CN110697275A - Embedded intelligent garbage classification system and working method thereof - Google Patents
Embedded intelligent garbage classification system and working method thereof Download PDFInfo
- Publication number
- CN110697275A CN110697275A CN201910820985.5A CN201910820985A CN110697275A CN 110697275 A CN110697275 A CN 110697275A CN 201910820985 A CN201910820985 A CN 201910820985A CN 110697275 A CN110697275 A CN 110697275A
- Authority
- CN
- China
- Prior art keywords
- garbage
- module
- picture
- redis database
- depth model
- 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
Images
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
-
- 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
Abstract
The invention relates to an embedded intelligent garbage classification system and a working method thereof, wherein the classification system comprises a photographing module, a picture preprocessing module, a redis database module, a compression depth model and an actuator, a picture obtained by photographing by the photographing module is transmitted to the redis database module after passing through the picture preprocessing module, the compression depth model is connected with the redis database module and used for identifying garbage types according to garbage picture information, the actuator is connected with the redis database module, and the actuator receives garbage type information through the redis database module and realizes classified throwing of garbage through different overturning directions. According to the invention, the shooting module shoots the garbage picture and transmits the garbage picture to the database, the embedded local compression depth model reads the garbage picture from the database and calculates and predicts the garbage category, the prediction result data is returned within two seconds, and the accuracy is over 85%.
Description
Technical Field
The invention relates to the technical field of garbage classification, in particular to an embedded intelligent garbage classification system and a working method thereof.
Background
In the cities of Beijing, Hangzhou and the like, the garbage classification test has been carried out for 14 years without obvious practical effect, and although garbage cans with different colors are placed at the doorway of a residential area, various garbage mixed and freely-distributed degradable garbage bags are also used for containing other garbage.
The problem of accurate classification accuracy is not solved, and the processing capacity of the rear-end equipment cannot meet the processing requirement due to the secondary mixing of the front end and the secondary mixing of the transfer. The reward stimulation to residents and clearing persons is small, the active classification of the young generation cannot be actuated, and effective reward stimulation cannot be implemented in garbage classification in public places.
If the front end is not classified or mixed in the transferring process, the treatment efficiency and the pollution level of the rear end are exponentially increased, the household garbage is classified at the rear end, the transferring cost is extremely high, and the front end classification can reduce the quantity of the household garbage to be transferred by 30-50%. Therefore, the development of intelligent garbage classification is of great significance.
Although the deep learning model for garbage classification can be deployed at the cloud end, a 4G network is required to ensure normal communication, so that the hardware cost is increased, and the intelligent garbage can cannot work normally in an area with poor 4G network signals.
Disclosure of Invention
The invention aims to provide an embedded intelligent garbage classification system and a working method thereof, which are used for solving the problems that garbage classification in the prior art is difficult to implement and a cloud garbage classification system needs to be in a normal network.
The invention provides an embedded intelligent garbage classification system which comprises a photographing module, a picture preprocessing module, a redis database module, a compression depth model and an actuator, wherein the photographing module is connected with the picture preprocessing module, the picture preprocessing module is connected with the redis database module, a picture obtained by photographing through the photographing module is transmitted to the redis database module after passing through the picture preprocessing module, the redis database module is used for information caching and information exchange, the compression depth model is connected with the redis database module, the compression depth model is used for identifying garbage types according to garbage picture information, the actuator is connected with the redis database module, and the actuator receives garbage type information through the redis database module and realizes classified throwing of garbage through different overturning directions.
Further, the compression depth model is a mobileNet compression depth model.
The invention also provides a working method of the embedded intelligent garbage classification system, which comprises the following steps:
(1) the shooting module shoots junk pictures and transmits the pictures to the picture preprocessing module, and the pictures are processed into RGB pictures with 224 × 224 pixels through opencv;
(2) the picture preprocessing module writes the preprocessed picture into a redis database module through a redis interface;
(3) the compressed depth model reads the picture information from the redis database module, then starts calculation, returns the sub-classification prediction result after the calculation is finished, and maps the sub-classification result to the recoverable garbage and other garbage categories according to the user-defined dictionary;
(4) the compression depth model returns the predicted garbage category information through a redis database and transmits the predicted garbage category information to an actuator;
(5) and the executor turns over in the corresponding direction according to the garbage category information to realize classified throwing of the garbage.
Further, the compression depth model performs convolution operation by using stride 2, and the number of channels of a convolution kernel is equal to the number of channels of the input feature map.
Further, the compressed depth model is trained and predicted using the tenserflow deep learning framework.
Further, the user-defined dictionary, which defines garbage models for the recyclable category and garbage models for the other categories, is loaded into the condensed depth model prior to step (1).
