CN109465809A - A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition - Google Patents
A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition Download PDFInfo
- Publication number
- CN109465809A CN109465809A CN201811542036.7A CN201811542036A CN109465809A CN 109465809 A CN109465809 A CN 109465809A CN 201811542036 A CN201811542036 A CN 201811542036A CN 109465809 A CN109465809 A CN 109465809A
- Authority
- CN
- China
- Prior art keywords
- image
- rubbish
- dimensional
- target
- recognition
- 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
- 239000010813 municipal solid waste Substances 0.000 title claims abstract description 84
- 238000012545 processing Methods 0.000 claims abstract description 35
- 238000012958 reprocessing Methods 0.000 claims abstract description 5
- 238000001914 filtration Methods 0.000 claims description 21
- 230000006870 function Effects 0.000 claims description 13
- 238000000034 method Methods 0.000 claims description 12
- 238000003709 image segmentation Methods 0.000 claims description 8
- 230000008569 process Effects 0.000 claims description 7
- 230000007246 mechanism Effects 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000012512 characterization method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 2
- 238000006116 polymerization reaction Methods 0.000 claims description 2
- 238000012805 post-processing Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000013500 data storage Methods 0.000 abstract 1
- 230000007613 environmental effect Effects 0.000 abstract 1
- 230000008685 targeting Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1679—Programme controls characterised by the tasks executed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/08—Indexing scheme for image data processing or generation, in general involving all processing steps from image acquisition to 3D model generation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Software Systems (AREA)
- Automation & Control Theory (AREA)
- Geometry (AREA)
- Artificial Intelligence (AREA)
- Computer Graphics (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to robot technology design fields, are related to a kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition.It is characterized in that including central processing unit, dsp processor, CCD camera, Video Decoder, image data memory, robot sorting structure in hardware design.The acquisition of rubbish target image is carried out by binocular CCD camera, it decodes and passes to dsp processor and do pretreatment and data storage, send rubbish image reprocessing and binocular stereo vision identification positioning that central processing unit carries out core the most to again, it obtains rubbish target three-dimensional information and compares preset model parameter, dsp processor is finally transmitted information to by central processing unit, DSP sends the classification that the sort operation realization rubbish of robot is completed in instruction.Designing system realization rubbish targeting accuracy positioning of the present invention and three-dimensional information obtain and obtain garbage classification data, play great meaning to city garbage specification treatment effeciency, raising social benefit and environmental protection aspect is accelerated.
Description
Technical field
The invention belongs to field of garbage disposal, spy is related to a kind of intelligent garbage based on binocular stereo vision fixation and recognition point
Class robot.
Background technique
As urban life increasingly automates, intelligence, intelligence is also done step-by-step in the classification processing of municipal refuse.It is right
Garbage reclamation recycles, and garbage as resource is made to have great importance.Present China's garbage classification processing is most of all to lean on manpower,
It handled with our desired intelligent classifications, realize that waste resources recycle also very big gap, so we are badly in need of in this way
A kind of Intelligent garbage classification robot for realizing identification, positioning.
Currently, garbage classification machine people mainly realizes its identification, positioning and sort operation, the accurate positioning of target is known
It is not particularly important.In existing patent and technical research, such as about based on image recognition technology vision positioning system and
Garbage classification machine people with infrared sensor, achieves more apparent progress in terms of intelligent recognition and positioning, but
In practical, the processing and sensor for only relying on two dimensional image are extremely difficult to the capture of target precisely to identify the requirement of positioning, and
The crawl sorting of robot is influenced afterwards.
Summary of the invention
In view of the above shortcomings, the present invention provide a kind of intelligence, high efficiency, pinpoint accuracy based on binocular tri-dimensional
Feel the garbage classification machine people of fixation and recognition.
The present invention adopts the following technical scheme:
A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition, including central processing unit, DSP processing
Device, two CCD cameras, Video Decoder, image data memory and mechanism for sorting, the output end of two CCD cameras respectively with
The input terminal of Video Decoder connects, and the output end of Video Decoder and the input terminal of dsp processor connect, central processing unit,
Image data memory and mechanism for sorting are connect with the output end of dsp processor respectively, the output end of central processing unit again with DSP
The input terminal of processor connects.
