CN107886136A - A kind of method based on depth learning technology identification goods medium species - Google Patents
A kind of method based on depth learning technology identification goods medium species Download PDFInfo
- Publication number
- CN107886136A CN107886136A CN201711364690.9A CN201711364690A CN107886136A CN 107886136 A CN107886136 A CN 107886136A CN 201711364690 A CN201711364690 A CN 201711364690A CN 107886136 A CN107886136 A CN 107886136A
- Authority
- CN
- China
- Prior art keywords
- deep learning
- goods
- medium
- picture
- vehicle
- 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
- 238000005516 engineering process Methods 0.000 title claims abstract description 26
- 238000000034 method Methods 0.000 title claims abstract description 12
- 238000013135 deep learning Methods 0.000 claims abstract description 56
- 239000002699 waste material Substances 0.000 claims description 14
- 239000010813 municipal solid waste Substances 0.000 claims description 8
- -1 dry loess Substances 0.000 claims description 5
- 239000004576 sand Substances 0.000 claims description 5
- 239000002689 soil Substances 0.000 claims description 5
- 239000004575 stone Substances 0.000 claims description 5
- 238000005034 decoration Methods 0.000 claims 1
- 239000002910 solid waste Substances 0.000 abstract description 9
- 210000005036 nerve Anatomy 0.000 abstract description 4
- 241000894007 species Species 0.000 description 12
- 230000000694 effects Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000004064 recycling Methods 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Processing Of Solid Wastes (AREA)
- Refuse Collection And Transfer (AREA)
Abstract
The invention discloses a kind of method based on depth learning technology identification goods medium species, comprise the following steps that：S1：Collect the goods picture for collecting and marked different medium attribute；S2：Goods medium picture is inputted to deep learning module, establishes sample pattern, and by constantly training amendment sample pattern；S3：The goods picture of captured in real-time is imported to degree study module, deep learning module exports result by sample pattern automatic recognition cargo medium species；The present invention is handled substantial amounts of goods picture based on depth learning technology by artificial nerve network model, establishes sample pattern, to the goods picture automatic identification media type of importing, so as to reach the purpose of the intelligent classification of solid waste.
Description
Technical field
The invention belongs to the goods media recognition technical field of engineering car, is related to a kind of based on deep learning technology identification
The method of goods medium species.
Background technology
As Urbanization in China paces are increasingly accelerated, building waste increases therewith.As shown by data, China's urban architecture
Rubbish has accounted for the 30%~40% of municipal refuse total amount, is the first big city rubbish source.It is expected that the year two thousand twenty, New Dwelling 30,000,000,000 is flat
Square rice, building waste are at least up to 50,000,000,000 tons.The data issued according to construction refuse resource industrial technology innovation strategic alliances shows
Show, Foreign Architecture waste treatment and utilization have higher technical threshold and industry standard, and this disposes profit to building castoff
The effect of guiding and supervising is played with industry.As building castoff regeneration techniques standard and various enforceable is daily paid much attention in USA and Europe
Operating specification.And the current building waste treatment industry in China lacks enforceable strict related specifications so that China's building waste
48%, the 63% of Singapore, the 80% of Hong Kong, Japanese 95% of recovery utilization rate 5% well below Britain.
Quality and effect to ensure construction refuse resource, first have to accomplish is that building waste is classified；
At this stage, also fully rely in industry and manually building waste is classified, then transport different processing mechanisms to respectively.But its
Middle human factor is larger, and building waste easily occurs and handles nonstandard situation, is unfavorable for management and the recovery profit of supervision department
It is smoothed out.
The content of the invention
The technical problems to be solved by the invention are the shortcomings that overcoming prior art, there is provided one kind is based on deep learning skill
The method that art identifies goods medium species, by artificial nerve network model, based on deep learning technology to substantial amounts of goods figure
Piece is handled, and establishes sample pattern, to the goods picture automatic identification media type of importing, so as to reach solid waste intelligence
The purpose of energyization classification.
In order to solve the above technical problems, present invention offer is a kind of to identify goods medium species based on deep learning technology
Method, comprise the following specific steps that：
S1：Collect the goods picture for collecting and marked different medium attribute；
S2：Goods medium picture is inputted, establishes sample pattern, and by constantly training, correcting sample pattern；
S3：The goods picture of captured in real-time is imported, by sample pattern automatic recognition cargo medium species, and exports result.