The technical scheme of the invention has the beneficial effects that:
according to the invention, a shooting module is used for shooting a garbage picture and transmitting the garbage picture to a database, an embedded local compression depth model is used for reading the garbage picture from the database and calculating and predicting the garbage category, prediction result data is returned within two seconds, and the accuracy is over 85%;
the redis database based on the memory performs information caching and information exchange with other local modules, and can return a prediction result under the acceptable time and accuracy, so that the related cost of the network is reduced, the edge calculation is realized, and the normal service can be maintained when the network is abnormal.
Drawings
FIG. 1 is a schematic diagram of an embedded intelligent garbage classification system according to the present invention;
FIG. 2 is a schematic diagram of a desktop intelligent garbage classification machine for use in an embedded intelligent garbage classification system;
in the drawings, the components represented by the respective reference numerals are listed below:
1-shooting module, 2-actuator and 3-throwing port.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention.
As shown in fig. 1, the embedded intelligent garbage classification system of the invention includes a photographing module, a picture preprocessing module, a redis database module, a compression depth model and an actuator, wherein the photographing module is connected with the picture preprocessing module, the picture preprocessing module is connected with the redis database module, a picture obtained by photographing by the photographing module is transmitted to the redis database module after passing through the picture preprocessing module, the redis database module is used for information caching and information exchange, the compression depth model is connected with the redis database module, the compression depth model is used for identifying garbage categories according to garbage picture information, the actuator is connected with the redis database module, the actuator receives garbage category information through the redis database module and realizes classified throwing of garbage through different overturning directions, and the compression depth model is a mobileNet compression depth model.
The working method of the embedded intelligent garbage classification system comprises the following steps:
training a mobileNet model by utilizing a tensoflow frame on an intelligent garbage classification desktop with a GPU based on 50 ten thousand garbage picture samples;
loading a trained mobileNet model based on an embedded linux system such as raspberry Pi 3B + deployment tensorflow, redis and other related frames;
a user-defined dictionary is loaded into the compaction depth model, which defines the sub-classifications of trash mapped to recyclable categories of trash and other categories of trash, such as beverage bottles, cans, paper, glass, etc., as recyclable categories.
The shooting module 1 is arranged at a position 40cm above the actuator 2 (garbage turnover plate), shoots a garbage picture input through the throwing port 3, the relative position is shown in fig. 2, the garbage picture is transmitted to the picture preprocessing module, the picture is processed into an RGB picture with 224 × 224 pixels through opencv, and the preprocessed picture is written into a redis database through a redis interface;
the compressed depth model reads the picture information from the redis database module, then starts calculation, performs convolution operation by adopting stride 2, the number of channels of a convolution kernel is equal to that of the input feature map, returns a sub-classification prediction result after the calculation is completed, and maps the sub-classification result to the recoverable garbage and other garbage categories according to a user-defined dictionary;
and returning the predicted garbage category information through a redis database, controlling the overturning direction of the actuator according to the result, and enabling the garbage to fall into a corresponding garbage can to realize garbage classification.
In conclusion, the junk pictures are shot through the shooting module and transmitted to the database, the embedded local compression depth model reads the junk pictures from the database and calculates and predicts the junk types, prediction result data are returned within two seconds, and the accuracy is over 85%; the redis database based on the memory performs information caching and information exchange with other local modules, and can return a prediction result under the acceptable time and accuracy, so that the related cost of the network is reduced, the edge calculation is realized, and the normal service can be maintained when the network is abnormal.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (6)
1. The utility model provides an embedded intelligent waste classification system, its characterized in that, includes module of shooing, picture preprocessing module, redis database module, compression depth model and executor, the module of shooing connects picture preprocessing module, picture preprocessing module connects redis database module, the picture that the module of shooing was obtained is transmitted to redis database module behind picture preprocessing module, redis database module is used for information cache and information exchange, compression depth model connection redis database module, the compression depth model is used for discerning the rubbish classification according to rubbish picture information, the executor is connected redis database module, the executor receives rubbish classification information and realizes the categorised throw of rubbish through different upset directions through redis database module.
2. The embedded intelligent garbage classification system of claim 1 wherein the compression depth model is a mobileNet compression depth model.
3. A working method of an embedded intelligent garbage classification system is characterized by comprising the following steps:
(1) the shooting module shoots junk pictures and transmits the pictures to the picture preprocessing module, and the pictures are processed into RGB pictures with 224 × 224 pixels through opencv;
(2) the picture preprocessing module writes the preprocessed picture into a redis database module through a redis interface;
(3) the compressed depth model reads the picture information from the redis database module, then starts calculation, returns the sub-classification prediction result after the calculation is finished, and maps the sub-classification result to the recoverable garbage and other garbage categories according to the user-defined dictionary;
(4) the compression depth model returns the predicted garbage category information through a redis database and transmits the predicted garbage category information to an actuator;
(5) and the executor turns over in the corresponding direction according to the garbage category information to realize classified throwing of the garbage.
4. The operating method of the embedded intelligent garbage classification system according to claim 3, wherein the compression depth model performs convolution operation by stride 2, and the number of channels of the convolution kernel is equal to the number of channels of the input feature map.
5. The method of claim 3, wherein the compressed depth model is trained and predicted using a tensoflow deep learning framework.
6. The method of claim 3, wherein the user-defined dictionary defining garbage models for recoverable categories and garbage models for other categories is loaded into the compaction depth model prior to step (1).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910820985.5A CN110697275A (en) | 2019-09-01 | 2019-09-01 | Embedded intelligent garbage classification system and working method thereof |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910820985.5A CN110697275A (en) | 2019-09-01 | 2019-09-01 | Embedded intelligent garbage classification system and working method thereof |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110697275A true CN110697275A (en) | 2020-01-17 |
Family
ID=69194115
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910820985.5A Pending CN110697275A (en) | 2019-09-01 | 2019-09-01 | Embedded intelligent garbage classification system and working method thereof |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110697275A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111709477A (en) * | 2020-06-16 | 2020-09-25 | 浪潮集团有限公司 | Method and tool for garbage classification based on improved MobileNet network |
CN113291661A (en) * | 2021-05-27 | 2021-08-24 | 浙江金实乐环境工程有限公司 | Automatic detect domestic waste collection box and system that letter sorting was classifyed |
CN113562356A (en) * | 2021-08-20 | 2021-10-29 | 阿尔飞思(昆山)智能物联科技有限公司 | Garbage classification box and classification method |
-
2019
- 2019-09-01 CN CN201910820985.5A patent/CN110697275A/en active Pending
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111709477A (en) * | 2020-06-16 | 2020-09-25 | 浪潮集团有限公司 | Method and tool for garbage classification based on improved MobileNet network |
CN113291661A (en) * | 2021-05-27 | 2021-08-24 | 浙江金实乐环境工程有限公司 | Automatic detect domestic waste collection box and system that letter sorting was classifyed |
CN113562356A (en) * | 2021-08-20 | 2021-10-29 | 阿尔飞思(昆山)智能物联科技有限公司 | Garbage classification box and classification method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110697275A (en) | Embedded intelligent garbage classification system and working method thereof | |
CN110210635A (en) | A kind of intelligent classification recovery system that can identify waste | |
CN204979922U (en) | Waste classification operation management system | |
CN105035600A (en) | Rubbish classification operation and management system and method | |
CN110119662A (en) | A kind of rubbish category identification system based on deep learning | |
CN111846693A (en) | Intelligent garbage classification putting supervision and control system based on real-name system | |
CN105872075B (en) | A method of internet of things equipment is mapped to smart city resource model | |
CN108009497A (en) | Image recognition monitoring method, system, computing device and readable storage medium storing program for executing | |
Hannan et al. | An automated solid waste bin level detection system using Gabor wavelet filters and multi-layer perception | |
CN111814750A (en) | Intelligent garbage classification method and system based on deep learning target detection and image recognition | |
CN105600218A (en) | Trash classification and intelligent recovery system and method | |
CN110659622A (en) | Detection method, device and system for garbage dumping | |
GB2402501A (en) | Method for controlling the performance of a target system | |
CN110956104A (en) | Method, device and system for detecting overflow of garbage can | |
CN101383814B (en) | Device and method implementing data access based on connection pool | |
CN109740553A (en) | A kind of image, semantic segmentation data screening method and system based on identification | |
CN103679534A (en) | Method of collecting and transporting household waste using two dimensional code interaction principle | |
CN105335758A (en) | Model identification method based on video Fisher vector descriptors | |
CN108957485A (en) | A kind of integrated managing and control system of satellite remote sensing picture processing transmission of application drive | |
CN107563939A (en) | A kind of environment sanitary cleaning vehicle remote control transfers method | |
CN111661509B (en) | Garbage recycling and classifying management system based on interaction of Internet and Internet of things | |
CN115457411A (en) | Unmanned inspection method and device for aviation oil pipeline and aviation oil safe operation and maintenance system | |
CN111806923B (en) | Intelligent resource recovery system | |
CN115330720A (en) | Unmanned aerial vehicle image earth surface solid waste detection model and detection method | |
CN114529834A (en) | Small target object detection method and detection device |
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: 20200117 |
|
WD01 | Invention patent application deemed withdrawn after publication |