Robot is as follows to the classification process of rubbish:
1) by the image of the two CCD camera acquisition rubbish targets in left and right, and collected digital image information is passed into video
Decoder；
2) pretreatment of dsp processor completion digital picture is passed to after Video Decoder decodes and passes to image data deposits
Reservoir is stored；
3) by dsp processor, treated that image information sends to that central processing unit carries out digital picture reprocessing stands with binocular again
Body vision identifies and positions, and completes to obtain the tool of rubbish target to the 3D vision matching of rubbish target and Studies About Reconstruction of Three-Dimension Information
Body three-dimensional dimension and three dimensional local information；
4) the final fixation and recognition information of rubbish target is finally transferred to dsp processor by central processing unit, DSP sends instruction
The classification of rubbish is completed to robot sorting manipulator arm.
The acquisition and pretreatment of rubbish target image and identifying and positioning for rubbish target three-dimensional information constitute rubbish mesh
Binocular Stereo Vision System is marked, rubbish target Binocular Stereo Vision System is the 3D vision based on cost rubbish objective function
Match, central processing unit constructs a three-dimensional cost collection according to cost function, passes through the local optimum and three-dimensional space to cost collection
The polymerization of parallax is realized in search, and is carried out three-dimensional filtering and operated to obtain the parallax information of two figures, after obtaining parallax, to disparity map
It is post-processed to obtain fine disparity map.
Dsp processor to rubbish target image pretreatment refer to first to rubbish target image carry out noise filtering after, then
Carry out image segmentation and edge detection.
The three-dimensional dimension and three dimensional local information of rubbish target are acquired under different location by two CCD cameras in left and right
One triangle of image and tested rubbish target configuration, according to the positional relationship and the rubbish of the CCD camera of known left and right two
The parallax that rubbish target point is formed in the digital picture of left and right, then acquisition is calculated by the identification of central processing unit.
The three-dimensional reconstruction of the rubbish target is the two images using same rubbish target different perspectives according to three-dimensional
The disparity map that vision matching obtains restores three-dimensional information according to triangle principle, has just obtained accurately three-dimensional dimension and three-dimensional seat
Mark, and obtained three-dimensional information is compared matching with the preset model data in deposit central processing unit, obtain classification knot
Fruit.
Noise filtering is using nonlinear filtering technique median filtering algorithm, the filtering algorithm are as follows:
Wherein N indicates that filter window includes the sum of pixel,Indicate the gray value of filter window center pixel,The row of expression
The gray value of i-th of pixel in sequence vector,It indicatesWeighting coefficient.
Image segmentation and edge detection are that Roberts operator is used to detect gray scale or image by basis of characterization of color
Middle color change position outstanding, the detection at image segmentation and edge is reached according to the size of variable gradient, Roberts operator
Calculation formula are as follows:
Due to digital picture one two-dimensional discrete type manifold of acquisition,Indicate a pixel in figureGradient
Sizes values,Respectively indicate the gradient of diagonal line both direction.
The 3D vision matching of rubbish target is an optimization object function to be constructed, to obtain not based on image pixel
With the corresponding relationship of rubbish target point same in multi-view image, the Stereo Matching Algorithm being polymerize by cost, construction is based on pixel
Following cost objective function:
Absolute error:
Square error:
WhereinIt is position in left imageThe image pixel intensities of point,It is position in right image
Image pixel intensities a little are set,It is parallax,It is corresponding matching cost function respectively.
The invention has the advantages that passing through the design of hardware and algorithm in the entire binocular stereo vision identifying system of patent
Reach the accurately fixation and recognition to target, has the characteristics that high efficiency, pinpoint accuracy, intelligentized.Defect is believed in number
It ceases in treatment process, due to obtain accurately target information and carry out three-dimensional reconstruction algorithm process design is complicated, also
Constantly system identification location algorithm is optimized to cope with different complex situations.
