CN110803406A - Intelligent classification dustbin based on degree of depth study - Google Patents
Intelligent classification dustbin based on degree of depth study Download PDFInfo
- Publication number
- CN110803406A CN110803406A CN201910993388.2A CN201910993388A CN110803406A CN 110803406 A CN110803406 A CN 110803406A CN 201910993388 A CN201910993388 A CN 201910993388A CN 110803406 A CN110803406 A CN 110803406A
- Authority
- CN
- China
- Prior art keywords
- garbage
- power supply
- module
- supply equipment
- classification
- 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
- B65F1/0053—Combination of several 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
- B65F1/14—Other constructional features; Accessories
-
- 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
- B65F2210/128—Data transmitting 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
- B65F2210/172—Solar cells
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Processing Of Solid Wastes (AREA)
Abstract
The invention discloses an intelligent classification dustbin based on deep learning, which comprises a dustbin body structure and an intelligent classification controller, wherein the dustbin body structure comprises a dustbin body, a rubbish containing chamber and a plurality of rubbish bins are arranged in the dustbin body, a movable bottom plate is arranged at the bottom of the containing chamber, when the movable bottom plate is opened, the containing chamber is communicated with a garbage bin positioned below the containing chamber, in an initial state, one garbage bin is positioned under the containing chamber, the movable bottom plate is in a closed state, the intelligent classification controller comprises an identification module, a control module, a wireless communication module, a power supply module, an APP module used for being installed in the intelligent terminal communication equipment, a garbage classification selection module and a driving module, the power supply module is connected with the control module, the control module is respectively connected with the identification module, the garbage classification selection module, the driving module and the wireless communication module, and the wireless communication module is connected with the APP module; the garbage classification method has the advantages of being capable of achieving automatic garbage classification, high in garbage classification accuracy and low in cost.
Description
Technical Field
The invention relates to a classification dustbin, in particular to an intelligent classification dustbin based on deep learning.
Background
China has a large population, and the quantity of the daily garbage generated every year is quite large. Public places are one of the main places for generating domestic garbage at present. At present, the conventional dustbin is mainly adopted by the dustbin arranged on the public places such as the roadside and the square. Traditional dustbin is including two garbage bins of parallel setting, and one of them garbage bin is used for collecting recoverable rubbish, and another garbage bin is used for collecting non-recoverable rubbish, and two garbage bins are marked with "recoverable rubbish" and "non-recoverable rubbish" tip words respectively, and the pedestrian confirms through the tip words that rubbish drops into in which garbage bin. However, although the pedestrian can clearly distinguish the two trash cans in the conventional trash bin, the pedestrian determines the trash type by subjective awareness when the trash is classified, and when the trash type is determined incorrectly, the trash is thrown into the wrong trash can. Because the cognition degree difference of people to garbage classification is very big at present, the garbage classification phenomenon often takes place, and traditional dustbin is not obvious to the effect of garbage classification from this.
With the development of the internet of things technology, the intelligent garbage can is also widely researched. The invention relates to an intelligent garbage can named BigBe11y invented by American BigBelly Solar corporation, which integrates Solar energy, an Internet of things and a high-efficiency compressor, wherein a Solar cell arranged at the top of the garbage can provides power for the electric appliance part of the garbage can to work, when the garbage can is quickly filled, the volume of garbage in the garbage can is compressed within 40 seconds through the high-efficiency compressor, and when the garbage in the garbage can is difficult to be compressed again, the garbage can is automatically networked to send the full information of the garbage can and the geographical position information of the garbage can to a garbage disposal center, and the garbage disposal center can recycle the garbage after receiving the information. Although this intelligent garbage bin intelligent degree is very high, not only can improve the timeliness of rubbish recovery and the utilization ratio of garbage bin, can also reduce the manpower and materials cost of rubbish recovery, but this garbage bin does not solve the unsafe problem of waste classification to the cost of manufacture is also very high.
Disclosure of Invention
The invention aims to solve the technical problem of providing an intelligent classification garbage can based on deep learning, which is high in garbage classification accuracy and low in cost.
The technical scheme adopted by the invention for solving the technical problems is as follows: an intelligent classification dustbin based on degree of depth learning, includes dustbin major structure and intelligent classification controller. The garbage can main body structure comprises a can body, a garbage containing chamber used for containing garbage thrown into the outside and a plurality of garbage bins used for containing classified garbage are arranged in the can body, the garbage bins are positioned below the garbage containing chamber and distributed along a circle, a movable bottom plate is arranged at the bottom of the containing chamber, when the movable bottom plate is opened, the containing chamber is communicated with the garbage bins positioned below the containing chamber, in an initial state, one garbage bin is positioned right below the containing chamber, and the movable bottom plate is in a closed state; the intelligent classification controller comprises an identification module, a control module, a wireless communication module, a power supply module, an APP module, a garbage classification selection module and a driving module, wherein the APP module, the garbage classification selection module and the driving module are installed in intelligent terminal communication equipment; the wireless communication module is used for realizing data interaction between the control module and the APP module, the power supply module is used for providing working voltage for the intelligent classification dustbin under the control of the control module, the identification module is used for identifying the type of garbage thrown into the garbage containing chamber and generating an identification result to be sent to the control module, the control module converts the received identification result into a classification signal which can be identified by the garbage classification selection module and sends the classification signal to the garbage classification selection module, the garbage classification selection module receives the classification signal sent by the control module and throws the garbage into a corresponding garbage bin after selecting the corresponding garbage bin according to the classification signal, the APP module is used for generating a signal for controlling the box body position to move and transmitting the signal to the control module through the wireless communication module, the control module generates mobile data which can be identified by the driving module based on the signal and sends the mobile data to the driving module, and the driving module drives the box body to move according to the received mobile data.
The identification module comprises an image acquisition module and a neural network identification module, the image acquisition module is used for acquiring images of garbage thrown into the garbage storage chamber in real time and transmitting the acquired images of the garbage to the neural network identification module for garbage type identification, the neural network identification module identifies the garbage images by using a convolutional neural network model and then determines the garbage types, the convolutional neural network model comprises 8 convolutional layers, 4 maximum pooling layers, 2 full-connection layers and one classification layer, the 8 convolutional layers are respectively called a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer, a fifth convolutional layer, a sixth convolutional layer, a seventh convolutional layer and an eighth convolutional layer, the 4 maximum pooling layers are respectively called a first maximum pooling layer, a second maximum pooling layer, a third maximum pooling layer and a fourth maximum pooling layer, the method comprises the following steps of referring 2 full-connection layers as a first full-connection layer and a second full-connection layer, wherein the activation functions of each convolution layer and each full-connection layer are Relu functions, the activation function of a classification layer is a Softmax function, an image acquisition module inputs acquired garbage images into a first convolution layer, the first convolution layer is used for carrying out first convolution filtering on the input garbage images to obtain initial garbage features, the second convolution layer is used for carrying out second convolution filtering on the initial garbage features to obtain second garbage features, a first maximum pooling layer is used for carrying out pooling sampling on the second garbage features to obtain first pooling sampling features, a third convolution layer is used for carrying out third convolution filtering on the first pooling sampling features to obtain third garbage features, and a fourth convolution layer is used for carrying out fourth convolution filtering on the third garbage features, obtaining a fourth garbage feature, wherein the second largest pooling layer is used for performing pooling sampling on the fourth garbage feature to obtain a second pooling sampling feature, the fifth convolutional layer is used for performing fifth convolution filtering on the second pooling sampling feature to obtain a fifth garbage feature, the sixth convolutional layer is used for performing sixth convolution filtering on the fifth garbage feature to obtain a sixth garbage feature, the third largest pooling layer is used for performing pooling sampling on the sixth garbage feature to obtain a third pooling sampling feature, the seventh convolutional layer is used for performing seventh convolution filtering on the third pooling sampling feature to obtain a seventh garbage feature, the eighth convolutional layer is used for performing eighth convolution filtering on the seventh garbage feature to obtain an eighth garbage feature, and the fourth largest pooling layer is used for performing pooling sampling on the eighth garbage feature, and obtaining a fourth pooling sampling feature, wherein the first full connection layer is used for carrying out first feature extraction on the fourth pooling sampling feature to obtain a deep garbage feature, the second full connection layer is used for carrying out second feature extraction on the deep garbage feature to obtain a final garbage feature, and the classification layer is used for outputting a classification result to the control module after the final garbage feature is identified and classified. In the structure, a plurality of convolution layers in the neural network identification module continuously extract deep features of the image, then the pooling layers are utilized to reduce network parameters, garbage category identification is realized, the automation degree is high, and classification is accurate.
The power supply module comprises a first power supply line, a second power supply line and a power supply selection module, the first power supply line adopts a 220V alternating current power supply mode, the second power supply line comprises a solar cell panel, first power supply equipment and second power supply equipment, and the first power supply equipment and the second power supply equipment are respectively realized by storage batteries; the power supply selection equipment divides 24 hours a day into two working time periods, namely a daytime working time period and a nighttime working time period; in the working time period in the daytime, the power supply selection equipment acquires weather information, if the weather is other weather except sunny, the first power supply line is selected for supplying power, if the weather is sunny, the second power supply line is preferentially adopted for supplying power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20%, the power supply selection equipment is switched to select the first power supply line for supplying power, the solar panel is controlled to sequentially charge the first power supply equipment and the second power supply equipment, if the electric quantity of the first power supply equipment is less than 20% and the electric quantity of the second power supply equipment is more than 20%, the second power supply equipment is adopted for supplying power, and the solar panel is controlled to charge the first power supply equipment, if the electric quantity of the first power supply equipment exceeds 20% and the electric quantity of the second power supply equipment is less than 20%, adopting the first power supply equipment to supply power and controlling the solar panel to charge the second power supply equipment, if the electric quantities of the first power supply equipment and the second power supply equipment both exceed 20%, preferentially adopting the first power supply equipment to supply power and judging whether the electric quantity of the second power supply equipment is full, and if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment is not full, controlling the solar panel to charge the second power supply equipment; in the night working time period, the power supply selection equipment preferentially adopts the second power supply circuit to supply power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20 percent, switching to the first power supply line for supplying power, if the electric quantity of the first power supply equipment is less than 20 percent and the electric quantity of the second power supply equipment exceeds 20 percent, the second power supply equipment is adopted to supply power, if the electric quantity of the first power supply equipment exceeds 20 percent and the electric quantity of the second power supply equipment is less than 20 percent, the first power supply equipment is adopted to supply power, and if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment both exceed 20%, preferentially adopting the first power supply equipment to supply power. This structure gathers, and power module adopts two kinds of power supply lines to realize, has guaranteed 24 hours uninterrupted power supply of dustbin, avoids dustbin outage stop work, sets up 20% electric quantity simultaneously and is the threshold value electric quantity, prevents that battery electric quantity from crossing low, influences battery life-span.
The garbage classification selection module comprises a garbage can scheduling module and a garbage can selection module, the garbage can selection module is connected with the control module, the garbage can selection module receives a garbage classification signal sent by the control module, and calculates to obtain a required rotation angle of a garbage can corresponding to the type of the garbage to be remained according to the type of the garbage can below the current garbage containing chamber, and sends the required rotation angle to the garbage can scheduling module, the garbage can scheduling module controls the garbage bin to rotate, the corresponding garbage bin rotates to the position under the garbage containing chamber, and then the bottom plate of the garbage containing chamber is driven to open to enable the garbage to fall into the corresponding garbage bin and then to be placed on the closed bottom plate, so that the garbage classification putting is completed.
The driving module comprises a chassis which is arranged at the bottom of the box body and can move up and down under stress, four pulleys which are arranged at the bottom of the box body, a lifting mechanism for driving the chassis to move up and down and a rotating mechanism for driving the pulleys to rotate, wherein the lifting mechanism and the rotating mechanism are respectively connected with the control module, four openings are arranged on the chassis, the chassis is positioned at the lowest position in an initial state, the four pulleys are embedded into the four openings in a one-to-one correspondence manner, at the moment, the box body cannot move, when the control module sends moving data to the driving module, firstly the lifting mechanism controls the chassis to move upwards to the highest position, at the moment, the four pulleys are separated from the chassis, and then the rotating mechanism drives the four pulleys to correspondingly rotate according to the moving data, the box body is moved, when the box body is moved in place, the lifting mechanism controls the base plate to descend to the lowest position, the four pulleys enter the base plate again, and the box body is fixed in position.
Compared with the prior art, the invention has the advantages that the intelligent classification garbage can is formed by the garbage can main body structure and the intelligent classification controller, the garbage can main body structure comprises a box body, a garbage containing chamber for containing externally thrown-in garbage and a plurality of garbage bins for containing classified garbage are arranged in the box body, the garbage bins are positioned below the garbage containing chamber and distributed along a circle, the bottom of the garbage containing chamber is a movable bottom plate, when the movable bottom plate is opened, the garbage containing chamber is communicated with the garbage bins positioned below the garbage containing chamber, in an initial state, one garbage bin is positioned right below the garbage containing chamber, the movable bottom plate is in a closed state, the intelligent classification controller comprises an identification module, a control module, a wireless communication module, a power supply module, an APP module for being installed in intelligent terminal communication equipment, a garbage classification selection module and a driving module, and the power supply module is connected with the control module, the control module is respectively connected with the identification module, the garbage classification selection module, the driving module and the wireless communication module, the wireless communication module is connected with the APP module, the wireless communication module is used for realizing data interaction between the control module and the APP module, the power supply module provides working voltage for the intelligent classification garbage can under the control of the control module, the identification module is used for identifying the type of garbage thrown into the garbage containing chamber and generating an identification result to be sent to the control module, the control module converts the received identification result into a classification signal which can be identified by the garbage classification selection module and sends the classification signal to the garbage classification selection module, the garbage classification selection module receives the classification signal sent by the control module and throws the garbage into a corresponding garbage bin after selecting the corresponding garbage bin according to the classification signal, and the APP module is used for generating a signal for controlling the position movement of the box body and transmitting the signal to the control module, the control module generates mobile data which can be identified by the driving module based on the signal and sends the mobile data to the driving module, and the driving module drives the box body to move according to the received mobile data.
Drawings
FIG. 1(a) is a schematic structural diagram of an intelligent classification garbage can of the present invention;
FIG. 1(b) is a top view of a garbage bin of the intelligent classification garbage can of the present invention;
FIG. 1(c) is a bottom view of the intelligent sorting bin of the present invention;
FIG. 2 is a block diagram of an intelligent classification controller of the intelligent classification garbage can according to the present invention;
FIG. 3 is a block diagram of the structure of the recognition module of the intelligent classification controller of the intelligent classification garbage can of the present invention;
FIG. 4 is a block diagram of the neural network recognition module of the intelligent classification controller of the intelligent classification garbage can according to the present invention;
FIG. 5 is a block diagram of a power supply module of the intelligent classification controller of the intelligent classification waste bin of the present invention;
FIG. 6 is a block diagram of a second power supply line of the intelligent classification controller of the intelligent classification dustbin of the present invention;
FIG. 7 is a connection block diagram of a wireless communication module and an APP module of an intelligent classification controller of the intelligent classification garbage can according to the present invention;
fig. 8 is a block diagram of a garbage classification selection module of the intelligent classification controller of the intelligent classification garbage can according to the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying examples.
Example (b): as shown in the figure, the intelligent classification dustbin based on deep learning comprises a dustbin main body structure and an intelligent classification controller. The garbage can main body structure comprises a can body 1, a garbage containing chamber for containing externally thrown-in garbage and a plurality of garbage bins 2 for containing classified garbage are arranged in the can body 1, the garbage bins 2 are positioned below the garbage containing chamber and distributed along a circle, a movable bottom plate is arranged at the bottom of the containing chamber, when the movable bottom plate is opened, the containing chamber is communicated with the garbage bins 2 positioned below the containing chamber, in an initial state, one garbage bin 2 is positioned right below the containing chamber, and the movable bottom plate is in a closed state; the intelligent classification controller comprises an identification module 3, a control module 4, a wireless communication module 5, a power supply module, an APP module for being installed in the intelligent terminal communication equipment, a garbage classification selection module 6 and a driving module 7, wherein the power supply module is connected with the control module 4, the control module 4 is respectively connected with the identification module 3, the garbage classification selection module 6, the driving module 7 and the wireless communication module 5, and the wireless communication module 5 is connected with the APP module; the wireless communication module 5 is used for realizing data interaction between the control module 4 and the APP module, the power supply module provides working voltage for the intelligent classification dustbin under the control of the control module 4, the recognition module 3 is used for recognizing the type of garbage thrown into the garbage containing chamber and generating a recognition result to send to the control module 4, the control module 4 converts the received recognition result into a classification signal recognizable by the garbage classification selection module 6 to send to the garbage classification selection module 6, the garbage classification selection module 6 receives the classification signal sent by the control module 4 and throws the garbage into the corresponding garbage bin 2 after selecting the corresponding garbage bin 2 according to the classification signal, the APP module is used for generating a signal for controlling the position movement of the dustbin body 1 and transmitting the signal to the control module 4 through the wireless communication module 5, the control module 4 generates mobile data recognizable by the drive module 7 based on the signal to send to the drive module 7, the driving module 7 drives the box body 1 to move according to the received movement data.
In this embodiment, the recognition module 3 includes an image collection module and a neural network recognition module, the image collection module is used for collecting images of garbage thrown into the garbage storage room in real time and transmitting the collected images of the garbage to the neural network recognition module for garbage type recognition, the neural network recognition module recognizes the garbage images by using a convolutional neural network model and then determines the garbage types, the convolutional neural network model includes 8 convolutional layers, 4 maximum pooling layers, 2 full-link layers and one classification layer, the 8 convolutional layers are respectively called a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer, a fifth convolutional layer, a sixth convolutional layer, a seventh convolutional layer and an eighth convolutional layer, the 4 maximum pooling layers are respectively called a first maximum pooling layer, a second maximum pooling layer, a third maximum pooling layer and a fourth maximum pooling layer, the method comprises the following steps of referring 2 full-connection layers as a first full-connection layer and a second full-connection layer, wherein the activation functions of each convolution layer and each full-connection layer are Relu functions, the activation function of a classification layer is a Softmax function, an image acquisition module inputs acquired garbage images into the first convolution layer, the first convolution layer is used for carrying out first convolution filtering on the input garbage images to obtain initial garbage features, the second convolution layer is used for carrying out second convolution filtering on the initial garbage features to obtain second garbage features, the first maximum pooling layer is used for carrying out pooling sampling on the second garbage features to obtain first pooled sampling features, the third convolution layer is used for carrying out third convolution filtering on the first pooled sampling features to obtain third garbage features, the fourth convolution layer is used for carrying out fourth convolution filtering on the third garbage features to obtain fourth garbage features, the second largest pooling layer is used for pooling sampling of fourth garbage features to obtain second pooling sampling features, the fifth convolutional layer is used for performing fifth convolution filtering on the second pooling sampling features to obtain fifth garbage features, the sixth convolutional layer is used for performing sixth convolution filtering on the fifth garbage features to obtain sixth garbage features, the third largest pooling layer is used for pooling sampling of the sixth garbage features to obtain third pooling sampling features, the seventh convolutional layer is used for performing seventh convolution filtering on the third pooling sampling features to obtain seventh garbage features, the eighth convolutional layer is used for performing eighth convolution filtering on the seventh garbage features to obtain eighth garbage features, the fourth largest pooling layer is used for pooling sampling of the eighth garbage features to obtain fourth pooling sampling features, and the first full-connection layer is used for performing first feature extraction on the fourth pooling sampling features, and the classification layer is used for identifying and classifying the final garbage features and outputting classification results to the control module 4. The neural network recognition module is realized by adopting a current mature model, a plurality of different garbage image templates and corresponding recognition results are input into the neural network recognition module for training before use, and the trained neural network recognition module can recognize similar garbage images through a self-learning function.
In this embodiment, the power supply module includes a first power supply line, a second power supply line and a power supply selection module, the first power supply line adopts a 220V alternating current power supply mode, the second power supply line includes a solar cell panel 8, a first power supply device and a second power supply device, and the first power supply device and the second power supply device are respectively realized by a storage battery 9; the power supply selection equipment divides 24 hours a day into two working time periods, namely a daytime working time period and a nighttime working time period; in the working time period in the daytime, the power supply selection equipment acquires weather information, if the weather is other weather except sunny days, the first power supply line is selected for supplying power, if the weather is sunny days, the second power supply line is preferentially adopted for supplying power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20%, the first power supply line is switched to be selected for supplying power, the solar panel 8 is controlled to charge the first power supply equipment and the second power supply equipment in sequence, if the electric quantity of the first power supply equipment is less than 20% and the electric quantity of the second power supply equipment is more than 20%, the second power supply equipment is adopted for supplying power, the solar panel 8 is controlled to charge the first power supply equipment, if the electric quantity of the first power, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment exceed 20%, preferentially adopting the first power supply equipment to supply power, judging whether the electric quantity of the second power supply equipment is full, and if not, controlling the solar panel 8 to charge the second power supply equipment; in the working time period at night, the power supply selection equipment preferentially selects a second power supply line to supply power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20%, the power supply is switched to the first power supply line, if the electric quantity of the first power supply equipment is less than 20%, and the electric quantity of the second power supply equipment is more than 20%, the second power supply equipment is adopted to supply power, if the electric quantity of the first power supply equipment is more than 20%, and the electric quantity of the second power supply equipment is less than 20%, the first power supply equipment is adopted to supply power, and if the electric quantity of the first power supply equipment and the electric quantity of the.
In this embodiment, the garbage classification selection module 6 includes a garbage can scheduling module and a garbage can selection module, the garbage can selection module is connected with the control module 4, the garbage can selection module receives the garbage classification signal sent by the control module 4, and calculates an angle at which the garbage can corresponding to the type of the garbage to be stored needs to rotate according to the type of the garbage can below the current garbage storage chamber, and sends the calculated angle to the garbage can scheduling module, the garbage can scheduling module controls the garbage bin 2 to rotate, the corresponding garbage bin 2 is rotated to a position right below the garbage storage chamber, and then the bottom plate of the garbage storage chamber is driven to open, so that the garbage falls into the corresponding garbage bin 2 and then is closed on the bottom plate, and classified garbage throwing is completed.
In this embodiment, the driving module 7 includes a chassis 10 disposed at the bottom of the box 1 and capable of moving up and down under stress, four pulleys 11 mounted at the bottom of the box 1, a lifting mechanism for driving the chassis 10 to move up and down and a rotating mechanism for driving the pulleys 11 to rotate, the lifting mechanism and the rotating mechanism are respectively connected with the control module 4, the chassis 10 is provided with four openings, in an initial state, the chassis 10 is at the lowest position, the four pulleys 11 are embedded into the four openings in a one-to-one correspondence manner, and at this time, the box 1 cannot move; when the staff sends the position movement instruction for wireless communication module through the APP module, wireless communication module sends the position movement instruction for control module 4, control module 4 generates according to the position movement instruction and removes when data send for drive module 7, at first elevating system control chassis 10 rebound to the highest position, four pulleys 11 break away from chassis 10 this moment, then four pulleys 11 of slewing mechanism drive rotate according to removing the data correspondence, realize the removal of box 1, when removing the back that targets in place, elevating system control chassis 10 descends to four pulleys 11 of lowest position and gets into chassis 10 once more, box 1 rigidity.
Claims (5)
1. The utility model provides an intelligent classification dustbin based on degree of depth study which characterized in that includes dustbin major structure and intelligent classification controller. The garbage can main body structure comprises a can body, a garbage containing chamber used for containing garbage thrown into the outside and a plurality of garbage bins used for containing classified garbage are arranged in the can body, the garbage bins are positioned below the garbage containing chamber and distributed along a circle, a movable bottom plate is arranged at the bottom of the containing chamber, when the movable bottom plate is opened, the containing chamber is communicated with the garbage bins positioned below the containing chamber, in an initial state, one garbage bin is positioned right below the containing chamber, and the movable bottom plate is in a closed state;
the intelligent classification controller comprises an identification module, a control module, a wireless communication module, a power supply module, an APP module, a garbage classification selection module and a driving module, wherein the APP module, the garbage classification selection module and the driving module are installed in intelligent terminal communication equipment;
the wireless communication module is used for realizing data interaction between the control module and the APP module, the power supply module is used for providing working voltage for the intelligent classification dustbin under the control of the control module, the identification module is used for identifying the type of garbage thrown into the garbage containing chamber and generating an identification result to be sent to the control module, the control module converts the received identification result into a classification signal which can be identified by the garbage classification selection module and sends the classification signal to the garbage classification selection module, the garbage classification selection module receives the classification signal sent by the control module and throws the garbage into a corresponding garbage bin after selecting the corresponding garbage bin according to the classification signal, the APP module is used for generating a signal for controlling the box body position to move and transmitting the signal to the control module through the wireless communication module, the control module generates mobile data which can be identified by the driving module based on the signal and sends the mobile data to the driving module, and the driving module drives the box body to move according to the received mobile data.
2. The intelligent classification garbage can based on deep learning of claim 1, wherein the recognition module comprises an image acquisition module and a neural network recognition module, the image acquisition module is used for acquiring images of garbage thrown into the garbage storage room in real time and transmitting the acquired images of the garbage to the neural network recognition module for garbage type recognition, the neural network recognition module utilizes a convolutional neural network model to determine garbage types after recognizing the garbage images, the convolutional neural network model comprises 8 convolutional layers, 4 maximum pooling layers, 2 full-link layers and one classification layer, the 8 convolutional layers are respectively called as a first convolutional layer, a second convolutional layer, a third convolutional layer, a fourth convolutional layer, a fifth convolutional layer, a sixth convolutional layer, a seventh convolutional layer and an eighth convolutional layer, the method comprises the steps of respectively calling 4 maximum pooling layers as a first maximum pooling layer, a second maximum pooling layer, a third maximum pooling layer and a fourth maximum pooling layer, calling 2 full-connection layers as a first full-connection layer and a second full-connection layer, wherein the activation function of each convolution layer and each full-connection layer is a Relu function, the activation function of a classification layer is a Softmax function, inputting a collected garbage image into a first convolution layer by an image acquisition module, the first convolution layer is used for carrying out first convolution filtering on the input garbage image to obtain an initial garbage feature, the second convolution layer is used for carrying out second convolution filtering on the initial garbage feature to obtain a second garbage feature, the first maximum pooling layer is used for carrying out pooling sampling on the second garbage feature to obtain a first pooling sampling feature, and the third convolution layer is used for carrying out third convolution filtering on the first pooling sampling feature, obtaining a third garbage feature, wherein the fourth convolutional layer is used for performing a fourth convolution filtering on the third garbage feature to obtain a fourth garbage feature, the second maximum pooling layer is used for performing pooling sampling on the fourth garbage feature to obtain a second pooling sampling feature, the fifth convolutional layer is used for performing a fifth convolution filtering on the second pooling sampling feature to obtain a fifth garbage feature, the sixth convolutional layer is used for performing a sixth convolution filtering on the fifth garbage feature to obtain a sixth garbage feature, the third maximum pooling layer is used for performing pooling sampling on the sixth garbage feature to obtain a third pooling sampling feature, the seventh convolutional layer is used for performing a seventh convolution filtering on the third pooling sampling feature to obtain a seventh garbage feature, and the eighth convolutional layer is used for performing an eighth convolution filtering on the seventh garbage feature, obtain the rubbish characteristic of eighth time, the biggest pooling layer of fourth be used for carrying out the pooling sampling to rubbish characteristic of eighth time, obtain the pooling sampling characteristic of fourth time, first full articulamentum be used for carrying out the first feature extraction to the pooling sampling characteristic of fourth time, obtain degree of depth rubbish characteristic, the full articulamentum of second be used for carrying out the second feature extraction to degree of depth rubbish characteristic, obtain final rubbish characteristic, the articulamentum be used for discerning classification back output classification result to final rubbish characteristic give control module.
