front-end image processing means;40, assistant images harvester in high precision;201, first group of collection
Camera lens;202, second group gathers camera lens;301, device is controlled;302, special visual processing apparatus;303, memorizer;3021, connect
Receive module;3022, identification module;3023, stereo matching module;3024, distance size computing module;3025, three-dimensional modeling mould
Block;3026, supersonic detector;3027, use processing device;304, speech recognition synthesizer;305, path planning
System.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with the embodiment of the present invention
Middle accompanying drawing, is clearly and completely described the technical scheme in the embodiment of the present invention, it is clear that described embodiment is only
It is a part of embodiment of the present invention rather than whole embodiments.Generally real with the present invention illustrated described in accompanying drawing herein
The assembly executing example can be arranged with various different configurations and design.Therefore, below to the present invention's provided in the accompanying drawings
The detailed description of embodiment is not intended to limit the scope of claimed invention, but is merely representative of the selected reality of the present invention
Execute example.Based on embodiments of the invention, the institute that those skilled in the art are obtained on the premise of not making creative work
There are other embodiments, broadly fall into the scope of protection of the invention.
Also rest on the video monitoring stage of planar in view of the monitoring being currently based on infrared night video camera, identify
Precision is relatively low, and identification and tracking to personage, vehicle etc. have some limitations.Based on this, the embodiment of the present invention provides
A kind of visual pattern processing equipment for overnight sight and system, it is possible to the vision monitoring space that will gather under overnight sight
Interior video image carries out three-dimensional modeling process, obtains the 3 D monitoring information of object, according to carrying out object security protection identification accordingly,
To carry out monitoring under safety monitoring, subdued light conditions measurement etc., its accuracy of identification is higher, at the same time it can also be according to above-mentioned three-dimensional prison
Control information carries out path planning so that automatic obstacle-avoiding walkaround inspection implemented by intelligent machine, is described below by embodiment.
With reference to Fig. 1, embodiments provide a kind of visual pattern processing equipment 100 for overnight sight, including:
Infrared light-emitting device 10, multidimensional image harvester 20 and front-end image processing means 30;
Infrared light-emitting device 10, for the vision monitoring spatial emission infrared light of multidimensional image harvester 20;
Multidimensional image harvester 20, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, right
Infrared light is focused imaging processing, obtains dynamic continuous print video image, is sent by video image to front-end image process dress
Put 30;
Front-end image processing means 30 is connected with multidimensional image harvester 20, for carrying out three-dimensional according to video image
Modeling, obtains the threedimensional model that the vision monitoring space of multidimensional image harvester 20 is corresponding, and obtains thing according to threedimensional model
The monitoring information of body;The following information of monitoring information at least object includes: kind of information, dimension information and spatial positional information.
The visual pattern processing equipment 100 that the embodiment of the present invention provides is mainly used under overnight sight;In view of night vision
Under environment, dark, now, the direct definition using multidimensional image harvester 20 to gather video image is poor, this
Bright embodiment carries out light filling beforehand through infrared light-emitting device 10, is i.e. filled to multidimensional image collection by infrared light-emitting device 10
Put the vision monitoring spatial emission infrared light of 20;The infrared light of arrival can be reflected by the object in above-mentioned visual space, and
Multidimensional image harvester 20 then collects the infrared light of the above-mentioned object reflection in its vision monitoring space, the infrared light that will collect
Line focus is processed into picture, is eventually converted into video information;In the embodiment of the present invention, above-mentioned multidimensional image harvester 20 is received in real time
Collecting the infrared light that the object in its vision monitoring space reflects, then line focus is processed into picture, obtains dynamic continuous print video figure
Picture.Wherein, above-mentioned infrared light-emitting device 10 can be infrared light board.
Above-mentioned front-end image processing means 30 is as a single processing means, and it can be by specific algorithm to upper
State the dynamic continuous print video image obtained to process, to form the 3-D view under night vision, enter from this 3-D view
Row identifies and location positioning accurately, in the hope of the true scene of real reduction, i.e. based on this 3-D view, can know that vision is supervised
The monitoring information of object in control space, in order to carry out security protection identification according to monitoring information, in order to follow-up with carrying out safety monitoring, secretly
Measurement etc. is monitored under optical condition;At the same time it can also be implement avoidance according to the intelligent machine that above-mentioned monitoring information system is corresponding;Its
In, (height such as people is believed for above-mentioned monitoring information at least kind of object information (such as personage, animal or equipment etc.), dimension information
Breath, the dimension information of object) and spatial positional information;The spatial positional information of above-mentioned object includes: between three-dimensional point cloud atlas, object
The information such as relative position, direction and distance.
