CN110589282A - Intelligent garbage classification method based on machine learning and automatic garbage sorting device - Google Patents
Intelligent garbage classification method based on machine learning and automatic garbage sorting device Download PDFInfo
- Publication number
- CN110589282A CN110589282A CN201910755986.6A CN201910755986A CN110589282A CN 110589282 A CN110589282 A CN 110589282A CN 201910755986 A CN201910755986 A CN 201910755986A CN 110589282 A CN110589282 A CN 110589282A
- Authority
- CN
- China
- Prior art keywords
- garbage
- sorted
- neural network
- network model
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/0033—Refuse receptacles; Accessories therefor specially adapted for segregated refuse collecting, e.g. receptacles with several compartments; Combination of receptacles
- B65F1/0053—Combination of several receptacles
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F1/00—Refuse receptacles; Accessories therefor
- B65F1/14—Other constructional features; Accessories
- B65F1/1484—Other constructional features; Accessories relating to the adaptation of receptacles to carry identification means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/138—Identification means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/152—Material detecting means
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B65—CONVEYING; PACKING; STORING; HANDLING THIN OR FILAMENTARY MATERIAL
- B65F—GATHERING OR REMOVAL OF DOMESTIC OR LIKE REFUSE
- B65F2210/00—Equipment of refuse receptacles
- B65F2210/176—Sorting means
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/10—Waste collection, transportation, transfer or storage, e.g. segregated refuse collecting, electric or hybrid propulsion
Abstract
The invention discloses an intelligent garbage classification method and an automatic garbage sorting device based on machine learning, which comprise the following steps: detecting the garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted; generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to the plurality of garbage storage boxes; and controlling the second rotating assembly to move according to a second control signal so as to drive the garbage to be sorted to rotate on the vertical surface, and pouring the garbage to be sorted into the corresponding garbage storage box. The intelligent garbage classification method can automatically identify and classify the garbage so as to solve the problem of puzzlement of people to throw the garbage, and in addition, the difficulty and the workload of manual sorting can be reduced, the labor cost and the time cost are saved, and the recovery and utilization efficiency of resources is improved.
Description
Technical Field
The invention belongs to the field of garbage treatment, and particularly relates to an intelligent garbage classification method and an automatic garbage sorting device based on machine learning.
Background
With the improvement of the living standard and the consumption standard of people, daily garbage is generated more and more while a large amount of resources are consumed and the production scale is improved. The data show that the clearing and transportation amount of the municipal refuse in the country is more than 2500 million tons in 1979, the clearing and transportation cost of the municipal refuse is 1.16 yuan/ton in 1996, the clearing and transportation cost is 4 times that in 1979, and the increase is more rapid after 2000 years. The traditional garbage treatment method is only simple stacking or landfill, and the treatment method can cause bad smell, deterioration of sanitation, soil pollution and water quality pollution, further influences the living environment and the body health of people, and has unreasonable consequences. The means for harmlessly treating garbage include deep landfilling, high-temperature incineration, composting and the like, but the cost for treating one ton of garbage is about several hundred yuan, so the cost for harmlessly treating garbage is high. In the face of the situation that garbage is growing day by day and the environment is worsening day by day, how to further utilize recoverable garbage through scientific management and scientific treatment of garbage classification, recycle garbage resources to the maximum extent, save garbage treatment cost, reduce harm of harmful garbage and improve living environment quality is one of the urgent problems which are commonly concerned by countries in the world at present.
In general, domestic waste is classified in foreign cities according to its composition and amount of waste produced, and in combination with the utilization of local waste resources and the manner of waste disposal. For example, germany is generally classified into paper, glass, metal, plastic, and the like; australia is generally classified as compostable waste, recoverable waste, non-recoverable waste; japan generally classifies plastic bottles, recyclable plastics, other plastics, resource waste, large-sized waste, combustible waste, incombustible waste, harmful waste, and the like.