The technical scheme that further limits of the present invention is：
The goods of different medium attribute includes building waste, stone, sand, dry loess, soil, mud and finishing rubbish in foregoing S1
Rubbish.
In foregoing S2, the training of sample pattern can be completed by server.
In foregoing S2, after server completes the training of sample pattern, the algorithm of sample pattern can be migrated to vehicle intelligent
Terminal and/or vehicle-mounted external equipment.
Further,
The present invention also provides a kind of system based on deep learning technology identification goods medium species, and system includes deep learning mould
Block, the input connection vehicle-mounted external equipment of deep learning module, its output end connection result reception device；Vehicle-mounted external equipment
For shooting real-time item picture, which kind of medium output recognition result is after deep learning module receives goods picture, is as a result connect
Receiving apparatus receives medium information.
The present invention also provides a kind of system based on deep learning technology identification goods medium species, and system includes depth
Practise module, the input connection vehicle-mounted external equipment of deep learning module, its output end connection server, server connection result
Reception device；Vehicle-mounted external equipment is used to shoot real-time item picture, and deep learning module exports identification after receiving goods picture
As a result for which kind of medium and medium information is sent to server, server medium information can be transmitted to result reception device.
The present invention also provides a kind of system based on deep learning technology identification goods medium species, and system includes depth
Practise module, the input connection vehicle intelligent terminal of deep learning module, vehicle intelligent terminal connection vehicle-mounted external equipment, depth
The output end connection result reception device of study module；Vehicle-mounted external equipment is used to shoot real-time item picture and send picture
To vehicle intelligent terminal, goods picture is forwarded to deep learning module by vehicle intelligent terminal, and deep learning module receives goods
Which kind of recognition result is exported after picture to be medium and send medium information to result reception device.
Aforementioned depth study module can be embedded in server and/or vehicle intelligent terminal is used.
Aforementioned result reception device is mobile intelligent terminal.
Foregoing mobile intelligent terminal includes smart mobile phone, pc or PAD.
The beneficial effects of the invention are as follows：
The present invention carries out key words sorting by inputting a large amount of different types of solid waste pictures, and to picture, such as：Build rubbish
Rubbish, stone, sand, dry loess, soil and mud, using existing artificial nerve network model, based on deep learning technology pair
Substantial amounts of solid waste picture is handled, and establishes sample pattern, so as to import the solid waste picture of captured in real-time
When, system is according to the media type of discarded object in existing sample pattern automatic identification picture, so as to reach solid waste intelligence
Energyization, the purpose of mechanized classification.Be advantageous to the recycling of building waste, building castoff treatment and use industry is played
The effect of guiding and supervising.
Brief description of the drawings
Fig. 1 is the principle flow chart of the present invention；
Fig. 2 is that system forms schematic diagram one；
Fig. 3 is that system forms schematic diagram two；
Fig. 4 is that system forms schematic diagram three.
Embodiment
The present invention program is further described by the following examples, wherein：
Deep learning module uses beautiful Tesla P4 deep learning processor of Leadtek/；
Vehicle intelligent terminal uses the sharp bright D5 types automobile travel recorders of Streamax/；
Vehicle-mounted external equipment uses the sharp bright infrared anti-riot dome-type high definition vehicle-mounted vidicons of C2 types of Streamax/；
Server is using association's ThinkSystem SR950GPU servers.
Embodiment 1
The present embodiment provides a kind of method based on deep learning technology identification goods medium species, as shown in figure 1, including as follows
Specific steps：
S1：Collect the goods picture for collecting and marked different medium attribute；
S2：Goods medium picture is inputted, establishes sample pattern, and by constantly training, correcting sample pattern；
S3：The goods picture of captured in real-time is imported, by sample pattern automatic recognition cargo medium species, and exports result.
The goods of different medium attribute includes building waste, stone, sand, dry loess, soil, mud and dress in foregoing S1
Repair rubbish.In foregoing S2, the training of sample pattern can be completed by server.In foregoing S2, server completes sample pattern
Training after, the algorithm of sample pattern can be migrated to vehicle intelligent terminal and/or vehicle-mounted external equipment.