Detailed description of the invention
Fig. 1 is target fixation and recognition hardware system basic framework of the invention；
Fig. 2 is the schematic diagram of binocular stereo vision robot target location technology；
Fig. 3 is Binocular Stereo Vision System algorithm process block diagram.
Specific embodiment
It is described further below with reference to specific embodiment of the structure chart to the invention, following specific embodiments are to use
It is of the invention to illustrate, any restrictions are not done to the present invention.
With reference to the accompanying drawing to the present invention is based on the Intelligent garbage classification robot of binocular stereo vision fixation and recognition do into
One step explanation.
With reference to the accompanying drawings 1, the garbage classification machine based on binocular stereo vision fixation and recognition of one embodiment of the present of invention
People is mainly sorted by central processing unit, dsp processor, two CCD cameras, Video Decoder, image data memory, robot
Structure and other assemblies composition.
In robot treatment process, the acquisition of target image is carried out by CCD camera first, in order to obtain space
The base image of scene, is decoded through Video Decoder, and passes to dsp processor and convert analog signals into digital signal
It is stored with the preliminary treatment to image and by image data memory, then sends central processing unit to again and counted
Word image reprocessing is identified and positioned with binocular stereo vision, is calculated the specific three-dimensional dimension of target and location information and is carried out
Preset model matching, finally transmits information to dsp processor by central processing unit again, and DSP sends instructions to robot sorting
Structure realizes that the sort operation of robot completes the classification of rubbish.
Binocular stereo vision garbage classification system in the present invention is equivalent to the eyes of people, in identification rubbish object target
When visually there is otherness, this visual difference is exactly based on to obtain target three-dimensional information, to compare point
Class processing.
In rubbish Image Acquisition, the image and measured target that two CCD cameras in left and right acquire under different location are constituted
One triangle, according to the parallax that the positional relationship of known left and right camera and the target point are formed in the digital picture of left and right,
The three-dimensional information of the available rubbish target is calculated by the identification of central processing unit algorithm.
In rubbish image procossing, first by filtering algorithm to image carry out noise filtering, after to noise processed after
Image carry out the detection of image segmentation and edge, complete pretreatment to image.
In digital picture reprocessing, central processing unit carries out 3D vision matching, passes through target each in two dimensional image
Relationship between element completes the calculating of parallax.And the three-dimensional information of target is established in stereo visual system, thus more precisely
Acquisition three-dimensional dimension and three-dimensional coordinate.
Executed in treatment process in processor, using DSP embedded processor to the accuracy of target identification processing compared with
Height, the required time is shorter, can satisfy high-precision and high efficiency quickly handles the requirement of image and feature extraction.
During robot sort operation, mechanical arm is manipulated by the command information obtained by processor to realize rubbish
Classify.
The schematic diagram of 2 Technology for Target Location with reference to the accompanying drawings, it is to pass through that Binocular Stereo Vision System institute, which carries out positioning to target,
Parallax is come what is completed, and two imagings in left and right are completed by CCD camera in figure, is equivalent to the binocular of people.Utilize left and right camera imaging
Visual difference to carry out target accurate coordinate positioning: rubbish target constitutes triangle with two cameras, it is known that two cameras are not
Same location information can be with according to the image of acquisition and by central processing unit and dsp processor to parallax data information processing
Obtain the specific three-dimensional dimension and three dimensional local information data of rubbish target.
With reference to the accompanying drawings in 3 Binocular Stereo Vision Systems, nonlinear filtering skill is used to the noise filtering of rubbish target image
Art median filtering algorithm.The algorithm overcomes the shortcomings that linear filtering algorithm blurred picture carefully accords with information to a certain extent, and
And to filtering out impulse disturbances noise and image scanning noise is maximally efficient.The statistics of image is not needed during actual operation
Characteristic, therefore operation time is greatly saved, while also bringing many convenience.The filtering algorithm are as follows:
(1)
(2)
Wherein N indicates that filter window includes the sum of pixel,Indicate the gray value of filter window center pixel,The row of expression
The gray value of i-th of pixel in sequence vector,It indicatesWeighting coefficient.After being filtered out to picture noise, then to acquisition
Rubbish image, which carries out image segmentation and edge detection, effectively to be identified target, use Roberts operator with color
Color change position outstanding in gray scale or image is detected for basis of characterization, and image point is reached according to the size of variable gradient
Cut the detection with edge, the calculation formula of Roberts operator are as follows:
Due to digital picture one two-dimensional discrete type manifold of acquisition,Indicate a pixel in figureGradient
Sizes values,Respectively indicate the gradient of diagonal line both direction.
After completing to the pretreatment of image, the storage of data is carried out by dsp processor, and sends central processing unit progress to
Identification, matching positioning and the reconstruction of final three-dimensional information.3D vision matching is also known as Stereo matching, the original substantially of Stereo matching
Reason is an optimization object function to be constructed, to obtain the correspondence of same target point in different perspectives image based on image pixel
Relationship, and parallax is calculated based on matched result, the result of Stereo matching directly determines the precision of binocular stereo vision.
Binocular stereo vision processing searches for mesh using epipolar geom etry constraint and similarity constraint, Stereo matching from large space
Search space is largely reduced during target, improves search efficiency.Epipolar geom etry constraint such as attached drawing 2, defines left figure
As the upper match point put on corresponding image right must be point-blank.Similarity constraint is according to similar on left images
Attribute, such as color, intensity profile, image moment information etc. carry out matching constraint.
3D vision matching constructs cost target letter pixel-based using the Stereo Matching Algorithm polymerizeing based on cost
Number.
Absolute error:(3)
Square error:(4)
WhereinIt is position in left imageThe image pixel intensities of point,It is position in right image
Image pixel intensities a little are set,It is parallax,It is corresponding matching cost function respectively.
The 3D vision matching based on cost objective function is carried out afterwards, and all possible parallax layers are calculated by central processing unit
The matching cost of pixel on secondary constructs a three-dimensional cost collection as disparity space image.Using local optimization methods in structure
It is scanned in the three-dimensional space made and to carrying out filtering operation on three-dimensional cost collection, the parallax after being polymerize.Obtain parallax
Later, disparity map is post-processed to obtain fine disparity map.
Three-dimensional reconstruction based on binocular stereo vision is to be regarded using the two images of same target different perspectives according to three-dimensional
Feel the disparity map that matching obtains, three-dimensional information is restored according to triangle principle, has just obtained accurately three-dimensional dimension and three-dimensional coordinate.
And obtained three-dimensional information is compared matching with the preset model data in deposit central processing unit, obtain classification results.
Central processing unit passes information to dsp processor and sends instructions to robot mechanism for sorting by it, and operating robotic arm is completed
Final garbage classification.Wherein, routine techniques can be used in robot mechanism for sorting, and sorting arm is cascaded structure and carries servo
The driving of motor realization arm.
Claims (9)
1. a kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition, it is characterised in that including central processing
Device, dsp processor, two CCD cameras, Video Decoder, image data memory and mechanism for sorting, two CCD cameras it is defeated
Outlet is connect with the input terminal of Video Decoder respectively, and the output end of Video Decoder and the input terminal of dsp processor connect, in
Central processor, image data memory and mechanism for sorting are connect with the output end of dsp processor respectively, the output of central processing unit
End is connect with the input terminal of dsp processor again.
2. the Intelligent garbage classification robot according to claim 1 based on binocular stereo vision fixation and recognition, feature
It is, robot is as follows to the classification process of rubbish:
1) by the image of the two CCD camera acquisition rubbish targets in left and right, and collected digital image information is passed into video
Decoder；
2) pretreatment of dsp processor completion digital picture is passed to after Video Decoder decodes and passes to image data deposits
Reservoir is stored；
3) by dsp processor, treated that image information sends to that central processing unit carries out digital picture reprocessing stands with binocular again
Body vision identifies and positions, and completes to obtain the tool of rubbish target to the 3D vision matching of rubbish target and Studies About Reconstruction of Three-Dimension Information
Body three-dimensional dimension and three dimensional local information；
4) the final fixation and recognition information of rubbish target is finally transferred to dsp processor by central processing unit, DSP sends instruction
The classification of rubbish is completed to robot sorting manipulator arm.
