CN105787506A - Method for assessing garbage classification based on image identification and two dimensional identification technology - Google Patents
Method for assessing garbage classification based on image identification and two dimensional identification technology Download PDFInfo
- Publication number
- CN105787506A CN105787506A CN201610056777.9A CN201610056777A CN105787506A CN 105787506 A CN105787506 A CN 105787506A CN 201610056777 A CN201610056777 A CN 201610056777A CN 105787506 A CN105787506 A CN 105787506A
- Authority
- CN
- China
- Prior art keywords
- rubbish
- garbage
- image
- quick response
- response code
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Abstract
The invention discloses a method for assessing garbage classification based on image identification and two dimensional identification technology. The method includes the following steps: (A) conducting large sample garbage classification image identifying training on a convolutional neural network algorithm; (B) using the two dimensional code to identify and scan ownership of the garbage; (3) acquiring a garbage image to be identified; (D) using the convolutional neural network algorithm that has been trained by the garbage classification big sample data in step A to identify the garbage image: (E) outputting a classification result of different garbage images. According to the invention, the method determines the residents who own the garbage based on the two dimensional code technology, and identifies garbage images by adopting the convolutional neural network algorithm, which substantially increases efficiency of garbage processing and brings clean and comfortable living environment.
Description
Technical field
The present invention relates to technology of garbage disposal, particularly relate to the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology.
Background technology
Changing Urban Garbage into Resources utilizes can provide huge economic benefit for national economy, and thus brings living environment cleaning and comfortable social benefit.And waste resources recycling, most important link seeks to rubbish through separating, classifying, then it is used according to its characteristic according to the rubbish separated, conventional garbage classification simply simply depends on artificial, manual operation efficiency is low and easily makes mistakes, it is impossible to meet the high request of Changing Urban Garbage into Resources.
Summary of the invention
For solving above-mentioned technical problem, it is an object of the invention to provide the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology, resident's refuse classification situation is evaluated, evaluation result and handling hook, promotes forming of resident's refuse classification consciousness and custom with this.
The technical solution used in the present invention is:
Based on the refuse classification evaluation methodology of image recognition Yu Quick Response Code identification technology, comprise the following steps: convolutional neural networks algorithm is carried out large sample refuse classification image recognition training by (A);(B) ownership of Quick Response Code mark and scanning rubbish is utilized;(C) rubbish image to be identified is gathered;(D) use the convolutional neural networks algorithm after refuse classification big-sample data training in step A that rubbish image is identified;(E) classification results of the different rubbish image of output.
Further, the described refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology also includes the step (F) the classification results calculating garbage disposal expense according to output after identifying.
Further, described step E also calculates institute's sorting rubbish error rate, garbage disposal expense=garbage weight * (1+ sorting rubbish error rate) * coefficient * unit waste disposal fee in step F.
Further, the described refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology also includes step (G) and feeds back the classification results of rubbish with handling to affiliated resident family and administration section.
Further, described step B includes: posts on refuse bag or prints Quick Response Code, and then scanning Quick Response Code confirms the resident family of rubbish ownership.
Further, described step C includes: (C1) utilizes machinery to open refuse bag;(C2) refuse bag inside rubbish is taken pictures, image is carried out pretreatment;(C3) algorithm inputs pretreated rubbish photo.
Beneficial effects of the present invention:
The refuse classification evaluation methodology of the present invention determines, based on planar bar code technology, the resident family that rubbish belongs to, and adopt convolutional neural networks algorithm that rubbish image is identified, with two dimensional image directly inputting for network, decrease the calculating processes such as complex characteristic extraction and data reconstruction, input picture and topology of networks can have well identical, feature extraction and pattern classification carry out simultaneously, and produce in training, weights are shared can largely reduce network training parameter, the adaptability making network structure is higher, it is greatly improved the efficiency of garbage disposal, thus bring living environment cleaning and comfortable social benefit.
Accompanying drawing explanation
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described further.
Fig. 1 is the flow chart of refuse classification evaluation methodology of the present invention.