A kind of visual pattern processing equipment 100 for overnight sight that the embodiment of the present invention provides, can embed intelligence
In machine 200, above-mentioned intelligent machine 200 can be to make an inspection tour intelligent machine or stationary machines under night vision;Embed for night vision
The fixing intelligent machine of the visual pattern processing equipment 100 of environment, it is possible to obtain 3 D monitoring by this equipment under night vision
Information, and identify the height of personage or the dimension information of object accordingly, survey in order to carry out monitoring under safety monitoring, subdued light conditions
Amount etc.;The intelligent machine of the mobile tour embedding the visual pattern processing equipment 100 for overnight sight equally can be in night vision
Under obtain 3 D monitoring information by this equipment, according to above-mentioned 3 D monitoring information automatic obstacle-avoiding make an inspection tour, the most also there is measurement
Function.
A kind of visual pattern processing equipment 100 for overnight sight that the embodiment of the present invention provides, and in prior art
Monitoring based on infrared night video camera also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, its base
In the light filling of infrared light-emitting device 10, dynamic continuous print video image in obtaining its vision monitoring space in real time, and pass through front end
Image processing apparatus 30 carries out real-time signal analysis to this video image, builds three dimensional local information model, obtains the three of object
Dimension monitoring information, according to carrying out object security protection identification accordingly, to carry out monitoring under safety monitoring, subdued light conditions measurement etc., it is known
Other precision is higher, at the same time it can also be carry out path planning according to above-mentioned 3 D monitoring information so that intelligent machine is implemented automatically to keep away
Barrier walkaround inspection.
Further, with reference to Fig. 2, the above-mentioned visual pattern processing equipment 100 for overnight sight, also include that high accuracy is auxiliary
Help image collecting device 40;
Assistant images harvester 40 in high precision, for gathering the infrared light of the object reflection in its vision monitoring space,
Infrared light is focused imaging processing, obtains dynamic continuous print high accuracy video image, high accuracy video image is sent extremely
Front-end image processing means 30;
Front-end image processing means 30 is additionally operable to, and according to high accuracy video image, threedimensional model is carried out calibration process,
Threedimensional model after calibration, and from the threedimensional model after calibration, obtain the monitoring information of object.
Concrete, by above-mentioned high accuracy assistant images harvester 40 Real-time Collection, multidimensional image harvester 20
The infrared light of the middle object reflection in vision monitoring space, and process through focal imaging, obtain dynamic continuous print high accuracy video
Image, and above-mentioned image is preset frame rate transmission to front-end image processing means 30 with second.Front-end image processing means 30
Then according to above-mentioned high accuracy video image, the threedimensional model that the video image gathered according to multidimensional image harvester 20 is set up
Carry out calibration process, the threedimensional model after the calibration obtained;And the threedimensional model after calibrating more can accurately show multidimensional image
Monitoring information in harvester 20 vision monitoring space.Wherein, second herein is preset frame rate and following first and presets frame
Frequency can be identical, it is also possible to different;In the embodiment of the present invention, arrange second and preset frame rate and following first and preset frame rate
Different.
In the embodiment of the present invention, above-mentioned high accuracy assistant images harvester 40 can be high accuracy photographic head, this shooting
Head can be with auto-focusing, and it is for by focusing parameter, to binocular difference image information, (i.e. multidimensional image harvester 20 gathers
Video image) process carry out correcting and supplementing;It addition, this system can also gather fine definition (high accuracy in other words)
Video image or picture;The figure that multidimensional image harvester 20 gathers is made up by the high accuracy vedio data gathered
Deficiency as data, it is possible to set up threedimensional model more accurately, and obtain monitoring information therein, and then believe according to this monitoring
Breath carries out object identification and/or implements avoidance, in order to carry out monitoring under safety monitoring or subdued light conditions measurement etc. so that identify
Precision is higher.
Further, with reference to Fig. 3, above-mentioned in the visual pattern processing equipment 100 of overnight sight, above-mentioned multidimensional image
Harvester 20 includes: monocular camera lens or binocular head;
Above-mentioned monocular camera lens, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, to infrared light
It is focused imaging processing, obtains dynamic continuous print video image, video image is sent to front end figure with the first predeterminated frequency
As processing means;
It addition, above-mentioned binocular head includes: first group gathers camera lens 201 and second group of collection camera lens 202;
First group gathers camera lens 201 and second group of collection camera lens 202, is respectively used to gather in corresponding vision monitoring space
Dynamically continuous print view data, all sends all of view data to front-end image processing means 30 with the first predeterminated frequency.
Concrete, two groups of collection camera lenses that above-mentioned binocular head includes are used to gather moving in self vision monitoring space
State continuous print video image, and send to front-end image processing means 30 with the first predeterminated frequency.Wherein, above-mentioned first frame is preset
Frequency can be configured in advance, as arranged 30 frames per second;To the above-mentioned first setting presetting frame rate in the embodiment of the present invention
It is not particularly limited.