In use, by the goods picture input server that largely marked different medium attribute, training server is to each
Mark is identified in the picture of kind media property, and server establishes sample pattern, and by constantly training, correcting sample pattern；
The algorithm of sample pattern is migrated in vehicle-mounted external equipment afterwards, passes through vehicle-mounted external equipment（The sharp bright C2 types of Streamax/ are red
Outer anti-riot dome-type high definition vehicle-mounted vidicon）The goods picture in container is shot, passes through the sample being implanted into vehicle-mounted external equipment
The algorithm of model, automatic identification is carried out to goods picture, draws thing medium species, and export recognition result.
Embodiment 2
The present embodiment also provides a kind of system based on deep learning technology identification goods medium species, as shown in Fig. 2 foregoing system
System includes deep learning module, the input connection vehicle-mounted external equipment of deep learning module, and its output end connection result receives
Device；Vehicle-mounted external equipment is used to shoot real-time item picture, and deep learning module exports recognition result after receiving goods picture
For which kind of medium, as a result reception device reception medium information.
Or as shown in figure 3, aforementioned system includes deep learning module, the input connection of deep learning module is vehicle-mounted
External device, its output end connection server, server connection result reception device；Vehicle-mounted external equipment is used to shoot real-time goods
Which kind of medium thing picture, deep learning module export recognition result as after receiving goods picture and send medium information to service
Medium information can be transmitted to result reception device by device, server.
Or as shown in figure 4, aforementioned system includes deep learning module, the input connection of deep learning module is vehicle-mounted
Intelligent terminal, vehicle intelligent terminal connection vehicle-mounted external equipment, the output end connection result reception device of deep learning module；Car
Carry external device to be used to shoot real-time item picture and send picture to vehicle intelligent terminal, vehicle intelligent terminal is by goods figure
Piece is forwarded to deep learning module, and deep learning module, which receives, exports recognition result is which kind of medium and by medium after goods picture
Information is sent to result reception device.
Aforementioned depth study module can be embedded in server and/or vehicle intelligent terminal is used.Aforementioned result receives dress
It is set to mobile intelligent terminal.Foregoing mobile intelligent terminal includes smart mobile phone, pc or PAD.
The present embodiment carries out key words sorting by inputting a large amount of different types of solid waste pictures, and to picture, such as：
Building waste, stone, sand, dry loess, soil and mud, using existing artificial nerve network model, based on deep learning
Technology is handled substantial amounts of solid waste picture, establishes sample pattern, so as to import the solid waste of captured in real-time
During thing picture, system is given up according to the media type of discarded object in existing sample pattern automatic identification picture so as to reach solid
Gurry is intelligent, the purpose of mechanized classification.Be advantageous to the recycling of building waste, to building castoff treatment and use row
Industry plays the effect of guiding and supervising.
The technological thought of above example only to illustrate the invention, it is impossible to protection scope of the present invention is limited with this, it is every
According to technological thought proposed by the present invention, any change done on the basis of technical scheme, the scope of the present invention is each fallen within
Within.
Claims (10)
- A kind of 1. method based on deep learning technology identification goods medium species, it is characterised in that comprise the following specific steps that：S1：Collect the goods picture for collecting and marked different medium attribute；S2：Goods medium picture is inputted, establishes sample pattern, and by constantly training, correcting sample pattern；S3：The goods picture of captured in real-time is imported, by sample pattern automatic recognition cargo medium species, and exports result.
- 2. the method according to claim 1 based on deep learning technology identification goods medium species, it is characterised in that institute Stating the goods of different medium attribute in S1 includes building waste, stone, sand, dry loess, soil, mud and decoration garbage.
- 3. the method according to claim 1 based on deep learning technology identification goods medium species, it is characterised in that institute State in S2, the training of sample pattern can be completed by server.
- 4. the method according to claim 3 based on deep learning technology identification goods medium species, it is characterised in that institute State in S2, after server completes the training of sample pattern, the algorithm of sample pattern can be migrated to vehicle intelligent terminal and/or car Carry external device.
- 5. the system according to claim 1 based on deep learning technology identification goods medium species, it is characterised in that institute Stating system includes deep learning module, the input connection vehicle-mounted external equipment of the deep learning module, the connection of its output end As a result reception device；The vehicle-mounted external equipment is used to shoot real-time item picture, and the deep learning module receives goods figure Export which kind of medium recognition result is after piece, the result reception device receives medium information.