Detailed description of the invention
As it is shown in figure 1, be the present invention refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology, comprise the following steps:
(A) convolutional neural networks algorithm is carried out large sample refuse classification image recognition training;
(B) ownership of Quick Response Code mark and scanning rubbish is utilized;This step B includes: posts on refuse bag or prints Quick Response Code, and then scanning Quick Response Code confirms the resident family of rubbish ownership.
(C) rubbish image to be identified is gathered;This step C includes: (C1) utilizes machinery to open refuse bag;(C2) refuse bag inside rubbish is taken pictures, image is carried out pretreatment;(C3) algorithm inputs pretreated rubbish photo.
(D) use the convolutional neural networks algorithm after refuse classification big-sample data training in step A that rubbish image is identified;
(E) classification results and the sorting rubbish error rate of different rubbish image are exported.
(F) garbage disposal expense, this garbage disposal expense=garbage weight * (1+ sorting rubbish error rate) * coefficient * unit waste disposal fee are calculated according to the classification results of output after identifying.
(G) classification results of feedback rubbish gives affiliated resident family and administration section with handling.
Wherein, the structure of above-mentioned convolutional neural networks algorithm is the perceptron of a kind of multilamellar, and every layer is made up of two dimensional surface, and each plane is made up of multiple independent neurons, comprises some simple units and complexity unit in network, is designated as C unit and S unit respectively.C unit condenses together composition convolutional layer, and S unit condenses together composition down-sampling layer.Input picture carries out convolution by wave filter with being biased, N number of characteristic pattern (N value can be manually set) is produced at C layer, then Feature Mapping figure is through summation, weighted value and biasing, obtains the Feature Mapping figure of S layer again through an activation primitive (generally selecting Sigmoid function).According to being manually set C layer and the quantity of S layer, above work circulates successively and carries out.Finally, down-sampling and output layer to most afterbody connect entirely, obtain last output.
The process of convolution: (be input picture at C1 layer with the trainable wave filter fx image inputted that deconvolutes, convolutional layer input afterwards is then the convolution characteristic pattern of preceding layer), by an activation primitive (what generally use is Sigmoid function), then add a biasing bx, obtain convolutional layer Cx.
The process of sub sampling includes: m pixel (m is manually set) summation of every neighborhood becomes a pixel, then passes through scalar Wx+1 weighting, is further added by biasing bx+1, then passes through activation primitive Sigmoid and produces Feature Mapping figure.Mapping from a plane to next plane can be regarded as makes convolution algorithm, and S layer is considered as fuzzy filter, serves the effect of Further Feature Extraction.Spatial resolution between hidden layer and hidden layer is successively decreased, and number of planes every layer contained is incremented by, and so can be used for detecting more characteristic information.For sub sampling layer, have N number of input feature vector figure, just have N number of output characteristic figure, simply each characteristic pattern size obtain corresponding change, concrete operation is following formula such as, down(in formula) represent down-sampling function.
down()+))
And the training process of convolutional neural networks is: convolutional neural networks is inherently a kind of mapping being input to output, it can learn the mapping relations between substantial amounts of input and output, without the accurate mathematical expression formula between any input and output.By known pattern, convolutional network being trained, network is just provided with the mapping ability between inputoutput pair.What convolutional neural networks performed is that the tutor having supervision trains, so sample set is if the vector of (input vector, desirable output vector) is to composition by shape.Convolutional neural networks training algorithm is similar to BP algorithm, is broadly divided into 4 steps, and this 4 step is divided into two stages:
1. communication process forward
1) from sample set, read (X, Y), X is inputted network.
2) corresponding actual output Op is calculated.
In this stage, information converts from input layer through successively, is sent to output layer, the input weight matrix dot product with every layer, obtains output result:
Op=Fn(... (F2 (F1 (XpW (1)) W (2)) ...) W (n)).
2. the back-propagation stage
1) reality output and the difference of desirable output are calculated;
2) send out back propagation by minimum error and adjust weight matrix.