Wherein, above-mentioned first group of collection camera lens 201 and second group of collection camera lens 202 all include a monocular camera lens or multiple
Monocular camera lens.
Further, with reference to Fig. 1~Fig. 3, above-mentioned in the visual pattern processing equipment 100 of overnight sight, front end figure
As processing means 30 includes: control device 301 and special visual processing apparatus 302;
Control device 301, for the algorithm controlling in special visual processing apparatus 302 processing procedure, and with algorithm pair
The result answered;
Special visual processing apparatus 302, under the control controlling device 301, extracts the sequential chart in video image
As data, consecutive image data are carried out three-dimensional modeling process, obtain threedimensional model, and from threedimensional model, obtain view data
The monitoring information of corresponding object, and the barrier integrated information of correspondence is obtained according to described monitoring information.
Concrete, in front-end image processing means 30, special visual processing apparatus 302 implement real computing
Journey, is controlled, by controlling device 301, algorithm and the corresponding operation result that special visual processing apparatus 302 is implemented.
Further, with reference to Fig. 3, above-mentioned in the visual pattern processing equipment 100 of overnight sight, front-end image processes
Device 30 also includes memorizer 303;
Control device 301 to be additionally operable to, control the data stream of memorizer 303 storage and corresponding data flow circuit;
Memorizer 303, under the control controlling device 301, it is dynamic that storage multidimensional image harvester 20 sends
Continuous print view data, in high precision the high accuracy video image of assistant images harvester 40 transmission, threedimensional model, and/or, school
Threedimensional model after standard, and, monitoring information and barrier integrated information.
Concrete, in front-end image processing means 30, also include memorizer 303;The storage merit implemented by memorizer 303
Can, controlled, by controlling device 301, path and the data stream that memorizer 303 carries out storing.
The video image gathered based on multidimensional image harvester 20, special visual processing apparatus 302 need to be by specifically
It is processed by algorithm, to set up corresponding threedimensional model;In the embodiment of the present invention, for multidimensional image harvester 20 it is
The situation of binocular head, corresponding, with reference to Fig. 4, special visual processing apparatus 302 includes:
Receiver module 3021, for receiving two groups of video images that binocular head sends, extracts two groups of video images respectively
In image feature data;
Identification module 3022, trains described image feature data for the method using the degree of depth to learn, obtains described video
The kind of information of the object in image;
Stereo matching module 3023, mates under different points of view two groups of image feature datas are carried out solid, and
It is corrected the two groups of image feature datas matching same position processing, obtains two groups of characteristics of image numbers in same position
According to parallax;
Distance size computing module 3024, for according to parallax and the attitude information of binocular head, calculate current location with
The distance of object in the vision monitoring space of binocular head;
Three-dimensional modeling module 3025, in the vision monitoring space according to current location and binocular head object away from
From, build threedimensional model, and obtain the spatial positional information of object in video image according to above-mentioned threedimensional model.
Concrete, every two field picture all includes a lot of pixels, receiver module 3021 after receiving two groups of video images,
Extract the characteristics of image of each pixel in each group of video image;The method being used the degree of depth to learn by identification module 3022 is trained
The above-mentioned image feature data extracted, to distinguish the kind of the object in video image;Two groups are regarded by stereo matching module 3022
Frequently in image, the characteristics of image of each pixel carries out Stereo matching, enters the two groups of image feature datas matching same position
Row correction process, obtains the characteristics of image of pixel in two groups of video images of same position.
Then, distance size computing module 3023 is according to the above-mentioned parallax obtained and the attitude information of binocular head, meter
Calculate the distance of current location and each object in the vision monitoring space of described binocular head, finally, three-dimensional modeling module
3024, according to all of distance calculated, can restore whole three in the vision monitoring space of multidimensional image harvester 20
Dimension scene, sets up corresponding threedimensional model, thus obtains the spatial positional information of the object in video image.
Further, with reference to Fig. 5, above-mentioned in the visual pattern processing equipment 100 of overnight sight, front-end image processes
Device 30 also includes: speech recognition synthesizer 304;
Speech recognition synthesizer 304, for receiving the voice messaging that user sends, is identified place to voice messaging
Reason, obtains the destination address of correspondence;And, according to destination address and barrier integrated information, synthesize corresponding with destination address
Suggestion voice, output suggestion voice is to point out.
In the embodiment of the present invention, user directly can say this by speech recognition synthesizer 304 and be used for overnight sight
The destination address of the intelligent machine 1 embedded by visual pattern processing equipment 100, speech recognition synthesizer 304 then according to
The destination address of the voice messaging synthesis correspondence at family, and according to destination address and barrier integrated information, synthesis and destination address
Corresponding suggestion voice, output suggestion voice is to point out, and such as, prompting intelligent machine 1 carries out left-hand rotation etc..
Further, with reference to Fig. 5, above-mentioned in the visual pattern processing equipment 100 of overnight sight, front-end image processes
Device 30 also includes: path planning subsystem 305;Path planning subsystem 305, is used for according to barrier integrated information original
Guidance path is adjusted or regenerates guidance path.
Concrete, path planning subsystem 305 is for according to the barrier integrated information received, entering original guidance path
Row sum-equal matrix or regenerate guidance path, with the intelligent machine embedding the visual pattern processing equipment 100 for overnight sight
Device 1 Real Time Obstacle Avoiding.
Further, with reference to Fig. 6, above-mentioned in the visual pattern processing equipment 100 of overnight sight, also include: wireless
Transport module;Special visual processing apparatus 302 includes: supersonic detector 3026 and use processing device 3027;
Supersonic detector 3026, for detecting the range information of barrier, by distance information transmission use processing
Device 3027;
Use processing device 3027 is for determining barrier integrated information according to video image and range information and defeated
Go out barrier integrated information;
Wireless transport module, for being sent to the remote control terminal 300 of correspondence by barrier integrated information.
A kind of visual pattern processing equipment 100 for overnight sight that the embodiment of the present invention provides, and in prior art
Monitoring based on infrared night video camera also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, its base
In the light filling of infrared light-emitting device 10, dynamic continuous print video image in obtaining its vision monitoring space in real time, and pass through front end
Image processing apparatus 30 carries out real-time signal analysis to this video image, builds three dimensional local information model, obtains object corresponding
Spatial positional information, information and the monitoring information of object such as such as three-dimensional point cloud atlas, distance, directional image segmentation result, and
Object identification is carried out so that accuracy of identification is higher with this.At the same time it can also be carry out path rule according to above-mentioned spatial positional information
Draw, so that the intelligent machine 200 of correspondence implements automatically to make an inspection tour avoidance.
With reference to Fig. 7, the embodiment of the present invention additionally provides a kind of intelligent machine 200, including intelligent machine body and above-mentioned use
Visual pattern processing equipment 100 in overnight sight;This visual pattern processing equipment embeds in intelligent machine body;
Intelligent machine 200, monitors the 3 D monitoring information of object in its vision monitoring space, and three-dimensional is supervised in real time
Control information stores, in order to follow-up carry out security protection identification;And/or, receive the voice messaging that user sends, voice messaging is entered
Row identifying processing, obtains the destination address of correspondence;And, the barrier integrated information in detection vision monitoring space, front, root
According to barrier integrated information, the walking path of planning purpose address, and control according to the walking path walking after planning, to implement
Avoidance.
Concrete, above-mentioned intelligent machine 200 can be to make an inspection tour intelligent machine or stationary machines under night vision;Embedding is used for
The fixing intelligent machine of the visual pattern processing equipment 100 of overnight sight, it is possible to obtain three-dimensional by this equipment under night vision
Spatial positional information and the monitoring height of personage and the dimension information etc. of object;Embed at the visual pattern of overnight sight
The intelligent machine of the mobile tour of reason equipment 100 can be maked an inspection tour with automatic obstacle-avoiding, the most also has measurement function, i.e. measures above-mentioned three
Dimension space positional information and the monitoring height of personage and the dimension information etc. of object.
A kind of intelligent machine 200 that the embodiment of the present invention provides, the prison based on infrared night video camera with prior art
Control also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, its light filling based on infrared light-emitting device, real
Time obtain in its vision monitoring space dynamically continuous print video image, and by front-end image processing means, this video image is entered
Row real-time signal analysis, builds three dimensional local information model, obtains the spatial positional information that object is corresponding, such as three-dimensional point cloud
Information and the monitoring information of object such as figure, distance, directional image segmentation result, and carry out object identification with this so that identify essence
Spend higher.At the same time it can also be carry out path planning according to above-mentioned spatial positional information, to implement automatically to make an inspection tour avoidance.
The embodiment of the present invention additionally provides a kind of visual pattern processing system for overnight sight, including: above-mentioned intelligence
Machine 200;Also include remote control terminal 300;
Remote control terminal 300, for receiving the 3 D monitoring information of intelligent machine 200 transmission and storing;And/or,
The spatial positional information of the object in threedimensional model according to intelligent machine 200 correspondence, computational intelligence machine 200 is relative to object
Distance, and according to calculate result planning intelligent machine 200 track route, and control intelligent machine 200 according to planning after
Track route travel, to implement avoidance;And/or, according to the track route after planning, the track route of intelligent machine 200 is entered
Row monitoring.
