CN115583448A

CN115583448A - Intelligent garbage putting and managing integrated platform and method

Info

Publication number: CN115583448A
Application number: CN202211268763.5A
Authority: CN
Inventors: 朱旭芳; 吴尽哲; 黄亮
Original assignee: Naval University of Engineering PLA
Current assignee: Naval University of Engineering PLA
Priority date: 2022-10-17
Filing date: 2022-10-17
Publication date: 2023-01-10

Abstract

The invention belongs to the technical field of garbage management, and discloses an intelligent garbage throwing management integrated platform and a method, wherein the intelligent garbage throwing management integrated platform is provided with an intelligent identification module for realizing image acquisition and image identification classification functions; the automatic throwing module receives the signal of the intelligent identification module and controls the steering engine to throw the garbage; the intelligent interconnection module is used for monitoring the garbage cans and performing man-machine interaction with an administrator; absorb solar energy through solar panel and convert solar energy into the electric energy, power module for whole system power supply. The invention can solve the problems of intelligent classification and automatic garbage throwing in one step; still possess Wi-Fi remote control function, can realize the interior data real-time transport of garbage bin through sensor and cell-phone APP's in the garbage bin being connected to help the administrator in time manage the garbage bin.

Description

Intelligent garbage putting and managing integrated platform and method

Technical Field

The invention belongs to the technical field of garbage management, and particularly relates to an intelligent garbage throwing management integrated platform and method.

Background

At present, according to data of development prospects and investment prediction analysis reports of Chinese domestic garbage treatment industry in 2017-2022 published by prospective industry research institutes, the clear volume of Chinese domestic garbage in 2011 reaches 1.64 hundred million tons, the year-by-year increase is 3.74%, the harmless treatment volume is 1.3 million tons, about 3330 million tons of municipal domestic garbage cannot be beneficially treated, and the clear volume of Chinese domestic garbage in 2016 is increased to 2.17 million tons. Therefore, classified recycling of garbage is a problem to be solved urgently at present, and intelligent garbage classification products are applied accordingly. The intelligent garbage classification refers to the steps that garbage throwing, a solar panel, a sensor, remote control, early warning, security and protection and the like are connected into a network, then unified management is achieved through Wi-Fi, data collection and equipment control are achieved, and a convenient and comfortable living environment is brought to a user.

The earliest intelligent garbage classification systems appeared in europe, and since the last 90 s, the development of intelligent garbage classification products was promoted by the investment of a great deal of manpower and financial resources by foreign science and technology companies such as IBM, panasonic, microsoft, samsung, and the like. At present, the foreign famous intelligent garbage classification system mainly comprises an Internet plus intelligent garbage putting management integrated platform, a French Eugene intelligent garbage putting management integrated platform, an intelligent garbage putting management integrated platform based on an STC89C52 single chip microcomputer, a Cambridge consultant intelligent garbage putting management integrated platform, an intelligent garbage can capable of realizing code scanning classification putting, an American BigBelly Solar intelligent garbage putting management integrated platform and the like.

The domestic awareness of intelligent garbage classification is relatively late, but with the rapid development of economy, users of intelligent garbage classification who have also become increasingly aware of the importance of the problem have focused primarily on southern cities. The reason is that apart from some conceptual factors, the most important is that the intelligent garbage classification system has incomplete functions, high price and limited practicability and popularization. Therefore, it is particularly necessary to develop an intelligent garbage classification product which is low in price, complete in function and capable of stably running for a long time.

In the class of well-known intelligent garbage classification products at home and abroad, most of the utilized technologies are mainly used for classifying the existing garbage samples, the final garbage putting step is still finished manually, the labor cost is not obviously reduced, and the application and popularization of the intelligent garbage putting management integrated platform are hindered. Although the individual invention technology considers the requirement of autonomous garbage throwing, the technology has obvious shortages in the aspects of remote control, energy endurance and the like.

Through the above analysis, the problems and defects of the prior art are as follows: in the prior art, the final garbage throwing step is still finished manually, the labor cost is not obviously reduced, and the application and popularization of the intelligent garbage throwing management integrated platform are hindered. Although the individual invention technology considers the requirement of autonomous garbage throwing, the technology has obvious shortages in the aspects of remote control, energy endurance and the like.

The difficulty in solving the above problems and defects is: how to carry out accurate and rapid intelligent identification on different types of garbage; how to automatically put in the intelligent identification module according to the signal of the intelligent identification module; how to increase the intelligent interconnection between the administrator and the equipment; how to use solar energy guarantees the long time duration of equipment.

The significance of solving the problems and the defects is as follows: the garbage classification management system integrates functions of garbage throwing, a solar panel, a sensor, remote control and the like, realizes unified management by utilizing Wi-Fi, achieves data collection and equipment control, is convenient for users to throw garbage in a classification mode, is beneficial to garbage collection of cleaners, and enables garbage classification management to be more convenient and faster.

Disclosure of Invention

Aiming at the problems in the prior art, the invention provides an intelligent garbage throwing management integrated platform and method.

The invention is realized in this way, an intelligent garbage putting management integrated platform, which comprises:

the intelligent identification module is used for realizing the functions of image acquisition and image identification and classification;

receiving an intelligent interconnection module which is used for monitoring each garbage can and performing man-machine interaction with an administrator;

the solar panel absorbs solar energy and converts the solar energy into electric energy, and the power supply module supplies power for the whole system.

Further, the platform further comprises:

each intelligent garbage throwing management integrated platform uploads self state information and position information to a cloud server through a network through an SIM card, a mobile phone APP client receives information of each garbage can from the cloud server and displays the positions and the state information of all the garbage cans on a map, and the mobile phone client plans a garbage can cleaning path according to the positions and the state information of the garbage cans and the positions and personnel conditions of cleaning stations;

the sensor module is combined with a networking technology, comprehensively processes the monitored physical information of the temperature, the humidity, the smoke concentration, the odor concentration and the garbage amount of the garbage can, judges whether the garbage can is full, peculiar smell and fire, and sends the information to the client;

sorting module, after intelligent recognition rubbish, the steering wheel automatic control module of management integration platform is put in to intelligence rubbish can carry out accurate input to recoverable rubbish, unrecoverable rubbish.

Furthermore, the intelligent identification module comprises an image signal acquisition module and a central image processing module, wherein the image signal acquisition module is a camera, acquires a real environment and transmits the acquired real environment into the central image processing module in a picture form; the central image processing module classifies the garbage placed on the platform, and finally transmits the result to the single chip microcomputer.

Further, the automatic module of puting in accepts intelligent recognition module's signal and control the steering wheel and put in rubbish specifically do:

receiving a garbage type recognition result of an upper-layer neural network through a communication interface, converting the garbage type recognition result into a driving control signal after being processed by a local single chip microcomputer, driving a stepping motor and a large-torque steering engine to complete the movement and dumping of garbage according to a certain time sequence, returning to a garbage collection position after the movement is completed, and waiting for a garbage processing signal; still possess information acquisition and state monitoring function simultaneously, send the inside information acquisition of garbage bin to upper monitoring network through the sensor.

Furthermore, the intelligent interconnection module is integrated by intelligent rubbish input management platform, high in the clouds server, managers and cell-phone APP and constitutes a whole.

Furthermore, the intelligent garbage throwing management integrated platform is provided with an SIM card, the SIM uploads the state information and the position information of the SIM card to the cloud server through the network, the mobile client receives the information of each garbage can from the cloud server and displays the positions and the state information of all the garbage cans on a map, and the mobile client carries out garbage can cleaning path planning according to the positions and the state information of the garbage cans and the positions and the personnel conditions of cleaning stations, so that workers can clean the garbage cans more quickly and efficiently.

Further, the state information includes temperature and garbage amount.

Further, power module absorbs solar energy through solar panel and converts solar energy into electric energy and specifically does as the power supply of whole system:

the sunlight maximum saturation automatic tracking device based on the photosensitive sensor is used for ensuring that the solar panel can absorb solar energy to the maximum extent; the problem of unstable power supply of the solar panel is solved by utilizing the special voltage-stabilizing adjusting device for the solar power supply, and a rechargeable lithium battery is added on a peripheral circuit of the solar panel; when the solar panel works, a part of resources can enter the lithium battery, and when the environment that the solar panel receives illumination is influenced by a series of low sunlight irradiation degree and low air visibility, the whole garbage disposal system still has the lithium battery as a power source.

Another object of the present invention is to provide a method for controlling an intelligent garbage throwing management integrated platform of the intelligent garbage throwing management integrated platform, including:

firstly, introducing a deep learning technology, establishing a garbage classification training set, training a binarization neural network for garbage classification on a CPU + GPU platform, deploying the network to a full-programmable system-on-chip PYNQ-Z1 based on Xilinx Zynq7020, classifying the thrown garbage into one of non-recyclable products, plastic products, metal products, paper products and harmful articles by using the trained convolution neural network, sending an instruction to a single chip microcomputer according to the condition, and controlling a garbage conveying device to move along a track to finish automatic throwing of the garbage;

secondly, combining the internet of things technology, comprehensively processing physical information such as the temperature, the humidity, the smoke concentration, the odor concentration and the garbage amount of the monitored garbage cans, judging whether the garbage cans are in the states of full cans, peculiar smell, fire and the like, sending the information to a client, developing an APP for intelligent garbage classification management, logging in the APP by managers, and knowing geographic information and state information of all the garbage cans in a film area in time;

finally, the whole system is powered by a solar panel, and the sunlight maximum saturation automatic tracking device based on the photosensitive sensor is used for ensuring that the sunlight irradiation maximum can be received at any time, and meanwhile, the system function is realized by using a low-power-consumption industrial personal computer and a development board.

Further, the garbage classification training set is specifically 5000 pieces of 7 types of garbage.

Further, the convolutional neural network specifically comprises the following processes:

inputting a color image with the size of 32 multiplied by 3, combining convolution layers with a 3 multiplied by 3 convolution, a 3 multiplied by 3 convolution and a pooling layer, repeating for 3 times, and finally adding two full-link layers containing 512 neurons and outputting 7 classification results; wherein the convolution kernels of the convolution layers are all 3 multiplied by 3, and the size of the pooling layer is 2 multiplied by 2; after the convolutional neural network training is finished, carrying out binarization on the network weight; meanwhile, an FPGA hardware structure is designed by means of a Xilinx HLS tool, a bit stream file is generated after the steps of compiling, integrating and the like, the FPGA structure is configured, and finally the binaryzation weight is uploaded into an on-chip memory of a PYNQ development board.

By combining all the technical schemes, the invention has the advantages and positive effects that:

the intelligent garbage throwing management integrated platform can solve the problems of intelligent classification and automatic throwing of garbage at one time. Moreover, this intelligent rubbish is put in and is managed integration platform still possesses Wi-Fi remote control function, can realize data real-time transport in the garbage bin through being connected of sensor and cell-phone APP in the garbage bin to help the administrator in time manage the garbage bin. In addition, the problem of endurance of the garbage can is solved; the invention provides an effective, energy-saving and low-cost energy source for the platform by using the solar panel.

The intelligent garbage throwing management integrated platform integrates four functions of intelligent garbage classification, autonomous garbage throwing, remote control, solar endurance and the like, and runs through the whole process from classified throwing to recycling treatment of garbage; meanwhile, the platform is low in price and stable in operation. Therefore, the method has certain popularization and use values.

Drawings

Fig. 1 is a schematic structural diagram of an intelligent garbage throwing management integrated platform provided in an embodiment of the present invention;

fig. 2 is a schematic view of an overall framework of the pipe feeding integrated platform provided by the embodiment of the invention.

Fig. 3 is a schematic diagram of a composition structure of an image acquisition and automatic identification module according to an embodiment of the present invention.

Fig. 4 is a schematic structural diagram of an automatic delivery module according to an embodiment of the present invention.

Fig. 5 is a schematic diagram of a system composition structure of an intelligent interconnection module according to an embodiment of the present invention.

Fig. 6 is a hardware system of an intelligent garbage putting and managing integrated platform provided in an embodiment of the present invention

Fig. 7 is a diagram of a convolutional neural network structure according to an embodiment of the present invention.

Fig. 8 is a schematic diagram of a condition monitoring and actuator design provided by an embodiment of the invention.

Fig. 9 is a partial object diagram of a power module according to an embodiment of the present invention.

Fig. 10 is a schematic structural diagram of a maximum light intensity capture device according to an embodiment of the present invention.

Fig. 11 is a circuit diagram of a maximum light intensity capturing device according to an embodiment of the present invention.

Fig. 12 is a flowchart of intelligent classification provided by the embodiment of the present invention.

Fig. 13 is a schematic structural diagram of an intelligent interconnection module system according to an embodiment of the present invention.

In FIG. 1: 1. an intelligent identification module; 2. an automatic delivery module; 3. an intelligent interconnection module; 4. and a power supply module.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Aiming at the problems in the prior art, the invention provides an intelligent garbage throwing management integrated platform and a control method thereof, and the invention is described in detail below with reference to the accompanying drawings.

A person skilled in the art of the integrated intelligent garbage throwing management platform provided by the present invention may also perform other steps, and the integrated intelligent garbage throwing management platform provided by the present invention in fig. 1 is only one specific embodiment.