The Chinese government has also carried out vigorous publicity and promotion on garbage classification in recent years, and the garbage classification management is formally started in the Shanghai 7/1 in 2019, so that wide reverberation at home and abroad is caused. The garbage in Shanghai can be classified into recoverable garbage, wet garbage, dry garbage and harmful garbage. The recyclable garbage comprises recyclable garbage such as paper, glass, metal and plastic, and aims to recycle resources and change waste into valuable. According to statistics, about 30-40% of domestic garbage can be recycled, 600 kg of diesel oil can be recycled from 1 ton of waste plastics, 1500 tons of waste paper can be recycled, forest trees for producing 1200 tons of paper can be saved from cutting, and aluminum blocks formed by melting one ton of pop cans can be recycled, which is equivalent to less mining of 20 tons of aluminum ores. Therefore, the method fully excavates the recyclable resources in the household garbage, and has very important values for resource utilization and environmental protection.
However, the current garbage classification method only separates recyclable garbage from non-recyclable garbage, and does not subdivide the types of the recyclable garbage. If the recyclable garbage is not subdivided, the recyclable garbage is classified and sorted once in a garbage disposal plant to classify different types of garbage, otherwise, the recyclable garbage cannot be effectively utilized because of different materials such as paper, glass, metal, plastics and the like and different subsequent treatment means. Simultaneously, in the rubbish transportation in-process, be convenient for transport in order to reduce storage space, rubbish has carried out compression treatment, a large amount of glass can be broken, and the letter sorting operation in a large amount of rubbish yards also often is through manual sorting, broken glass or other rubbish also cause personnel's injury easily, the degree of difficulty and the work load of letter sorting work have been increased, undoubtedly be to people's quality of quality and patience's very big test, and the great variety also can make the dustbin kind infinitely increase, cause the degree of difficulty that people are more and operate.
In view of the above, overcoming the drawbacks of the prior art is an urgent problem in the art.
Disclosure of Invention
Aiming at the defects or improvement requirements of the prior art, the invention provides an intelligent garbage classification method and an automatic garbage sorting device based on machine learning, aiming at automatically identifying and classifying garbage so as to solve the problem of people about garbage throwing, and in addition, the difficulty and workload of manual sorting can be reduced, the labor cost and the time cost are saved, and the resource recovery and utilization efficiency is improved.
In order to achieve the above object, according to one aspect of the present invention, there is provided an intelligent garbage classification method based on machine learning, the intelligent garbage classification method being applied to an automatic garbage sorting device, the automatic garbage sorting device including a garbage can body, a first rotating assembly and a second rotating assembly, the garbage can body having a plurality of garbage storage boxes therein, the intelligent garbage classification method including:
detecting garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted;
generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to a plurality of garbage storage boxes; and controlling the second rotating assembly to move according to the second control signal so as to drive the garbage to be sorted to rotate on a vertical surface, and pouring the garbage to be sorted into a corresponding garbage storage box.
Preferably, be provided with the conductor detection sensor on the rubbish automatic sorting device, treat to sort rubbish and detect, acquire with treat to sort rubbish assorted detected signal, in order to confirm treat the type of sorting rubbish includes:
detecting the garbage to be sorted through the conductor detection sensor to obtain an electrical property detection signal;
judging whether the electrical detection signal exceeds a first preset threshold value or not;
if the electrical property detection signal exceeds the first preset threshold value, the garbage to be sorted is metal garbage;
and if the electrical property detection signal does not exceed the first preset threshold value, the garbage to be sorted is nonmetal garbage.
Preferably, still be provided with stress sensor on the rubbish automatic sorting device, if electrical property detected signal does not exceed first preset threshold value, then still include after waiting to sort rubbish is non-metal waste:
detecting the garbage to be sorted through the stress sensor to obtain a hardness detection signal;
judging whether the hardness detection signal exceeds a second preset threshold value or not;
and if the hardness detection signal exceeds the second preset threshold value, the garbage to be sorted is glass garbage.
Preferably, the intelligent garbage classification method further comprises:
if the hardness detection signal does not exceed the second preset threshold, judging whether the hardness detection signal exceeds a third preset threshold, wherein the third preset threshold is smaller than the second preset threshold;
if the hardness detection signal exceeds the third preset threshold value, the garbage to be sorted is paper garbage;
and if the hardness detection signal does not exceed the third preset threshold value, the garbage to be sorted is plastic garbage.