Convolutional neural networks is mainly used in identifying that displacement, convergent-divergent and other form distort indeformable two dimensional image.By the feature detection layer of convolutional neural networks by training, owing to the neuron weights on same characteristic plane are identical, so network can collateral learning, this special construction shared with local weight has the superiority of uniqueness in speech recognition and image processing method mask so that it is layout is more closely similar to biological neural network.The more general neutral net of convolutional neural networks has the following advantages in image recognition:
1) with two dimensional image directly inputting for network, the calculating processes such as complex characteristic extraction and data reconstruction are decreased.
2) input picture and topology of networks can have well identical.
3) feature extraction and pattern classification carry out simultaneously, and produce in training.
4) weights are shared and can largely be reduced network training parameter, are that the adaptability of network structure is higher.
The foregoing is only the preferred embodiments of the present invention, the present invention is not limited to above-mentioned embodiment, broadly falls within protection scope of the present invention as long as realizing the technical scheme of the object of the invention with essentially identical means.
Claims (6)
1. based on the refuse classification evaluation methodology of image recognition Yu Quick Response Code identification technology, it is characterised in that comprise the following steps: convolutional neural networks algorithm is carried out large sample refuse classification image recognition training by (A);(B) ownership of Quick Response Code mark and scanning rubbish is utilized;(C) rubbish image to be identified is gathered;(D) use the convolutional neural networks algorithm after refuse classification big-sample data training in step A that rubbish image is identified;(E) classification results of the different rubbish image of output.
2. the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology according to claim 1, it is characterised in that: it also includes step (F) and calculates garbage disposal expense according to the classification results of output after identifying.
3. the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology according to claim 2, it is characterized in that: described step E also calculates institute's sorting rubbish error rate, garbage disposal expense=garbage weight * (1+ sorting rubbish error rate) * coefficient * unit waste disposal fee in step F.
4. the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology according to claim 3, it is characterised in that: it also includes step (G) and feeds back the classification results of rubbish with handling to affiliated resident family and administration section.
5. the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology according to claim 1, it is characterised in that described step B includes: posts on refuse bag or prints Quick Response Code, then scanning Quick Response Code confirms the resident family of rubbish ownership.
6. the refuse classification evaluation methodology based on image recognition Yu Quick Response Code identification technology according to claim 5, it is characterised in that described step C includes: (C1) utilizes machinery to open refuse bag;(C2) refuse bag inside rubbish is taken pictures, image is carried out pretreatment;(C3) algorithm inputs pretreated rubbish photo.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610056777.9A CN105787506A (en) | 2016-01-26 | 2016-01-26 | Method for assessing garbage classification based on image identification and two dimensional identification technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610056777.9A CN105787506A (en) | 2016-01-26 | 2016-01-26 | Method for assessing garbage classification based on image identification and two dimensional identification technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105787506A true CN105787506A (en) | 2016-07-20 |
Family
ID=56402458
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610056777.9A Pending CN105787506A (en) | 2016-01-26 | 2016-01-26 | Method for assessing garbage classification based on image identification and two dimensional identification technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105787506A (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106373112A (en) * | 2016-08-31 | 2017-02-01 | 北京比特大陆科技有限公司 | Image processing method, image processing device and electronic equipment |
CN106874954A (en) * | 2017-02-20 | 2017-06-20 | 佛山市络思讯科技有限公司 | The method and relevant apparatus of a kind of acquisition of information |
CN106875061A (en) * | 2017-02-20 | 2017-06-20 | 佛山市络思讯科技有限公司 | Method and relevant apparatus that a kind of destination path determines |
CN107054936A (en) * | 2017-03-23 | 2017-08-18 | 广东数相智能科技有限公司 | A kind of refuse classification prompting dustbin and system based on image recognition |
CN108520193A (en) * | 2018-03-27 | 2018-09-11 | 康体佳智能科技(深圳)有限公司 | Quick Response Code identifying system based on neural network and recognition methods |
CN108932510A (en) * | 2018-08-20 | 2018-12-04 | 贵州宜行智通科技有限公司 | A kind of rubbish detection method and device |
WO2019000929A1 (en) * | 2017-06-30 | 2019-01-03 | 京东方科技集团股份有限公司 | Garbage sorting and recycling method, garbage sorting equipment, and garbage sorting and recycling system |
CN109299793A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
CN109299792A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
CN109299790A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
CN110059767A (en) * | 2019-04-28 | 2019-07-26 | 宿迁海沁节能科技有限公司 | One kind identifying classification processing deep learning method based on the super relevant rubbish of time convolution |
CN110119662A (en) * | 2018-03-29 | 2019-08-13 | 王胜春 | A kind of rubbish category identification system based on deep learning |
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 |
CN110210635A (en) * | 2019-06-05 | 2019-09-06 | 周皓冉 | A kind of intelligent classification recovery system that can identify waste |
CN110473130A (en) * | 2019-07-30 | 2019-11-19 | 五邑大学 | A kind of garbage classification evaluation method, device and storage medium based on deep learning |
US11881019B2 (en) | 2018-09-20 | 2024-01-23 | Cortexia Sa | Method and device for tracking and exploiting at least one environmental parameter |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102000688A (en) * | 2010-10-12 | 2011-04-06 | 贺俊 | Household garbage separate collection method and system |
CN201997485U (en) * | 2011-04-12 | 2011-10-05 | 朱建功 | Garbage auto-separation device |
US20140050397A1 (en) * | 2011-04-01 | 2014-02-20 | Envac Optibag Ab | Method and system for identifying waste containers based on pattern |
CN104624505A (en) * | 2015-01-16 | 2015-05-20 | 同济大学 | Waste plastic separating method and system based on image recognition |
CN204507883U (en) * | 2015-04-05 | 2015-07-29 | 西安航空学院 | A kind of garbage classification system |
CN104850858A (en) * | 2015-05-15 | 2015-08-19 | 华中科技大学 | Injection-molded product defect detection and recognition method |
-
2016
- 2016-01-26 CN CN201610056777.9A patent/CN105787506A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102000688A (en) * | 2010-10-12 | 2011-04-06 | 贺俊 | Household garbage separate collection method and system |
US20140050397A1 (en) * | 2011-04-01 | 2014-02-20 | Envac Optibag Ab | Method and system for identifying waste containers based on pattern |
CN201997485U (en) * | 2011-04-12 | 2011-10-05 | 朱建功 | Garbage auto-separation device |
CN104624505A (en) * | 2015-01-16 | 2015-05-20 | 同济大学 | Waste plastic separating method and system based on image recognition |
CN204507883U (en) * | 2015-04-05 | 2015-07-29 | 西安航空学院 | A kind of garbage classification system |
CN104850858A (en) * | 2015-05-15 | 2015-08-19 | 华中科技大学 | Injection-molded product defect detection and recognition method |
Non-Patent Citations (1)
Title |
---|
黄伦等: "智能垃圾分拣系统的模拟与实现", 《机电工程》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106373112A (en) * | 2016-08-31 | 2017-02-01 | 北京比特大陆科技有限公司 | Image processing method, image processing device and electronic equipment |
CN106373112B (en) * | 2016-08-31 | 2020-08-04 | 北京比特大陆科技有限公司 | Image processing method and device and electronic equipment |
CN106874954A (en) * | 2017-02-20 | 2017-06-20 | 佛山市络思讯科技有限公司 | The method and relevant apparatus of a kind of acquisition of information |
CN106875061A (en) * | 2017-02-20 | 2017-06-20 | 佛山市络思讯科技有限公司 | Method and relevant apparatus that a kind of destination path determines |
CN107054936A (en) * | 2017-03-23 | 2017-08-18 | 广东数相智能科技有限公司 | A kind of refuse classification prompting dustbin and system based on image recognition |
WO2019000929A1 (en) * | 2017-06-30 | 2019-01-03 | 京东方科技集团股份有限公司 | Garbage sorting and recycling method, garbage sorting equipment, and garbage sorting and recycling system |
CN109201514A (en) * | 2017-06-30 | 2019-01-15 | 京东方科技集团股份有限公司 | Waste sorting recycle method, garbage classification device and classified-refuse recovery system |
US11446706B2 (en) | 2017-06-30 | 2022-09-20 | Beijing Boe Technology Development Co., Ltd. | Trash sorting and recycling method, trash sorting device, and trash sorting and recycling system |
CN109201514B (en) * | 2017-06-30 | 2019-11-08 | 京东方科技集团股份有限公司 | Waste sorting recycle method, garbage classification device and classified-refuse recovery system |
CN108520193A (en) * | 2018-03-27 | 2018-09-11 | 康体佳智能科技(深圳)有限公司 | Quick Response Code identifying system based on neural network and recognition methods |
CN110119662A (en) * | 2018-03-29 | 2019-08-13 | 王胜春 | A kind of rubbish category identification system based on deep learning |
CN108932510A (en) * | 2018-08-20 | 2018-12-04 | 贵州宜行智通科技有限公司 | A kind of rubbish detection method and device |
CN109299790A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
CN109299792A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
CN109299793A (en) * | 2018-08-21 | 2019-02-01 | 杭州复杂美科技有限公司 | Garbage classification motivational techniques and device, equipment and storage medium |
US11881019B2 (en) | 2018-09-20 | 2024-01-23 | Cortexia Sa | Method and device for tracking and exploiting at least one environmental parameter |
CN110059767A (en) * | 2019-04-28 | 2019-07-26 | 宿迁海沁节能科技有限公司 | One kind identifying classification processing deep learning method based on the super relevant rubbish of time convolution |
CN110210635A (en) * | 2019-06-05 | 2019-09-06 | 周皓冉 | A kind of intelligent classification recovery system that can identify waste |
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 |
CN110473130A (en) * | 2019-07-30 | 2019-11-19 | 五邑大学 | A kind of garbage classification evaluation method, device and storage medium based on deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105787506A (en) | Method for assessing garbage classification based on image identification and two dimensional identification technology | |
CN105772407A (en) | Waste classification robot based on image recognition technology | |
CN111368896B (en) | Hyperspectral remote sensing image classification method based on dense residual three-dimensional convolutional neural network | |
CN109584248B (en) | Infrared target instance segmentation method based on feature fusion and dense connection network | |
CN111310862B (en) | Image enhancement-based deep neural network license plate positioning method in complex environment | |
CN107220657B (en) | A kind of method of high-resolution remote sensing image scene classification towards small data set | |
CN106682697A (en) | End-to-end object detection method based on convolutional neural network | |
CN113065558A (en) | Lightweight small target detection method combined with attention mechanism | |
CN106650690A (en) | Night vision image scene identification method based on deep convolution-deconvolution neural network | |
Ioannidis et al. | A recurrent graph neural network for multi-relational data | |
CN108549893A (en) | A kind of end-to-end recognition methods of the scene text of arbitrary shape | |
CN108463876A (en) | Simulation output is generated for sample | |
CN111444924B (en) | Method and system for detecting plant diseases and insect pests and analyzing disaster grade | |
CN112733950A (en) | Power equipment fault diagnosis method based on combination of image fusion and target detection | |
CN109711401A (en) | A kind of Method for text detection in natural scene image based on Faster Rcnn | |
CN111696092B (en) | Defect detection method and system based on feature comparison and storage medium | |
CN107016413A (en) | A kind of online stage division of tobacco leaf based on deep learning algorithm | |
CN111178121B (en) | Pest image positioning and identifying method based on spatial feature and depth feature enhancement technology | |
CN109815945A (en) | A kind of respiratory tract inspection result interpreting system and method based on image recognition | |
CN107103308A (en) | A kind of pedestrian's recognition methods again learnt based on depth dimension from coarse to fine | |
CN108764244A (en) | Potential target method for detecting area based on convolutional neural networks and condition random field | |
CN110490262A (en) | Image processing model generation method, image processing method, device and electronic equipment | |
CN107330907B (en) | A kind of MRF image partition methods of combination deep learning shape prior | |
CN112528782A (en) | Underwater fish target detection method and device | |
CN115937774A (en) | Security inspection contraband detection method based on feature fusion and semantic interaction |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination |