A garbage classification system based on visual recognition technology and its control method

Abstract

The present invention proposes a garbage classification system based on visual recognition technology and a control method thereof. The system includes a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module. Through the scheme of the present invention, the visual recognition technology is used to make garbage classification more intelligent, more accurate and more efficient.

Description

Technical Field

[0001] The present invention relates to the field of machine vision technology, and in particular to a garbage classification system based on visual recognition technology and a control method thereof. Background Art

[0002] With the increase of domestic garbage, the environment has been polluted, the ecological environment has been damaged, and resources have been wasted. To this end, the country has successively implemented a number of policies to promote garbage classification, solve the pollution and waste caused by domestic garbage, and realize the recycling of garbage. However, the current garbage classification and recycling mostly rely on manual work, and there are problems such as low garbage classification efficiency and inaccurate classification. Therefore, there is an urgent need for a garbage classification solution based on visual recognition technology to improve the efficiency of garbage classification, avoid the influence of subjective factors in manual classification, and reduce the consumption of human resources, which will play a positive role in garbage recycling. Summary of the invention

[0003] The present invention is based on the above-mentioned problems and proposes a garbage classification system and a control method thereof based on visual recognition technology. Through the scheme of the present invention, visual recognition technology is used to make garbage classification smarter, more accurate and more efficient.

[0004] In view of this, one aspect of the present invention proposes a garbage classification system based on visual recognition technology, including: a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module; wherein,

[0005] The cloud server is configured as follows:

[0006] Acquire historical garbage data, and process the historical garbage data to obtain garbage image sample data and garbage description data;

[0007] Based on machine vision technology, the first convolutional neural network is trained using the garbage image sample data, and is corrected using the garbage description data to obtain a first garbage recognition model;

[0008] Sending the first garbage identification model to the Internet of Things server;

[0009] The user terminal is configured to: send the registration information input by the first user to the Internet of Things server;

[0010] The IoT server is configured as follows:

[0011] Receive the registration information, establish a first account for the first user and assign the first user a corresponding smart garbage room;

[0012] Acquire first family characteristic information of a first family corresponding to the first user;

[0013] generating first family portrait information according to the first family characteristic information;

[0014] Sending the first garbage recognition model to the smart garbage room;

[0015] Acquire first garbage identification data of the first household;

[0016] generating a first household garbage model of the first household using the first garbage identification data and the first household portrait information;

[0017] The smart garbage room is configured as follows:

[0018] receiving first login information of the first user, and sending the first login information to the Internet of Things server for verification by using the communication module;

[0019] Receiving verification pass information sent by the Internet of Things server;

[0020] Using the monitoring module to monitor whether a first garbage object is dropped;

[0021] When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data;

[0022] Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module;

[0023] Receive a first processing instruction from the Internet of Things server;

[0024] Processing the first garbage data according to the first processing instruction to obtain second garbage data;

[0025] Sending the second garbage data to the Internet of Things server;

[0026] The IoT server is configured as follows:

[0027] Using the first garbage identification model and the first household garbage model, identifying the second garbage data to obtain second garbage identification data;

[0028] sending the second garbage identification data to the smart garbage room;

[0029] The smart garbage room is configured as follows:

[0030] The first garbage object is classified using the classification module according to the second garbage identification data.

[0031] Another aspect of the present invention provides a garbage classification control method based on visual recognition technology, which is applied to a garbage classification system based on visual recognition technology. The garbage classification system based on visual recognition technology includes a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, and a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module. The garbage classification control method based on visual recognition technology includes:

[0032] The cloud server obtains historical garbage data, and processes the historical garbage data to obtain garbage image sample data and garbage description data;

[0033] Based on machine vision technology, the first convolutional neural network is trained using the garbage image sample data, and is corrected using the garbage description data to obtain a first garbage recognition model;

[0034] Sending the first garbage identification model to the Internet of Things server;

[0035] The user terminal sends the registration information input by the first user to the Internet of Things server;

[0036] The Internet of Things server receives the registration information, establishes a first account for the first user, and allocates the corresponding smart garbage room to the first user;

[0037] Acquire first family characteristic information of a first family corresponding to the first user;

[0038] generating first family portrait information according to the first family characteristic information;

[0039] Sending the first garbage recognition model to the smart garbage room;

[0040] Acquire first garbage identification data of the first household;

[0041] generating a first household garbage model of the first household using the first garbage identification data and the first household portrait information;

[0042] The smart garbage room receives the first login information of the first user, and uses the communication module to send the first login information to the Internet of Things server for verification;

[0043] Receiving verification pass information sent by the Internet of Things server;

[0044] Using the monitoring module to monitor whether a first garbage object is dropped;

[0045] When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data;

[0046] Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module;

[0047] Receive a first processing instruction from the Internet of Things server;

[0048] Processing the first garbage data according to the first processing instruction to obtain second garbage data;

[0049] Sending the second garbage data to the Internet of Things server;

[0050] The Internet of Things server identifies the second garbage data by using the first garbage identification model and the first household garbage model to obtain second garbage identification data;

[0051] sending the second garbage identification data to the smart garbage room;

[0052] The smart garbage room classifies the first garbage objects using the classification module according to the second garbage identification data.

[0053] Optionally, the cloud server acquires historical garbage data, and processes the historical garbage data to obtain garbage image sample data and garbage description data, including:

[0054] The cloud server segments the historical junk data to obtain segmented historical junk data;

[0055] Distributing the segmented historical junk data and the data processing model to a plurality of the IoT servers;

[0056] Receiving clean data sent by the plurality of IoT servers after cleaning, classifying and marking the historical garbage data, and fusing the clean data to obtain total clean data;

[0057] The total clean data is processed to obtain the garbage image sample data and the garbage description data.

[0058] Optionally, when the first garbage object is placed, the step of activating the first collection module to collect garbage data to obtain first garbage data includes:

[0059] The first acquisition module sends a first motion request instruction including a first motion duration, a first motion speed, a first motion angle, and a first motion trajectory to the control processing module;

[0060] The control processing module sends the first motion request instruction to the first motion module of the smart garbage room to drive the first garbage object to move in the placement area of ​​the smart garbage room;

[0061] The first acquisition module acquires first image data and first point cloud data of the first garbage object when the first garbage object moves, and sends the first image data and the first point cloud data to the Internet of Things server;

[0062] The Internet of Things server determines whether the first image data and the first point cloud data meet preset requirements;

[0063] If the preset requirements are met, the first image data and the first point cloud data are used as the first junk data;

[0064] If the preset requirements are not met, a first adjustment instruction for the first acquisition module and a second adjustment instruction for the first motion module are sent to the control processing module;

[0065] The control processing module adjusts the working parameters of the first acquisition module and the first motion module according to the first adjustment instruction and the second adjustment instruction respectively to acquire second image data and second point cloud data;

[0066] The second image data and the second point cloud data are used as the first garbage data.

[0067] Optionally, the first family characteristic information includes at least family member information, consumption habit information, eating habit information, hygiene habit information, health status information and education status information of the first family; the step of generating the first family portrait information according to the first family characteristic information includes:

[0068] Acquire a preset family portrait model, and determine a plurality of corresponding family basic attribute data and a first association relationship between the family basic attribute data according to the family portrait model;

[0069] extracting first family basic attribute data from the first family characteristic information according to the family basic attribute data and the first association relationship;

[0070] Inputting the first family basic attribute data into the family portrait model to obtain family portrait information of the first family;

[0071] Acquire a preset personnel portrait model, and determine corresponding multiple personnel basic attribute data and a second association relationship between the personnel basic attribute data according to the personnel portrait model;

[0072] Extracting first basic attribute data of each family member in the first family from the first family characteristic information according to the basic attribute data of the individuals and the second association relationship;

[0073] inputting the first basic attribute data of each family member in the first family into the person portrait model respectively to obtain separate person portrait information of each family member in the first family;

[0074] Using the family portrait information and the person portrait information as the first family portrait information;

[0075] The step of generating a first household garbage model of the first household by using the first garbage identification data and the first household portrait information includes:

[0076] generating an overall garbage model of the first household using the first garbage identification data and the household portrait information;

[0077] Generate a separate personal garbage model for each family member in the first family by using the first garbage identification data and the separate personal portrait information of each family member in the first family respectively;

[0078] The overall garbage model and the personal garbage model are used as the first household garbage model.

[0079] Optionally, the step of the Internet of Things server identifying the second garbage data by using the first garbage identification model and the first household garbage model to obtain second garbage identification data includes:

[0080] The Internet of Things server determines current resident member data of the first household, and determines a corresponding current first household garbage model from the first household garbage model according to the current resident member data;

[0081] Modifying the first garbage recognition model using the current first household garbage model to obtain a first household garbage recognition model;

[0082] The second garbage data is identified by using the first household garbage identification model to obtain the second garbage identification data.

[0083] Optionally, after the step of classifying the first garbage objects using the classification module according to the second garbage identification data, the smart garbage room further includes:

[0084] The Internet of Things server sends a first acquisition instruction to the first acquisition module;

[0085] The first collection module parses the first collection instruction to obtain a first collection model, and collects data on the first garbage object according to the first collection model to obtain first data;

[0086] Sending the first data to the Internet of Things server;

[0087] The Internet of Things server generates health advice data for the first family based on the first data and the first family portrait information, and sends the health advice data to the user terminal, and / or sends the health advice data to the output module of the smart garbage room for output.

[0088] Optionally, after the IoT server generates the recommended data for the first family according to the first data and the first family portrait information, the method further includes:

[0089] The Internet of Things server generates garbage classification publicity data of the first family according to the first data and the first family portrait information;

[0090] Generate garbage classification knowledge data and garbage pre-processing recommended measures based on the garbage classification publicity data;

[0091] The garbage classification knowledge data and the garbage pre-processing recommended measures are sent to the user terminal, and / or sent to the output module of the smart garbage room for output.

[0092] Optionally, after the step of classifying the first garbage objects using the classification module according to the second garbage identification data, the smart garbage room further includes:

[0093] Transferring the first classified garbage objects to corresponding packaging areas in order according to their categories;

[0094] The packaging module prints corresponding category labels and garbage information on different packages in turn, and puts the corresponding garbage into the packages for packaging;

[0095] The control processing module sends the package identification of the package, the category label and the junk information to the Internet of Things server.

[0096] Optionally, after the step of the control processing module sending the package identification of the package, the category label and the junk information to the Internet of Things server, the step further includes:

[0097] The Internet of Things server determines the corresponding unmanned transportation terminal to transfer the packaged garbage according to the package identification, the category label and the garbage information.