Preferably, a processor and an image acquisition device are arranged on the automatic garbage sorting device, the processor is connected with the image acquisition device, a neural network model is built in the processor, the garbage to be sorted is detected, and a detection signal matched with the garbage to be sorted is acquired to determine the type of the garbage to be sorted, including:
acquiring a first detection image of the garbage to be sorted through an image acquisition device;
inputting the first detection image into the neural network model, and processing the first detection image through the neural network model to determine the type of the garbage to be sorted.
Preferably, the neural network model building process includes:
acquiring various types of garbage images;
sequentially inputting the garbage images into the neural network model, and processing the garbage images through the neural network model to obtain classification results corresponding to the garbage images;
judging whether the classification result is consistent with the actual category of the garbage image;
if the neural network model is consistent with the neural network model, constructing the neural network model in the processor;
if not, adjusting weights of all layers in the neural network model according to a gradient descent method, re-training the garbage image according to the adjusted neural network model until a classification result output by the neural network model is consistent with the actual class of the garbage image, and constructing the final neural network model in the processor.
Preferably, the neural network model comprises: an input layer, a convolutional layer, a plurality of separable convolutional layers, a pooling layer, a full-link layer, a flexibility maximizing layer, and an output layer;
the input layer is used for receiving first detection image, the convolution layer is used for right first detection image carries out the convolution filtration for the first time, obtains initial rubbish characteristic, separable convolution layer is used for right initial rubbish characteristic carries out multiple convolution operation, obtains deep level rubbish characteristic, the pooling layer is used for sampling rubbish characteristic, it is right that the full articulamentum is used for rubbish characteristic is purified, flexible maximize layer is used for carrying out normalization processing to rubbish characteristic, the output layer is used for exporting the type that rubbish corresponds.
Preferably, the intelligent garbage classification method further comprises:
after the first detection image is obtained, continuously obtaining a second detection image of the garbage to be sorted through the image acquisition device, wherein the first detection image and the second detection image have different image obtaining angles;
inputting the second detection image into the neural network model, and processing the second detection image through the neural network model;
and judging and determining the type of the garbage to be sorted according to the detection result of the first detection image and the detection result of the second detection image.
Preferably, the automatic garbage sorting device comprises a baffle plate and a guide channel, the baffle plate is used for bearing the garbage to be sorted, the guide channel is used for guiding the garbage to be sorted onto the baffle plate, and the intelligent garbage sorting method further comprises the following steps:
acquiring the position of a baffle, and judging whether the position of the baffle corresponds to the position of the guide channel;
if the position of the baffle plate does not correspond to the position of the guide channel, closing the opening of the guide channel;
if the position of the baffle corresponds to the position of the guide channel, judging whether the baffle is provided with the garbage to be sorted;
if so, closing the opening of the guide channel;
if not, the opening of the guide channel is opened.
According to another aspect of the present invention, there is provided an automatic sorting apparatus for garbage, comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions programmed to perform the intelligent garbage classification method of the present invention.
Generally, compared with the prior art, the technical scheme of the invention has the following beneficial effects: the invention provides an intelligent garbage classification method and an automatic garbage sorting device based on machine learning, wherein the intelligent garbage classification method comprises the following steps: detecting the garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted; generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to the plurality of garbage storage boxes; and controlling the second rotating assembly to move according to a second control signal so as to drive the garbage to be sorted to rotate on the vertical surface, and pouring the garbage to be sorted into the corresponding garbage storage box. The intelligent garbage classification method can automatically identify and classify the garbage so as to solve the problem of puzzlement of people to throw the garbage, and in addition, the difficulty and the workload of manual sorting can be reduced, the labor cost and the time cost are saved, and the recovery and utilization efficiency of resources is improved.
Furthermore, the present embodiment determines the type of garbage by using a lightweight neural network model, has the characteristics of fast calculation speed, low complexity and small memory space occupied by the model, is convenient to be implemented on a miniaturized low-cost microprocessor quickly and conveniently, and saves the deployment cost.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required to be used in the embodiments of the present invention will be briefly described below. It is obvious that the drawings described below are only some embodiments of the invention, and that for a person skilled in the art, other drawings can be derived from them without inventive effort.