As shown in fig. 1, an intelligent garbage putting management integrated platform provided in an embodiment of the present invention includes:

and the intelligent identification module 1 is used for realizing the functions of image acquisition and image identification and classification.

And the automatic throwing module 2 receives the signal of the intelligent identification module and controls the steering engine to throw the garbage.

And the intelligent interconnection module 3 is used for monitoring the garbage cans and performing man-machine interaction with an administrator.

And the power supply module 4 absorbs solar energy through the solar panel to convert the solar energy into electric energy and supplies power to the whole system.

The intelligent identification module provided by the embodiment of the invention comprises an image signal acquisition module and a central image processing module, wherein the image signal acquisition module is a camera, acquires a real environment and transmits the acquired real environment into the central image processing module in a picture form; the central image processing module classifies the garbage placed on the platform, and finally transmits the result to the single chip microcomputer.

The automatic throwing module provided by the embodiment of the invention receives the signal of the intelligent identification module and controls the steering engine to throw garbage, and the specific process comprises the following steps:

receiving a garbage type recognition result of an upper-layer neural network through a communication interface, converting the garbage type recognition result into a driving control signal after being processed by a local single chip microcomputer, driving a stepping motor and a large-torque steering engine to finish the movement and dumping actions of garbage according to a certain time sequence, returning to a garbage collection position after the actions are finished, and waiting for a garbage processing signal; still possess information acquisition and state monitoring function simultaneously, send the upper monitoring network with the inside information acquisition of garbage bin through the sensor.

The intelligent interconnection module provided by the embodiment of the invention is a whole formed by an intelligent garbage putting management integrated platform, a cloud server, a manager and a mobile phone APP. Every intelligent rubbish puts in management integration platform is from taking an SIM card, SIM can pass through the network with the state information and the positional information of self and upload the high in the clouds server, mobile client can receive the information of each garbage bin from the high in the clouds server, and show the position and the state information (like temperature, rubbish volume etc.) of all garbage bins on the map, mobile client can carry out garbage bin clearance route planning according to the position and the state information of garbage bin and the position and the personnel condition of clearance website, the staff of being convenient for clears up the garbage bin more fast more efficiently.

The power module provided by the embodiment of the invention converts solar energy into electric energy by absorbing the solar energy through the solar panel to supply power to the whole system, and specifically comprises the following steps:

the sunlight maximum saturation automatic tracking device based on the photosensitive sensor is used for ensuring that the solar panel can absorb solar energy to the maximum extent; utilize special steady voltage adjusting device of solar power supply to solve solar panel's the unstable problem of power supply, add chargeable lithium cell on solar panel peripheral circuit, when solar panel during operation, have partly resource can get into the lithium cell, run into a series of influences solar panel such as sunshine shines the degree low, air visibility low and accept the environment of illumination, whole refuse treatment system still has the lithium cell as power source.

The technical solution of the present invention will be described in detail with reference to the following specific examples.

1 System protocol analysis

FIG. 2 is a general schematic view of an integrated frame of an intelligent multi-barrel tube feeding platform. The platform needs to have the functions of object identification, automatic putting, information sensing, mobile phone APP management detection, solar power supply, alarming and the like. According to the functions of the platform, the system is divided into three major submodules, namely an intelligent identification module, an automatic delivery module and an intelligent interconnection module. The intelligent identification module mainly has the functions of image acquisition and image classification; the automatic throwing module is mainly used for receiving the signal of the intelligent identification module and controlling the steering engine to throw the garbage; the intelligent interconnection module is responsible for monitoring each garbage can and performing human-computer interaction with an administrator.

In the embodiment of the invention, each intelligent garbage throwing and managing integrated platform is provided with an SIM card, the SIM can upload the state information and the position information of the SIM card to a cloud server through a network, a mobile phone APP client can receive the information of each garbage can from the cloud server and display the positions and the state information (such as temperature, garbage amount and the like) of all the garbage cans on a map, and the mobile phone client can plan the garbage can cleaning path according to the positions and the state information of the garbage cans and the positions and the personnel conditions of cleaning stations, so that workers can clean the garbage cans more quickly and efficiently.

In the embodiment of the invention, the sensor module combines the internet of things technology to comprehensively process the monitored physical information of the garbage bin such as temperature, humidity, smoke concentration, odor concentration, garbage amount and the like, judges whether the garbage bin is in the states of full garbage bin, peculiar smell, fire and the like, sends the information to the client,

in the embodiment of the invention, after the sorting module intelligently identifies garbage, the automatic control module of the steering engine of the intelligent garbage throwing management integrated platform can accurately throw recoverable garbage and unrecoverable garbage. In addition, when inflammable and explosive objects are identified, the platform can send out an alarm to ensure safety.

1.1 image acquisition and intelligent identification module

The function that the module needs to complete is object identification and output of the identification result. The system is designed based on a PYNQ development system board and mainly comprises an image signal acquisition module and a central image processing module, wherein the image signal acquisition module is a camera and is used for acquiring a real environment as shown in figure 3,

the images are transmitted into a central image processing module in a picture mode, the central image processing module classifies the garbage placed on the platform, and finally, the results are transmitted into the single chip microcomputer.

1.2 the automatic module of puting in of steering wheel receives upper neural network's rubbish kind discernment result through communication interface, converts the drive control signal into after local singlechip is handled, according to certain chronogenesis drive step motor and big moment of torsion steering wheel completion rubbish and removes and the action of empting, gets back to rubbish collection position after the action is accomplished, waits for the refuse treatment signal. Still possess information acquisition and state monitoring function simultaneously, send the upper monitoring network through sensor with the inside information acquisition of garbage bin, the structural composition is as shown in figure 4, through off-line neural network FPGA and Arduino's coordinated operation, realizes the drive to two L298N step motor and the drive of sensors such as temperature, humidity, quality, conflagration. Thereby realizing various functions.

1.3 Intelligent interconnection Module

The intelligent interconnection module is integrated by intelligent rubbish input management platform, high in the clouds server, managers and cell-phone APP and constitutes a whole. Every intelligent rubbish puts in management integration platform is from taking an SIM card, SIM can upload the high in the clouds server with the state information and the positional information of self through the network, mobile client can receive the information of each garbage bin from the high in the clouds server, and show the position and the state information (like temperature, rubbish volume etc.) of all garbage bins on the map, mobile client can carry out garbage bin clearance route planning according to the position and the state information of garbage bin and the position and the personnel condition of clearance website, the staff of being convenient for clears up the garbage bin more fast more efficiently. The working principle of the system is shown in fig. 5, the server data is transmitted to the router and then transmitted to the network to manage the state data of each trash can, and workers can know the real-time state of the trash can according to the information of the mobile phone client.

The above is an introduction to the general framework design of the intelligent multi-barrel casting and management integrated platform, and the hardware system design and the software system design of the platform are respectively introduced below,

2 hardware System design

The hardware system is mainly as shown in fig. 6, the design of the hardware system part is mainly performed on the PYNQ development board system and the ARDUINO UNO REV3 system, and the hardware system is as shown in fig. 6 according to the functional requirements. The hardware system comprises a PYNQ image acquisition and identification module, an automatic control module, a steering engine, a driver, a sensor, a mobile phone and the like. Arduino is used as a core controller to control the driving of two L298N stepping motors and the driving of sensors for temperature, humidity, quality, fire and the like. The neural network module is mainly used for intelligently distinguishing different types of garbage. The sensor module is interconnected with the mobile phone client, so that real-time dynamic management of the garbage can is realized.

2.1 identification Module

The acquisition and identification module adopts a PYNQ development board, which can enable an embedded programmer to fully play the function of an Xilinx Zynq All Programmable SoC (APSoC) without designing a Programmable logic circuit. The user can program using Python and the code can be developed and tested directly on PYNQ.

The design of automatic identification mainly comprises deep learning, establishment of a classification sample library, automatic identification and release. The garbage classification of the system is used for classifying and identifying garbage based on a deep learning technology according to the picture information acquired by the camera. Therefore, the invention establishes a large-scale sample library which comprises 7 kinds of garbage, wherein the total amount of the garbage samples is 5000, and the garbage samples are mainly divided into four categories, namely non-recyclable products, harmful articles, paper products and plastic/metal products.

The convolutional neural network for garbage recognition of the present invention is shown in fig. 7, the input image size is 32 × 32 × 3 color image, the convolution layer combination (3 × 3 convolution, pooling layer) is repeated 3 times, and finally two full link layers containing 512 neurons are added, and 7 classification results are output. Wherein the convolution kernels of the convolution layers are all 3 x 3, and the size of the pooling layer is 2 x' 2. And after the convolutional neural network training is finished, carrying out binarization on the network weight. Meanwhile, an FPGA hardware structure is designed by using a Xilinx HLS tool, a bit stream file is generated after the steps of compiling, synthesizing and the like, the FPGA structure is configured, and finally the binary weight is uploaded into an on-chip memory of a PYNQ development board.

The input image size is 32 × 32 × 3 color image, the convolution layer combination (3 × 3 convolution, pooling layer) is repeated 3 times, and finally two full-link layers containing 512 neurons are added, and 7 classification results are output. Wherein the convolution kernels of the convolution layers are all 3 x 3, and the size of the pooling layer is 2 x' 2. And after the convolutional neural network training is finished, carrying out binarization on the network weight. Meanwhile, an FPGA hardware structure is designed by using a Xilinx HLS tool, a bit stream file is generated after the steps of compiling, synthesizing and the like, the FPGA structure is configured, and finally the binary weight is uploaded into an on-chip memory of a PYNQ development board.

2.2 image information acquisition Module

The PYNQ adopts an Ubuntu operating system, and the image acquisition module adopts an rmoncam S907 USB camera supporting a UVC protocol as image acquisition equipment. Focal length of the lens: 6mm; and (3) outputting resolution: 640 x 480; pixel: 130 ten thousand.

2.3 automatic control module of steering engine

The steering engine automatic control S9810 super large torque steering engine (including the drive), two h bridge structure step motor drive plates of L298n, 42BYGH step motor, arduino UNO R3 control panel, multiple monitoring sensor (sensor such as temperature, humidity, quality, conflagration, battery residual capacity detection). The power supply voltage is 12V, the maximum power does not exceed 40W, the standby power is less than 1W, and the endurance time is more than one week. The system has the functions of uploading local monitoring data, controlling an actuating mechanism and switching standby working modes. The condition monitoring and actuator design is shown in fig. 8. The partial structure is shown in figure 9.

2.4 Power supply Module

In the design of the system, aiming at the dispersion problem of solar energy, the sunlight maximum saturation automatic tracking device based on the photosensitive sensor is designed, and the absorption of the solar panel to the solar energy can be ensured to the maximum extent. The structure of the capturing device is shown in fig. 10, which is a mechanical structure part of the intelligent garbage can, the garbage can is powered by a solar panel (1), when people throw garbage in, the type of the garbage is intelligently judged, a garbage can cover (8) is opened by driving a supporting rod (2) through a motor, and then the garbage is thrown in. The circuit design is shown in fig. 11.

In addition, this system still utilizes the special steady voltage adjusting device of solar power supply to solve solar panel's unstable problem of supplying power, adds chargeable lithium cell on solar panel peripheral circuit, and when solar panel during operation, some resource can get into the lithium cell, even if meet a series of influences solar panel and receive the environment of illumination such as sunshine low, air visibility low, whole refuse treatment system still has the lithium cell as the power source.

3 software system scheme design

The software system of the platform is mainly divided into three modules, namely, image acquisition and automatic identification, intelligent interconnection and data management.

3.1 image acquisition, automatic identification

The image acquisition and the automatic identification are an automatic process, and after a user presses a release key, the integrated platform can complete the identification and release processes through serial port communication. The flow is shown in fig. 12, which is a process of intelligent recognition of spam types. Firstly, reading the image, and then comparing the image with the image of the sample library to realize the identification of the type of the garbage image, thereby intelligently outputting the detection result. And finishing the program after the recognition is finished. This figure is a basic introduction of the mobile phone client interface. The method comprises the steps of opening a mobile phone client, entering a welcome page, and then entering a main interface, wherein the main interface mainly comprises a map interface, a data interface and an administrator interface. The map interface can display the position of the garbage can in two modes of a satellite map and a common map. The data interface can observe physical information such as temperature, humidity, quality and the like of the garbage can.

3.2 Intelligent interconnection Module

The APP is C/S-structured system software developed by Android studio serving as a development platform and java language programming, and a system design scheme is shown in FIG. 13.

3.3 data management Module

The present platform mainly involves the following data: the garbage can uploads the position data and the state data of the garbage can to the cloud server through the Internet, and the mobile phone client records and uploads the online data and the working data of users and managers. The detailed data content of the database is shown in table 1.

TABLE 1 database details

The technical scheme of the invention is described in detail by combining experiments.

The effects of the four functions, namely intelligent garbage classification, automatic garbage feeding, remote control and solar endurance are introduced respectively.

Intelligent garbage classification effect

The images in the test set are tested, and the results show that the designed garbage classification system based on the convolutional neural network can rapidly complete the classification of the target and count the time of partial test images, as can be seen from table 2, the average classification time is 1.6 mus, so the classification speed can approximately reach 625 images per second, and the requirement of real-time classification can be met.

Table 2 partial test image classification time

Automatic garbage throwing effect

After the rubbish is intelligently identified, the steering engine automatic control module of the intelligent rubbish throwing management integrated platform can accurately throw in recoverable rubbish and non-recoverable rubbish. In addition, when inflammable and explosive objects are identified, the platform can send out an alarm to ensure safety.

Remote control effect

The intelligent interconnection master control system is combined with the wireless transmission module, various data can be sent to the mobile phone terminal in real time under the Wi-Fi environment, and the state information of all the garbage cans in the parcel can be looked up by clicking the garbage can icons in the drawing. After the administrator finishes cleaning the garbage can, the administrator presses the mechanical switch, and the system can pop out the garbage bag and automatically pack the garbage bag.

The APP has the following specific functions:

and checking the garbage can information in real time. Intelligent rubbish is put in management integration platform terminal information and can be transmitted to cell-phone APP through the GPRS module, and the staff can look over garbage bin position and state information through cell-phone APP in real time, carries corresponding instrument according to the state of garbage bin (have the abnormal smell, whether full bucket, whether catch fire) and goes to the position and clear up. The state of the trash can is changed, and the color of the icon displayed by the trash can is changed accordingly. Clicking a garbage can icon to display information of the garbage can; if the trash can catches fire, clicking a report button to inform workers of processing; clicking any position of the map, and hiding the information of the garbage can;

and pushing the message in real time. If the worker watches the mobile phone interface for a long time and observes the state change of the garbage can, eyestrain is easy to cause. Therefore, this APP can carry out the suggestion of unusual information message of state according to the state information of garbage bin, and the staff of being convenient for uses. And (6) planning the path. In order to clear up the garbage bin more efficiently, more rationally, swiftly, this APP carries out reasonable path planning according to the position of staff in the piece district and the state and the positional information of garbage bin, and the staff of being convenient for has the garbage bin of abnormal state in the shortest time processing. The APP can be switched between a satellite map and a common map. In addition, for the administrator's management of the trash can, the APP may display real-time traffic plans.

The whole system adopts solar panel power supply, and based on the sunshine maximum saturation automatic tracking device of photosensitive sensor, select industrial computer and the development board of low-power consumption for use simultaneously to and add multiple modes such as chargeable lithium cell on solar panel peripheral circuit and guarantee long-time stable continuation of journey of intelligent rubbish input management integration platform.

The working principle of the invention is as follows: firstly, introducing a deep learning technology, establishing 5000 garbage classification training sets containing 7 types of garbage, training a binarization neural network for garbage classification on a CPU + GPU platform, deploying the network to a full-programmable system-on-chip PYNQ-Z1 based on Xilinx Zynq7020, classifying the thrown garbage into one of non-recyclable products, plastic products, metal products, paper products and harmful products by using the trained convolution neural network, and accordingly sending an instruction to a single chip microcomputer to control a garbage conveying device to move along a track to finish automatic throwing of the garbage; secondly, combining the internet of things technology, comprehensively processing physical information such as the temperature, the humidity, the smoke concentration, the odor concentration and the garbage amount of the monitored garbage cans, judging whether the garbage cans are in the states of full cans, peculiar smell, fire and the like, sending the information to a client, developing an APP for intelligent garbage classification management, and enabling managers to log in the APP to know the geographic information and the state information of all the garbage cans in a film area in time; finally, the whole system is powered by a solar panel, the sunlight maximum saturation automatic tracking device based on the photosensitive sensor guarantees that the sunlight irradiation maximum can be received at any time, and meanwhile, the industrial personal computer and the development board with low power consumption are selected to realize the system function.

In the description of the present invention, "a plurality" means two or more unless otherwise specified; the terms "upper", "lower", "left", "right", "inner", "outer", "front", "rear", "head", "tail", and the like, indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, are merely for convenience in describing the present invention and to simplify the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed in a particular orientation, and be operated, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first," "second," "third," and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

It should be noted that the embodiments of the present invention can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus of the present invention and its modules may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, or software executed by various types of processors, or a combination of hardware circuits and software, e.g., firmware.

The above description is only for the purpose of illustrating the present invention and the appended claims are not to be construed as limiting the scope of the invention, which is intended to cover all modifications, equivalents and improvements that are within the spirit and scope of the invention as defined by the appended claims.

Claims

1. The utility model provides an intelligence rubbish is put in management integration platform which characterized in that, intelligence rubbish is put in management integration platform includes:

the automatic throwing module receives the signal of the intelligent identification module and controls the steering engine to throw the garbage;

the intelligent interconnection module is used for monitoring the garbage cans and performing man-machine interaction with an administrator;

absorb solar energy through solar panel and convert solar energy into the electric energy, power module for whole system power supply.

2. The intelligent garbage placement management integration platform of claim 1, wherein said platform further comprises:

the sensor module is combined with a networking technology, comprehensively processes the monitored physical information of the temperature, the humidity, the smoke concentration, the odor concentration and the garbage amount of the garbage can, judges whether the garbage can is in a full-can state, a peculiar smell state and a fire state, and sends the information to the client;

and the sorting module is used for intelligently identifying garbage, and the steering engine automatic control module of the intelligent garbage throwing management integrated platform can throw in recoverable garbage and non-recoverable garbage.

3. The intelligent integrated platform for garbage throwing management as claimed in claim 1, wherein the intelligent recognition module comprises an image signal acquisition module and a central image processing module, the image signal acquisition module is a camera for acquiring a real environment and transmitting the acquired image into the central image processing module in the form of a picture; the central image processing module classifies the garbage placed on the platform, and finally transmits the result to the single chip microcomputer.

4. The intelligent garbage throwing management integrated platform of claim 1, wherein the automatic throwing module receives the signal of the intelligent recognition module and controls a steering engine to throw garbage specifically as follows:

5. The intelligent garbage throwing management integrated platform according to claim 1, wherein the intelligent interconnection module is formed by the intelligent garbage throwing management integrated platform, a cloud server, a manager and a mobile phone APP into a whole.

6. The intelligent trash throwing management integrated platform of claim 4, wherein the intelligent trash throwing management integrated platform is provided with an SIM card, the SIM uploads the state information and the position information of the SIM card to a cloud server through a network, the mobile phone client receives the information of each trash can from the cloud server and displays the positions and the state information of all the trash cans on a map, and the mobile phone client performs trash can cleaning path planning according to the positions and the state information of the trash cans and the positions and the personnel conditions of cleaning stations, so that workers can clean the trash cans more quickly and efficiently;

the state information comprises temperature and garbage amount.

7. The intelligent garbage throwing management integrated platform of claim 1, wherein the power module absorbs solar energy through the solar panel to convert the solar energy into electric energy to supply power to the whole system specifically comprises:

8. An intelligent garbage putting and managing integrated platform control method according to any one of claims 1 to 7, wherein the intelligent garbage putting and managing integrated platform control method comprises the following steps:

(1) Establishing a garbage classification training set by using a deep learning method, training a binarization neural network for garbage classification on a CPU + GPU platform, deploying the network to a full programmable system-on-chip PYNQ-Z1 based on Xilinx Zynq7020, classifying the thrown garbage into one of non-recyclable products, plastic products, metal products, paper products and harmful products by using the trained convolution neural network, sending an instruction to a single chip microcomputer according to the classification training set, and controlling a garbage conveying device to move along a track to finish automatic throwing of the garbage;

(2) The method comprises the steps that physical information such as temperature, humidity, smoke concentration, odor concentration and garbage amount of a garbage can obtained through monitoring is comprehensively processed by combining the Internet of things, whether the garbage can is in a full can state, peculiar smell state, fire state and the like is judged, the information is sent to a client side, an APP for intelligent garbage classification management is developed at the same time, a manager logs in the APP, and geographic information and state information of all the garbage cans in a film area are known in time;

(3) Adopt solar panel power supply, the biggest saturation degree automatic tracking device of sunshine based on photosensitive sensor to guarantee can both receive local sunshine at any time and shine the maximum value, select industrial computer and the development board of low-power consumption to realize system's function simultaneously for use.

9. The intelligent garbage putting management integrated platform control method according to claim 8, wherein the garbage classification training set is 5000 pieces of 7 types of garbage.

10. The intelligent garbage throwing management integrated platform control method according to claim 8, wherein the convolutional neural network comprises the following specific processes:

inputting a color image with the size of 32 multiplied by 3, repeating 3 times the convolution layer combination of 3 multiplied by 3 convolution, 3 multiplied by 3 convolution and the pooling layer, finally adding two full-link layers containing 512 neurons, and outputting 7 classification results; wherein the convolution kernels of the convolution layers are all 3 multiplied by 3, and the size of the pooling layer is 2 multiplied by 2; after the training of the convolutional neural network is finished, carrying out binaryzation on the network weight; meanwhile, an FPGA hardware structure is designed by using a Xilinx HLS tool, a bit stream file is generated after the steps of compiling, synthesizing and the like, the FPGA structure is configured, and finally the binary weight is uploaded into an on-chip memory of a PYNQ development board.