- 6. the system according to claim 1 based on deep learning technology identification goods medium species, it is characterised in that institute Stating system includes deep learning module, the input connection vehicle-mounted external equipment of the deep learning module, the connection of its output end Server, the server connection result reception device；The vehicle-mounted external equipment is used to shoot real-time item picture, the depth Which kind of medium degree study module exports recognition result as after receiving goods picture and sends medium information to server, the clothes Medium information can be transmitted to result reception device by business device.
- 7. the system according to claim 1 based on deep learning technology identification goods medium species, it is characterised in that institute Stating system includes deep learning module, the input connection vehicle intelligent terminal of the deep learning module, the vehicle intelligent Terminal connects vehicle-mounted external equipment, the output end connection result reception device of the deep learning module；The vehicle-mounted external is set It is ready for use on shooting real-time item picture and sends picture to vehicle intelligent terminal, the vehicle intelligent terminal turns goods picture It is sent to deep learning module, the deep learning module, which receives, exports recognition result is which kind of medium and by medium after goods picture Information is sent to result reception device.
- 8. the system according to claim 5 based on deep learning technology identification goods medium species, it is characterised in that institute State that deep learning module can be embedded in server and/or vehicle intelligent terminal is used.
- 9. the system based on deep learning technology identification goods medium species according to claim 5,6 or 7, its feature exist In the result reception device is mobile intelligent terminal.
- 10. the system according to claim 9 based on deep learning technology identification goods medium species, it is characterised in that The mobile intelligent terminal includes smart mobile phone, pc or PAD.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711364690.9A CN107886136A (en) | 2017-12-18 | 2017-12-18 | A kind of method based on depth learning technology identification goods medium species |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711364690.9A CN107886136A (en) | 2017-12-18 | 2017-12-18 | A kind of method based on depth learning technology identification goods medium species |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107886136A true CN107886136A (en) | 2018-04-06 |
Family
ID=61771798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711364690.9A Pending CN107886136A (en) | 2017-12-18 | 2017-12-18 | A kind of method based on depth learning technology identification goods medium species |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107886136A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647671A (en) * | 2018-06-28 | 2018-10-12 | 武汉市哈哈便利科技有限公司 | A kind of optical indicia visual identity method and the self-service cabinet based on this method |
| CN108734435A (en) * | 2018-08-14 | 2018-11-02 | 惠龙易通国际物流股份有限公司 | Source of goods automatic auditing method based on big data analysis and system |
| CN108940919A (en) * | 2018-06-14 | 2018-12-07 | 华东理工大学 | Garbage classification machine people based on wireless transmission and deep learning |
| CN109493091A (en) * | 2018-11-07 | 2019-03-19 | 合肥友高物联网标识设备有限公司 | A kind of method and system of the physical security based on two dimensional code |
| CN110443119A (en) * | 2019-06-25 | 2019-11-12 | 中车工业研究院有限公司 | Cargo state recognition methods and device in compartment |
| CN111151368A (en) * | 2020-01-09 | 2020-05-15 | 珠海格力电器股份有限公司 | Garbage treatment method, system, storage medium and garbage treatment equipment |
| US10832435B1 (en) | 2019-04-26 | 2020-11-10 | Caterpillar Inc. | Determining payload carrier volume using a neural network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150189239A1 (en) * | 2013-12-27 | 2015-07-02 | Nuctech Company Limited | Fluoroscopic inspection method, device and storage medium for automatic classification and recognition of cargoes |
| CN105013718A (en) * | 2015-07-31 | 2015-11-04 | 泉州装备制造研究所 | System for sorting blocky solid construction waste on basis of multiple detection modes |
| CN105575119A (en) * | 2015-12-29 | 2016-05-11 | 大连楼兰科技股份有限公司 | Road condition climate deep learning and recognition method and apparatus |
| JP6209717B1 (en) * | 2016-06-02 | 2017-10-11 | サインポスト株式会社 | Information processing system, information processing method, and program |
-
2017