Fig. 1 is a schematic structural diagram of an automatic garbage sorting device according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of another automatic garbage sorting device provided by the embodiment of the invention;
fig. 3 is a schematic structural diagram of a rotating mechanism according to an embodiment of the present invention;
fig. 4 is a schematic structural diagram of another rotating mechanism provided in the embodiment of the present invention;
fig. 5 is a schematic structural diagram of a rotating mechanism according to an embodiment of the present invention after a baffle plate rotates on a vertical plane;
fig. 6 is a schematic view of a connection structure of a first connecting rod and a first bushing according to an embodiment of the present invention;
fig. 7 is a schematic flowchart of an intelligent garbage classification method based on machine learning according to an embodiment of the present invention;
fig. 8 is a schematic flow chart of acquiring a detection signal of garbage to be sorted according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of a neural network according to an embodiment of the present invention;
fig. 10 is a schematic flow chart of another method for acquiring a detection signal of garbage to be sorted according to an embodiment of the present invention;
fig. 11 is a schematic structural diagram of an automatic garbage sorting device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In addition, the technical features involved in the embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.
Example 1:
in order to clearly illustrate the intelligent garbage classification method according to the present invention, an automatic garbage sorting apparatus to which the intelligent garbage classification method is applied will be first described, with reference to fig. 1 and 2, the automatic garbage sorting apparatus includes: the garbage can comprises a garbage can body 1 and a rotating mechanism 2, wherein the rotating mechanism 2 is arranged in the garbage can body 1, a garbage throwing port 11 is formed in the garbage can body 1, a plurality of garbage storage boxes 12 are arranged in the garbage can body 1, and the garbage storage boxes 12 are used for containing garbage of corresponding types. The number of the garbage storage boxes 12 is not particularly limited, and may be designed according to actual situations.
In practical application, the garbage input port 11 may be disposed on an upper cover of the garbage can body 1 (as shown in fig. 1). In a preferred embodiment, the garbage input opening 11 can also be arranged on the side surface of the garbage can body 1, for example, the garbage input opening 11 is arranged above the side surface of the garbage can body 1, so as to prevent rainwater from entering the garbage can body 1. Of course, in other embodiments, the position of the garbage throwing port 11 is not specifically limited, and may be designed according to actual situations.
Under the practical application scene, the rubbish automatic sorting device of this embodiment is mainly applied to recoverable rubbish, and at present, recoverable rubbish mainly falls into four types: paper, plastic, glass and metal, and correspondingly, in an alternative scheme, the number of the garbage storage boxes 12 can be four, and as shown in fig. 1, the four garbage storage boxes 12 are distributed in a grid manner inside the garbage box body 1.
The rotating mechanism 2 includes a first rotating assembly 21 and a second rotating assembly 22, the first rotating assembly 21 is connected with the second rotating assembly 22, and a baffle 224 is disposed on the second rotating assembly 22.
In practical use, after the garbage is thrown into the garbage can body 1 from the garbage throwing port 11, the garbage falls onto the baffle 224, the first rotating assembly 21 is used for driving the baffle 224 to rotate on a horizontal plane, the position of the baffle 224 relative to the plurality of garbage storage boxes 12 is switched, and the second rotating assembly 22 is used for driving the baffle 224 to rotate on a vertical plane, so that the garbage is poured into the corresponding garbage storage box 12. In the view of fig. 2, the horizontal plane refers to the plane where the opening of the bin 12 is located, and the vertical plane refers to one of the planes perpendicular to the horizontal plane.
The garbage storage boxes 12 arranged in the automatic garbage sorting device are used for storing different types of garbage, after the garbage is thrown into the garbage box body 1, the garbage firstly falls onto the baffle 224, then the type of the garbage is determined according to the scheme in the prior art, and the baffle 224 is driven to move under the action of the rotating mechanism 2, so that the garbage is thrown into the corresponding garbage storage boxes 12, the automatic classification of the garbage is realized, and the labor and the cost are saved.