3. the Intelligent garbage classification robot according to claim 2 based on binocular stereo vision fixation and recognition, feature
It is, the acquisition and pretreatment of rubbish target image and identifying and positioning for rubbish target three-dimensional information constitute rubbish target pair
Item stereo vision system, rubbish target Binocular Stereo Vision System are the 3D vision matchings based on cost rubbish objective function,
Central processing unit constructs a three-dimensional cost collection according to cost function, passes through the local optimum and three-dimensional space search to cost collection
It realizes the polymerization of parallax, and carries out three-dimensional filtering and operate to obtain the parallax information of two figures, after obtaining parallax, disparity map is carried out
Post-processing obtains fine disparity map.
4. the Intelligent garbage classification robot according to claim 3 based on binocular stereo vision fixation and recognition, feature
Be, dsp processor to rubbish target image pretreatment refer to first to rubbish target image carry out noise filtering after, then carry out
Image segmentation and edge detection.
5. the Intelligent garbage classification robot according to claim 4 based on binocular stereo vision fixation and recognition, feature
It is, the three-dimensional dimension and three dimensional local information of rubbish target are acquired under different location by two CCD cameras in left and right
One triangle of image and tested rubbish target configuration, according to the positional relationship and the rubbish of the CCD camera of known left and right two
The parallax that target point is formed in the digital picture of left and right, then acquisition is calculated by the identification of central processing unit.
6. the Intelligent garbage classification robot according to claim 5 based on binocular stereo vision fixation and recognition, feature
It is, the three-dimensional reconstruction of the rubbish target, is to be regarded using the two images of same rubbish target different perspectives according to three-dimensional
Feel the disparity map that matching obtains, three-dimensional information restored according to triangle principle, has just obtained accurately three-dimensional dimension and three-dimensional coordinate,
And obtained three-dimensional information is compared matching with the preset model data in deposit central processing unit, obtain classification results.
7. the Intelligent garbage classification robot according to claim 6 based on binocular stereo vision fixation and recognition, feature
It is, noise filtering is using nonlinear filtering technique median filtering algorithm, the filtering algorithm are as follows:
(1)
(2)
Wherein N indicates that filter window includes the sum of pixel,Indicate the gray value of filter window center pixel,The row of expression
The gray value of i-th of pixel in sequence vector,It indicatesWeighting coefficient.
8. the Intelligent garbage classification robot according to claim 7 based on binocular stereo vision fixation and recognition, feature
It is, image segmentation and edge detection are that Roberts operator is used to detect in gray scale or image by basis of characterization of color
Color change position outstanding reaches the detection at image segmentation and edge, the meter of Roberts operator according to the size of variable gradient
Calculate formula are as follows:
Due to digital picture one two-dimensional discrete type manifold of acquisition,Indicate a pixel in figureGradient it is big
Small value,Respectively indicate the gradient of diagonal line both direction.
9. the Intelligent garbage classification robot according to claim 8 based on binocular stereo vision fixation and recognition, feature
It is, the 3D vision matching of rubbish target is an optimization object function to be constructed, to obtain different views based on image pixel
The corresponding relationship of same rubbish target point in the image of angle, the Stereo Matching Algorithm being polymerize by cost, construction are pixel-based such as
Lower cost objective function:
Absolute error:
Square error:
WhereinIt is position in left imageThe image pixel intensities of point,It is position in right image
Image pixel intensities a little are set,It is parallax,It is corresponding matching cost function respectively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811542036.7A CN109465809A (en) | 2018-12-17 | 2018-12-17 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811542036.7A CN109465809A (en) | 2018-12-17 | 2018-12-17 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109465809A true CN109465809A (en) | 2019-03-15 |
Family
ID=65675950
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811542036.7A Pending CN109465809A (en) | 2018-12-17 | 2018-12-17 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109465809A (en) |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110116415A (en) * | 2019-06-12 | 2019-08-13 | 中北大学 | A kind of Bottle & Can class rubbish identification sorting machine people based on deep learning |
| CN110223350A (en) * | 2019-05-23 | 2019-09-10 | 汕头大学 | A kind of building blocks automatic sorting method and system based on binocular vision |
| CN110342134A (en) * | 2019-07-23 | 2019-10-18 | 珠海市一微半导体有限公司 | A kind of garbage classification identifying system and its method based on binocular vision |
| CN110648364A (en) * | 2019-09-17 | 2020-01-03 | 华侨大学 | Multi-dimensional space solid waste visual detection positioning and identification method and system |
| CN110755002A (en) * | 2019-04-04 | 2020-02-07 | 苏州科睿信飞智能科技有限公司 | Intelligent multifunctional outdoor cleaning robot |
| CN111017429A (en) * | 2019-11-20 | 2020-04-17 | 重庆特斯联智慧科技股份有限公司 | Community garbage classification method and system based on multi-factor fusion |
| CN111428595A (en) * | 2020-03-13 | 2020-07-17 | 五邑大学 | Underwater visual garbage cleaning robot and operation method thereof |
| CN111445368A (en) * | 2020-03-16 | 2020-07-24 | 五邑大学 | Garbage classification method, device and equipment based on machine vision and deep learning |
| CN111746728A (en) * | 2020-06-17 | 2020-10-09 | 重庆大学 | Novel overwater cleaning robot based on reinforcement learning and control method |
| CN112270673A (en) * | 2020-11-11 | 2021-01-26 | 珠海格力智能装备有限公司 | Method and apparatus for treating garbage |
| CN112678364A (en) * | 2020-11-11 | 2021-04-20 | 浙江机电职业技术学院 | Garbage can |
| CN113001557A (en) * | 2021-02-21 | 2021-06-22 | 浙江博城机器人科技有限公司 | Garbage sorting robot with binocular vision recognition |
| CN113297221A (en) * | 2021-05-25 | 2021-08-24 | 山东万国云大数据科技有限公司 | Data center monitoring system based on three-dimensional model |
| CN113378627A (en) * | 2021-04-23 | 2021-09-10 | 晋城鸿智纳米光机电研究院有限公司 | Automatic garbage classification method and device based on DSP chip and single chip microcomputer control |
| CN113441421A (en) * | 2021-07-22 | 2021-09-28 | 北京信息科技大学 | Automatic garbage classification system and method |
| CN113610883A (en) * | 2021-04-30 | 2021-11-05 | 新驱动重庆智能汽车有限公司 | Point cloud processing system and method, computer device, and storage medium |
| CN113866627A (en) * | 2021-11-23 | 2021-12-31 | 广东电网有限责任公司 | Automatic wiring device for electrical test of high-voltage circuit breaker |
| CN116228890A (en) * | 2023-05-05 | 2023-06-06 | 深圳市拓普泰克技术股份有限公司 | Intelligent control system based on oiling equipment |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012192473A (en) * | 2011-03-15 | 2012-10-11 | Seibu Electric & Mach Co Ltd | Stereoscopically viewing robot picking device |
| CN103136750A (en) * | 2013-01-30 | 2013-06-05 | 广西工学院 | Stereo matching optimization method of binocular visual system |
| CN103934211A (en) * | 2014-04-30 | 2014-07-23 | 重庆环视科技有限公司 | Stereoscopic vision-based three-dimensional product size sorting system |
| CN107972026A (en) * | 2016-10-25 | 2018-05-01 | 深圳光启合众科技有限公司 | Robot, mechanical arm and its control method and device |
| CN108053449A (en) * | 2017-12-25 | 2018-05-18 | 北京工业大学 | Three-dimensional rebuilding method, device and the binocular vision system of binocular vision system |
| CN108480239A (en) * | 2018-02-10 | 2018-09-04 | 浙江工业大学 | Workpiece quick sorting method based on stereoscopic vision and device |
| CN108607819A (en) * | 2018-04-25 | 2018-10-02 | 重庆邮电大学 | Material sorting system and method |
| CN108876855A (en) * | 2018-05-28 | 2018-11-23 | 哈尔滨工程大学 | A kind of sea cucumber detection and binocular visual positioning method based on deep learning |
| CN108942929A (en) * | 2018-07-10 | 2018-12-07 | 广州供电局有限公司 | The method and device of mechanical arm positioning crawl based on binocular stereo vision |
| CN209289269U (en) * | 2018-12-17 | 2019-08-23 | 中北大学 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
-
2018
- 2018-12-17 CN CN201811542036.7A patent/CN109465809A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012192473A (en) * | 2011-03-15 | 2012-10-11 | Seibu Electric & Mach Co Ltd | Stereoscopically viewing robot picking device |
| CN103136750A (en) * | 2013-01-30 | 2013-06-05 | 广西工学院 | Stereo matching optimization method of binocular visual system |
| CN103934211A (en) * | 2014-04-30 | 2014-07-23 | 重庆环视科技有限公司 | Stereoscopic vision-based three-dimensional product size sorting system |
| CN107972026A (en) * | 2016-10-25 | 2018-05-01 | 深圳光启合众科技有限公司 | Robot, mechanical arm and its control method and device |
| CN108053449A (en) * | 2017-12-25 | 2018-05-18 | 北京工业大学 | Three-dimensional rebuilding method, device and the binocular vision system of binocular vision system |
| CN108480239A (en) * | 2018-02-10 | 2018-09-04 | 浙江工业大学 | Workpiece quick sorting method based on stereoscopic vision and device |
| CN108607819A (en) * | 2018-04-25 | 2018-10-02 | 重庆邮电大学 | Material sorting system and method |
| CN108876855A (en) * | 2018-05-28 | 2018-11-23 | 哈尔滨工程大学 | A kind of sea cucumber detection and binocular visual positioning method based on deep learning |
| CN108942929A (en) * | 2018-07-10 | 2018-12-07 | 广州供电局有限公司 | The method and device of mechanical arm positioning crawl based on binocular stereo vision |
| CN209289269U (en) * | 2018-12-17 | 2019-08-23 | 中北大学 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110755002A (en) * | 2019-04-04 | 2020-02-07 | 苏州科睿信飞智能科技有限公司 | Intelligent multifunctional outdoor cleaning robot |
| CN110223350A (en) * | 2019-05-23 | 2019-09-10 | 汕头大学 | A kind of building blocks automatic sorting method and system based on binocular vision |
| CN110116415A (en) * | 2019-06-12 | 2019-08-13 | 中北大学 | A kind of Bottle & Can class rubbish identification sorting machine people based on deep learning |
| CN110342134A (en) * | 2019-07-23 | 2019-10-18 | 珠海市一微半导体有限公司 | A kind of garbage classification identifying system and its method based on binocular vision |
| CN110648364A (en) * | 2019-09-17 | 2020-01-03 | 华侨大学 | Multi-dimensional space solid waste visual detection positioning and identification method and system |
| CN110648364B (en) * | 2019-09-17 | 2022-09-30 | 华侨大学 | Multi-dimensional space solid waste visual detection positioning and identification method and system |
| CN111017429A (en) * | 2019-11-20 | 2020-04-17 | 重庆特斯联智慧科技股份有限公司 | Community garbage classification method and system based on multi-factor fusion |
| CN111428595A (en) * | 2020-03-13 | 2020-07-17 | 五邑大学 | Underwater visual garbage cleaning robot and operation method thereof |
| CN111428595B (en) * | 2020-03-13 | 2021-10-26 | 五邑大学 | Underwater visual garbage cleaning robot and operation method thereof |
| CN111445368B (en) * | 2020-03-16 | 2023-03-17 | 五邑大学 | Garbage classification method, device and equipment based on machine vision and deep learning |
| CN111445368A (en) * | 2020-03-16 | 2020-07-24 | 五邑大学 | Garbage classification method, device and equipment based on machine vision and deep learning |
| CN111746728A (en) * | 2020-06-17 | 2020-10-09 | 重庆大学 | Novel overwater cleaning robot based on reinforcement learning and control method |
| CN112678364A (en) * | 2020-11-11 | 2021-04-20 | 浙江机电职业技术学院 | Garbage can |