[0098] By adopting the technical solution of the present invention, the garbage classification system based on visual recognition technology is provided with a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module. By adopting the solution of the present invention, the visual recognition technology is used to make garbage classification more intelligent, more accurate and more efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] Figure 1 is a schematic block diagram of a garbage classification system based on visual recognition technology provided by an embodiment of the present invention;

[0100] Figure 2 It is a flow chart of a garbage classification control method based on visual recognition technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0101] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0102] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0103] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0104] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0105] Refer to the following Figure 1 to Figure 2 To describe a garbage classification system and a control method thereof based on visual recognition technology provided according to some embodiments of the present invention.

[0106] like Figure 1 As shown, an embodiment of the present invention provides a garbage classification system based on visual recognition technology, including: a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module; wherein,

[0107] The cloud server is configured as follows:

[0108] Obtain historical junk data, and process the historical junk data to obtain junk image sample data and junk description data; the historical junk data can be data that is identified and labeled / described by an artificial intelligence algorithm or manually, and specifically can be identified / labeled on a distributed server / terminal.

[0109] Based on machine vision technology, the first convolutional neural network is trained using the junk image sample data, and is corrected using the junk description data to obtain a first junk recognition model; that is, the cloud server trains the first convolutional neural network using a large amount of junk image sample data, and corrects the first convolutional neural network using the junk description data to obtain a first junk recognition model.

[0110] The first garbage identification model is sent to the Internet of Things server; that is, the cloud server sends the universal first garbage identification model to all Internet of Things servers within its management scope.

[0111] The user terminal is configured to: send the registration information input by the first user to the Internet of Things server;

[0112] The user terminal can download and install the corresponding application / client to register the user, enter the user's geographical location or community name, user identity information and other information to generate registration information, and send the registration information to the Internet of Things server corresponding to the user's geographical location or community name (the Internet of Things server with a management relationship); the user terminal can also initiate a registration request after communicating with the smart garbage room, and the smart garbage room will send the registration request to the corresponding Internet of Things server. After the user terminal establishes a connection with the Internet of Things server, it sends the registration information for registration.

[0113] The IoT server is configured as follows:

[0114] The registration information is received, a first account is established for the first user, and the corresponding smart garbage room is allocated to the first user; the Internet of Things server performs user registration management and saves relevant registration data, which can not only ensure data security but also reduce the operating pressure of the cloud server.

[0115] Acquire first family characteristic information of a first family corresponding to the first user;

[0116] The first family portrait information is generated based on the first family characteristic information to express the first family's individuality (such as what kind of items they like, how they like to deal with garbage, etc.), regularity (such as using another item while using another item, discarding another item after a basically fixed period of time after discarding a certain item, etc.), and periodicity (such as discarding the same item at basically fixed intervals).

[0117] Sending the first garbage recognition model to the smart garbage room to quickly obtain recognition data of a new user when placing garbage for the first time, so as to provide data support for the subsequent establishment of a household garbage model for the new user;

[0118] The first garbage identification data of the first family is obtained; that is, the garbage put into the smart garbage room by the first family is preliminarily identified using the first garbage identification model to obtain the first garbage identification data.

[0119] generating a first household garbage model of the first household using the first garbage identification data and the first household portrait information;

[0120] The first household waste model includes some predictions / determinations of the individuality, regularity, periodicity, etc. of the waste generated by the first household to assist in identifying the waste placed by the first household.

[0121] The smart garbage room is configured as follows:

[0122] receiving first login information of the first user, and sending the first login information to the Internet of Things server for verification by using the communication module;

[0123] Receiving verification pass information sent by the Internet of Things server;

[0124] The monitoring module is used to monitor whether the first garbage object is thrown in; the monitoring module can monitor whether the first garbage object is thrown in by infrared sensing, ultrasonic sensing, pressure sensing, image sensing and the like.

[0125] When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data;

[0126] Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module;

[0127] The Internet of Things server is also configured as:

[0128] receiving the first identification request information (including at least the data volume and data type of the first garbage data), parsing the first identification request information, and judging whether the first garbage data is to be preprocessed by the Internet of Things server or the smart garbage room according to the data volume to be identified parsed from the first identification request information and in combination with its own workload (when the workload of the Internet of Things server exceeds a preset threshold, the smart garbage room performs preprocessing), and generating a corresponding first processing instruction when the smart garbage room needs to perform data preprocessing;

[0129] The smart garbage room is configured as follows:

[0130] Receive a first processing instruction from the Internet of Things server;

[0131] Processing the first garbage data according to the first processing instruction to obtain second garbage data;

[0132] Sending the second garbage data to the IoT server, and having the IoT server with stronger data processing function and higher algorithm execution efficiency perform subsequent garbage identification operations;

[0133] The IoT server is configured as follows:

[0134] Using the first garbage identification model and the first household garbage model, identifying the second garbage data to obtain second garbage identification data;

[0135] sending the second garbage identification data to the smart garbage room;

[0136] The smart garbage room is configured as follows:

[0137] The first garbage object is classified using the classification module according to the second garbage identification data.

[0138] In this embodiment, the IoT server manages all intelligent garbage rooms, user terminals and other terminals within a park / community / street (administrative division unit), while a cloud server manages all IoT servers, garbage disposal stations and other terminals within a city / province. The cloud server uses the IoT servers within its management scope to clean, classify and mark historical garbage data, improving processing efficiency and resource utilization; in particular, distributing pre-trained data processing models to multiple IoT servers ensures the uniformity of the data processing methods / standards of all IoT servers, greatly facilitating subsequent data fusion.

[0139] The technical solution of this embodiment is adopted, and the system includes a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module. Through the solution of the present invention, the visual recognition technology is used to make garbage classification more intelligent, more accurate and more efficient.

[0140] It should be known that Figure 1 The block diagram of the garbage classification system based on visual recognition technology is for illustration only, and the number of modules / terminals shown therein does not limit the protection scope of the present invention.

[0141] See also Figure 2 Another embodiment of the present invention provides a garbage classification control method based on visual recognition technology, which is applied to a garbage classification system based on visual recognition technology. The garbage classification system based on visual recognition technology includes a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module. The garbage classification control method based on visual recognition technology includes:

[0142] The cloud server obtains historical garbage data, and processes the historical garbage data to obtain garbage image sample data and garbage description data;

[0143] In this step, historical junk data can be identified and labeled / described by an artificial intelligence algorithm or manually, and specifically can be identified / labeled on a distributed server / terminal.

[0144] Based on machine vision technology, the first convolutional neural network is trained using the garbage image sample data, and is corrected using the garbage description data to obtain a first garbage recognition model;

[0145] In this step, the cloud server trains the first convolutional neural network using a large amount of garbage image sample data, and uses the garbage description data to correct the first convolutional neural network to obtain a first garbage recognition model.

[0146] Sending the first garbage identification model to the Internet of Things server;

[0147] In this step, the cloud server sends the universal first garbage identification model to all IoT servers within its management scope.

[0148] The user terminal sends the registration information input by the first user to the Internet of Things server;

[0149] In this step, the user terminal can download and install the corresponding application / client to perform user registration, enter the user's geographical location or community name, user identity information and other information to generate registration information, and send the registration information to the Internet of Things server corresponding to the user's geographical location or community name (the Internet of Things server with a management relationship); the user terminal can also initiate a registration request after communicating with the smart garbage room, and the smart garbage room will send the registration request to the corresponding Internet of Things server. After the user terminal establishes a connection with the Internet of Things server, it sends the registration information for registration.

[0150] The Internet of Things server receives the registration information, establishes a first account for the first user, and allocates the corresponding smart garbage room to the first user;

[0151] In this step, the IoT server performs user registration management and saves relevant registration data, which can not only ensure data security but also reduce the operating pressure of the cloud server.

[0152] The Internet of Things server obtains first family characteristic information of a first family corresponding to the first user;

[0153] The Internet of Things server generates first family portrait information according to the first family characteristic information;

[0154] In this step, first family portrait information is generated based on the first family characteristic information, which is used to express some characteristics of the first family, such as personalization (such as what kind of items they like, how they like to deal with garbage, etc.), regularity (such as using a certain item while using another item, discarding another item after a basically fixed time after discarding a certain item, etc.), and periodicity (such as discarding the same item at a basically fixed interval of time).

[0155] The IoT server sends the first garbage identification model to the smart garbage room to quickly obtain identification data of a new user when placing garbage for the first time, so as to provide data support for the subsequent establishment of a household garbage model for the new user;

[0156] The Internet of Things server obtains first garbage identification data of the first household;

[0157] In this step, the first garbage identification model can be used to perform preliminary identification on the garbage put into the smart garbage room by the first family to obtain first garbage identification data.

[0158] The IoT server generates a first household garbage model of the first household using the first garbage identification data and the first household portrait information;

[0159] In this step, the first garbage identification data and the first family portrait information are used to generate a first household garbage model for the first household. The first household garbage model includes some predictions / determinations on the personalization, regularity, periodicity, etc. of the garbage generated by the first household to assist in identifying the garbage disposed of by the first household.

[0160] The smart garbage room receives the first login information of the first user, and uses the communication module to send the first login information to the Internet of Things server for verification;

[0161] The smart garbage room receives the verification pass information sent by the Internet of Things server;

[0162] The smart garbage room uses the monitoring module to monitor whether a first garbage object is placed;

[0163] In this step, the monitoring module can monitor whether the first garbage object is thrown in through infrared sensing, ultrasonic sensing, pressure sensing, image sensing and the like.

[0164] When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data;

[0165] In this step, the first acquisition module may be a camera module, a lidar scanning module, etc.

[0166] Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module;

[0167] The Internet of Things server receives the first identification request information (including at least the data volume and data type of the first garbage data), parses the first identification request information, and determines whether the Internet of Things server or the smart garbage room pre-processes the first garbage data according to the data volume to be identified parsed from the first identification request information and its own workload (when the workload of the Internet of Things server exceeds a preset threshold, the smart garbage room performs pre-processing), and generates a corresponding first processing instruction when the smart garbage room needs to perform data pre-processing;

[0168] Receiving the first processing instruction sent by the Internet of Things server;

[0169] Processing the first garbage data according to the first processing instruction to obtain second garbage data;

[0170] Sending the second garbage data to the IoT server, and having the IoT server with stronger data processing function and higher algorithm execution efficiency perform subsequent garbage identification operations;

[0171] The Internet of Things server identifies the second garbage data by using the first garbage identification model and the first household garbage model to obtain second garbage identification data;

[0172] sending the second garbage identification data to the smart garbage room;

[0173] The smart garbage room classifies the first garbage objects using the classification module according to the second garbage identification data.

[0174] It can be understood that in the embodiments of the present invention, the Internet of Things server manages all smart garbage rooms, user terminals and other terminals within a park / community / street (administrative area division unit), while a cloud server manages all Internet of Things servers, garbage disposal stations and other terminals within a city / province.

[0175] The technical solution of this embodiment uses visual recognition technology to make garbage classification smarter, more accurate and more efficient.

