A garbage sorting terminal based on an Internet of Things and a platform implementation method thereof

Abstract

The present application relates to waste management technical field, especially to a kind of garbage sorting terminal based on Internet of Things and platform implementation method thereof.It includes pre-sorting processing unit, for receiving the garbage of putting, and identifying garbage type obtains large garbage and bagged garbage, bagged garbage is broken bag, and the bagged garbage is air selected to obtain to be layered garbage and air selected garbage, according to size filter layering to be layered garbage, and obtain layered garbage;Multi-sensing sorting unit, for the layered garbage is sequentially placed to the garbage of multi-sensing sorting unit, and the garbage type is sequentially handled and identified, and according to garbage type is classified to obtain to be compressed garbage and rigid garbage;Compression unit, for respectively receiving to be compressed garbage and air selected garbage and compressing, and obtaining compressed garbage;Recycling unit, for respectively receiving compressed garbage, rigid garbage and large garbage.Effective classification of garbage is realized and guiding transport vehicle efficient processing recovery.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of waste management, in particular to a garbage sorting terminal based on the Internet of Things and a platform implementation method thereof. BACKGROUND

[0002] With the continuous acceleration of urbanization, the amount of urban garbage is increasing day by day, and garbage disposal has become the focus of attention of the whole society. Among various places, the complexity of community garbage is particularly prominent. Its types are various, materials are different, and dry and wet mixing, which makes the sorting and classification of community garbage face great challenges. Compared with the garbage in construction sites, office areas and schools, the source of community garbage is more extensive, including daily life garbage of residents, commercial garbage, etc. The nature and characteristics of these garbage are different, which increases the difficulty of sorting and classification.

[0003] At present, intelligent garbage sorting systems based on image recognition are gradually applied in the field of garbage disposal. The working principle of the system is to take pictures of garbage with the help of color camera technology, and then automatically identify the material, type and specific location of the target object through image processing algorithm. These information will be sent to the subsequent work equipment, and the terminal equipment will carry out targeted grabbing after receiving the instruction, so as to realize intelligent garbage sorting. In this process, image recognition technology plays a key role, which is the key to realize intelligent sorting.

[0004] However, the existing artificial intelligence technology has some obvious problems in garbage sorting. First of all, these technologies may encounter "fraud", which leads to the inability to achieve comprehensive and accurate classification of garbage. For example, when the discarded objects appear broken, dirty, deformed, etc., ordinary image recognition technology often fails to accurately identify the type of garbage, which seriously reduces the effectiveness of intelligent classification. Secondly, for some special garbage, such as rice dumpling leaves, cosmetics, etc., or whether there is water in the beverage bottle, or the serious extrusion deformation, and the different classification rules in different places, image recognition technology will also encounter difficulties in application. The existence of these problems makes the intelligent garbage sorting system not satisfactory in practical application.

[0005] Intelligent garbage recycling is a comprehensive system engineering, which needs strong technical support. In the aspect of garbage recognition, high-precision recognition technology is needed to accurately judge the type and nature of garbage; in the aspect of garbage classification, efficient classification algorithm is needed to quickly classify garbage; in the aspect of garbage data processing information collection, reliable data processing system is needed to collect and process garbage related data in time, and in the aspect of building Internet of Things platform, garbage type quantity position and other related information need to be collected, and transportation route and recycling method need to be optimized, etc. There is a lack of related research in the above aspects.

[0006] The most important problem to be solved by the garbage sorting terminal is the effective sorting of garbage with different characteristics. In this process, the garbage needs to be dispersed, identified and picked, etc. However, due to the complexity of mixed garbage, the above three steps cannot be implemented one by one, that is, a complete dispersion, identification and picking process can only sort out one type of garbage, and each operation period needs to be redesigned to adapt to the operation process due to the different properties of the garbage.

[0007] It is of great significance to study how to realize efficient garbage collection and management, focus on the garbage sorting intelligent terminal, ensure the miniaturization, intelligence and low cost of the terminal, effectively identify all types of garbage, ensure low noise and no odor during the sorting process, realize efficient interconnection of the intelligent sorting terminal and the platform by using the Internet of Things technology, and complete a series of work such as transportation, treatment and recycling after garbage sorting. SUMMARY

[0008] In order to solve the above-mentioned problems, the application provides a garbage sorting terminal based on the Internet of Things and an implementation method of the platform.

[0009] In a first aspect, the application provides a garbage sorting terminal based on the Internet of Things, which adopts the following technical scheme:

[0010] A garbage sorting terminal based on the Internet of Things, comprising:

[0011] A pre-sorting processing unit is used for receiving the put-in garbage and identifying the garbage type to obtain large garbage and bagged garbage, breaking the bag of the bagged garbage, air separating the broken bag garbage to obtain layering garbage and air separated garbage, filtering and layering the layering garbage according to the size to obtain layered garbage;

[0012] A multi-sensing sorting unit is used for layering the layered garbage to the garbage of the multi-sensing sorting unit, layering processing and identifying the garbage type, and classifying according to the garbage type to obtain compressed garbage and rigid garbage;

[0013] A compression unit is used for respectively receiving the compressed garbage and the air separated garbage and compressing to obtain compressed garbage;

[0014] A recycling unit is used for respectively receiving the compressed garbage, the rigid garbage and the large garbage.

[0015] Further, the pre-sorting processing unit comprises a bag breaking mechanism, the bag breaking mechanism comprises a bag breaking cylinder, the bag breaking cylinder has a centrifugal section and a bag breaking section; the centrifugal section is provided with a spiral channel for making the bagged garbage rotate and centrifuge; the bag breaking section is provided with a bag breaking shaft along the axis of the bag breaking cylinder, the bag breaking shaft and the bag breaking cylinder form a bag breaking cavity, and the bag breaking shaft is uniformly provided with spiral blades.

[0016] Further, the pre-sorting processing unit comprises a layering mechanism for layering the broken bag garbage, the layering mechanism comprises a barrel-shaped separation barrel, a plurality of groups of movable supporting rods arranged at different depths of the separation barrel from bottom to top for carrying and filtering garbage of different sizes, and the movable supporting rods from top to bottom are used for classifying and filtering the garbage to be layered from large to small, and after the classification and filtering, the movable supporting rods are sequentially lowered from bottom to top to complete the feeding of the layered garbage.

[0017] Further, the multi-sensing sorting unit comprises a sorting mechanism arranged below the layering mechanism, the sorting mechanism comprises a sorting chamber rotating in a horizontal plane, a hanging arm hung above the sorting chamber and provided with a plurality of garbage type identification sensors, a rotating lever arranged on the side wall of the sorting chamber, and a passage corresponding to the gate valve and the gate valve arranged at the bottom of the sorting chamber for guiding the corresponding identified type of garbage.

[0018] Further, the garbage type identification sensor comprises an infrared photoelectric sensor for distinguishing plastic and glass, and a metal sensor for distinguishing metal.

[0019] Further, the pre-sorting processing unit further comprises a first identification module for identifying bagged garbage and large garbage, and the first identification module is a visual sensor.

[0020] Further, the multi-sensing sorting unit further comprises a second identification module for identifying the garbage type, and the second identification module is a visual sensor, and the second identification module identifies again after the garbage type identification sensor identifies.

[0021] Further, the visual sensor is a visual sensor based on a YOLOv5 deep learning model, and a deformable convolution algorithm is introduced.

[0022] The second aspect is a garbage sorting platform implementation method based on the Internet of Things, comprising:

[0023] S1. Obtain the position information of the garbage sorting terminal;

[0024] S2. Obtain the collection data of the garbage sorting terminal, the collection data comprising the large garbage quantity data identified by the first identification module, the garbage type data and quantity data of each garbage type identified by the garbage type sensor through radio frequency identification, and the image data of the garbage type identified by the second identification module;

[0025] S3. Transmit the position information and the collection data to the Internet of Things application platform through the basic network, and perform data integration, distributed processing and storage on the data;

[0026] S4. Guide the garbage cleaning and transportation vehicles to process and recycle the garbage through analysis of the data.

[0027] Further, the basic network includes a mobile communication network, a wired broadband, a wireless network and a satellite communication network.

[0028] In summary, the present application has the following beneficial technical effects:

[0029] 1. The garbage sorting terminal based on the Internet of Things is proposed, a sorting scheme suitable for the whole chain of household garbage is proposed, and the garbage is identified and classified through the close cooperation of the pre-sorting processing unit, the multi-sensing sorting unit and the identification module. Whether it is for common garbage, or for garbage with damage, dirt, deformation and the like, the present application can effectively identify and accurately classify. The efficiency of garbage sorting is improved, manual intervention is reduced, the accuracy of classification is significantly improved, and subsequent garbage treatment is facilitated.

[0030] 2. The garbage sorting terminal based on the Internet of Things is proposed, and the bag breaking mechanism and the layering mechanism in the pre-sorting processing unit play an important role in realizing garbage dispersion and layering. They can break the bag and layer the garbage, so that the garbage is fully dispersed. After layering according to the size of the garbage, the garbage can be processed layer by layer, greatly improving the accuracy of garbage identification and the efficiency of garbage sorting, improving the efficiency and quality of the whole garbage treatment process. In addition, the synergistic effect of the deep learning algorithm of the intelligent garbage sorting terminal and the embedded sensor enables the terminal to more intelligently cope with various situations, reduces the misjudgment rate, and can more accurately classify the garbage.

[0031] 3. The garbage sorting platform based on the Internet of Things is proposed, which realizes efficient interconnection of intelligent sorting terminals and platforms by using Internet of Things technology. In terms of optimizing cleaning and transportation management, the present application obtains the position information of the garbage sorting terminal and comprehensive collection data, and efficiently transmits, processes and analyzes these data, which can provide scientific and reasonable guidance for garbage cleaning and transportation vehicles, so that they can more reasonably arrange the transportation route and recycling method. This not only improves the efficiency of garbage cleaning and transportation, reduces transportation costs, but also realizes the optimization and upgrading of garbage cleaning and transportation management, and provides strong support for the standardization and scientization of urban garbage treatment. The present application reduces the pollution of garbage to the environment, and at the same time, improves the recycling rate of resources, has good environmental protection benefits and sustainable development significance. BRIEF DESCRIPTION OF DRAWINGS

[0032] Figure 1 is a schematic diagram of the garbage sorting terminal of embodiment 1 of the present application;

[0033] Figure 2 is a structural schematic diagram of the layering mechanism of embodiment 1 of the present application;

[0034] Figure 3is a structural schematic diagram of a sorting mechanism of embodiment 1 of the present application;

[0035] Figure 4 is a flow chart of an implementation method of a garbage sorting platform based on the Internet of Things of embodiment 2 of the present application;

[0036] Figure 5 is a schematic diagram of a 4-layer Internet of Things system architecture of embodiment 2 of the present application; DETAILED DESCRIPTION

[0037] The present application will be further described in detail below with reference to the accompanying drawings.

[0038] Embodiment 1

[0039] With reference to Figure 1 , the garbage sorting terminal based on the Internet of Things of the present embodiment comprises:

[0040] A pre-sorting processing unit is configured to receive the put-in garbage, identify the garbage type to obtain large garbage and bagged garbage, break the bag of the bagged garbage, perform air separation on the broken-bag garbage to obtain layering-to-be garbage and air-separated garbage, filter and layer the layering-to-be garbage according to size to obtain layered garbage, and so on.

[0041] A multi-sensing sorting unit is configured to sequentially drop the layered garbage to the garbage of the multi-sensing sorting unit, sequentially process and identify the garbage type, and classify according to the garbage type to obtain compression-to-be garbage and rigid garbage.

[0042] A compression unit is configured to receive the compression-to-be garbage and the air-separated garbage respectively and compress them to obtain compressed garbage.

[0043] A recycling unit is configured to receive the compressed garbage, the rigid garbage and the large garbage respectively.

[0044] In the pre-sorting processing unit, paper, plastic film and other light-weight garbage can be screened out by mechanical air separation, and these garbage directly enters the compression unit for compression processing.

[0045] The pre-sorting processing unit comprises a bag breaking mechanism, the bag breaking mechanism comprises a bag breaking cylinder, the bag breaking cylinder has a centrifugal section and a bag breaking section; a spiral channel is arranged in the centrifugal section to make the bagged garbage perform rotational centrifugal motion; a bag breaking shaft is arranged along the axis of the bag breaking cylinder, a bag breaking cavity is formed between the bag breaking shaft and the bag breaking cylinder, and spiral blades are uniformly arranged on the bag breaking shaft.

[0046] Garbage is generally classified as waste paper, including newspapers, periodicals, books and packaging paper; plastic, including plastic bags, plastic packaging, cups and beverage bottles; glass, including glass bottles, mirrors and light bulbs; kitchen garbage, including leftovers, bones and fruit peels; and hazardous waste, including batteries.

[0047] Referring to Figure 2 , the pre-sorting processing unit comprises a layering mechanism for layering broken bag garbage, the layering mechanism comprises a barrel-shaped separation barrel, a plurality of groups of movable supporting rods arranged at different depth positions of the separation barrel from bottom to top for carrying and filtering garbage of different sizes, and the movable supporting rods from top to bottom are used for classifying and filtering the garbage to be layered from large to small, and after the classified filtering, the movable supporting rods are sequentially lowered from bottom to top to complete the feeding of the layered garbage.

[0048] Referring to Figure 3 , the multi-sensing sorting unit comprises a sorting mechanism arranged below the layering mechanism, the sorting mechanism comprises a sorting chamber rotating in a horizontal plane, a hanging arm hung above the sorting chamber and provided with a plurality of garbage type identification sensors, a rotating lever arranged on the side wall of the sorting chamber, and a passage corresponding to the gate valve and the gate valve arranged at the bottom of the sorting chamber for guiding the corresponding identified type of garbage.

[0049] The garbage type identification sensor comprises an infrared photoelectric sensor for distinguishing plastic and glass, and a metal sensor for distinguishing metal.

[0050] The pre-sorting processing unit further comprises a first identification module for identifying bagged garbage and large garbage, and the first identification module is a visual sensor.

[0051] The multi-sensing sorting unit further comprises a second identification module for identifying garbage types, and the second identification module is a visual sensor, and the second identification module identifies again after the garbage type identification sensor identifies.

[0052] In the sorting chamber, a plurality of garbage type identification sensors and visual sensors with a deep learning model are required to participate in garbage classification, wherein the optical sensor is used to sort out plastic and glass garbage by refractive index, the plastic enters the compression unit for treatment, the magnetic separation and eddy current electric separation are used to sort out metal garbage, the metal garbage enters the recycling box, the glass is a rigid garbage and cannot be compressed, and enters the recycling box, other garbage after removing the rigid garbage enters the compression unit for compression, and waste water is discharged into the municipal waste water pipeline.

[0053] The visual sensor is a visual sensor based on a YOLOv5 deep learning model, and a deformable convolution algorithm is introduced.

[0054] YOLOv5 network structure diagram, mainly composed of Backbone, Neck and Head. Backbone is the backbone network, used to extract image features. As the feature extraction continues to deepen, some local information of the image will disappear. By using the Neck network to fuse the feature maps of different network levels, more rich feature information of the image can be obtained, and these processed features are input into the Head layer to better classify and locate. Neck uses a multi-scale feature fusion structure to fuse deep and shallow feature maps. This dual design method can not only transmit strong semantic information, but also convey strong positioning information, effectively improving the algorithm performance. However, the feature extraction of different material garbage needs to rely on the cooperation of multiple sensors. In this project, the sensors to be used have been determined, including infrared photoelectric sensors for distinguishing plastic and glass, metal sensors for distinguishing metal and other materials, and electronic scales for measuring garbage quality. After the garbage is sorted by the sensors, the detection algorithm is used to further identify the garbage, so as to realize more accurate garbage sorting.

[0055] After the garbage enters the garbage sorting terminal, it first enters the pre-sorting processing unit and is initially identified by the image intelligent recognition algorithm. Large recyclable garbage is directly transported to the recycling unit, while household bagged garbage enters the bag breaking mechanism after breaking the bag. After processing, the garbage is layered and dispersed. Light garbage can be sorted by air selection and compressed into a recycling barrel.

[0056] The main purpose of the layering mechanism is to disperse the garbage as much as possible to facilitate the subsequent identification and picking process. However, due to the complexity of mixed garbage materials, shapes, dryness and wetness, the degree of dispersion of the garbage becomes a prerequisite for the success of garbage sorting. The garbage is preliminarily pre-sorted by its size. Although this sorting process cannot directly classify the garbage by material, it can layer and disperse the garbage by size. Large-sized garbage is dispersed in the upper layer of the separation barrel, while debris and wet garbage are filtered to the lowermost layer. Although this method cannot solve the classification problem, it can pre-separate the garbage, thereby ensuring that the subsequent garbage identification and picking process will not be affected by occlusion and layering, and providing conditions for subsequent operations.

[0057] The barrel wall of the separation barrel is made of stainless steel material. During the collection of garbage, the supporting rod is opened to complete the functions of supporting and filtering garbage. During the process of putting garbage into the sorting chamber, the bottom supporting rod is opened according to the instruction to put garbage. After the garbage is sorted in the sorting chamber, the upper supporting rod is opened according to the instruction to put garbage, and the putting function is completed in turn. This design ensures that the garbage is sorted in the sorting chamber in a small amount and dispersed state as much as possible.

[0058] The remaining garbage will pass through the layering mechanism into the sorting chamber of the multi-sensing sorting unit. In order to ensure the odor-free requirement, the sorting mechanism is designed as a closed cavity structure, that is, the garbage entering the sorting chamber will be isolated from the outside air to ensure that there is no odor transmission. The sorting chamber is equipped with various material sensors and intelligent image sensors to classify and sort garbage of different materials. The garbage passing through the sorting chamber will be processed and transported according to plastic, glass, metal and other garbage respectively.

[0059] The sorting chamber preferably integrates multiple mechanical drive units: first, the sorting chamber has a rotating function to better disperse the garbage pre-processed by the separation barrel for easy identification and picking; second, the side wall of the sorting chamber has a rotating lever to collect the identified garbage to the designated position; finally, the bottom of the sorting chamber has a movable gate valve to guide the sorted garbage into the corresponding collection channel. In addition, the upper part of the sorting chamber is hung with a variety of sensor arms. The arm is fixed to the bracket, not to the side wall of the sorting chamber. Although this design increases the difficulty of mechanical design and assembly, it can effectively avoid the problem of reduced sensor recognition accuracy caused by the movement of the sorting chamber. The arm integrates various material sensors and photoelectric sensors to realize the identification function.

[0060] The intelligent terminal for garbage identification mainly includes two stages: initial identification of the garbage delivery link and multi-sensing identification of the sorting unit. In the initial identification stage, intelligent image recognition technology is mainly used to determine whether the garbage is bagged or large-sized. This stage only needs to identify the garbage into two categories, so only the intelligent image recognition technology is used to reduce the cost of equipment and operation.

[0061] In the multi-sensor identification stage of the sorting unit, the garbage needs to be classified in detail, and each type of garbage is prone to damage and contamination. Therefore, the strategy of mainly using intelligent image recognition and supplementing with multi-sensor identification is adopted. The attention mechanism module is introduced to optimize the network in the intelligent image classification algorithm. Considering the fact that garbage is prone to deformation in recycling, the deformable convolution method is introduced to further improve the algorithm. On the other hand, the physical parameters of the garbage are obtained through various material sensors to verify the accuracy of image recognition as auxiliary data. For example, when classifying bottle-type garbage, the camera can easily identify the object as a bottle and classify the garbage as recyclable, but it is difficult to quickly and accurately identify the material, such as whether it is metal, making it difficult to achieve accurate garbage classification. However, the integration of multi-sensor in the classification process can ensure accurate classification of garbage with high precision and speed. However, relying solely on sensor identification of garbage types can correctly identify the material of the garbage, but cannot avoid some special cases, such as two types of plastic bottles, empty bottles and bottles filled with water, the former being recyclable and the latter being non-recyclable. The sensor used in the intelligent terminal device in the embodiment can distinguish between the two based on image information.

[0062] The intelligent garbage sorting terminal of the embodiment is mainly built by intelligent visual sensors. The built-in deep learning algorithm is trained through a large amount of images, including water bottles, glass, food packaging boxes and various recyclable waste. The image database contains various states of these recyclable materials, such as intact, indentation or crushed. This makes the intelligent garbage sorting terminal more accurate in classification, more efficient in learning new materials, and better adapted to packaging design and lighting changes. The embedded sensors use computer vision to scan rapidly moving objects and distinguish them based on color, texture, shape, size and material. All this data is input into the monitoring camera of the intelligent terminal, which can monitor the activity and performance of the waste and send real-time notifications, allowing users to stay up-to-date on the latest information, including potential device issues and hazards. The sorting terminal of the embodiment has a detection time of not more than 5s; the system effective image acquisition frame rate is not less than 1 frame / s; the system effective recognition range is not less than 1m2 effective area; the system sorting accuracy is 70% —80%; the object false detection rate is not higher than 10%; and the garbage sorting based on image recognition is 2 times that of manual sorting. 2

[0063] Embodiment 2

[0064] With reference to Figure 4 , the embodiment provides an implementation method of a garbage sorting platform based on the Internet of Things, comprising:

[0065] S1. Obtain the location information of the garbage sorting terminal; ​

[0066] S2. Obtain collection data of the garbage sorting terminal, the collection data including large garbage quantity data identified by the first identification module, garbage category data and quantity data of each garbage category identified by the garbage category sensor through radio frequency identification, and image data of the garbage category identified by the second identification module;

[0067] S3. Transmit the position information and the collection data to an Internet of Things application platform through a basic network, and perform data integration, distributed processing and storage on the data;

[0068] S4. Guide garbage cleaning and transportation vehicles to process and recycle garbage through analysis of the data.

[0069] The basic network includes a mobile communication network, a wired broadband, a wireless network and a satellite communication network.

[0070] Referring to Figure 5 The main research contents of the Internet of Things for the platform construction and data detection of the intelligent garbage sorting process are as follows: the Internet of Things platform is a garbage sorting intelligent platform based on a four-layer Internet of Things system architecture. The architecture is from bottom to top, respectively, a sensing and identification layer, a network construction layer, a management service layer and a comprehensive application layer, that is, the Internet of Things senses (thorough perception, data collection and data acquisition), connects (interconnection, information interaction and information sharing), knows (in-depth intelligence, data analysis and comprehensive summary) and controls (guides practice). The sensing and identification layer, as the basis of the upper structure, is located at the bottom end of the Internet of Things system architecture.

[0071] The sensing and identification layer is the “tentacle” of the Internet of Things to perceive the physical world, and is the link connecting the information world, including a large number of information generation devices. The network construction layer is the peripheral nervous system of the Internet of Things, which is used for the transmission of sensing information. Its main function is to connect the lower layer sensing and identification layer devices to the Internet, and to transmit the lower layer data to the upper layer service. The key is to optimize and transform the application characteristics of the Internet of Things to form a network of system perception. The core content of the network construction layer mainly involves 3G, 4G, WIFI, WIMX, Bluetooth, Zigbee, NFC and other communication technologies. The management service layer is the central nervous system of the Internet of Things, which is used for information analysis and processing. It is mainly responsible for organizing large-scale data safely and efficiently to realize data storage, query, analysis and processing, and provides a powerful support platform for industry applications. The key is to solve the problems of how to store data, how to retrieve data and how to use data. The comprehensive application layer is equivalent to the human brain, which provides users with rich intelligent services based on sensing data. The key is to accurately match data and real transactions, closely link data content with specific content of various transactions, and realize the combination of data and business applications.

[0072] The four-layer Internet of Things system is closely related to the garbage sorting platform based on the Internet of Things.

[0073] Information perception and data collection: The information perception layer contains a large number of information generation devices that can perceive various information during the garbage sorting process, such as the type, quantity, and location of garbage, providing a data foundation for the garbage sorting platform.

[0074] Information transmission and interconnection: The network construction layer connects the information perception layer devices to the Internet, enabling information transmission. In garbage sorting, the data of various sorting devices and sensors can be transmitted to the management service layer in a timely manner, ensuring the real-time and accuracy of information.

[0075] Data analysis and processing: The management service layer is responsible for the secure and efficient organization, storage, query, analysis, and processing of large-scale data. In garbage sorting, the collected garbage information can be analyzed to optimize the sorting process and improve efficiency, while providing decision support for garbage transportation, processing, and recycling.

[0076] Application implementation and intelligent services: The comprehensive application layer applies Internet of Things technology to various aspects of garbage sorting, providing users with a wide range of intelligent services. Through data analysis, the rational arrangement and optimization of garbage cleaning and transportation vehicles can be achieved, reducing transportation costs and energy consumption; effective transportation, processing, and recycling of sorted garbage can be achieved.

[0077] In this embodiment, the Internet of Things communication technology mainly uses wireless network technology, such as WIFI technology, wireless metropolitan area networks, and wireless wide area networks, such as GSM mobile communication systems and satellite communication systems, as the preferred method of information transmission. Finally, the Internet of Things is a data-centric network, so data storage and management become the core of Internet of Things application design. The computing and service technology of the Internet of Things is mainly responsible for the calculation and processing of perceived information, and is the supporting force for realizing the service and application value system of the Internet of Things. The sorting platform will use database technology, data fusion technology, data mining technology, and computing technology to realize the operation, management, and control of the platform.

[0078] The above are preferred embodiments of the present application, and are not intended to limit the scope of protection of the present application. Therefore, any equivalent changes made in terms of structure, shape, and principle based on the present application should be covered within the scope of protection of the present application.

Claims

1. An Internet of Things-based garbage sorting terminal, comprising: a pre-sorting processing unit for receiving deposited garbage and identifying garbage types to obtain bulk garbage and bagged garbage, breaking the bag of the bagged garbage, air sorting the broken bag garbage to obtain layering garbage and air sorted garbage, filtering and layering the layering garbage according to size to obtain layered garbage; a multi-sensing sorting unit for sequentially depositing the layered garbage to the garbage of the multi-sensing sorting unit, layer by layer, processing and identifying the garbage type, and classifying according to the garbage type to obtain compressed garbage and rigid garbage; a compression unit for receiving and compressing the compressed garbage and the air sorted garbage respectively to obtain compressed garbage; a recycling unit for receiving the compressed garbage, the rigid garbage and the bulk garbage respectively; the pre-sorting processing unit comprises a layering mechanism for layering the broken bag garbage, the layering mechanism comprises a barrel-shaped separation barrel, a plurality of groups of movable support rods for carrying and filtering garbage of different sizes are arranged at different depth positions of the separation barrel from bottom to top, the movable support rods from top to bottom are used for grading and filtering the layering garbage from large to small, and the movable support rods sequentially deposit from bottom to top after grading and filtering to complete the depositing of the layering garbage; the multi-sensing sorting unit comprises a sorting mechanism arranged below the layering mechanism, the sorting mechanism comprises a sorting chamber rotating in a horizontal plane, a boom hung above the sorting chamber and provided with a plurality of garbage type identification sensors, a rotating lever arranged on the side wall of the sorting chamber, and a passage corresponding to the gate valve for introducing the corresponding identified type of garbage; the pre-sorting processing unit further comprises a first identification module for identifying bagged garbage and bulk garbage, the first identification module is a visual sensor, the multi-sensing sorting unit further comprises a second identification module for identifying garbage type, the second identification module is a visual sensor, and the second identification module identifies again after the garbage type identification sensor identifies.

2. The Internet of Things-based garbage sorting terminal according to claim 1, wherein the pre-sorting processing unit comprises a bag breaking mechanism, the bag breaking mechanism comprises a bag breaking cylinder, the bag breaking cylinder has a centrifugal section and a bag breaking section; the centrifugal section is provided with a spiral channel for making the bagged garbage rotate and centrifuge; the bag breaking section is provided with a bag breaking shaft along the axis of the bag breaking cylinder, the bag breaking shaft and the bag breaking cylinder form a bag breaking cavity, and the bag breaking shaft is uniformly provided with spiral blades.

3. The Internet of Things-based garbage sorting terminal according to claim 1, wherein the garbage type identification sensor comprises an infrared photoelectric sensor for distinguishing plastic and glass, and a metal sensor for distinguishing metal.

4. The Internet of Things-based garbage sorting terminal according to claim 1, wherein the visual sensor is a visual sensor based on a YOLOv5 deep learning model and introduces a deformable convolution algorithm.

5. An Internet of Things-based garbage sorting platform implementation method based on the Internet of Things-based garbage sorting terminal according to any one of claims 1-4, comprising: S1. Obtain the location information of the garbage sorting terminal; S2. Obtain collection data of the garbage sorting terminal, wherein the collection data comprises large garbage quantity data identified by a first identification module, garbage type data and quantity data of each garbage type identified by a garbage type sensor through radio frequency identification, and image data of the garbage type identified by a second identification module; S3. Transmit the position information and the collection data to an Internet of Things application platform through a basic network, and perform data integration, distributed processing and storage on the data; S4. Guide garbage cleaning and transportation vehicles to process and recycle garbage through analysis of the data.

6. The implementation method of the garbage sorting platform based on the Internet of Things according to claim 5, wherein the basic network comprises a mobile communication network, a wired broadband network, a wireless network and a satellite communication network.

Images