CN109974853A - Based on the multispectral compound detection of bionical sensation target and tracking - Google Patents
Based on the multispectral compound detection of bionical sensation target and tracking Download PDFInfo
- Publication number
- CN109974853A CN109974853A CN201811628870.8A CN201811628870A CN109974853A CN 109974853 A CN109974853 A CN 109974853A CN 201811628870 A CN201811628870 A CN 201811628870A CN 109974853 A CN109974853 A CN 109974853A
- Authority
- CN
- China
- Prior art keywords
- target
- tracking
- multispectral
- bionical
- holder
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 150000001875 compounds Chemical class 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 title claims abstract description 31
- 230000035807 sensation Effects 0.000 title claims abstract description 14
- 230000033001 locomotion Effects 0.000 claims abstract description 33
- 230000004438 eyesight Effects 0.000 claims abstract description 23
- 230000003287 optical effect Effects 0.000 claims abstract description 12
- 230000007246 mechanism Effects 0.000 claims abstract description 5
- 230000004807 localization Effects 0.000 claims abstract description 4
- 238000004088 simulation Methods 0.000 claims abstract description 4
- 241000288906 Primates Species 0.000 claims abstract description 3
- 230000003137 locomotive effect Effects 0.000 claims abstract description 3
- 210000001508 eye Anatomy 0.000 claims description 39
- 230000000007 visual effect Effects 0.000 claims description 7
- 210000005252 bulbus oculi Anatomy 0.000 claims description 4
- 230000004044 response Effects 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 2
- 239000000284 extract Substances 0.000 abstract description 2
- 238000000034 method Methods 0.000 description 15
- 238000003384 imaging method Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 239000011664 nicotinic acid Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000000701 chemical imaging Methods 0.000 description 2
- 238000011217 control strategy Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 210000001525 retina Anatomy 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012407 engineering method Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000011089 mechanical engineering Methods 0.000 description 1
- 150000002730 mercury Chemical class 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 210000003739 neck Anatomy 0.000 description 1
- 235000001968 nicotinic acid Nutrition 0.000 description 1
- 230000004297 night vision Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/28—Investigating the spectrum
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/28—Investigating the spectrum
- G01J3/2823—Imaging spectrometer
- G01J2003/2826—Multispectral imaging, e.g. filter imaging
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
- Studio Devices (AREA)
Abstract
The invention discloses a kind of based on the multispectral compound detection of bionical sensation target and tracking, it include: simulation primate head eye coordination locomotory mechanism, build multispectral compound bionical vision system, including the macro dynamic holder of neck level-one, eye dual-ratio holder, thermal infrared camera, Visible Light Camera;Set the macro dynamic holder movement threshold η of neck level-oneNWith eye dual-ratio holder movement threshold ηE, the optical axis rotational angle β of multispectral compound bionical vision system is introduced, according to the stage multispectral compound bionical vision system of control of optical axis rotational angle β, realizes head eye coordination motion control;Coarse localization and accurate positioning are carried out respectively using the feature that infrared image and visible light respectively extract moving target, the state of new frame moving target is estimated using KCF tracking, and decision condition of the ratio peak value as tracking result confidence level is introduced, realize the real-time online detection and tracking of moving target.
Description
Technical field
The present invention relates to a kind of based on the multispectral compound detection of bionical sensation target and tracking, is related to bionical machine
Device people's technical field, can be used for service robot technology and field of intelligent monitoring.
Background technique
In recent years, it is constantly merged with bionics with machine vision, bionical vision robotics are just with unprecedented
Speed advances and realizes breakthrough, uses for reference opthalmic optics' system imaging mechanism and constructs intelligent bionic eye, realizes moving target
Real-time detection and tracking are the focus on research direction of bionical vision robotics, in intelligent video monitoring and service robot
The related fieldss such as system have a wide range of applications.All the time, lot of domestic and foreign scholar is respectively from different perspectives and direction
Many explorations and research have been carried out to bionical vision robot's Target Tracking System.