| CN112270673A (en) * | 2020-11-11 | 2021-01-26 | 珠海格力智能装备有限公司 | Method and apparatus for treating garbage |
| CN113001557A (en) * | 2021-02-21 | 2021-06-22 | 浙江博城机器人科技有限公司 | Garbage sorting robot with binocular vision recognition |
| CN113378627A (en) * | 2021-04-23 | 2021-09-10 | 晋城鸿智纳米光机电研究院有限公司 | Automatic garbage classification method and device based on DSP chip and single chip microcomputer control |
| CN113610883B (en) * | 2021-04-30 | 2022-04-08 | 新驱动重庆智能汽车有限公司 | Point cloud processing system and method, computer device, and storage medium |
| CN113610883A (en) * | 2021-04-30 | 2021-11-05 | 新驱动重庆智能汽车有限公司 | Point cloud processing system and method, computer device, and storage medium |
| CN113297221A (en) * | 2021-05-25 | 2021-08-24 | 山东万国云大数据科技有限公司 | Data center monitoring system based on three-dimensional model |
| CN113441421A (en) * | 2021-07-22 | 2021-09-28 | 北京信息科技大学 | Automatic garbage classification system and method |
| CN113441421B (en) * | 2021-07-22 | 2022-12-13 | 北京信息科技大学 | Automatic garbage classification system and method |
| CN113866627A (en) * | 2021-11-23 | 2021-12-31 | 广东电网有限责任公司 | Automatic wiring device for electrical test of high-voltage circuit breaker |
| CN116228890A (en) * | 2023-05-05 | 2023-06-06 | 深圳市拓普泰克技术股份有限公司 | Intelligent control system based on oiling equipment |
| CN116228890B (en) * | 2023-05-05 | 2023-08-11 | 深圳市拓普泰克技术股份有限公司 | Intelligent control system based on oiling equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109465809A (en) | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition | |
| CN209289269U (en) | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition | |
| CN104063702B (en) | Three-dimensional gait recognition based on shielding recovery and partial similarity matching | |
| CN106709950B (en) | Binocular vision-based inspection robot obstacle crossing wire positioning method | |
| CN105574527B (en) | A kind of quick object detecting method based on local feature learning | |
| CN112700499B (en) | Deep learning-based visual positioning simulation method and system in irradiation environment | |
| CN107833181B (en) | Three-dimensional panoramic image generation method based on zoom stereo vision | |
| CN103093191B (en) | A kind of three dimensional point cloud is in conjunction with the object identification method of digital image data | |
| CN107248159A (en) | A kind of metal works defect inspection method based on binocular vision | |
| CN101443817B (en) | Method and device for determining correspondence, preferably for the three-dimensional reconstruction of a scene | |
| CN111445368B (en) | Garbage classification method, device and equipment based on machine vision and deep learning | |
| CN106228528B (en) | A kind of multi-focus image fusing method based on decision diagram and rarefaction representation | |
| CN109685141A (en) | A kind of robotic article sorting visible detection method based on deep neural network | |
| CN109766873B (en) | Pedestrian re-identification method based on hybrid deformable convolution | |
| CN112067233B (en) | Six-degree-of-freedom motion capture method for wind tunnel model | |
| CN103530599A (en) | Method and system for distinguishing real face and picture face | |
| CN105809626A (en) | Self-adaption light compensation video image splicing method | |
| CN110084243A (en) | It is a kind of based on the archives of two dimensional code and monocular camera identification and localization method | |
| CN110334727B (en) | Intelligent matching detection method for tunnel cracks | |
| CN101510295B (en) | Design method for machine vision system based on PCIe and Vision Assistan | |
| CN113901874A (en) | Tea tender shoot identification and picking point positioning method based on improved R3Det rotating target detection algorithm | |
| CN108182704A (en) | Localization method based on Shape context feature | |
| CN111325828A (en) | Three-dimensional face acquisition method and device based on three-eye camera | |
| Mao et al. | Power transmission line image segmentation method based on binocular vision and feature pyramid network | |
| CN115601423A (en) | Edge enhancement-based round hole pose measurement method in binocular vision scene |
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 |