CN110342134A - A kind of garbage classification identifying system and its method based on binocular vision - Google Patents
A kind of garbage classification identifying system and its method based on binocular vision Download PDFInfo
- Publication number
- CN110342134A CN110342134A CN201910664400.5A CN201910664400A CN110342134A CN 110342134 A CN110342134 A CN 110342134A CN 201910664400 A CN201910664400 A CN 201910664400A CN 110342134 A CN110342134 A CN 110342134A
- Authority
- CN
- China
- Prior art keywords
- rubbish
- ccd camera
- garbage classification
- access road
- ccd
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/0033—Refuse receptacles; Accessories therefor specially adapted for segregated refuse collecting, e.g. receptacles with several compartments; Combination of receptacles
- B65F1/0053—Combination of several receptacles
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/0033—Refuse receptacles; Accessories therefor specially adapted for segregated refuse collecting, e.g. receptacles with several compartments; Combination of receptacles
- B65F2001/008—Means for automatically selecting the receptacle in which refuse should be placed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/138—Identification means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/176—Sorting means
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/10—Waste collection, transportation, transfer or storage, e.g. segregated refuse collecting, electric or hybrid propulsion
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Mechanical Engineering (AREA)
- Image Processing (AREA)
Abstract
The present invention discloses a kind of garbage classification identifying system and its method based on binocular vision, the garbage classification identifying system includes sorting guiding mechanism, rubbish detection access road, local side processor and two CCD cameras, the two CCD cameras are in setting up and down in the vertical direction of rubbish detection access road, rubbish detects inlet passage and sorts guiding mechanism, the input terminal of the output end connection local side processor of the two CCD cameras, there are electrical connections with local side processor for sorting guiding mechanism.Binocular camera, especially CCD camera is arranged at the vertical up and down direction of rubbish detection access road in the present invention, realizes the detection and tracking of the target rubbish moved to vertical direction, reduces system complexity, so that faster by the identification classification of throwing rubbish.
Description
Technical field
The present invention relates to the technical fields of garbage disposal, more particularly to a kind of garbage classification based on binocular vision to identify system
System and a kind of garbage classification recognition methods based on binocular vision.
Background technique
Number of patent application 201810319315 .0 " automatic classification dustbin " patent provides a kind of automatic sorting rubbish
Case, comprising: casing frame, several dustbins, electronic control system, camera and bracket are provided with several in casing frame
A dustbin, each dustbin upper end are equipped with top cover, and top cover opens or closes it by switch thereon, in the side of dustbin
Door leaf is housed on wall, bracket is housed in the surface of casing frame, at least one camera, each camera are housed on bracket
It is connected with electronic control system with each switch, each camera will be to be placed into the rubbish image transmitting electron in dustbin
Control system, electronic control system pass through image recognition, and the top cover of corresponding dustbin is opened or closed by switching;The patent
This is accordingly automatically opened to recognize the need for the type of the rubbish abandoned using the image recognition technology in electronic control system
The top cover of class rubbish, which can only be put into the dustbin being opened by user, after rubbish is thrown away corresponding dustbin, top cover
It is automatically closed.
Although which is able to achieve automatic classification, but need user to hold rubbish waiting system and identify and classify,
After completing identification and classification, it just can be carried out rubbish and abandon, user experience is poor, does not meet universal straight of user of public arena
Connect the use habit for abandoning rubbish.Therefore, it after user throws rubbish toward dustbin, needs that dustbin is allowed to realize quickly and easily
Automatic identification rubbish.
Summary of the invention
In view of the problems of the existing technology, to this, we provide a kind of garbage classification identifying system based on binocular vision
And its method, specific technical solution are as follows:
A kind of garbage classification identifying system based on binocular vision, the garbage classification identifying system include sorting guiding mechanism,
It is characterized in that, the garbage classification identifying system further include: rubbish detects access road, local side processor and two CCD camera shootings
Head, the two CCD cameras are in setting up and down, rubbish detection inlet passage in the vertical direction of rubbish detection access road
Guiding mechanism is sorted, the input terminal of the output end connection local side processor of the two CCD cameras sorts guiding mechanism and this
There are electrical connections for ground terminal processor.Technical solution setting at the vertical up and down direction of rubbish detection access road is double
Mesh camera, especially CCD camera realize the detection and tracking of the target rubbish moved to vertical direction, and it is multiple to reduce system
Polygamy, so that faster by the identification classification of throwing rubbish.