Cone cell and rod-shaped in human eye retina is simulated by Northwestern Univ USA (Northwestern University)
Cell distribution situation combines multiple institutes and has made a hemispherical retina detector, for the first time realize variable resolution curved surface at
Picture.The human simulations eye image collection process such as US Army night vision electronic sensor research institute Vizgaitis devise two waveband
The target following task of round-the-clock big visual field may be implemented in the infrared optics image capturing system of more visual fields.The king of Zhejiang University
Xuan Yin et al. is proposed based on flexible drive parallel connected bionic human eyes structure, is sensed using bionic pneumatic muscle actuator, CMOS
Device and three dimension acceleration sensor have made Prosthetic Hand visual perception system.Control Science and Engineering institute, Hebei University of Technology
Yang Fucai et al. proposes to realize steady infrared object tracking based on sparse coding histogram feature and disturbance perception mould
The infrared object tracking method of type.Portugal Universidade de Coimbra technical college Henriques et al. is in order to improve track algorithm
Calculating speed, propose a kind of nuclear phase and close filter tracks algorithm, acquired using circular matrix in target peripheral region positive and negative
Sample can satisfy target tracking algorism requirement of real-time using ridge regression training objective detector.Shandong University's electrician
Journey institute Li Hao etc. simulates the motion mode of human eye ball, develops bionical ophthalmically acceptable four-degree-of-freedom stepper motor driving control system,
The basic exercise function of bionical eyeball may be implemented.School of Mechanical Engineering, Dalian University of Technology Liu Yi et al. apish head,
Eye, neck motion mode devise Prosthetic Hand and people's neck vision system, realize robot static object location tasks.However, mesh
The research of preceding bionical mobile robot target detection tracking is mostly to obtain target using left and right cameras based on engineering method
Single channel image is respectively processed, and it is shared not carry out effective image information, and right and left eyes and neck lack effectively
Coordination linkage mechanism is not able to achieve maximally utilizing for image information.
In consideration of it, the advantage and disadvantage herein for visible images and infrared image in target detection and tracking application, it will
The validity feature information progress that the two is extracted is compound, and uses head eye coordination Motion Control Strategies, realizes to moving target
Real-time online detection and tracking purpose.
Summary of the invention
The present invention is difficult to for single-range imaging sensor in illumination and the changed complex dynamic environment of temperature
The lower accuracy for guaranteeing target detection and tracking discloses one kind in conjunction with visible images and infrared image itself imaging characteristic
Based on the multispectral compound detection of bionical sensation target and tracking.This method simulates human eye cone cell and rhabdocyte
Imaging characteristic, the target effective information progress that infrared and visible images are detected is compound, and using head eye coordination movement control
System strategy, realizes the real-time online detection and tracking to moving target.
For achieving the above object, the technical scheme is that it is a kind of based on multispectral compound bionical vision mesh
Mark detection and tracking, comprising the following steps:
Step 1, simulation primate head eye coordination locomotory mechanism, build multispectral compound bionical vision system, wrap
Include the macro dynamic holder of neck level-one, eye dual-ratio holder, thermal infrared camera, Visible Light Camera, head eye coordination control system;
Step 2, the setting macro dynamic holder movement threshold η of neck level-oneNWith eye dual-ratio holder movement threshold ηE, introduce
The optical axis rotational angle β of multispectral compound bionical vision system is multispectral multiple according to the stage control of optical axis rotational angle β
The bionical vision system closed realizes head eye coordination motion control;
Step 3, the position using thermal infrared camera coarse localization target in image coordinate system, when target is sat in image
When marking the central area of system, using the exact position of Visible Light Camera positioning target, and pass through KCF tracking estimation new one
The motion state of frame target;Otherwise, the shooting angle of thermal infrared camera is adjusted by head eye coordination control system, repeats step
3,;
Step 4 introduces ratio peak value, determines the confidence level of the tracking result of the motion state of new frame target,
The motion state that confidence level meets target in the new frame image of decision condition is fed back into head eye coordination control system, is completed
Otherwise the real-time tracking of moving target repeats step 3.
Further, the macro dynamic holder of neck level-one, eye dual-ratio holder are 4DOF cloud in the step 1
Platform.