3. The intelligent classification dustbin based on deep learning of claim 1, wherein the power supply module comprises a first power supply circuit, a second power supply circuit and a power supply selection module, the first power supply circuit adopts a 220V alternating current power supply mode, the second power supply circuit comprises a solar cell panel, a first power supply device and a second power supply device, and the first power supply device and the second power supply device are respectively realized by storage batteries;
the power supply selection equipment divides 24 hours a day into two working time periods, namely a daytime working time period and a nighttime working time period;
in the working time period in the daytime, the power supply selection equipment acquires weather information, if the weather is other weather except sunny, the first power supply line is selected for supplying power, if the weather is sunny, the second power supply line is preferentially adopted for supplying power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20%, the power supply selection equipment is switched to select the first power supply line for supplying power, the solar panel is controlled to sequentially charge the first power supply equipment and the second power supply equipment, if the electric quantity of the first power supply equipment is less than 20% and the electric quantity of the second power supply equipment is more than 20%, the second power supply equipment is adopted for supplying power, and the solar panel is controlled to charge the first power supply equipment, if the electric quantity of the first power supply equipment exceeds 20% and the electric quantity of the second power supply equipment is less than 20%, adopting the first power supply equipment to supply power and controlling the solar panel to charge the second power supply equipment, if the electric quantities of the first power supply equipment and the second power supply equipment both exceed 20%, preferentially adopting the first power supply equipment to supply power and judging whether the electric quantity of the second power supply equipment is full, and if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment is not full, controlling the solar panel to charge the second power supply equipment;
in the night working time period, the power supply selection equipment preferentially adopts the second power supply circuit to supply power, at the moment, the power supply selection equipment detects the electric quantity of the first power supply equipment and the second power supply equipment in real time, if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment are both less than 20 percent, switching to the first power supply line for supplying power, if the electric quantity of the first power supply equipment is less than 20 percent and the electric quantity of the second power supply equipment exceeds 20 percent, the second power supply equipment is adopted to supply power, if the electric quantity of the first power supply equipment exceeds 20 percent and the electric quantity of the second power supply equipment is less than 20 percent, the first power supply equipment is adopted to supply power, and if the electric quantity of the first power supply equipment and the electric quantity of the second power supply equipment both exceed 20%, preferentially adopting the first power supply equipment to supply power.
4. The intelligent classification dustbin based on deep learning of claim 1, wherein the garbage classification selection module comprises a garbage can scheduling module and a garbage can selection module, the garbage can selection module is connected with the control module, the garbage can selection module receives a garbage classification signal sent by the control module, calculates an angle, required to rotate, of a garbage can corresponding to a type of garbage to be kept according to a type of the garbage can below a current garbage storage chamber, and sends the calculated angle to the garbage can scheduling module, the garbage can scheduling module controls the rotation of the garbage can, rotates the corresponding garbage can to a position right below the garbage storage chamber, and then drives a bottom plate of the garbage storage chamber to open so that the garbage falls into the corresponding garbage can and then is placed on a closed bottom plate, thereby completing garbage classification putting.
5. The intelligent classification garbage can based on deep learning as claimed in claim 1, wherein the driving module comprises a chassis which is arranged at the bottom of the can body and can move up and down under force, four pulleys which are arranged at the bottom of the can body, a lifting mechanism for driving the chassis to move up and down and a rotating mechanism for driving the pulleys to rotate, the lifting mechanism and the rotating mechanism are respectively connected with the control module, the chassis is provided with four openings, in an initial state, the chassis is at the lowest position, and the four pulleys are embedded into the four openings in a one-to-one correspondence manner, so that the can body cannot move;
when the control module sends the movement data to the driving module, firstly the lifting mechanism controls the chassis to move upwards to the highest position, at the moment, the four pulleys are separated from the chassis, then the rotating mechanism drives the four pulleys to correspondingly rotate according to the movement data, so that the box body moves, after the box body moves in place, the lifting mechanism controls the chassis to descend to the lowest position, the four pulleys enter the chassis again, and the position of the box body is fixed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910993388.2A CN110803406A (en) | 2019-10-18 | 2019-10-18 | Intelligent classification dustbin based on degree of depth study |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910993388.2A CN110803406A (en) | 2019-10-18 | 2019-10-18 | Intelligent classification dustbin based on degree of depth study |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110803406A true CN110803406A (en) | 2020-02-18 |
Family
ID=69488663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910993388.2A Pending CN110803406A (en) | 2019-10-18 | 2019-10-18 | Intelligent classification dustbin based on degree of depth study |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110803406A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708823A (en) * | 2020-08-18 | 2020-09-25 | 腾讯科技（深圳）有限公司 | Abnormal social account identification method and device, computer equipment and storage medium |
| CN112046962A (en) * | 2020-08-20 | 2020-12-08 | 浙江始祖鸟科技有限公司 | Classifiable voice control intelligent garbage can |
| CN112046963A (en) * | 2020-08-20 | 2020-12-08 | 浙江始祖鸟科技有限公司 | Anticorrosive sortable garbage bin of intelligence |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011006253A (en) * | 2009-06-29 | 2011-01-13 | Wataru Hasegawa | Garbage sorting chute for home |
| CN107082207A (en) * | 2017-06-17 | 2017-08-22 | 陈钦先 | A kind of energy saving and environmental protection garbage case with sterilizing function |
| CN107359689A (en) * | 2017-08-28 | 2017-11-17 | 安徽汉威电子有限公司 | A kind of intelligent solar charging system |
| CN108182455A (en) * | 2018-01-18 | 2018-06-19 | 齐鲁工业大学 | A kind of method, apparatus and intelligent garbage bin of the classification of rubbish image intelligent |
| CN108190317A (en) * | 2018-03-01 | 2018-06-22 | 苏州欧蒂华电子有限公司 | A kind of dustbin of energy guiding movement |
| CN108341184A (en) * | 2018-03-01 | 2018-07-31 | 安徽省星灵信息科技有限公司 | A kind of intelligent sorting dustbin |
| CN109335371A (en) * | 2018-09-30 | 2019-02-15 | 上海檀楠信息科技有限公司 | Intelligent classification dustbin |
| CN109606991A (en) * | 2019-01-11 | 2019-04-12 | 郑州大学 | Intelligent garbage bin and refuse classification method based on deep learning |
| 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 |
| CN109809060A (en) * | 2019-03-12 | 2019-05-28 | 广东工业大学 | A kind of automatic sorting garbage bin and its management system |
| CN110210635A (en) * | 2019-06-05 | 2019-09-06 | 周皓冉 | A kind of intelligent classification recovery system that can identify waste |
| CN110228673A (en) * | 2019-04-17 | 2019-09-13 | 华南师范大学 | A kind of intelligent classification dustbin |
-
2019
- 2019-10-18 CN CN201910993388.2A patent/CN110803406A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2011006253A (en) * | 2009-06-29 | 2011-01-13 | Wataru Hasegawa | Garbage sorting chute for home |
| CN107082207A (en) * | 2017-06-17 | 2017-08-22 | 陈钦先 | A kind of energy saving and environmental protection garbage case with sterilizing function |
| CN107359689A (en) * | 2017-08-28 | 2017-11-17 | 安徽汉威电子有限公司 | A kind of intelligent solar charging system |
| CN108182455A (en) * | 2018-01-18 | 2018-06-19 | 齐鲁工业大学 | A kind of method, apparatus and intelligent garbage bin of the classification of rubbish image intelligent |
| CN108190317A (en) * | 2018-03-01 | 2018-06-22 | 苏州欧蒂华电子有限公司 | A kind of dustbin of energy guiding movement |
| CN108341184A (en) * | 2018-03-01 | 2018-07-31 | 安徽省星灵信息科技有限公司 | A kind of intelligent sorting dustbin |
| CN109335371A (en) * | 2018-09-30 | 2019-02-15 | 上海檀楠信息科技有限公司 | Intelligent classification dustbin |
| 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 |
| CN109606991A (en) * | 2019-01-11 | 2019-04-12 | 郑州大学 | Intelligent garbage bin and refuse classification method based on deep learning |
| CN109809060A (en) * | 2019-03-12 | 2019-05-28 | 广东工业大学 | A kind of automatic sorting garbage bin and its management system |
| CN110228673A (en) * | 2019-04-17 | 2019-09-13 | 华南师范大学 | A kind of intelligent classification dustbin |
| CN110210635A (en) * | 2019-06-05 | 2019-09-06 | 周皓冉 | A kind of intelligent classification recovery system that can identify waste |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708823A (en) * | 2020-08-18 | 2020-09-25 | 腾讯科技（深圳）有限公司 | Abnormal social account identification method and device, computer equipment and storage medium |
| CN112046962A (en) * | 2020-08-20 | 2020-12-08 | 浙江始祖鸟科技有限公司 | Classifiable voice control intelligent garbage can |
| CN112046963A (en) * | 2020-08-20 | 2020-12-08 | 浙江始祖鸟科技有限公司 | Anticorrosive sortable garbage bin of intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110803406A (en) | Intelligent classification dustbin based on degree of depth study | |
| WO2020206862A1 (en) | Automatic sorting system | |
| CN109606991B (en) | Intelligent garbage can and garbage classification method based on deep learning | |
| CN109809060A (en) | A kind of automatic sorting garbage bin and its management system | |
| CN106395186A (en) | Intelligent urban waste bin | |
| CN103287770B (en) | Underground garbage classified collection system | |
| CN201105894Y (en) | Multifunctional environment protection dustbin | |
| CN206485830U (en) | A kind of intelligent garbage bin | |
| CN103991660A (en) | Intelligent weighing sorting dustbin | |
| CN103754527A (en) | Intelligent garbage can and management system thereof | |
| CN107745893B (en) | Interactive intelligent cleaning dustbin | |
| CN112046962B (en) | Classifiable voice control intelligent garbage can | |
| CN105416916A (en) | Trash can with automatic classification function | |
| CN210853860U (en) | Intelligent garbage can | |
| CN112320138A (en) | Indoor movable garbage classification device | |
| CN205366713U (en) | Automatic sortable garbage bin | |
| CN112429431A (en) | Automatic induction type intelligent garbage can control system | |
| CN108275395B (en) | Solar garbage throwing device and control method thereof | |
| CN214297580U (en) | Intelligent garbage recognition and classification device based on internet | |
| CN113800146A (en) | Independent power supply intelligent classification dustbin | |
| CN204161894U (en) | Intelligent dustbin | |
| CN215708916U (en) | Intelligent recognition garbage can | |
| CN215437993U (en) | Intelligent garbage classification device based on vision | |
| CN215827540U (en) | Garbage classification collection system based on voice recognition | |
| CN204726970U (en) | A kind of intelligent sound garbage collection system based on Internet of Things |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200218 |