Further, the vertical direction of rubbish detection access road is provided with vertical bracket, the vertical bracket it is upper
The vertical visual angle of the CCD camera of the CCD camera and its downside setting of side setting is on the inside of rubbish detection access road
Same perpendicular on, and the vertical bracket upside setting CCD camera and its downside setting CCD camera do not locate
In orthogonal posture.The camera that the technical solution is arranged using on the upside of the fixed vertical direction of the vertical bracket and its downside, it is real
When capture throw from top to bottom into rubbish detection access road target rubbish image, vertical visual angle obtain target rubbish
Image meet the motion feature of target rubbish, which is based on above-mentioned binocular vision structure and preferably reacts
Visual effect in true environment.
Further, the local side processor simultaneously connect the vertical bracket upside setting CCD camera and
The CCD camera that its downside is arranged detects the dynamic of the target rubbish to fall inside access road for rubbish described in synchronous acquisition
The CCD of state image, CCD camera and its downside setting then in conjunction with the upside setting for the vertical bracket being obtained ahead of time takes the photograph
The CCD being arranged as the positional distance information between head, the pixel coordinate system where controlling the CCD camera that upside is arranged toward downside
Pixel coordinate system where camera translates conversion, so that the coordinate origin of the two is unified, then sits in the unified pixel of origin
Mark fastens processing dynamic image to obtain the two dimensional motion information of target rubbish, and corresponding with rubbish identification sorting parameter library
Identification model is matched, and finally sends garbage sorting instruction to sorting guiding mechanism according to matching result;Wherein, the two dimension
Motion information includes motion state, forms of motion and the speed of variation of the target rubbish, for identifying the target rubbish
Physical state;The identification model of rubbish identification sorting parameter library and its inside is all stored in the local side processor.The skill
Art scheme determines the precision of binocular vision identification whereabouts rubbish according to matched result, and is guaranteeing the same of image recognition precision
When, reduce the complexity of system.
Further, it is provided with infrared array sensor on the inside of the open edge of the rubbish detection access road, it is infrared
Sensor array and the local side processor, which exist, to be electrically connected, for triggering described two CCD camera synchronous averaging works
Make.Improve the intelligence degree of entire identifying system.
A kind of garbage classification recognition methods based on the garbage classification identifying system, the garbage classification recognition methods packet
Include: rubbish described in synchronous acquisition detects the dynamic image of the target rubbish to fall inside access road;In conjunction with the institute being obtained ahead of time
State the positional distance information between the CCD camera of the upside setting of vertical bracket and its CCD camera of downside setting, control
Pixel coordinate system translation where the CCD camera that pixel coordinate system where the CCD camera of upside setting is arranged toward downside
Conversion, so that the coordinate origin of the two is unified;Processing dynamic image is fastened in the unified pixel coordinate of origin to obtain target
The two dimensional motion information of rubbish, and matched with corresponding identification model in rubbish identification sorting parameter library;It is tied according to matching
The identification work of fruit completion target rubbish.The technical solution determines the essence of binocular vision identification whereabouts rubbish according to matched result
Degree, and while guaranteeing image recognition precision, reduce the complexity of system.
Detailed description of the invention
Fig. 1 is a kind of knot for garbage classification identifying system based on binocular vision that first embodiment of the invention provides
Structure schematic diagram.
Fig. 2 is a kind of knot for garbage classification identifying system based on binocular vision that second embodiment of the invention provides
Structure schematic diagram.
Fig. 3 is a kind of knot for garbage classification identifying system based on binocular vision that third embodiment of the invention provides
Structure schematic diagram.
Fig. 4 is a kind of flow chart of garbage classification recognition methods based on binocular vision provided in an embodiment of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to each reality of the invention
The mode of applying is explained in detail.However, it will be understood by those skilled in the art that in each embodiment of the present invention,
Claims are proposed in embodiment and do not have disclosed technical detail in order to make reader more fully understand the application.But
Even if the application also may be implemented and respectively weigh without these technical details and various changes and modifications based on the following respective embodiments
Benefit requires technical solution claimed.
First embodiment of the invention provides a kind of garbage classification identifying system based on binocular vision, as shown in Figure 1, rubbish
Rubbish classifying and identifying system includes that rubbish detection access road 101, sorting guiding mechanism 103, local side processor (do not indicate in figure
Out), CCD camera 1021 and CCD camera 1022.Vertical side of the two CCD cameras in rubbish detection access road 101
To in setting up and down, the binocular camera of the vertical direction setting of composition rubbish detection access road 101, the two CCD camera shooting
The input terminal of the output end connection local side processor of head, is used for transmission what CCD camera 1021 and CCD camera 1022 acquired
Binocular vision image to local side processor carries out rubbish identification classification;Rubbish detects the connection sorting Guiding machine of access road 101
Structure 103, sorting guiding mechanism 103 and local side processor, which exist, to be electrically connected, and sorting guiding mechanism 103 receives at local side
Garbage sorting operation is executed after managing the garbage classification result of device output.Garbage classification identifying system is assembled to any by the present embodiment
In dustbin, taken the photograph by the way that binocular camera, especially CCD are arranged at the vertical up and down direction that rubbish detects access road 101
As head, the detection and tracking of the target rubbish moved to vertical direction are realized, system complexity is reduced, so that by throwing rubbish
The speed of identification classification is accelerated.
In second embodiment of the invention, as shown in Fig. 2, the present embodiment is on the basis of first embodiment of the invention
On, in the vertical direction of rubbish detection access road 101, vertical bracket 104, the upside setting of the vertical bracket 102 are set
CCD camera 1021 and its downside setting CCD camera 1022 vertical visual angle on same perpendicular, CCD camera shooting
First 1021 and its downside setting the relatively fixed installation of CCD camera 1022, be located at rubbish detection access road 101 and be open
The inside of lower section, current embodiment require that in the CCD camera for physically guaranteeing CCD camera 1021 and its downside setting
1022 optical axis is parallel, is conducive to the conversion for completing the coordinate system during subsequent processing image data.So with the prior art
It compares, the CCD camera 1021 of the upside setting of the vertical bracket 104 and its CCD camera 1022 of downside setting are not at
Orthogonal posture.Although sacrificing the angular field of view of binocular camera, it is to speed up the processing speed of the target area of the image of acquisition
Degree.
In the present embodiment, the CCD that the local side processor connects the upside setting of the vertical bracket 104 simultaneously takes the photograph
As first 1021 and its CCD camera 1022 of downside setting, for subordinate in the detection access road 101 of rubbish described in synchronous acquisition
The dynamic image of the target rubbish fallen visually has otherness in the target rubbish that identification is fallen;Based on the otherness, knot
Close the CCD camera 1021 of the upside setting for the vertical bracket being obtained ahead of time and its CCD camera 1022 of downside setting
Between positional distance information, the CCD that the pixel coordinate system where the CCD camera 1021 that control upside is arranged is arranged toward downside
Pixel coordinate system where camera 1022 translates conversion, so that the coordinate origin of the two is unified, then the picture unified in origin
Dynamic image is handled on plain coordinate system to obtain the two dimensional motion information of target rubbish, then identifies sorting parameter library with rubbish again
In corresponding identification model carry out values match, garbage sorting is finally sent to sorting guiding mechanism 103 according to matching result and is referred to
It enables.The present embodiment is captured in real time using the upside of the fixed vertical direction of the vertical bracket 104 and its camera of downside setting
The image of the target rubbish into rubbish detection access road 101 is thrown from top to bottom, thus the target rubbish obtained on vertical visual angle
The image of rubbish meets the motion feature of target rubbish, and it is preferably anti-that which is based on above-mentioned binocular vision structure
Answer the visual effect in true environment.
As shown in figure 4, being based on previous embodiment, the embodiment of the present invention provides a kind of based on aforementioned garbage classification identifying system
Garbage classification recognition methods, specifically include:
Step S201, after rubbish, which is thrown, enters rubbish detection access road 101, CCD camera 1021 and CCD are controlled
Camera 1022 is synchronous to open collecting work, and different perspectives from vertical direction acquires in the rubbish detection access road 101
The target rubbish image that subordinate falls, subsequently into step S202.
Step S202, the target rubbish acquired with the CCD camera 1022 that the downside of the vertical bracket 104 is arranged
The corresponding dynamic image is template image, subsequently into step S203.
Step S203, it is taken the photograph in conjunction with the CCD that the upside for the vertical bracket for pre-entering the local side processor is arranged
As the positional distance information between first 1021 and its CCD camera 1022 that is arranged of downside, the local side processor calculated
The transformational relation of the corresponding pixel coordinate system of above-mentioned two CCD camera acquisition image, including spin matrix R and translation matrix
T, subsequently into step S204.
Step S204, the pixel coordinate system where the CCD camera 1021 that control upside is arranged is taken the photograph toward the CCD that downside is arranged
Pixel coordinate system as where first 1022 translates conversion, so that the coordinate origin of the two is unified logical to rubbish detection entrance
At the same position in road, subsequently into step S205.The conversion of the corresponding coordinate system of above-mentioned two CCD camera acquisition image
The original of pixel coordinate system where relationship, the i.e. origin of pixel coordinate system where CCD camera 1021 and CCD camera 1022
Vector position relationship between point.Since the optical axis of the CCD camera 1021 in second embodiment of the invention and its downside are set
The optical axis for the CCD camera 1022 set is parallel, so, the figure of two CCD cameras acquisition under second embodiment of the invention
As the conversion of the coordinate system at place is only simple translation transformation, and the subsequent mesh searched on CCD camera picture
The centroid position efficiency for marking rubbish is higher, to accelerate the speed of garbage classification identification.