Further, the stage control of the multispectral compound bionical vision system of the step 2 specifically includes three ranks
Section:
1)0<β<ηN, blinkpunkt is completed by the macro dynamic holder of neck level-one at this time and shifts task, rotational angle ηN=β, is not required to
Eye dual-ratio holder is wanted to participate in movement;
2)ηN<β<ηE, the macro dynamic holder of neck level-one and eye dual-ratio holder are needed at this time while movement is to complete to infuse
Viewpoint shifts task, the macro dynamic cloud platform rotation angle η of neck level-oneN, eye dual-ratio cloud platform rotation angle ηE=β-ηN;
3)ηE< β needs the macro dynamic holder of neck level-one and eye dual-ratio holder at this time while movement is to complete to watch attentively
Point transfer task, neck rotation angle ηN=β-ηE, Rotation of eyeball angle ηE。
Further, the step 3 specifically includes:
Big visual field low-resolution image is acquired using thermal infrared camera, CENTRIST feature is extracted, utilizes trained line
Property SVM classifier roughly determine position of the target in image coordinate system;
It is that △ x calculates bionical vision using the margin of error of known camera focus f and target's center's point slip chart principal point
The system optical axis needs rotational angle β,
Head eye coordination control system is maintained at target according to the stage driving two-stage cloud platform rotation of optical axis rotational angle β
The central area of infrared image;
Small field of view high-definition picture is acquired using Visible Light Camera, extracts the HOG feature of current frame image, determines fortune
The exact position of moving-target estimates the state of new frame moving target using KCF tracking;
Further, the step 4 specifically includes:
Introducing ratio peak value s determines that s value is got over to the confidence level of the tracking result of the motion state of new frame target
Show that tracking result reliability is higher greatly,
In formula, f (z) is the response of new frame Image Classifier, and ф is the response distribution map centered on maximum value
20% region, μфAnd σфMean value and standard deviation in respectively region ф;
Ratio peak value is standardized:
O in formula, after standardizationtValue is between [0.1], stFor the ratio peak value of t frame;
Given threshold θ is that classifier relocates decision condition, i.e., as detection of classifier peak value OtLess than reorientation decision threshold
When value θ, illustrate that tracking result is unreliable, needs to relocate target using infrared image CENTRIST feature;When classifier is examined
Survey peak value OtWhen greater than reorientation decision threshold θ, illustrates that tracking result is effective, the HOG feature of detection target is extracted, from detection
To target around choose positive negative sample, training set is added and simultaneously updates SVM classifier.
The motion state that confidence level meets target in the new frame image of decision condition is fed back into head eye coordination control system
System, completes the real-time tracking of moving target.
The beneficial effects of the present invention are: method of the invention is by collected infrared and visible light multi-modality images information
Carry out it is compound, and using head eye coordination Motion Control Strategies carry out object detecting and tracking, can solve single band image object
There is detecting and tracking failure under the complex situations such as the visual field out, partial occlusion and similar purpose interference in detecting and tracking method
Problem.Experiments verify that compared to conventional one-channel tracking image target method, the target following side based on multispectral image
13.6% and 7.8% has been respectively increased in method success rate, has stronger robustness.
Detailed description of the invention
Fig. 1 is multispectral compound bionical vision system figure;
Fig. 2 is double-view field multispectral imaging optical system diagram;
Fig. 3 is the accurate positioning figure of target under small field of view;
Fig. 4 is the coarse localization figure of target under big visual field.
Specific embodiment
Below in conjunction with attached drawing, technical scheme in the embodiment of the invention is clearly and completely described.
In order to verify the present invention is based on the accuracy of multispectral compound bionical sensation target detection and tracking and
The validity of active vision control theory, is made a concrete analysis of by following tests.
Build multispectral compound bionical vision system as shown in Figure 1, comprising: model Raven-640-Analog
Belgian Xenics thermal infrared camera, the colored industrial phase of Imax Corp., Daheng Mercury series of model MER-310-12UC
Machine, the free bionical Vision Table of the two-stage 4 independently built, after the identification of metering institute, China, rotating left and right precision is ± 2.4 ',
Upper and lower pitching precision is ± 0.6 ', using STM32F4 development board as horizontal stage electric machine control module, matches company, Sentos ZYNQ-
Acquisition and processing module of the 7000 series ZedBoard embedded boards as infrared image, processor host frequency 766MHz,
It is run in Ubuntu14.04 system.Acquisition and processing module of the master control PC machine as visible images, processor intel
(R) Core(TM)i7-4790K [email protected] ° of neck movement threshold value η N=, eye motion threshold value η are set in experiment
E=45 °, relocate decision threshold θ=0.55.The double-view field multispectral imaging of the multispectral compound bionical vision system
Optical system diagram is as shown in Figure 2.
In order to be quantitatively evaluated to based on multispectral compound bionical sensation target detection with tracking, with center
The quantitative assessing index of location error and success rate figure as tracking accuracy.Center error of the invention is defined as tracking
Target shows the average Euclidean distance of box center and picture centre origin, and success rate figure is defined as in detection mesh
It marks in video, the frame number that center error meets threshold requirement accounts for the percentage of all frame numbers of video.It for convenience, will be more
The compound bionical sensation target detection of spectrum and tracking and visible light or thermal infrared single channel target detection tracking method into
Visible images and infrared image are adjusted to 640 × 480pixels size in experiment, it is seen that light figure by row comparative analysis
As selecting HOG feature to carry out target detection, as shown in figure 3, thermal infrared images selects CENTRIST feature to carry out target detection,
As shown in figure 4, choosing 50pixels as error threshold, respectively in lab, tri- scenes of corridor and meeting room
Under be compared analysis, experimental result is as shown in table 1:
1 visible images of table and infrared image target tracking accuracy quantitative comparison
As shown in Table 1, it compared to visible images or infrared image single channel target detection tracking method is based on, is based on
13.6% and 7.8% has been respectively increased in multispectral compound bionical sensation target detecting and tracking method success rate.Visible images
Detecting and tracking unsuccessfully mainly appears on partial occlusion and out in the case of the visual field, and infrared image detecting and tracking unsuccessfully mainly appears on
In the case that similar purpose interferes (such as display or radiator) and dimensional variation.Being compared by three kinds of method operation times can be with
Find out, is that ratio is based on based on multispectral compound detecting and tracking method due to reducing the search range of visible images
Thermal infrared images detecting and tracking method slightly has time-consuming, but is considerably less than based on the detecting and tracking method of visible images
Evaluation time can satisfy target following real-time demand substantially.
Described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Based on the present invention
In embodiment, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, belongs to the scope of the present invention.
Claims (5)
1. a kind of based on the multispectral compound detection of bionical sensation target and tracking, which comprises the following steps:
Step 1, simulation primate head eye coordination locomotory mechanism, build multispectral compound bionical vision system, including neck
The macro dynamic holder of departmental level, eye dual-ratio holder, thermal infrared camera, Visible Light Camera, head eye coordination control system;
Step 2, the setting macro dynamic holder movement threshold η of neck level-oneNWith eye dual-ratio holder movement threshold ηE, introduce mostly light
The optical axis rotational angle β for composing compound bionical vision system is imitated according to the stage control of optical axis rotational angle β is multispectral compound
Raw vision system, realizes head eye coordination motion control;
Step 3, the position using thermal infrared camera coarse localization target in image coordinate system, when target is in image coordinate system
When central area, using the exact position of Visible Light Camera positioning target, and new frame target is estimated by KCF tracking
Motion state;Otherwise, the shooting angle of thermal infrared camera is adjusted by head eye coordination control system, repeats step 3,;
Step 4 introduces ratio peak value, determines the confidence level of the tracking result of the motion state of new frame target, can
The motion state that reliability meets target in the new frame image of decision condition feeds back to head eye coordination control system, completes movement mesh
Otherwise target real-time tracking repeats step 3.
2. as described in claim 1 a kind of based on the multispectral compound detection of bionical sensation target and tracking, feature
It is, the macro dynamic holder of neck level-one, eye dual-ratio holder are 4DOF holder in the step 1.
3. as described in claim 1 a kind of based on the multispectral compound detection of bionical sensation target and tracking, feature
It is, the stage control of the multispectral compound bionical vision system of step 2 specifically includes three phases:
1)0<β<ηN, blinkpunkt is completed by the macro dynamic holder of neck level-one at this time and shifts task, rotational angle ηN=β, does not need eye
Dual-ratio holder participates in movement;
2)ηN<β<ηE, the macro dynamic holder of neck level-one and eye dual-ratio holder are needed at this time while movement turns to complete blinkpunkt
Shifting task, the macro dynamic cloud platform rotation angle η of neck level-oneN, eye dual-ratio cloud platform rotation angle ηE=β-ηN;
3)ηE< β needs the macro dynamic holder of neck level-one and eye dual-ratio holder while movement at this time to complete blinkpunkt transfer
Task, neck rotation angle ηN=β-ηE, Rotation of eyeball angle ηE。
4. as described in claim 1 a kind of based on the multispectral compound detection of bionical sensation target and tracking, feature
It is, the step 3 specifically includes:
Big visual field low-resolution image is acquired using thermal infrared camera, CENTRIST feature is extracted, utilizes trained Linear SVM
Classifier determines roughly position of the target in image coordinate system;
It is that △ x calculates bionical vision system using the margin of error of known camera focus f and target's center's point slip chart principal point
The optical axis needs rotational angle β,
Head eye coordination control system makes target be maintained at infrared according to the stage driving two-stage cloud platform rotation of optical axis rotational angle β
The central area of image;
Small field of view high-definition picture is acquired using Visible Light Camera, the HOG feature of current frame image is extracted, determines moving target
Exact position, the state of new frame moving target is estimated using KCF tracking.
5. as described in claim 1 a kind of based on the multispectral compound detection of bionical sensation target and tracking, feature
It is, the step 4 specifically includes:
Introducing ratio peak value s, determines the confidence level of the tracking result of the motion state of new frame target, the bigger table of s value
Bright tracking result reliability is higher,
In formula, f (z) is the response of new frame Image Classifier, and ф is 20% of the response distribution map centered on maximum value
Region, μфAnd σфMean value and standard deviation in respectively region ф;
Ratio peak value is standardized:
O in formula, after standardizationtValue is between [0.1], stFor the ratio peak value of t frame;
Given threshold θ is that classifier relocates decision condition, i.e., as detection of classifier peak value OtWhen less than reorientation decision threshold θ,
Illustrate that tracking result is unreliable, needs to relocate target using infrared image CENTRIST feature;When detection of classifier peak value
OtWhen greater than reorientation decision threshold θ, illustrates that tracking result is effective, the HOG feature of detection target is extracted, from the target detected
Surrounding chooses positive negative sample, and training set is added and simultaneously updates SVM classifier.
The motion state that confidence level meets target in the new frame image of decision condition is fed back into head eye coordination control system, it is complete
At the real-time tracking of moving target.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811628870.8A CN109974853B (en) | 2018-12-28 | 2018-12-28 | Bionic visual target detection and tracking method based on multispectral composition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811628870.8A CN109974853B (en) | 2018-12-28 | 2018-12-28 | Bionic visual target detection and tracking method based on multispectral composition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109974853A true CN109974853A (en) | 2019-07-05 |
CN109974853B CN109974853B (en) | 2020-12-04 |
Family
ID=67076545
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811628870.8A Active CN109974853B (en) | 2018-12-28 | 2018-12-28 | Bionic visual target detection and tracking method based on multispectral composition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109974853B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111292376A (en) * | 2020-02-13 | 2020-06-16 | 北京理工大学 | Visual target tracking method of bionic retina |
CN112936348A (en) * | 2021-03-04 | 2021-06-11 | 哈尔滨工业大学 | Sensor integration device for sensing RGB-D-T information of intelligent robot |
CN113033891A (en) * | 2021-03-19 | 2021-06-25 | 河北水熠木丰工程技术有限责任公司 | High pier bridge health monitoring method and device |
CN115278055A (en) * | 2022-06-24 | 2022-11-01 | 维沃移动通信有限公司 | Shooting method, shooting device and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903279A (en) * | 2014-03-21 | 2014-07-02 | 上海大学 | Parallel tracking system and method based on bionic binocular vision onboard platform |
CN106646694A (en) * | 2016-11-02 | 2017-05-10 | 北京信息科技大学 | Bionic vision imaging technology based on visible light and near-infrared rays |
CN108616744A (en) * | 2017-01-23 | 2018-10-02 | 上海爱观视觉科技有限公司 | A kind of bionical binocular vision calibration system and calibration method |
-
2018
- 2018-12-28 CN CN201811628870.8A patent/CN109974853B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103903279A (en) * | 2014-03-21 | 2014-07-02 | 上海大学 | Parallel tracking system and method based on bionic binocular vision onboard platform |
CN106646694A (en) * | 2016-11-02 | 2017-05-10 | 北京信息科技大学 | Bionic vision imaging technology based on visible light and near-infrared rays |
CN108616744A (en) * | 2017-01-23 | 2018-10-02 | 上海爱观视觉科技有限公司 | A kind of bionical binocular vision calibration system and calibration method |
Non-Patent Citations (2)
Title |
---|
FRANCESCHINI NICOLAS ET AL.: "Neuromimetic Robots inspired by Insect Vision", 《MINING SMARTNESS FROM NATURE》 * |
宋言明: "基于仿人眼自适应调节的多光谱视觉图像处理方法", 《红外与激光工程》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111292376A (en) * | 2020-02-13 | 2020-06-16 | 北京理工大学 | Visual target tracking method of bionic retina |
CN111292376B (en) * | 2020-02-13 | 2023-06-30 | 北京理工大学 | Visual target tracking method of bionic retina |
CN112936348A (en) * | 2021-03-04 | 2021-06-11 | 哈尔滨工业大学 | Sensor integration device for sensing RGB-D-T information of intelligent robot |
CN113033891A (en) * | 2021-03-19 | 2021-06-25 | 河北水熠木丰工程技术有限责任公司 | High pier bridge health monitoring method and device |
CN113033891B (en) * | 2021-03-19 | 2022-01-18 | 河北水熠木丰工程技术有限责任公司 | High pier bridge health monitoring method and device |
CN115278055A (en) * | 2022-06-24 | 2022-11-01 | 维沃移动通信有限公司 | Shooting method, shooting device and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109974853B (en) | 2020-12-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109308693A (en) | By the target detection and pose measurement list binocular vision system of a ptz camera building | |
CN107909061B (en) | Head posture tracking device and method based on incomplete features | |
CN107423729B (en) | Remote brain-like three-dimensional gait recognition system oriented to complex visual scene and implementation method | |
CN106339006B (en) | A kind of method for tracking target and device of aircraft | |
CN107105147B (en) | A kind of bionical super-resolution imaging sensor and imaging method | |
US8005264B2 (en) | Method of automatically detecting and tracking successive frames in a region of interesting by an electronic imaging device | |
CN108205658A (en) | Detection of obstacles early warning system based on the fusion of single binocular vision | |
CN109974853A (en) | Based on the multispectral compound detection of bionical sensation target and tracking | |
CN110268444A (en) | A kind of number of people posture tracing system for transcranial magnetic stimulation diagnosis and treatment | |
CN108731587A (en) | A kind of the unmanned plane dynamic target tracking and localization method of view-based access control model | |
CN108932477A (en) | A kind of crusing robot charging house vision positioning method | |
CN110458025A (en) | A kind of personal identification and localization method based on binocular camera | |
CN110334701A (en) | Collecting method based on deep learning and multi-vision visual under the twin environment of number | |
CN105631859A (en) | Three-degree of freedom bionic stereo vision system | |
CN114119739A (en) | Binocular vision-based hand key point space coordinate acquisition method | |
CN110065075A (en) | A kind of spatial cell robot external status cognitive method of view-based access control model | |
Cong et al. | Weakly supervised 3d multi-person pose estimation for large-scale scenes based on monocular camera and single lidar | |
CN107564037A (en) | A kind of multirobot detection and tracking based on local feature | |
Qi et al. | Alpine skiing tracking method based on deep learning and correlation filter | |
CN117156298A (en) | Low-speed small target perception method and system based on artificial insect composite bionics | |
CN115797397B (en) | Method and system for all-weather autonomous following of robot by target personnel | |
CN207529394U (en) | A kind of remote class brain three-dimensional gait identifying system towards under complicated visual scene | |
WO2023103145A1 (en) | Head pose truth value acquisition method, apparatus and device, and storage medium | |
CN113670268B (en) | Binocular vision-based unmanned aerial vehicle and electric power tower distance measurement method | |
CN113361333B (en) | Non-contact type riding motion state monitoring method and 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |