CN113182194A - Intelligent automatic garbage classification system and classification method based on AI (Artificial Intelligence) recognition technology - Google Patents

Abstract

The invention provides an AI recognition technology-based intelligent automatic garbage classification system and method. The system comprises a garbage coarse collecting device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine. The garbage coarse grouping device obtains garbage with a preset grouping size and transmits the garbage to the horizontal throwing belt; when the pressure sampling parameters meet preset conditions, intercepting the garbage with the preset packet size through a garbage intercepting device, and moving a screening frame to enable the intercepted garbage with the preset packet size to enter the screening frame; the working parameters of the garbage coarse feeding device, the garbage coarse grouping device and the garbage screening device are determined by the artificial intelligence engine based on the weight and the type of the garbage left in the screening frame. The invention also discloses an AI identification technology-based intelligent automatic garbage classification method realized based on the system and a nonvolatile storage medium for realizing the method.

Description

Intelligent automatic garbage classification system and classification method based on AI (Artificial Intelligence) recognition technology

Technical Field

The invention belongs to the technical field of artificial intelligence and garbage classification, and particularly relates to an intelligent automatic garbage classification system and method based on an AI (artificial intelligence) recognition technology, and a nonvolatile storage medium for realizing the method.

Background

With the rapid development of social economy and the great improvement of the consumption level of resident materials, the production quantity of domestic garbage of residents in China is rapidly increased, the hidden environmental danger is increasingly prominent, and the new type of urban development is a restriction factor. At present, about 90% of cities in China adopt landfill to treat domestic garbage of residents, and because the components of the domestic garbage of the residents are increasingly complicated, high-concentration percolation generated by a landfill method is difficult to effectively treat, so that the ecological environment of regions near a plurality of landfill sites is seriously damaged.

In order to effectively alleviate the problem, the garbage classification is a key treatment link, and can obviously reduce the garbage treatment cost, improve the garbage recycling value and make the best use of the garbage. However, the variety of the household garbage is numerous and various, and many people do not know the accurate category of the garbage, so that a certain random throwing behavior is caused, and the cost of garbage disposal is increased.

The utilization can implement extrusion and dehydration to rubbish based on the categorised recovery system of artificial intelligence technique, can reduce the volume of rubbish like this, is favorable to reducing the categorised recovery working strength of rubbish, improves work efficiency, can avoid rubbish secondary pollution's problem to appear when the operation rubbish in addition, through the cyclic utilization of classification treatment city domestic waste, can also realize rubbish.

Chinese patent CN111137602B proposes an artificial intelligence garbage classification system, which includes an artificial intelligence classification recognition engine. The artificial intelligence classification and identification engine is used for identifying and classifying the garbage conveyed by the plurality of conveyor belts, and each conveyor belt is provided with a weight sensor and a humidity sensor; the weight sensor is used for detecting the weight of the garbage on the corresponding conveyor belt, and the humidity sensor is used for detecting the humidity of the garbage on the corresponding conveyor belt; if the weight values of all the garbage on the plurality of conveyor belts accord with the Gaussian distribution, keeping the rotating speed and the horizontal direction moving speed of the measuring hopper in the process of reciprocating along the first horizontal direction unchanged, and otherwise, adjusting the rotating speed and/or the horizontal direction moving speed of the measuring hopper in the process of reciprocating along the first horizontal direction, so that the weight values of all the garbage on the plurality of conveyor belts accord with the Gaussian distribution.

The chinese patent application with application number CN202010512980.9 proposes a recyclable garbage image classification method based on artificial intelligence, which comprises: step S1: deploying a hardware environment, and establishing a connection among the camera, the development board main body and the external display equipment; step S2: shooting pictures after detecting garbage by a camera; step S3: preprocessing the pictures, comparing the definition of the shot pictures according to a preset definition judgment standard, and screening out a picture with the highest definition; step S4: carrying out background segmentation on the screened picture to segment out objects contained in the picture; step S5: putting the segmented object pictures into a convolutional neural network for picture classification, and outputting corresponding probability distribution; step S6: and judging the probability distribution and making corresponding measures. The invention improves the efficiency of the user for classifying the garbage, thereby improving the probability of correctly classifying the garbage.

However, most of the garbage classification systems in the prior art emphasize that the garbage is accurately classified from the source of the collected garbage, including image recognition, combined sensing recognition and the like, so that the garbage classification cost is greatly increased and the garbage classification systems cannot be popularized because the garbage classification boxes need to be configured with high-precision recognition; and other garbage classification treatment and collection are more and completely dependent on the accuracy of garbage classification collected by the garbage can.

However, in practice, it is found that the garbage classification recognition work itself cannot be accurately developed due to different consciousness of garbage classification of different people and different garbage classification standards of different regions, and if accuracy is excessively pursued, the garbage classification cannot be practically popularized. How to obtain classifiable garbage to a greater extent in the later processing under the condition that the classification of garbage generated by a source end is inaccurate or even the garbage is mixed becomes a technical problem to be solved firstly.

Disclosure of Invention

In order to solve the technical problems, the invention provides an intelligent automatic garbage classification system and method based on an AI (Artificial intelligence) identification technology. The system comprises a garbage coarse collecting device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine. The garbage coarse grouping device obtains garbage with a preset grouping size and transmits the garbage to the horizontal throwing belt; when the pressure sampling parameters meet preset conditions, intercepting the garbage with the preset packet size through a garbage intercepting device, and moving a screening frame to enable the intercepted garbage with the preset packet size to enter the screening frame; the working parameters of the garbage coarse feeding device, the garbage coarse grouping device and the garbage screening device are determined by the artificial intelligence engine based on the weight and the type of the garbage left in the screening frame. The invention also discloses an AI identification technology-based intelligent automatic garbage classification method realized based on the system and a nonvolatile storage medium for realizing the method.

Specifically, the technical scheme of the invention comprises the following three aspects:

in a first aspect of the invention, an intelligent automatic garbage classification system based on an AI identification technology is provided, and the classification system comprises a garbage rough collection device, a garbage rough delivery device, a garbage rough grouping device, a garbage screening device and an artificial intelligence engine.

The garbage rough collecting device is used for collecting garbage generated by a plurality of garbage collecting points, and each garbage collecting point generates at least two types of garbage;

the garbage rough throwing device comprises a horizontal throwing belt with a first length, and the first end of the horizontal throwing belt is connected to the garbage rough collecting device through an inclined transmission belt with a second length; the garbage coarse grouping device acquires garbage with a preset grouping size from the garbage coarse collecting device based on a coarse grouping parameter, and transmits the garbage with the preset grouping size to the horizontal throwing belt through the inclined transmission belt;

the garbage screening device comprises at least one garbage intercepting device and a plurality of screening frames with adjustable screening sizes;

when the garbage with the preset grouping size moves on the horizontal throwing belt, a plurality of pressure sampling parameters are obtained through the pressure sampling sensor;

intercepting the garbage with the preset packet size through the garbage intercepting device when the pressure sampling parameter meets a preset condition, and moving the screening box to enable the intercepted garbage with the preset packet size to enter the screening box;

and the coarse grouping parameter, the screening size, and the predetermined condition are determined by the artificial intelligence engine based on the weight and kind of the garbage remaining in the screening box.

Preferably, the system further comprises a blower disposed between the horizontal casting belt and the screening frame;

as a further preference, the system further comprises a plurality of refuse transfer silos; the waste transfer bin receives waste delivered through the second end of the horizontal launch belt.

In the second aspect of the invention, the invention also provides an intelligent automatic garbage classification method based on AI identification technology, which is realized based on an intelligent automatic garbage classification system comprising a garbage coarse collection device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine,

based on the hardware device, the method comprises the following steps:

s800: initializing coarse grouping parameters, screening sizes and pressure sampling threshold values;

s810: determining a grouping size value based on the coarse grouping parameter, and acquiring a grouping garbage pile from the garbage coarse aggregation device based on the grouping size value;

s820: the grouped garbage is piled on the end surface of the first end of the horizontal throwing belt of the garbage rough throwing device;

s830: acquiring pressure sampling parameters generated by the grouped garbage piles passing through the pressure sampling sensors on the horizontal throwing belt through a plurality of pressure sampling sensors arranged on the horizontal throwing belt;

s840: judging whether the pressure sampling parameter is larger than the pressure sampling threshold value;

if yes, go to the next step, otherwise, go to step S870;

s850: starting a garbage intercepting device of the garbage screening device, intercepting the grouped garbage piles passing through the pressure sampling sensor, and moving a screening frame of the garbage screening device to enable the intercepted grouped garbage piles to enter the screening frame;

s860: acquiring the weight and the type of the garbage entering the screening frame, adjusting the coarse grouping parameter, the screening size and the pressure sampling threshold value based on the weight and the type of the garbage, and returning to the step S810;

s870: receiving the grouped garbage piles from the second end of the horizontal throwing belt through the garbage transfer bucket, and returning to the step S810.

The above method of the present invention can be automatically executed by program instructions through a terminal device, especially an image processing terminal device, including a mobile terminal, a desktop terminal, a server cluster, and the like, which includes a processor and a memory, and therefore, in a third aspect of the present invention, there is also provided a non-volatile computer-readable storage medium having computer program instructions stored thereon; the program instructions are executed by an image terminal processing device comprising a processor and a memory for implementing all or part of the steps of the method of the second aspect.

Compared with the prior art, the garbage classification method can at least perform accurate identification and classification on dry and wet garbage and mixed garbage with obvious weight difference; meanwhile, the invention can treat the garbage which is not classified uniformly and accurately at the garbage putting source, the technical scheme of the invention can be regarded as a primary classification stage in the garbage classification process, and the classification output result of the garbage classification stage can not be used as the final classification result, but can provide a good basis for the subsequent further accurate classification; furthermore, the present invention is also directed to not only performing accurate classification, but also accumulating learning data of garbage separation through primary classification, thereby better training the artificial intelligence engine.

Further advantages of the invention will be apparent in the detailed description section in conjunction with the drawings attached hereto.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

FIG. 1 is a block diagram of an AI-recognition based intelligent automatic garbage classification system according to an embodiment of the present invention

FIG. 2 is a schematic view of the installation position of a part of the system shown in FIG. 1

FIG. 3 is a schematic diagram of the operational data control of the artificial intelligence engine of the system of FIG. 1

FIG. 4 is a main flowchart of an AI-recognition-technology-based intelligent automatic garbage classification method according to an embodiment of the present invention

FIG. 5 is a schematic diagram of a further preferred embodiment of the method of FIG. 4

FIG. 6 is a schematic structural diagram of a terminal device for implementing the method shown in FIG. 5

Detailed Description

The invention is further described with reference to the following drawings and detailed description.

Fig. 1 is a block diagram of an intelligent automatic garbage classification system based on AI recognition technology according to an embodiment of the present invention.

In the structure of fig. 1, the intelligent automatic garbage classification system based on the AI identification technology includes a garbage coarse collecting device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device, and an artificial intelligence engine.

The garbage rough collecting device is used for collecting garbage generated by a plurality of garbage collecting points, works independently and is not controlled or influenced by the artificial intelligence engine;

the garbage rough collecting device can be a garbage collecting vehicle, the garbage collecting vehicle comprises at least two garbage collecting buckets, and the garbage collecting buckets collect the first type of garbage and the second type of garbage from a plurality of garbage collecting points.

According to the actual situation of the current garbage classification, each garbage collection point is at least provided with two garbage collection boxes of different categories. For convenience of practice, classification criteria for categories used are different from place to place, but at least two categories are used.

As an exemplary two-class classification, each waste collection point is provided with a dry waste collection bin and a wet waste collection bin;

as another illustrative two-class classification, each garbage collection point is provided with a recyclable garbage collection bin and a non-recyclable garbage collection bin.

Of course, there are many classification criteria in practice, but all can be considered as further refinements based on the existing two classifications.

The other modular devices of the system of fig. 1 are in communication with the artificial intelligence engine through which the operational status of the other modular devices is adjusted.

The other modules comprise a garbage coarse feeding device, a garbage coarse grouping device and a garbage screening device, and the working state comprises feeding working parameters of the garbage coarse feeding device, coarse grouping working parameters of the garbage coarse grouping device and screening parameters of the garbage screening device;

next, on the basis of fig. 1, a layout of the devices of the system shown in fig. 1 and the above-mentioned operating parameters will be further described with reference to fig. 2.

Referring to fig. 2, the garbage rough collecting device includes a first garbage collecting bin and a second garbage collecting bin; and collecting the first type of garbage into the first type of garbage collection bucket, and collecting the second type of garbage into the second type of garbage collection bucket.

The garbage rough throwing device comprises a horizontal throwing belt with a first length and an inclined conveying belt with a second length; the first end of the horizontal throwing belt is connected to the garbage rough collecting device through an inclined transfer belt;

referring to fig. 2, the upper end of the inclined conveying belt is connected with a garbage coarse grouping device, the lower end of the inclined conveying belt is connected with a horizontal throwing belt, and the horizontal throwing belt is provided with a plurality of pressure sampling sensors.

The working parameters of the garbage rough throwing device comprise the moving speed of the horizontal throwing belt and the pressure sampling threshold value of the pressure sampling sensor;

the action parameters of the garbage coarse grouping device comprise coarse grouping parameters, and the garbage heap grouping size can be determined based on the coarse grouping parameters;

therefore, based on the coarse grouping parameter, the garbage coarse grouping device can acquire garbage with a preset grouping size from the garbage coarse collecting device and transfer the garbage with the preset grouping size to the horizontal throwing belt through the inclined transfer belt.

The garbage screening device comprises at least one garbage intercepting device and a plurality of screening frames with adjustable screening sizes, and therefore working parameters of the garbage screening device comprise screening sizes of the screening frames.

The coarse grouping parameter, the screening size and the predetermined condition are determined by the artificial intelligence engine based on the weight and kind of the garbage remaining in the screening box, i.e., the working state of the other module devices is adjusted by the artificial intelligence engine.

In a specific implementation manner, the garbage intercepting device comprises at least one mechanical arm, the height of the mechanical arm is the same as that of the upper surface of the conveying belt, and the mechanical arm is used for executing the grouping garbage intercepting and removing.

Further, as a further preferred, in the system of fig. 2, the system further includes a blower disposed between the horizontal casting belt and the screening frame.

Further, the system still including set up in the collection transfer plate of screening frame below, it includes a plurality of gravity sensor to collect the transfer plate, through gravity sensor obtains being located collect the rubbish weight on the transfer plate.

Further, the system also comprises a plurality of garbage transfer buckets; the waste transfer bin receives waste delivered through the second end of the horizontal launch belt.

It should be noted that fig. 2 only shows a part of the apparatus or a part of the structure, and does not represent all the components of the whole solution. For example, the screening frame should further have corresponding supporting devices and rotating or moving devices, the inclined conveying belt, the horizontal throwing belt should further have corresponding connecting conveyor belts, the trash transferring barrel should further have corresponding transferring and moving devices, and the like, which are not described in the present invention.

Next, the operation principle and the control flow of the system or the apparatus shown in fig. 1-2 will be described with reference to fig. 3.

In fig. 2, the garbage rough sorting device obtains garbage of a predetermined grouping size from the garbage rough collecting device based on a rough grouping parameter, and transfers the garbage of the predetermined grouping size to the horizontal throwing belt through the inclined transfer belt;

the garbage screening device comprises at least one garbage intercepting device and a plurality of screening frames with adjustable screening sizes;

when the garbage with the preset grouping size moves on the horizontal throwing belt, a plurality of pressure sampling parameters are obtained through the pressure sampling sensor;

intercepting the garbage with the preset packet size through the garbage intercepting device when the pressure sampling parameter meets a preset condition, and moving the screening box to enable the intercepted garbage with the preset packet size to enter the screening box.

As a further preference, the garbage rough collecting device comprises a first garbage collecting barrel and a second garbage collecting barrel;

collecting the first type of garbage into the first type of garbage collection bucket, and collecting the second type of garbage into the second type of garbage collection bucket;

and the garbage coarse grouping device alternately obtains garbage with a preset grouping size from the first garbage collection bucket or the second garbage collection bucket.

The garbage intercepting device comprises at least one mechanical arm, and the height of the mechanical arm is the same as that of the upper surface of the conveying belt;

when the pressure sampling parameters obtained by the pressure sampling sensor meet preset conditions, the garbage intercepting device starts the mechanical arm to intercept garbage of the preset grouping size, and simultaneously moves the screening frame to enable the garbage of the preset grouping size to enter the screening frame.

The predetermined condition includes determining whether the pressure sampling parameter is greater than a pressure sampling threshold.

In one aspect, the blower is activated at a first power when the pressure sampling parameter satisfies a predetermined condition; otherwise, the blower is started at a second power.

The artificial intelligence engine acquires the weight and the category of the garbage entering the screening box, and adjusts any one or more of the grouping size, the screening size and the predetermined condition based on the weight and the category of the garbage;

the artificial intelligence engine adjusts the power of the blower based on the acquired weight of the refuse on the collection transfer plate.

The artificial intelligence engine can carry out self-learning adjustment based on the feedback signal of the final garbage classification result, so that adjustment parameters are obtained, and the working parameters are adjusted. There are many realizable ways for the artificial intelligence self-learning model in the field, such as giving inputs and outputs, performing iterative loops and parameter tuning after determining the parameters of the initial artificial intelligence model, etc.;

as an illustrative example, if the weight of the refuse entering the screening frame is below a set weight threshold, the pressure sampling threshold should be increased so that most of the refuse is transferred to the refuse transfer bin through the second end of the horizontal chute.

Of course, the more specific parameter setting may be that after the coarse packet parameters, the screening size and the pressure sampling threshold are initialized, the artificial intelligence engine self-learns and trains and then dynamically adjusts and determines, which is not specifically limited by the present invention, and those skilled in the art may implement the self-learning and training model by using various known technologies such as machine learning, neural networks (CNN, RNN), deep learning, and deep belief networks (DRN).

On the basis of fig. 1-3, referring to fig. 4-5, the main steps of the intelligent automatic garbage classification method based on the AI identification technology are given.

In one aspect, the methods described in fig. 4-5 may be implemented based on the embodiments described in fig. 1-3; on the other hand, the method can be realized based on any intelligent garbage automatic classification system comprising a garbage coarse collecting device, a garbage coarse putting device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine.

In a specific implementation, referring to fig. 4, the method includes the following steps:

s800: initializing coarse grouping parameters, screening sizes and pressure sampling threshold values;

s810: determining a grouping size value based on the coarse grouping parameter, and acquiring a grouping garbage pile from the garbage coarse aggregation device based on the grouping size value;

s820: the grouped garbage is piled on the end surface of the first end of the horizontal throwing belt of the garbage rough throwing device;

s830: acquiring pressure sampling parameters generated by the grouped garbage piles passing through the pressure sampling sensors on the horizontal throwing belt through a plurality of pressure sampling sensors arranged on the horizontal throwing belt;

s840: judging whether the pressure sampling parameter is larger than the pressure sampling threshold value;

if yes, go to the next step, otherwise, go to step S870;

s850: starting a garbage intercepting device of the garbage screening device, intercepting the grouped garbage piles passing through the pressure sampling sensor, and moving a screening frame of the garbage screening device to enable the intercepted grouped garbage piles to enter the screening frame;

s860: acquiring the weight and the type of the garbage entering the screening frame, adjusting the coarse grouping parameter, the screening size and the pressure sampling threshold value based on the weight and the type of the garbage, and returning to the step S810;

s870: receiving the grouped garbage piles from the second end of the horizontal throwing belt through the garbage transfer bucket, and returning to the step S810.

In each of the above embodiments, when the garbage transferring bucket receives the grouped garbage pile from the second end of the horizontal throwing belt, the height of the inlet end surface of the garbage transferring bucket and the height of the second end surface of the horizontal throwing belt have a difference larger than a preset height; the preset height difference is not less than the height difference between the screening frame and the horizontal throwing belt.

It should be noted that the lower collection and transfer plate extends all the way to the underside of the waste transport bin, so that the grouped waste piles can be screened by the blower during the process of falling into the waste transport bin, and then fall into the lower collection and transfer plate.

It is clear that the presence of the above-mentioned height difference enables the blower to function better in all conditions.

Preferably, referring to fig. 5, at least one blower is disposed between the screening frame and the horizontal casting belt;

when the pressure sampling parameter is not larger than the pressure sampling threshold value in the step S840, turning on the blower at a first power;

when the pressure sampling parameter is judged to be larger than the pressure sampling threshold in the step S840, turning on the blower at a second power;

the first power is lower than the second power.

Obviously, the power adjusting method can intelligently adapt to the weight condition of the grouped garbage piles on the horizontal conveyor belt, so that the grouped garbage piles can be screened by the air blower whether falling into the screening frame or the garbage transfer barrel, and then fall into the collecting and transferring plate below.

The above-described methods of fig. 4-5 may be performed automatically by program instructions through a terminal device comprising a processor and a memory, in particular an image processing terminal device, including a mobile terminal, a desktop terminal, a server cluster, and the like.

As an example, referring to fig. 6, the terminal device comprises a memory and a processor, the memory and the processor are connected through a bus, the memory is used for storing program codes, and the processor executes the program codes, so as to realize part or all of the steps of the foregoing steps S800-S870.

The technical scheme of the invention can be regarded as a primary classification stage in the garbage classification process, and although the classification output result of the primary classification stage may not be used as the final classification result, the primary classification stage can provide a good basis for subsequent further accurate classification;

although accurate classification is not required, the learning data of garbage separation is accumulated through primary classification, so that the artificial intelligence engine is trained better, and the garbage classification method can at least perform accurate identification and classification on dry and wet garbage and mixed garbage with obvious weight difference; meanwhile, the invention can treat the garbage which is not classified accurately and has uneven classification at the garbage putting source, and has better applicability and popularization.

Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.

Claims (10)

1. An AI identification technology-based intelligent automatic garbage classification system comprises a garbage coarse collection device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine,

the method is characterized in that:

the garbage rough collecting device is used for collecting garbage generated by a plurality of garbage collecting points, each garbage point generates at least two types of garbage, and the garbage rough collecting device collects the garbage generated by the garbage collecting points into a first type of garbage and a second type of garbage;

the garbage rough throwing device comprises a horizontal throwing belt with a first length, and the first end of the horizontal throwing belt is connected to the garbage rough collecting device through an inclined transmission belt with a second length;

the horizontal casting belt is provided with a plurality of pressure sampling sensors;

the garbage coarse grouping device acquires garbage with a preset grouping size from the garbage coarse collecting device based on a coarse grouping parameter, and transmits the garbage with the preset grouping size to the horizontal throwing belt through the inclined transmission belt;

the garbage screening device comprises at least one garbage intercepting device and a plurality of screening frames with adjustable screening sizes;

when the garbage with the preset grouping size moves on the horizontal throwing belt, a plurality of pressure sampling parameters are obtained through the pressure sampling sensor;

intercepting the garbage with the preset packet size through the garbage intercepting device when the pressure sampling parameter meets a preset condition, and moving the screening box to enable the intercepted garbage with the preset packet size to enter the screening box;

and the coarse grouping parameter, the screening size, and the predetermined condition are determined by the artificial intelligence engine based on the weight and kind of the garbage remaining in the screening box.

2. The intelligent automatic garbage classification system based on AI identification technology as claimed in claim 1 wherein:

the garbage rough collecting device comprises a first garbage collecting barrel and a second garbage collecting barrel;

collecting the first type of garbage into the first type of garbage collection bucket, and collecting the second type of garbage into the second type of garbage collection bucket;

and the garbage coarse grouping device alternately obtains garbage with a preset grouping size from the first garbage collection bucket or the second garbage collection bucket.

3. The intelligent automatic garbage classification system based on AI identification technology as claimed in claim 1 wherein:

a plurality of pressure sampling sensors are arranged at the bottom of the conveying belt of the horizontal casting belt;

the garbage intercepting device comprises at least one mechanical arm, and the height of the mechanical arm is the same as that of the upper surface of the conveying belt;

when the pressure sampling parameters obtained by the pressure sampling sensor meet preset conditions, the garbage intercepting device starts the mechanical arm to intercept garbage of the preset grouping size, and simultaneously moves the screening frame to enable the garbage of the preset grouping size to enter the screening frame.

4. The intelligent automatic garbage classification system based on AI identification technology as claimed in claim 1 wherein:

the system also comprises an air blower arranged between the horizontal casting belt and the screening frame;

and when the pressure sampling parameter meets a preset condition, starting the air blower.

5. The AI-recognition technology-based intelligent automatic garbage classification system of claim 4, wherein:

the system is characterized by further comprising a collecting and transferring plate arranged below the screening frame, wherein the collecting and transferring plate comprises a plurality of gravity sensors, and the gravity sensors acquire the weight of the garbage on the collecting and transferring plate.

6. The AI-recognition technology-based intelligent automatic garbage classification system of claim 4, wherein:

the artificial intelligence engine acquires the weight and the category of the garbage entering the screening box, and adjusts any one or more of the grouping size, the screening size and the predetermined condition based on the weight and the category of the garbage;

and the artificial intelligence engine adjusts the power of the blower based on the acquired weight of the garbage on the collection transfer plate.

7. The intelligent automatic garbage classification system based on AI identification technology as claimed in any one of claims 1-6 wherein:

the system further comprises a plurality of refuse transfer silos;

the waste transfer bin receives waste delivered through the second end of the horizontal launch belt.

8. An AI identification technology-based intelligent automatic garbage classification method is realized based on an intelligent automatic garbage classification system comprising a garbage coarse collection device, a garbage coarse feeding device, a garbage coarse grouping device, a garbage screening device and an artificial intelligence engine, and is characterized by comprising the following steps of:

s800: initializing coarse grouping parameters, screening sizes and pressure sampling threshold values;

s810: determining a grouping size value based on the coarse grouping parameter, and acquiring a grouping garbage pile from the garbage coarse aggregation device based on the grouping size value;

s820: the grouped garbage is piled on the end surface of the first end of the horizontal throwing belt of the garbage rough throwing device;

s830: acquiring pressure sampling parameters generated by the grouped garbage piles passing through the pressure sampling sensors on the horizontal throwing belt through a plurality of pressure sampling sensors arranged on the horizontal throwing belt;

s840: judging whether the pressure sampling parameter is larger than the pressure sampling threshold value;

if yes, go to the next step, otherwise, go to step S870;

s850: starting a garbage intercepting device of the garbage screening device, intercepting the grouped garbage piles passing through the pressure sampling sensor, and moving a screening frame of the garbage screening device to enable the intercepted grouped garbage piles to enter the screening frame;

s860: acquiring the weight and the type of the garbage entering the screening frame, adjusting the coarse grouping parameter, the screening size and the pressure sampling threshold value based on the weight and the type of the garbage, and returning to the step S810;

s870: receiving the grouped garbage piles from the second end of the horizontal throwing belt through the garbage transfer bucket, and returning to the step S810.

9. The intelligent automatic garbage classification method based on the AI identification technology as recited in claim 8, wherein:

when the garbage transferring barrel receives the grouped garbage pile from the second end of the horizontal throwing belt, the height of the inlet end face of the garbage transferring barrel and the height of the second end face of the horizontal throwing belt have a height difference larger than a preset height difference;

the preset height difference is not less than the height difference between the screening frame and the horizontal throwing belt.

10. The AI recognition technology-based intelligent automatic garbage classification method according to claim 9, characterized in that:

at least one blower is arranged between the screening frame and the horizontal throwing belt;

when the pressure sampling parameter is not larger than the pressure sampling threshold value in the step S840, turning on the blower at a first power;

when the pressure sampling parameter is judged to be larger than the pressure sampling threshold in the step S840, turning on the blower at a second power;

the first power is lower than the second power.

Images