Step S205, control that the CCD camera 1021 of the upside setting of the vertical bracket 104 acquires with step S202
Belong to the corresponding each frame image of the same target rubbish and each frame template image of synchronous acquisition and make pixel difference, then by picture
Plain difference is transformed into world coordinate system, so that two dimensional motion information is obtained, subsequently into step S206.The step is first chosen CCD and is taken the photograph
The pixel value of the target garbage shares in template image acquired as first 1022, as the first pixel value;CCD is chosen again to take the photograph
The pixel value of the target garbage shares as described in the dynamic image of first 1021 acquisition is then transferred to institute as the second pixel value
State local side processor.Here the target garbage shares are the same position center-of-mass coordinates on two CCD camera pictures
Place;Then the first pixel value is controlled by the local side processor and subtracts the second pixel value acquisition pixel value difference, as same institute
State motion change feature of the target rubbish under pixel coordinate system.The picture noise or background for avoiding the target rubbish are being converted
The error applied after to world coordinate system influences the actual movement change amount of target rubbish.Wherein, the two dimensional motion letter
Breath includes motion state, forms of motion and the speed of variation of the target rubbish, for identifying the physics of the target rubbish
State.
The present embodiment is obtained in the vertical upper and lower multi-view image of union using binocular vision (CCD cameras of upper and lower sides)
The distance of the corresponding center pixel of same target rubbish, is then based on two CCD camera relative positions of same perpendicular
Relationship, which establishes numerical value conversion relationship, reduces the complexity of system while guaranteeing image recognition precision.
Step S206, the two dimensional motion information of aforementioned acquisition and rubbish are identified sorting parameter library by the described local side processor
In corresponding identification model matched, when successful match then enters step S207;If matching is unsuccessful, the local side
Processor sends cloud server for the image for matching the corresponding target rubbish of unsuccessful two dimensional motion information and carries out
Artificial to screen, then the recognition result of the target rubbish is sent back to the garbage classification identifying system again by cloud server, then
The recognition result is updated into rubbish identification sorting parameter library in corresponding identification model by the local side processor, so that
The image of the garbage classification identifying system study unsuccessful target rubbish of automatic identification two dimensional motion information matches, so
After enter step S207.
Step S207, the described local side processor is completed the type identification of the target rubbish according to matching result and is returned
Then class sends garbage sorting instruction to sorting guiding mechanism, rubbish identification sorting parameter library includes being based on pixel-parameters
Identification model include: image feature model, color characteristic model, Facial Features model, form feature model.The present embodiment is logical
Motion state, forms of motion and the speed of variation that Image Matching parameter obtains rubbish are crossed, identifies the physical state of rubbish, passes through
Rubbish identifies water content, ingredient, the viscosity of rubbish launching change in shape, flowing and the speed of movement that power generates, and passes through bullet
Property variation frequency identify rubbish in rubber, jelly, raw meat, cold cuts, rice;Rubbish is distinguished according to the Color characteristics parameters of rubbish
Plastics, paper, ceramics, food and metal in rubbish, Facial Features parameter distinguish animal, plant, battery and the light bulb in rubbish, shape
Shape characteristic parameter distinguishes bottle, tank, plate and the film in rubbish.
The present embodiment calculates two CCD based on same perpendicular using binocular vision (CCD cameras of upper and lower sides)
Camera relative positional relationship establishes numerical value conversion relationship, closes to obtain the corresponding of same target point in different perspectives image
System, then operation it is vertical above and below multi-view image in the corresponding center pixel of same target rubbish distance, guaranteeing that image knows
While other precision, the complexity of system is reduced.
In third embodiment of the invention, as shown in figure 3, the present embodiment is on the basis of second embodiment of the invention
On, infrared array sensor 105, infrared sensor battle array are provided on the inside of the open edge of the rubbish detection access road 101
The sensor of column 105 is evenly distributed in the open edge of rubbish detection access road 101, and its infrared probe both facing to
The opening center of the rubbish detection access road 101, infrared array sensor 105 and the local side processor are (in figure not
Indicate) there is electric connection, access road 101 is detected into the rubbish or is equipped with for detecting whether there is rubbish to be thrown
In the dustbin of the garbage classification identifying system, when having detected that rubbish enters the garbage classification identifying system, triggering
The image of CCD camera 1021 and 1022 synchronous averaging of camera acquisition target rubbish.In order to more accurately know to rubbish
Not to classify, the infrared array sensor 105 can also be to the ruler for the rubbish for being put into the garbage classification identifying system
Very little range is detected, the more comprehensively identification for the garbage classification identifying system to the target rubbish.Also it improves
The intelligence degree of the garbage classification identifying system.