- 2017-12-18 CN CN201711364690.9A patent/CN107886136A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150189239A1 (en) * | 2013-12-27 | 2015-07-02 | Nuctech Company Limited | Fluoroscopic inspection method, device and storage medium for automatic classification and recognition of cargoes |
| CN105013718A (en) * | 2015-07-31 | 2015-11-04 | 泉州装备制造研究所 | System for sorting blocky solid construction waste on basis of multiple detection modes |
| CN105575119A (en) * | 2015-12-29 | 2016-05-11 | 大连楼兰科技股份有限公司 | Road condition climate deep learning and recognition method and apparatus |
| JP6209717B1 (en) * | 2016-06-02 | 2017-10-11 | サインポスト株式会社 | Information processing system, information processing method, and program |
Non-Patent Citations (2)
| Title |
|---|
| OLIVER BATCHELOR: "Object Recognition by Stochastic Metric Learning" * |
| 周 慧: "危化品公路运输 调度优化与预警技术研究" * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108940919A (en) * | 2018-06-14 | 2018-12-07 | 华东理工大学 | Garbage classification machine people based on wireless transmission and deep learning |
| CN108647671A (en) * | 2018-06-28 | 2018-10-12 | 武汉市哈哈便利科技有限公司 | A kind of optical indicia visual identity method and the self-service cabinet based on this method |
| CN108647671B (en) * | 2018-06-28 | 2023-12-22 | 武汉市哈哈便利科技有限公司 | Optical identification visual identification method and unmanned sales counter based on same |
| CN108734435A (en) * | 2018-08-14 | 2018-11-02 | 惠龙易通国际物流股份有限公司 | Source of goods automatic auditing method based on big data analysis and system |
| CN109493091A (en) * | 2018-11-07 | 2019-03-19 | 合肥友高物联网标识设备有限公司 | A kind of method and system of the physical security based on two dimensional code |
| US10832435B1 (en) | 2019-04-26 | 2020-11-10 | Caterpillar Inc. | Determining payload carrier volume using a neural network |
| CN110443119A (en) * | 2019-06-25 | 2019-11-12 | 中车工业研究院有限公司 | Cargo state recognition methods and device in compartment |
| CN111151368A (en) * | 2020-01-09 | 2020-05-15 | 珠海格力电器股份有限公司 | Garbage treatment method, system, storage medium and garbage treatment equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107886136A (en) | A kind of method based on depth learning technology identification goods medium species | |
| CN106960154A (en) | A kind of rogue program dynamic identifying method based on decision-tree model | |
| CN102489450B (en) | Intelligent separation method and system | |
| CN104063686B (en) | Crop leaf diseases image interactive diagnostic system and method | |
| CN107358194A (en) | A kind of violence sorting express delivery determination methods based on computer vision | |
| CN104156729B (en) | A kind of classroom demographic method | |
| CN110844403B (en) | Garbage classification identification method and device based on target identification | |
| CN105184812A (en) | Target tracking-based pedestrian loitering detection algorithm | |
| CN107798304B (en) | Method for rapidly auditing video | |
| CN108985363A (en) | A kind of cracks in reinforced concrete bridge classifying identification method based on RBPNN | |
| CN111340443A (en) | Refined garbage classification full-flow processing method | |
| CN107704882A (en) | A kind of kinds of laundry recognition methods and system based on digital image processing techniques | |
| CN110110732A (en) | A kind of intelligence inspection algorithm for kitchen after food and drink | |
| CN110427869A (en) | A kind of distal end visual selection recognition methods for garbage disposal | |
| CN109509299A (en) | A kind of automatic vending machine with recognition of face | |
| CN103544692A (en) | Blind detection method for tamper with double-compressed JPEG (joint photographic experts group) images on basis of statistical judgment | |
| CN103824074A (en) | Crowd density estimation method based on background subtraction and texture features and system | |
| CN107516315A (en) | A kind of development machine based on machine vision is slagged tap monitoring method | |
| CN105898722A (en) | Discriminating method and device for abnormal short messages, and electronic device | |
| CN113922950B (en) | Electronic waste recycling transaction system based on block chain technology | |
| CN110096952A (en) | A kind of artificial intelligence image processing system and its method for production line | |
| CN104780386B (en) | Video tamper detection method based on LOF Co Forest algorithms | |
| Li et al. | The machine recognition for population feature of wheat images based on BP neural network | |
| CN206991382U (en) | A kind of moving vehicle detection image processing system | |
| CN106623361B (en) | System and method for recycling urban waste |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180406 |
|
| WD01 | Invention patent application deemed withdrawn after publication |