[0176] In some possible implementations of the present invention, the cloud server obtains historical garbage data, and the step of processing the historical garbage data to obtain garbage image sample data and garbage description data includes:

[0177] The cloud server segments the historical junk data to obtain segmented historical junk data;

[0178] Distributing the segmented historical junk data and the data processing model to a plurality of the IoT servers;

[0179] Receiving clean data sent by the plurality of IoT servers after cleaning, classifying and marking the historical garbage data, and fusing the clean data to obtain total clean data;

[0180] The total clean data is processed to obtain the garbage image sample data and the garbage description data.

[0181] In this embodiment, the cloud server uses the IoT servers within its management scope to clean, classify and mark historical garbage data, thereby improving processing efficiency and resource utilization; in particular, the pre-trained data processing model is distributed to multiple IoT servers, ensuring the uniformity of the data processing methods / standards of all IoT servers, greatly facilitating subsequent data fusion.

[0182] In order to ensure the accuracy of identification, in some possible implementations of the present invention, when the first garbage object is thrown, the step of enabling the first collection module to collect garbage data to obtain the first garbage data includes:

[0183] The first acquisition module sends a first motion request instruction including at least a first motion duration, a first motion speed, a first motion angle, and a first motion trajectory to the control processing module;

[0184] The control processing module sends the first motion request instruction to the first motion module of the smart garbage room to drive the first garbage object to move in the delivery area of ​​the smart garbage room so that the first collection module can collect as detailed data as possible from multiple angles to ensure the accuracy of recognition;

[0185] The first acquisition module acquires first image data and first point cloud data of the first garbage object when the first garbage object moves, and sends the first image data and the first point cloud data to the Internet of Things server;

[0186] Since only when the collected data meets the preset requirements can the accuracy of the recognition result be better guaranteed, the Internet of Things server determines whether the first image data and the first point cloud data meet the preset requirements according to the preset junk data standard model;

[0187] If the preset requirements are met, the first image data and the first point cloud data are used as the first junk data;

[0188] If the preset requirements are not met, a first adjustment instruction for the first acquisition module (such as an instruction for adjusting shooting parameters, scanning power, acquisition frequency and other parameters) and a second adjustment instruction for the first motion module (such as an instruction for adjusting the second motion duration, second motion speed, second motion angle, second motion trajectory and other parameters of the first motion module) are sent to the control processing module;

[0189] The control processing module adjusts the working parameters of the first acquisition module and the first motion module according to the first adjustment instruction and the second adjustment instruction respectively to acquire second image data and second point cloud data;

[0190] The second image data and the second point cloud data are used as the first garbage data.

[0191] In this embodiment, by collecting the point cloud data and image data of the first garbage object, it is ensured that the collected data can most realistically represent the first garbage object.

[0192] In some possible implementations of the present invention, the first family characteristic information at least includes family member information (such as family population, maximum / minimum age, generational structure, etc.), consumption habit information, eating habit information, hygiene habit information, health status information, and education status information of the first family; the step of generating the first family portrait information according to the first family characteristic information includes:

[0193] Acquire a preset family portrait model, and determine a plurality of corresponding family basic attribute data and a first association relationship between the family basic attribute data according to the family portrait model;

[0194] In this step, because the basic family attribute data determined according to the family portrait model may be different from the description in the first family characteristic information, the first association relationship between the basic family attribute data is extracted according to the semantic recognition algorithm to facilitate the accuracy and comprehensiveness of the subsequent extraction of the first family basic attribute data from the first family characteristic information based on the association relationship.

[0195] extracting first family basic attribute data from the first family characteristic information according to the family basic attribute data and the first association relationship;

[0196] Inputting the first family basic attribute data into the family portrait model to obtain family portrait information of the first family;

[0197] Acquire a preset personnel portrait model, and determine corresponding multiple personnel basic attribute data and a second association relationship between the personnel basic attribute data according to the personnel portrait model;

[0198] In this step, because the basic attribute data of the personnel determined according to the personnel portrait model may be different from the description in the first family characteristic information, the second association relationship between the basic attribute data of the personnel is extracted according to the semantic recognition algorithm to facilitate the accuracy and comprehensiveness of the subsequent extraction of the basic attribute data of the first person from the first family characteristic information based on the association relationship.

[0199] Extracting first basic attribute data of each family member in the first family from the first family characteristic information according to the basic attribute data of the individuals and the second association relationship;

[0200] inputting the first basic attribute data of each family member in the first family into the person portrait model respectively to obtain separate person portrait information of each family member in the first family;

[0201] Using the family portrait information and the person portrait information as the first family portrait information;

[0202] The step of generating a first household garbage model of the first household by using the first garbage identification data and the first household portrait information includes:

[0203] generating an overall garbage model of the first household using the first garbage identification data and the household portrait information;

[0204] Generate a separate personal garbage model for each family member in the first family by using the first garbage identification data and the separate personal portrait information of each family member in the first family respectively;

[0205] The overall garbage model and the personal garbage model are used as the first household garbage model.

[0206] It is understandable that in practice there may be situations where family members do not live in this residence periodically (for example, a family member periodically goes on business trips or students return to school, etc.). In order to ensure high-precision identification of garbage generated by the family, in this embodiment, family portrait information of the first family and separate personal portrait information of the family members are generated. Combined with the first garbage identification data, an overall garbage model of the first family and separate personal garbage models of each family member in the first family are generated. The overall garbage model and the personal garbage model are used as the first family garbage model. According to the periodic characteristics of family members, appropriate parts can be selected from the first family garbage model to assist in garbage identification.

[0207] In some possible implementations of the present invention, the step of the Internet of Things server identifying the second garbage data using the first garbage identification model and the first household garbage model to obtain second garbage identification data includes:

[0208] The Internet of Things server determines current resident member data of the first household, and determines a corresponding current first household garbage model from the first household garbage model according to the current resident member data;

[0209] Modifying the first garbage recognition model using the current first household garbage model (such as modifying the corresponding weights in the model) to obtain a first household garbage recognition model;

[0210] The second garbage data is identified by using the first household garbage identification model to obtain the second garbage identification data.

[0211] It can be understood that in this embodiment, the Internet of Things server determines the current resident member data of the first family, and determines the corresponding current first family garbage model from the first family garbage model based on the current resident member data, thereby ensuring the accuracy of the model; the first garbage identification model is modified using the current first family garbage model (for example, modifying the corresponding weights in the model) to obtain a more accurate first family garbage identification model.

[0212] In some possible implementations of the present invention, after the step of classifying the first garbage object using the classification module according to the second garbage identification data, the smart garbage room further includes:

[0213] The Internet of Things server sends a first acquisition instruction to the first acquisition module;

[0214] The first acquisition module parses the first acquisition model from the first acquisition instruction (such as a health data acquisition model, a garbage disposal method acquisition model, etc., and including the first acquisition model in the first acquisition instruction can also better ensure the efficiency of the first acquisition module and the quality of the collected data), and collects data on the first garbage object according to the first acquisition model to obtain first data;

[0215] Sending the first data to the Internet of Things server;

[0216] The Internet of Things server generates health advice data for the first family based on the first data and the first family portrait information, and sends the health advice data to the user terminal, and / or sends the health advice data to the output module of the smart garbage room for output.

[0217] It is understandable that people's lifestyle / habit information can be determined from garbage information (such as favorite items / foods, used medicines, etc. can be analyzed from product packaging, and eating habits can be analyzed from kitchen waste, etc.), so that the health advice data of the first family can be generated in combination with the first family portrait information to provide users with a better experience. Preferably, non-sensitive health advice data is output by the smart garbage room for the current logged-in user, and sensitive health advice data is sent to the user terminal. It should be noted that the first collection model includes operation instructions for collecting data on the first garbage object, for example, the health data collection model includes operation instructions on how to collect data on item packaging bags / boxes, kitchen waste, etc., and data reflecting health status can be collected according to the operation instructions; for another example, the garbage disposal method collection model includes operation instructions on how to collect data on the user's garbage disposal method (such as whether to pre-classify, whether to perform corresponding special treatment on items such as batteries and glass, etc.), and data reflecting the user's garbage disposal method can be collected according to the operation instructions.

[0218] In order to promote garbage classification / processing knowledge in a targeted manner and better implement garbage classification, in some possible implementations of the present invention, after the IoT server generates the recommended data for the first family according to the first data and the first family portrait information, it also includes:

[0219] The IoT server generates the garbage classification publicity data of the first household based on the first data (such as cutting, tearing and other processing trace data, garbage packaging data / pre-classification data, etc.), the first household portrait information, and the first household garbage model;

[0220] Generate garbage classification knowledge data and garbage pre-processing recommended measures based on the garbage classification publicity data;

[0221] The garbage classification knowledge data and the garbage pre-processing recommended measures are sent to the user terminal, and / or sent to the output module of the smart garbage room for output.

[0222] In some possible implementations of the present invention, after the step of classifying the first garbage object using the classification module according to the second garbage identification data, the smart garbage room further includes:

[0223] Transferring the first classified garbage objects to corresponding packaging areas in order according to their categories;

[0224] The packaging module prints corresponding category labels and garbage information on different packages in turn, and puts the corresponding garbage into the packages for packaging;

[0225] The control processing module sends the package identification of the package, the category label and the junk information to the Internet of Things server.

[0226] It can be understood that in order to facilitate the transportation and management of classified garbage, in this embodiment, the classified second garbage objects are transported to the corresponding packaging area in the smart garbage room in sequence according to category; the packaging module prints the corresponding category label and garbage information on each package in sequence through the printing unit, and packs the corresponding garbage into the package printed with the corresponding category label according to category; the control processing module sends the package identification, the category label and the garbage information of the package to the Internet of Things server for statistics and generation of work tasks.

[0227] In order to ensure the high efficiency of the transportation of the classified garbage, in some possible implementations of the present invention, after the step of the control processing module sending the package identification of the package, the category label and the garbage information to the Internet of Things server, the control processing module further includes:

[0228] The Internet of Things server determines the corresponding unmanned transportation terminal to transfer the packaged garbage based on the package identification, the category label and the garbage information. Specifically, it can be: the Internet of Things server generates first transportation task information based on the package identification, the category label and the garbage information, and obtains the transportation terminal attribute information and the first task table (tasks that each unmanned transportation terminal needs to handle) of all the unmanned transportation terminals, and determines the corresponding appropriate (first) unmanned transportation terminal from the unmanned transportation terminals based on the first transportation task information, the first task table and the transportation terminal attribute information (such as determining the type (such as unmanned aircraft, unmanned ship or unmanned vehicle, etc.) and number of unmanned transportation terminals based on the type of garbage transported, the amount / volume of garbage, whether it is a special item, the terrain of the garbage treatment station and the surrounding traffic conditions, etc.); the Internet of Things server sends the first transportation task information to the appropriate (first) unmanned transportation terminal, which transfers the packaged garbage.

[0229] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0230] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0231] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0232] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0233] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0234] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0235] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0236] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0237] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions without departing from the spirit and scope of the present invention, and can make various changes and modifications, including the combination of the above-mentioned different functions and implementation steps, including software and hardware implementation methods, all of which are within the scope of protection of the present invention.

Claims

1. A garbage classification system based on visual recognition technology, characterized in that: include: A user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room includes a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module; wherein, The cloud server is configured as follows: Acquire historical garbage data, and process the historical garbage data to obtain garbage image sample data and garbage description data; Based on machine vision technology, the first convolutional neural network is trained using the garbage image sample data, and is corrected using the garbage description data to obtain a first garbage recognition model; Sending the first garbage identification model to the Internet of Things server; The user terminal is configured to: send the registration information input by the first user to the Internet of Things server; The IoT server is configured as follows: Receive the registration information, establish a first account for the first user and assign the first user a corresponding smart garbage room; Acquire first family characteristic information of a first family corresponding to the first user; generating first family portrait information according to the first family characteristic information; Sending the first garbage recognition model to the smart garbage room; Acquire first garbage identification data of the first household; generating a first household garbage model of the first household using the first garbage identification data and the first household portrait information; The smart garbage room is configured as follows: receiving first login information of the first user, and sending the first login information to the Internet of Things server for verification by using the communication module; Receiving verification pass information sent by the Internet of Things server; Using the monitoring module to monitor whether a first garbage object is dropped; When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data; Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module; Receive a first processing instruction from the Internet of Things server; Processing the first garbage data according to the first processing instruction to obtain second garbage data; Sending the second garbage data to the Internet of Things server; The IoT server is configured as follows: Using the first garbage identification model and the first household garbage model, identifying the second garbage data to obtain second garbage identification data; sending the second garbage identification data to the smart garbage room; The smart garbage room is configured as follows: classifying the first garbage object using the classification module according to the second garbage identification data; The step of acquiring historical garbage data and processing the historical garbage data to obtain garbage image sample data and garbage description data includes: The cloud server segments the historical junk data to obtain segmented historical junk data; Distributing the segmented historical junk data and the data processing model to a plurality of the IoT servers; Receiving clean data sent by the plurality of IoT servers after cleaning, classifying and marking the historical garbage data, and fusing the clean data to obtain total clean data; Processing the total clean data to obtain the garbage image sample data and the garbage description data; Wherein, when the first garbage object is placed, the step of activating the first collection module to collect garbage data to obtain first garbage data includes: The first acquisition module sends a first motion request instruction including a first motion duration, a first motion speed, a first motion angle, and a first motion trajectory to the control processing module; The control processing module sends the first motion request instruction to the first motion module of the smart garbage room to drive the first garbage object to move in the placement area of ​​the smart garbage room; The first acquisition module acquires first image data and first point cloud data of the first garbage object when the first garbage object moves, and sends the first image data and the first point cloud data to the Internet of Things server; The Internet of Things server determines whether the first image data and the first point cloud data meet preset requirements; If the preset requirements are met, the first image data and the first point cloud data are used as the first junk data; If the preset requirements are not met, a first adjustment instruction for the first acquisition module and a second adjustment instruction for the first motion module are sent to the control processing module; The control processing module adjusts the working parameters of the first acquisition module and the first motion module according to the first adjustment instruction and the second adjustment instruction respectively to acquire second image data and second point cloud data; The second image data and the second point cloud data are used as the first garbage data.

2. A garbage classification control method based on visual recognition technology, characterized in that: The invention is applied to a garbage classification system based on visual recognition technology, the garbage classification system based on visual recognition technology comprises a user terminal, an intelligent garbage room, an Internet of Things communication gateway, an Internet of Things server connected to the intelligent garbage room through the Internet of Things communication gateway, a cloud server and an unmanned transportation terminal connected to the Internet of Things server through the Internet of Things communication gateway respectively; the intelligent garbage room comprises a monitoring module, a first acquisition module, a control processing module, a communication module for receiving and sending data, a classification module and a packaging module, and the garbage classification control method based on visual recognition technology comprises: The cloud server obtains historical garbage data, and processes the historical garbage data to obtain garbage image sample data and garbage description data; Based on machine vision technology, the first convolutional neural network is trained using the garbage image sample data, and is corrected using the garbage description data to obtain a first garbage recognition model; Sending the first garbage identification model to the Internet of Things server; The user terminal sends the registration information input by the first user to the Internet of Things server; The Internet of Things server receives the registration information, establishes a first account for the first user, and allocates the corresponding smart garbage room to the first user; Acquire first family characteristic information of a first family corresponding to the first user; generating first family portrait information according to the first family characteristic information; Sending the first garbage recognition model to the smart garbage room; Acquire first garbage identification data of the first household; generating a first household garbage model of the first household using the first garbage identification data and the first household portrait information; The smart garbage room receives the first login information of the first user, and uses the communication module to send the first login information to the Internet of Things server for verification; Receiving verification pass information sent by the Internet of Things server; Using the monitoring module to monitor whether a first garbage object is dropped; When the first garbage object is thrown, the first collection module is activated to collect garbage data to obtain first garbage data; Generate first identification request information by using the control processing module, and send the first identification request information to the Internet of Things server through the communication module; Receive a first processing instruction from the Internet of Things server; Processing the first garbage data according to the first processing instruction to obtain second garbage data; Sending the second garbage data to the Internet of Things server; The Internet of Things server identifies the second garbage data by using the first garbage identification model and the first household garbage model to obtain second garbage identification data; sending the second garbage identification data to the smart garbage room; The smart garbage room classifies the first garbage objects using the classification module according to the second garbage identification data; The cloud server acquires historical garbage data, and processes the historical garbage data to obtain garbage image sample data and garbage description data, including: The cloud server segments the historical junk data to obtain segmented historical junk data; Distributing the segmented historical junk data and the data processing model to a plurality of the IoT servers; Receiving clean data sent by the plurality of IoT servers after cleaning, classifying and marking the historical garbage data, and fusing the clean data to obtain total clean data; Processing the total clean data to obtain the garbage image sample data and the garbage description data; Wherein, when the first garbage object is placed, the step of activating the first collection module to collect garbage data to obtain first garbage data includes: The first acquisition module sends a first motion request instruction including a first motion duration, a first motion speed, a first motion angle, and a first motion trajectory to the control processing module; The control processing module sends the first motion request instruction to the first motion module of the smart garbage room to drive the first garbage object to move in the placement area of ​​the smart garbage room; The first acquisition module acquires first image data and first point cloud data of the first garbage object when the first garbage object moves, and sends the first image data and the first point cloud data to the Internet of Things server; The Internet of Things server determines whether the first image data and the first point cloud data meet preset requirements; If the preset requirements are met, the first image data and the first point cloud data are used as the first junk data; If the preset requirements are not met, a first adjustment instruction for the first acquisition module and a second adjustment instruction for the first motion module are sent to the control processing module; The control processing module adjusts the working parameters of the first acquisition module and the first motion module according to the first adjustment instruction and the second adjustment instruction respectively to acquire second image data and second point cloud data; The second image data and the second point cloud data are used as the first garbage data.

3. The garbage classification control method based on visual recognition technology according to claim 2 is characterized in that: The first family characteristic information at least includes family member information, consumption habit information, eating habit information, hygiene habit information, health status information and education status information of the first family; the step of generating the first family portrait information according to the first family characteristic information includes: Acquire a preset family portrait model, and determine a plurality of corresponding family basic attribute data and a first association relationship between the family basic attribute data according to the family portrait model; extracting first family basic attribute data from the first family characteristic information according to the family basic attribute data and the first association relationship; Inputting the first family basic attribute data into the family portrait model to obtain family portrait information of the first family; Acquire a preset personnel portrait model, and determine corresponding multiple personnel basic attribute data and a second association relationship between the personnel basic attribute data according to the personnel portrait model; Extracting first basic attribute data of each family member in the first family from the first family characteristic information according to the basic attribute data of the individuals and the second association relationship; inputting the first basic attribute data of each family member in the first family into the person portrait model respectively to obtain separate person portrait information of each family member in the first family; Using the family portrait information and the person portrait information as the first family portrait information; The step of generating a first household garbage model of the first household by using the first garbage identification data and the first household portrait information includes: generating an overall garbage model of the first household using the first garbage identification data and the household portrait information; Generate a separate personal garbage model for each family member in the first family by using the first garbage identification data and the separate personal portrait information of each family member in the first family respectively; The overall garbage model and the personal garbage model are used as the first household garbage model.

4. The garbage classification control method based on visual recognition technology according to claim 3 is characterized in that: The step of the Internet of Things server identifying the second garbage data by using the first garbage identification model and the first household garbage model to obtain second garbage identification data includes: The Internet of Things server determines current resident member data of the first household, and determines a corresponding current first household garbage model from the first household garbage model according to the current resident member data; Modifying the first garbage recognition model using the current first household garbage model to obtain a first household garbage recognition model; The second garbage data is identified by using the first household garbage identification model to obtain the second garbage identification data.

5. The garbage classification control method based on visual recognition technology according to claim 4 is characterized in that: After the step of classifying the first garbage objects by using the classification module according to the second garbage identification data, the smart garbage room further includes: The Internet of Things server sends a first acquisition instruction to the first acquisition module; The first acquisition module parses the first acquisition instruction to obtain a first acquisition model, and performs data acquisition on the first garbage object according to the first acquisition model to obtain first data; Sending the first data to the Internet of Things server; The Internet of Things server generates health advice data for the first family based on the first data and the first family portrait information, and sends the health advice data to the user terminal, and / or sends the health advice data to the output module of the smart garbage room for output.

6. The garbage classification control method based on visual recognition technology according to claim 5 is characterized in that: After the Internet of Things server generates the recommended data for the first family according to the first data and the first family portrait information, the method further includes: The Internet of Things server generates garbage classification publicity data of the first family according to the first data and the first family portrait information; Generate garbage classification knowledge data and garbage pre-processing recommended measures based on the garbage classification publicity data; The garbage classification knowledge data and the garbage pre-processing recommended measures are sent to the user terminal, and / or sent to the output module of the smart garbage room for output.

7. The garbage classification control method based on visual recognition technology according to claim 6 is characterized in that: After the step of classifying the first garbage objects by using the classification module according to the second garbage identification data, the smart garbage room further includes: Transferring the first classified garbage objects to corresponding packaging areas in order according to their categories; The packaging module prints corresponding category labels and garbage information on different packages in turn, and puts the corresponding garbage into the packages for packaging; The control processing module sends the package identification of the package, the category label and the junk information to the Internet of Things server.

8. The garbage classification control method based on visual recognition technology according to claim 7 is characterized in that: After the step of the control processing module sending the package identification of the package, the category label and the junk information to the Internet of Things server, the method further includes: The Internet of Things server determines the corresponding unmanned transportation terminal to transfer the packaged garbage according to the package identification, the category label and the garbage information.

Images