In previous embodiment, the opening of the sorting guiding mechanism is connected with rubbish detection access road, is used for
Rubbish is received to enter in the refuse container combined by heterogeneity type rubbish, when the local side processor completes the target
After the type identification of rubbish, sends corresponding garbage sorting and instruct to the sorting guiding mechanism, the sorting guiding mechanism will
The target rubbish is put into the refuse container of corresponding identification types.Routine techniques can be used in the sorting guiding mechanism, including
There are by cascaded structure and carry the garbage classification dispensing bucket of the sorting arm that servo is constituted or motor driven movement, these points
The forms of motion for picking guiding mechanism includes rotary motion and linear motion, depending on the kind quantity of rubbish, garbage classification and is divided
The kind of choosing is how many and the classification of kind requires and quantity and corresponding rubbish receive the quantity and its arrangement mode of container.
Why binocular camera in previous embodiment selects CCD camera to acquire to throw and enter rubbish detection entrance
The rubbish in channel is the fast response time because of CCD camera, and suitable to scan the target rubbish to hurtle down, pattern distortion is small,
It without image retention, and obtains and contains much information, complicated rubbish image can be handled;In addition, CCD camera shock resistance and vibration, performance are steady
Fixed, the service life is long, is suitble to be installed on the inside of dustbin execution sampling in (including on the inside of rubbish detection access road) this environment and appoints
Business.The small power consumption of CCD camera simultaneously, operating voltage is low, so that the overall work power consumption of the garbage classification identifying system drops
It is low.
Above embodiments be only it is sufficiently open is not intended to limit the present invention, it is all based on the inventive subject matter of the present invention, without creating
Property labour equivalence techniques feature replacement, should be considered as the application exposure range.
Claims (5)
1. a kind of garbage classification identifying system based on binocular vision, which includes sorting guiding mechanism,
It is characterized in that, the garbage classification identifying system further include: rubbish detection access road, local side processor and two CCD take the photograph
As head, the two CCD cameras are in setting up and down, rubbish detection access road company in the vertical direction of rubbish detection access road
The reduction of fractions to a common denominator picks guiding mechanism, the input terminal of the output ends of the two CCD cameras connection local side processor, sorting guiding mechanism with
There are electrical connections for local side processor.
2. garbage classification identifying system according to claim 1, which is characterized in that the rubbish detects the vertical of access road
Direction is provided with vertical bracket, and the CCD camera of the CCD camera and its downside setting of the upside setting of the vertical bracket erects
Angle is looked at straight on the same perpendicular on the inside of rubbish detection access road, and the CCD of the upside setting of the vertical bracket
Camera and its CCD camera of downside setting are not at orthogonal posture.
3. garbage classification identifying system according to claim 2, which is characterized in that the local side processor connects institute simultaneously
The CCD camera of the upside setting of vertical bracket and its CCD camera of downside setting are stated, for the inspection of rubbish described in synchronous acquisition
The dynamic image for surveying the target rubbish to fall inside access road, then in conjunction with being set on the upside of the vertical bracket being obtained ahead of time
Positional distance information between the CCD camera set and its CCD camera of downside setting, the CCD camera shooting that control upside is arranged
Pixel coordinate system translation conversion where the CCD camera that pixel coordinate system where head is arranged toward downside, so that the seat of the two
Mark system origin is unified, then fastens processing dynamic image in the unified pixel coordinate of origin to obtain the two dimensional motion letter of target rubbish
Breath, and matched with corresponding identification model in rubbish identification sorting parameter library, finally it is oriented to according to matching result to sorting
Mechanism sends garbage sorting instruction;
Wherein, the two dimensional motion information includes motion state, forms of motion and the speed of variation of the target rubbish, is used for
Identify the physical state of the target rubbish;The identification model of rubbish identification sorting parameter library and its inside is all stored in described
Ground terminal processor.
4. according to claim 1 or garbage classification identifying system described in claim 2, which is characterized in that the rubbish detect into
Infrared array sensor is provided on the inside of the open edge in mouth channel, infrared array sensor and the local side processor exist
It is electrically connected, for triggering described two CCD camera synchronous averaging work.