The embodiment of the present invention provide a kind of visual pattern processing system for overnight sight, with prior art in based on
The monitoring of infrared night video camera also rests on the video monitoring stage of planar, and accuracy of identification is relatively low to be compared, above-mentioned intelligence
Machine 200 can light filling based on infrared light-emitting device, obtain in real time in its vision monitoring space dynamically continuous print video image, and
By front-end image processing means 30, this video image is carried out real-time signal analysis, build three dimensional local information model, obtain
The spatial positional information that object is corresponding, information and the monitoring of object such as such as three-dimensional point cloud atlas, distance, directional image segmentation result
Information, and carry out object identification with this so that accuracy of identification is higher.At the same time it can also be carry out according to above-mentioned spatial positional information
Path planning, to implement automatically to make an inspection tour avoidance.Further, system can also control above-mentioned intelligent machine by remote control terminal 300
200 or above-mentioned intelligent machine 200 is monitored, it is possible to preferably object be identified and control intelligent machine 200
Implement avoidance.
In embodiment provided by the present invention, it should be understood that disclosed equipment and system, can be by other side
Formula realizes.Device embodiment described above is only that schematically such as, the division of described unit, the most only one are patrolled
Volume function divides, and actual can have other dividing mode when realizing, the most such as, multiple unit or assembly can in conjunction with or can
To be integrated into another system, or some features can be ignored, or does not performs.Another point, shown or discussed each other
Coupling direct-coupling or communication connection can be the INDIRECT COUPLING by some communication interfaces, device or unit or communication link
Connect, can be electrical, machinery or other form.
The described unit illustrated as separating component can be or may not be physically separate, shows as unit
The parts shown can be or may not be physical location, i.e. may be located at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected according to the actual needs to realize the mesh of the present embodiment scheme
's.
It addition, each functional unit in the embodiment that the present invention provides can be integrated in a processing unit, it is possible to
Being 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, permissible
It is stored in a computer read/write memory medium.Based on such understanding, technical scheme is the most in other words
The part contributing prior art or the part of this technical scheme can embody with the form of software product, this meter
Calculation machine software product is stored in a storage medium, including some instructions with so that a computer equipment (can be individual
People's 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-Only Memory), random access memory are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic disc or CD.
It should also be noted that similar label and letter represent similar terms, therefore, the most a certain Xiang Yi in following accompanying drawing
Individual accompanying drawing is defined, then need not it be defined further and explains in accompanying drawing subsequently, additionally, term " the
One ", " second ", " the 3rd " etc. are only used for distinguishing and describe, and it is not intended that instruction or hint relative importance.
It is last it is noted that the detailed description of the invention of embodiment described above, the only present invention, in order to the present invention to be described
Technical scheme, be not intended to limit, protection scope of the present invention is not limited thereto, although with reference to previous embodiment to this
Bright it is described in detail, it will be understood by those within the art that: any those familiar with the art
In the technical scope that the invention discloses, the technical scheme described in previous embodiment still can be modified or can be light by it
It is readily conceivable that change, or wherein portion of techniques feature is carried out equivalent;And these are revised, change or replace, do not make
The essence of appropriate technical solution departs from the spirit and scope of embodiment of the present invention technical scheme.All should contain the protection in the present invention
Within the scope of.Therefore, protection scope of the present invention should be as the criterion with described scope of the claims.
Claims (10)
1. the visual pattern processing equipment for overnight sight, it is characterised in that including: infrared light-emitting device, multi-dimensional map
As harvester and front-end image processing means;
Described infrared light-emitting device, for the vision monitoring spatial emission infrared light of described multidimensional image harvester;
Described multidimensional image harvester, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, to institute
State infrared light and be focused imaging processing, obtain dynamic continuous print video image, described video image is sent to front-end image
Processing means;
Described front-end image processing means is connected with described multidimensional image harvester, for carrying out according to described video image
Three-dimensional modeling, obtains the threedimensional model that the vision monitoring space of described multidimensional image harvester is corresponding, and according to described three-dimensional
Model obtains the monitoring information of object;The following information of described monitoring information at least object includes: kind of information, dimension information and
Spatial positional information.
Visual pattern processing equipment for overnight sight the most according to claim 1, it is characterised in that also include high-precision
Degree assistant images harvester;
Described high accuracy assistant images harvester, for gathering the infrared light of the object reflection in its vision monitoring space, right
Described infrared light is focused imaging processing, obtains dynamic continuous print high accuracy video image, by described high accuracy video image
Send to front-end image processing means;
Described front-end image processing means is additionally operable to, and carries out described threedimensional model at calibration according to described high accuracy video image
Reason, the threedimensional model after being calibrated, and from the threedimensional model after described calibration, obtain the monitoring information of described object.
Visual pattern processing equipment for overnight sight the most according to claim 2, it is characterised in that described multi-dimensional map
As harvester includes: monocular camera lens or binocular head;Described binocular head includes: first group gathers camera lens and second group and adopt
Collection camera lens;
Described monocular camera lens, the infrared light of the object reflection in its vision monitoring space of Real-time Collection, to described infrared light
It is focused imaging processing, obtains dynamic continuous print video image, described video image is sent to front with the first predeterminated frequency
End image processing apparatus;
Described first group gathers camera lens and described second group of collection camera lens, is respectively used to gather moving in corresponding vision monitoring space
State continuous print view data, all sends all of view data to front-end image processing means with described first predeterminated frequency.
Visual pattern processing equipment for overnight sight the most according to claim 3, it is characterised in that described front end figure
As processing means includes: control device, special visual processing apparatus and memorizer;
Described control device, for controlling algorithm in described special visual processing apparatus processing procedure and corresponding with described algorithm
Result;And, control the data stream of described memorizer storage and corresponding data flow circuit;
Described consecutive image data, under the control of described control device, are carried out three by described special visual processing apparatus
Dimension modeling processes, and obtains threedimensional model, and obtains the monitoring letter of object corresponding to described view data from described threedimensional model
Breath, and the barrier integrated information of correspondence is obtained according to described monitoring information;
Described memorizer, under the control of described control device, stores the dynamic of described multidimensional image harvester transmission
Continuous print view data, in high precision the high accuracy video image of assistant images harvester transmission, threedimensional model, and/or, calibration
After threedimensional model, and, described monitoring information and described barrier integrated information.
Visual pattern processing equipment for overnight sight the most according to claim 4, it is characterised in that described special regard
Feel that processing means includes:
Receiver module, for receiving two groups of video images that described binocular head sends, extracts video image described in two groups respectively
In image feature data;
Identification module, trains described image feature data for the method using the degree of depth to learn, obtains in described video image
The kind of information of object;
Stereo matching module, mates under different points of view two groups of image feature datas are carried out solid, and to matching
Two groups of image feature datas of same position are corrected processing, and obtain regarding of in same position two groups of image feature datas
Difference;
Distance size computing module, for according to described parallax and the attitude information of described binocular head, calculate current location with
The distance of object in the vision monitoring space of described binocular head;
Three-dimensional modeling module, the distance of object, structure in the vision monitoring space according to current location and described binocular head
Build threedimensional model, and obtain the spatial positional information of object in described video image according to described threedimensional model.
Visual pattern processing equipment for overnight sight the most according to claim 5, it is characterised in that described front end figure
As processing means also includes: speech recognition synthesizer;
Described speech recognition synthesizer, for receiving the voice messaging that user sends, is identified place to described voice messaging
Reason, obtains the destination address of correspondence;And, according to described destination address and described barrier integrated information, synthesis and described mesh
Suggestion voice corresponding to address, export described suggestion voice to point out.
Visual pattern processing equipment for overnight sight the most according to claim 6, it is characterised in that described front end figure
As processing means also includes: path planning subsystem;
Described path planning subsystem, for being adjusted or weight original guidance path according to described barrier integrated information
Newly-generated guidance path.
Visual pattern processing equipment for overnight sight the most according to claim 3, it is characterised in that described first group
Gather camera lens and described second group of collection camera lens all includes a monocular camera lens or multiple monocular camera lens.
9. an intelligent machine, it is characterised in that include described in intelligent machine body and any one of claim 1~8 for
The visual pattern processing equipment of overnight sight;Described visual pattern processing equipment embeds in described intelligent machine body;
Described intelligent machine, monitors the 3 D monitoring information of object in its vision monitoring space in real time, and by described three-dimensional
Monitoring information stores, in order to follow-up carry out security protection identification;And/or, receive the voice messaging that user sends, to described voice
Information is identified processing, and obtains the destination address of correspondence;And, the barrier in detection vision monitoring space, front is comprehensively believed
Breath, according to described barrier integrated information, plans the walking path of described destination address, and controls according to the walking road after planning
Walk in footpath, to implement avoidance.
10. the visual pattern processing system for overnight sight, it is characterised in that including: described in the claims 9
Intelligent machine;Also include remote control terminal;
Described remote control terminal, for receiving the 3 D monitoring information of described intelligent machine transmission and storing;And/or, root
According to the spatial positional information of the object in the threedimensional model that described intelligent machine is corresponding, calculate described intelligent machine relative to described
The distance of object, and according to calculate result plan described intelligent machine track route, and control described intelligent machine according to
Track route after planning travels, to implement avoidance;And/or, according to the row to described intelligent machine of the track route after planning
Walk route to be monitored.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610580656.4A CN106162144A (en) | 2016-07-21 | 2016-07-21 | A kind of visual pattern processing equipment, system and intelligent machine for overnight sight |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610580656.4A CN106162144A (en) | 2016-07-21 | 2016-07-21 | A kind of visual pattern processing equipment, system and intelligent machine for overnight sight |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106162144A true CN106162144A (en) | 2016-11-23 |
Family
ID=58059440
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610580656.4A Pending CN106162144A (en) | 2016-07-21 | 2016-07-21 | A kind of visual pattern processing equipment, system and intelligent machine for overnight sight |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106162144A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107545247A (en) * | 2017-08-23 | 2018-01-05 | 北京伟景智能科技有限公司 | Three-dimensional cognitive approach based on binocular identification |
CN107562050A (en) * | 2017-08-29 | 2018-01-09 | 广东工业大学 | A kind of method and system of robot environment-identification |
CN107766833A (en) * | 2017-10-30 | 2018-03-06 | 惠州市德赛西威汽车电子股份有限公司 | A kind of harvester and method of monocular vision ADAS systematic samples |
CN109040612A (en) * | 2018-08-29 | 2018-12-18 | 百度在线网络技术(北京)有限公司 | Image processing method, device, equipment and the storage medium of target object |
CN109074490A (en) * | 2018-07-06 | 2018-12-21 | 深圳前海达闼云端智能科技有限公司 | Path detection method, related device and computer readable storage medium |
CN110445832A (en) * | 2019-06-30 | 2019-11-12 | 安徽宇艳智能机械科技有限公司 | A kind of environment self-adaption regulating system of view-based access control model conduction |
CN111309032A (en) * | 2020-04-08 | 2020-06-19 | 江苏盛海智能科技有限公司 | Autonomous obstacle avoidance method and control end of unmanned vehicle |
CN111631927A (en) * | 2020-05-25 | 2020-09-08 | 浪潮金融信息技术有限公司 | Blind person aided navigation scheme based on visual information |
US11004232B2 (en) | 2019-07-12 | 2021-05-11 | Mediatek Inc. | Method and apparatus for performing object detection according to bounding box distribution that is determined by considering lens configuration |
CN113487727A (en) * | 2021-07-14 | 2021-10-08 | 广西民族大学 | Three-dimensional modeling system, device and method |
CN114088059A (en) * | 2020-07-29 | 2022-02-25 | 珠海星客合创科技有限公司 | Map information acquisition method based on intelligent street lamp and construction method of environment map |
CN116708683A (en) * | 2023-08-01 | 2023-09-05 | 文博安全科技有限公司 | Digital automatic acquisition system and method for wall painting |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102932638A (en) * | 2012-11-30 | 2013-02-13 | 天津市电视技术研究所 | 3D video monitoring method based on computer modeling |
CN103986917A (en) * | 2014-06-03 | 2014-08-13 | 中科融通物联科技无锡有限公司 | Multi-angle thermal image monitoring system |
CN104504748A (en) * | 2014-12-03 | 2015-04-08 | 中国科学院遥感与数字地球研究所 | Unmanned aerial vehicle oblique photographing infrared three-dimensional imaging system and modeling method thereof |
CN104834381A (en) * | 2015-05-15 | 2015-08-12 | 中国科学院深圳先进技术研究院 | Wearable device for sight focus positioning and sight focus positioning method |
JP2015198260A (en) * | 2014-03-31 | 2015-11-09 | アイホン株式会社 | Monitoring camera system |
CN105554470A (en) * | 2016-01-16 | 2016-05-04 | 上海图漾信息科技有限公司 | Depth data monitoring system |
-
2016
- 2016-07-21 CN CN201610580656.4A patent/CN106162144A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102932638A (en) * | 2012-11-30 | 2013-02-13 | 天津市电视技术研究所 | 3D video monitoring method based on computer modeling |
JP2015198260A (en) * | 2014-03-31 | 2015-11-09 | アイホン株式会社 | Monitoring camera system |
CN103986917A (en) * | 2014-06-03 | 2014-08-13 | 中科融通物联科技无锡有限公司 | Multi-angle thermal image monitoring system |
CN104504748A (en) * | 2014-12-03 | 2015-04-08 | 中国科学院遥感与数字地球研究所 | Unmanned aerial vehicle oblique photographing infrared three-dimensional imaging system and modeling method thereof |
CN104834381A (en) * | 2015-05-15 | 2015-08-12 | 中国科学院深圳先进技术研究院 | Wearable device for sight focus positioning and sight focus positioning method |
CN105554470A (en) * | 2016-01-16 | 2016-05-04 | 上海图漾信息科技有限公司 | Depth data monitoring system |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107545247B (en) * | 2017-08-23 | 2020-05-12 | 北京伟景智能科技有限公司 | Stereo cognition method based on binocular recognition |
CN107545247A (en) * | 2017-08-23 | 2018-01-05 | 北京伟景智能科技有限公司 | Three-dimensional cognitive approach based on binocular identification |
CN107562050B (en) * | 2017-08-29 | 2021-03-16 | 广东工业大学 | Method and system for robot to recognize environment |
CN107562050A (en) * | 2017-08-29 | 2018-01-09 | 广东工业大学 | A kind of method and system of robot environment-identification |
CN107766833A (en) * | 2017-10-30 | 2018-03-06 | 惠州市德赛西威汽车电子股份有限公司 | A kind of harvester and method of monocular vision ADAS systematic samples |
CN109074490A (en) * | 2018-07-06 | 2018-12-21 | 深圳前海达闼云端智能科技有限公司 | Path detection method, related device and computer readable storage medium |
WO2020006764A1 (en) * | 2018-07-06 | 2020-01-09 | 深圳前海达闼云端智能科技有限公司 | Path detection method, related device, and computer readable storage medium |
CN109040612A (en) * | 2018-08-29 | 2018-12-18 | 百度在线网络技术(北京)有限公司 | Image processing method, device, equipment and the storage medium of target object |
CN109040612B (en) * | 2018-08-29 | 2020-07-28 | 百度在线网络技术(北京)有限公司 | Image processing method, device and equipment of target object and storage medium |
CN110445832A (en) * | 2019-06-30 | 2019-11-12 | 安徽宇艳智能机械科技有限公司 | A kind of environment self-adaption regulating system of view-based access control model conduction |
US11004232B2 (en) | 2019-07-12 | 2021-05-11 | Mediatek Inc. | Method and apparatus for performing object detection according to bounding box distribution that is determined by considering lens configuration |
TWI738406B (en) * | 2019-07-12 | 2021-09-01 | 聯發科技股份有限公司 | Method and apparatus for object detection |
CN111309032A (en) * | 2020-04-08 | 2020-06-19 | 江苏盛海智能科技有限公司 | Autonomous obstacle avoidance method and control end of unmanned vehicle |
CN111631927A (en) * | 2020-05-25 | 2020-09-08 | 浪潮金融信息技术有限公司 | Blind person aided navigation scheme based on visual information |
CN114088059A (en) * | 2020-07-29 | 2022-02-25 | 珠海星客合创科技有限公司 | Map information acquisition method based on intelligent street lamp and construction method of environment map |
CN113487727A (en) * | 2021-07-14 | 2021-10-08 | 广西民族大学 | Three-dimensional modeling system, device and method |
CN113487727B (en) * | 2021-07-14 | 2022-09-02 | 广西民族大学 | Three-dimensional modeling system, device and method |
CN116708683A (en) * | 2023-08-01 | 2023-09-05 | 文博安全科技有限公司 | Digital automatic acquisition system and method for wall painting |
CN116708683B (en) * | 2023-08-01 | 2023-10-10 | 文博安全科技有限公司 | Digital automatic acquisition system and method for wall painting |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106162144A (en) | A kind of visual pattern processing equipment, system and intelligent machine for overnight sight | |
CN108229366B (en) | Deep learning vehicle-mounted obstacle detection method based on radar and image data fusion | |
US11204417B2 (en) | Selective attention mechanism for improved perception sensor performance in vehicular applications | |
CN106197452A (en) | A kind of visual pattern processing equipment and system | |
CN111958592B (en) | Image semantic analysis system and method for transformer substation inspection robot | |
CA2950791C (en) | Binocular visual navigation system and method based on power robot | |
CN104036488B (en) | Binocular vision-based human body posture and action research method | |
CN100369487C (en) | Object detection device, object detection server, and object detection method | |
CN110147706A (en) | The recognition methods of barrier and device, storage medium, electronic device | |
CN106647742A (en) | Moving path planning method and device | |
CN108427438A (en) | Flight environment of vehicle detection method, device, electronic equipment and storage medium | |
WO2015180021A1 (en) | Pruning robot system | |
CN107223261A (en) | Man-machine hybrid decision method and device | |
JP2005268847A (en) | Image generating apparatus, image generating method, and image generating program | |
CN105760846A (en) | Object detection and location method and system based on depth data | |
CN109547769A (en) | A kind of highway communication dynamic 3 D number scene acquisition building system and its working method | |
CN206611521U (en) | A kind of vehicle environment identifying system and omni-directional visual module based on multisensor | |
CN105608417A (en) | Traffic signal lamp detection method and device | |
CN109213138A (en) | A kind of barrier-avoiding method, apparatus and system | |
CN109035307B (en) | Set area target tracking method and system based on natural light binocular vision | |
US20220230458A1 (en) | Recognition and positioning device and information conversion device | |
EP4354401A1 (en) | Method and system of detecting obstacle elements with a visual aid device | |
CN107607939B (en) | Optical target tracking and positioning radar device based on real map and image | |
CN116416291A (en) | Electronic fence automatic generation method, real-time detection method and device | |
JP2020149186A (en) | Position attitude estimation device, learning device, mobile robot, position attitude estimation method, and learning method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161123 |
|
RJ01 | Rejection of invention patent application after publication |