5. a kind of garbage classification recognition methods based on any one of Claims 1-4 garbage classification identifying system, feature
It is, which includes:
Rubbish described in synchronous acquisition detects the dynamic image of the target rubbish to fall inside access road;
In conjunction with the vertical bracket being obtained ahead of time upside be arranged CCD camera and its downside be arranged CCD camera it
Between positional distance information, the CCD camera that is arranged toward downside of pixel coordinate system where the CCD camera that control upside is arranged
The pixel coordinate system at place translates conversion, so that the coordinate origin of the two is unified;
Processing dynamic image is fastened in the unified pixel coordinate of origin to obtain the two dimensional motion information of target rubbish, and and rubbish
Corresponding identification model is matched in identification sorting parameter library;
The identification work of target rubbish is completed according to matching result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910664400.5A CN110342134B (en) | 2019-07-23 | 2019-07-23 | Garbage classification and identification system and method based on binocular vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910664400.5A CN110342134B (en) | 2019-07-23 | 2019-07-23 | Garbage classification and identification system and method based on binocular vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110342134A true CN110342134A (en) | 2019-10-18 |
CN110342134B CN110342134B (en) | 2023-06-09 |
Family
ID=68179767
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910664400.5A Active CN110342134B (en) | 2019-07-23 | 2019-07-23 | Garbage classification and identification system and method based on binocular vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110342134B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111086797A (en) * | 2019-11-25 | 2020-05-01 | 徐州立方机电设备制造有限公司 | Dustbin with recognition function |
CN111179938A (en) * | 2019-12-26 | 2020-05-19 | 安徽仁昊智能科技有限公司 | Speech recognition garbage classification system based on artificial intelligence |
CN112232246A (en) * | 2020-10-22 | 2021-01-15 | 深兰人工智能(深圳)有限公司 | Garbage detection and classification method and device based on deep learning |
CN113159343A (en) * | 2021-04-28 | 2021-07-23 | 深圳市阳谷医疗***有限公司 | Medical waste purification treatment system based on shared database |
CN115866392A (en) * | 2022-11-25 | 2023-03-28 | 深圳市智百威科技发展有限公司 | AI object visual identification system and method |
Citations (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001256477A (en) * | 2000-03-13 | 2001-09-21 | Ecchandesu:Kk | Information collecting device |
CN101527046A (en) * | 2009-04-28 | 2009-09-09 | 青岛海信数字多媒体技术国家重点实验室有限公司 | Motion detection method, device and system |
US20100232683A1 (en) * | 2009-03-11 | 2010-09-16 | Omron Corporation | Method For Displaying Recognition Result Obtained By Three-Dimensional Visual Sensor And Three-Dimensional Visual Sensor |
CN102313536A (en) * | 2011-07-21 | 2012-01-11 | 清华大学 | Method for barrier perception based on airborne binocular vision |
CN102589516A (en) * | 2012-03-01 | 2012-07-18 | 长安大学 | Dynamic distance measuring system based on binocular line scan cameras |
CN102592117A (en) * | 2011-12-30 | 2012-07-18 | 杭州士兰微电子股份有限公司 | Three-dimensional object identification method and system |
JP2013152219A (en) * | 2011-12-26 | 2013-08-08 | Japan Traffic Management Technology Association | Speed measurement system, speed measurement method and program |
JP2013201714A (en) * | 2012-03-26 | 2013-10-03 | Central Research Institute Of Electric Power Industry | Moving-object image discrimination device and moving-object image discrimination method |
CN105227828A (en) * | 2015-08-25 | 2016-01-06 | 努比亚技术有限公司 | Filming apparatus and method |
CN105550670A (en) * | 2016-01-27 | 2016-05-04 | 兰州理工大学 | Target object dynamic tracking and measurement positioning method |
CN105772407A (en) * | 2016-01-26 | 2016-07-20 | 耿春茂 | Waste classification robot based on image recognition technology |
CN106485735A (en) * | 2015-09-01 | 2017-03-08 | 南京理工大学 | Human body target recognition and tracking method based on stereovision technique |
CN106960210A (en) * | 2017-03-23 | 2017-07-18 | 上海视可电子科技有限公司 | The method and apparatus of target detection |
JP6364101B1 (en) * | 2017-01-31 | 2018-07-25 | 森ビル株式会社 | Air monitoring device, air monitoring method and program |
JP2018193204A (en) * | 2017-05-19 | 2018-12-06 | 新明和工業株式会社 | Garbage collection vehicle and person recognition device provided on the same |
CN109089102A (en) * | 2018-09-05 | 2018-12-25 | 华南智能机器人创新研究院 | A kind of robotic article method for identifying and classifying and system based on binocular vision |
CN109308693A (en) * | 2018-08-29 | 2019-02-05 | 北京航空航天大学 | By the target detection and pose measurement list binocular vision system of a ptz camera building |
CN109465809A (en) * | 2018-12-17 | 2019-03-15 | 中北大学 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
CN109665232A (en) * | 2018-12-18 | 2019-04-23 | 南京阿尔飞思智能物联科技有限公司 | A kind of device of the open air life automatic garbage sorting based on artificial intelligence and compression |
US20190122378A1 (en) * | 2017-04-17 | 2019-04-25 | The United States Of America, As Represented By The Secretary Of The Navy | Apparatuses and methods for machine vision systems including creation of a point cloud model and/or three dimensional model based on multiple images from different perspectives and combination of depth cues from camera motion and defocus with various applications including navigation systems, and pattern matching systems as well as estimating relative blur between images for use in depth from defocus or autofocusing applications |
CN109867071A (en) * | 2019-03-28 | 2019-06-11 | 柳州市妇幼保健院 | A kind of intelligent medical refuse sorting vehicle and its classification method |
CN110032278A (en) * | 2019-03-29 | 2019-07-19 | 华中科技大学 | A kind of method for recognizing position and attitude, the apparatus and system of human eye attention object |
-
2019
- 2019-07-23 CN CN201910664400.5A patent/CN110342134B/en active Active
Patent Citations (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001256477A (en) * | 2000-03-13 | 2001-09-21 | Ecchandesu:Kk | Information collecting device |
US20100232683A1 (en) * | 2009-03-11 | 2010-09-16 | Omron Corporation | Method For Displaying Recognition Result Obtained By Three-Dimensional Visual Sensor And Three-Dimensional Visual Sensor |
CN101527046A (en) * | 2009-04-28 | 2009-09-09 | 青岛海信数字多媒体技术国家重点实验室有限公司 | Motion detection method, device and system |
CN102313536A (en) * | 2011-07-21 | 2012-01-11 | 清华大学 | Method for barrier perception based on airborne binocular vision |
JP2013152219A (en) * | 2011-12-26 | 2013-08-08 | Japan Traffic Management Technology Association | Speed measurement system, speed measurement method and program |
CN102592117A (en) * | 2011-12-30 | 2012-07-18 | 杭州士兰微电子股份有限公司 | Three-dimensional object identification method and system |
CN102589516A (en) * | 2012-03-01 | 2012-07-18 | 长安大学 | Dynamic distance measuring system based on binocular line scan cameras |
JP2013201714A (en) * | 2012-03-26 | 2013-10-03 | Central Research Institute Of Electric Power Industry | Moving-object image discrimination device and moving-object image discrimination method |
CN105227828A (en) * | 2015-08-25 | 2016-01-06 | 努比亚技术有限公司 | Filming apparatus and method |
CN106485735A (en) * | 2015-09-01 | 2017-03-08 | 南京理工大学 | Human body target recognition and tracking method based on stereovision technique |
CN105772407A (en) * | 2016-01-26 | 2016-07-20 | 耿春茂 | Waste classification robot based on image recognition technology |
CN105550670A (en) * | 2016-01-27 | 2016-05-04 | 兰州理工大学 | Target object dynamic tracking and measurement positioning method |
JP6364101B1 (en) * | 2017-01-31 | 2018-07-25 | 森ビル株式会社 | Air monitoring device, air monitoring method and program |
CN106960210A (en) * | 2017-03-23 | 2017-07-18 | 上海视可电子科技有限公司 | The method and apparatus of target detection |
US20190122378A1 (en) * | 2017-04-17 | 2019-04-25 | The United States Of America, As Represented By The Secretary Of The Navy | Apparatuses and methods for machine vision systems including creation of a point cloud model and/or three dimensional model based on multiple images from different perspectives and combination of depth cues from camera motion and defocus with various applications including navigation systems, and pattern matching systems as well as estimating relative blur between images for use in depth from defocus or autofocusing applications |
JP2018193204A (en) * | 2017-05-19 | 2018-12-06 | 新明和工業株式会社 | Garbage collection vehicle and person recognition device provided on the same |
CN109308693A (en) * | 2018-08-29 | 2019-02-05 | 北京航空航天大学 | By the target detection and pose measurement list binocular vision system of a ptz camera building |
CN109089102A (en) * | 2018-09-05 | 2018-12-25 | 华南智能机器人创新研究院 | A kind of robotic article method for identifying and classifying and system based on binocular vision |
CN109465809A (en) * | 2018-12-17 | 2019-03-15 | 中北大学 | A kind of Intelligent garbage classification robot based on binocular stereo vision fixation and recognition |
CN109665232A (en) * | 2018-12-18 | 2019-04-23 | 南京阿尔飞思智能物联科技有限公司 | A kind of device of the open air life automatic garbage sorting based on artificial intelligence and compression |
CN109867071A (en) * | 2019-03-28 | 2019-06-11 | 柳州市妇幼保健院 | A kind of intelligent medical refuse sorting vehicle and its classification method |
CN110032278A (en) * | 2019-03-29 | 2019-07-19 | 华中科技大学 | A kind of method for recognizing position and attitude, the apparatus and system of human eye attention object |
Non-Patent Citations (5)
Title |
---|
万燕;周增刚;李晓园;: "基于双目视觉的车辆检测和车速测定" * |
张建;李宗南;张楠;谢静;贺立源;: "基于实测数据的作物三维信息获取与重建方法研究进展", 华中农业大学学报 * |
林森;殷新勇;唐延东;: "双目视觉立体匹配技术研究现状和展望" * |
白中浩;王鹏辉;李智强;: "基于Stixel-world及特征融合的双目立体视觉行人检测" * |
鲍振强;李艾华;崔智高;袁梦;: "深度学习在视觉定位与三维结构恢复中的研究进展", 激光与光电子学进展 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111086797A (en) * | 2019-11-25 | 2020-05-01 | 徐州立方机电设备制造有限公司 | Dustbin with recognition function |
CN111179938A (en) * | 2019-12-26 | 2020-05-19 | 安徽仁昊智能科技有限公司 | Speech recognition garbage classification system based on artificial intelligence |
CN112232246A (en) * | 2020-10-22 | 2021-01-15 | 深兰人工智能(深圳)有限公司 | Garbage detection and classification method and device based on deep learning |
CN113159343A (en) * | 2021-04-28 | 2021-07-23 | 深圳市阳谷医疗***有限公司 | Medical waste purification treatment system based on shared database |
CN115866392A (en) * | 2022-11-25 | 2023-03-28 | 深圳市智百威科技发展有限公司 | AI object visual identification system and method |
Also Published As
Publication number | Publication date |
---|---|
CN110342134B (en) | 2023-06-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110342134A (en) | A kind of garbage classification identifying system and its method based on binocular vision | |
US9403278B1 (en) | Systems and methods for detecting and picking up a waste receptacle | |
US12006141B2 (en) | Systems and methods for detecting waste receptacles using convolutional neural networks | |
Milan et al. | Semantic segmentation from limited training data | |
CA3100569C (en) | Ship identity recognition method base on fusion of ais data and video data | |
CN106447680B (en) | The object detecting and tracking method that radar is merged with vision under dynamic background environment | |
Sock et al. | Multi-view 6D object pose estimation and camera motion planning using RGBD images | |
CN109923583A (en) | A kind of recognition methods of posture, equipment and moveable platform | |
CN110689519B (en) | Fog drop deposition image detection system and method based on yolo network | |
CN106339657B (en) | Crop straw burning monitoring method based on monitor video, device | |
CN114693661A (en) | Rapid sorting method based on deep learning | |
WO2021179596A1 (en) | Garbage classification method and smart household appliance | |
CN111161295B (en) | Dish image background stripping method | |
CN114359552A (en) | Instrument image identification method based on inspection robot | |
CN113947770A (en) | Method for identifying object placed in different areas of intelligent cabinet | |
CN116161357A (en) | Intelligent networking dustbin system and rubbish throwing monitoring method | |
CN114506591A (en) | Intelligent garbage can design method based on image classification technology | |
CN110533995A (en) | A kind of applied mathematics probability statistics compiling device | |
WO2023068929A1 (en) | Automated bin-picking based on deep learning | |
CN112364727B (en) | Image recognition-based junk information acquisition system | |
CN109493371A (en) | A kind of quadrotor drone pedestrian tracting method of view-based access control model | |
CN115661745A (en) | Target classification method based on 3D laser radar detection and image analysis | |
CN114851194A (en) | Object picking method and device of cleaning robot and related equipment and storage medium thereof | |
CN211002883U (en) | Garbage classification recognition system based on binocular vision | |
CN208125940U (en) | Multi-level various dimensions stereo intelligent safe examination 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 | ||
CB02 | Change of applicant information |
Address after: 519000 2706, No. 3000, Huandao East Road, Hengqin new area, Zhuhai, Guangdong Applicant after: Zhuhai Yiwei Semiconductor Co.,Ltd. Address before: Room 105-514, No.6 Baohua Road, Hengqin New District, Zhuhai City, Guangdong Province Applicant before: AMICRO SEMICONDUCTOR Co.,Ltd. |
|
CB02 | Change of applicant information | ||
GR01 | Patent grant | ||
GR01 | Patent grant |