Specifically, as shown in fig. 3, the first rotating assembly 21 includes a first connecting rod 211, a first sleeve 212 and a first gear assembly 213, the first sleeve 212 is movably sleeved on the first connecting rod 211, the first gear assembly 213 is connected to the first sleeve 212, and the second rotating assembly 22 is fixedly connected to the first sleeve 212.
In an optional scheme, the first connecting rod 211 is fixed relative to the trash can body 1, and the first connecting rod 211 is fixedly connected with the upper cover of the trash can body 1, and the first connecting rod 211 is separated relative to the bottom cover of the trash can body 1, so that in actual use, when the upper cover of the trash can body 1 is taken down, the rotating mechanism 2 is taken out along with the upper cover, and the trash in the trash storage box 12 can be taken out conveniently. In another alternative, the first link 211 is fixedly connected to the bottom cover of the dustbin body 1, and the first link 211 is separated from the upper cover of the dustbin body 1, so as to facilitate taking out the garbage in the garbage storage bin 12 after the upper cover of the dustbin body 1 is removed.
The first gear assembly 213 is used to drive the first shaft sleeve 212 to rotate around the first connecting rod 211, and since the second rotating assembly 22 is fixed on the first shaft sleeve 212, when the first shaft sleeve 212 rotates, the baffle 224 on the second rotating assembly 22 can be driven to rotate on the horizontal plane.
Further, as shown in fig. 4, the first gear assembly 213 includes a first main gear 2131, a first driving gear 2132 and a first motor 2133, wherein the first driving gear 2132 is connected with the first main gear 2131 in a meshing manner, the first motor 2133 is connected with the first driving gear 2132, and the first motor 2133 is disposed on the first bushing 212; the first main gear 2131 is fixedly sleeved on the first link 211.
The first motor 2133 is configured to drive the first driving gear 2132 to rotate, and the first driving gear 2132 rotates along the first main gear 2131, so as to drive the first shaft sleeve 212 to rotate around the first link 211, where the baffle 224 may rotate in a range of 0 degree to 360 degrees on a horizontal plane.
In this embodiment, the first main gear 2131 is fixedly connected to the first link 211, the first motor 2133 is fixedly disposed on the first sleeve 212, and when the first motor 2133 drives the first driving gear 2132 to rotate, the first main gear 2131 provides a running track, so that the first driving gear 2132 rotates along the first main gear 2131, and the first sleeve 212 is driven to rotate around the first link 211.
In an alternative, as shown in fig. 6, a ball 3 is disposed between the first shaft sleeve 212 and the first link 211, a bearing connection is formed between the first shaft sleeve 212 and the first link 211, and the first shaft sleeve 212 can rotate along the first link 211 under the action of an external force.
In this embodiment, as shown in fig. 3, the second rotating assembly 22 includes a second connecting rod 221, a second bushing 222 and a second gear assembly 223, the second bushing 222 is disposed on the first bushing 212, the second connecting rod 221 is movably connected to the second bushing 222, wherein the second connecting rod 221 and the second bushing 222 may also form a bearing connection in the manner of fig. 5. The second gear assembly 223 is coupled with the second link 221, and the baffle 224 is disposed at an end of the second link 221; the second gear assembly 223 is used for driving the second connecting rod 221 to rotate, so as to drive the baffle 224 to rotate on a vertical plane.
Specifically, as shown in fig. 4, the second gear assembly 223 includes a second main gear 2231, a second driving gear 2232 and a second motor 2233, the second motor 2233 is disposed on the second bushing 222, the second main gear 2231 is fixedly sleeved on the second connecting rod 221, and the second main gear 2231 and the second driving gear 2232 are in meshed connection; the second motor 2233 is configured to drive the second driving gear 2232 to rotate, and the second driving gear 2232 drives the second main gear 2231 to rotate, so as to drive the baffle 224 at the end of the second connecting rod 221 to rotate on the vertical plane.
In this embodiment, the first link 211 and the second link 221 are disposed to intersect, for example, the first link 211 and the second link 221 are disposed at 90 degrees to each other, wherein a horizontal plane may be a plane perpendicular to the first link 211, and a vertical plane may be a plane perpendicular to the second link 221.
The second main gear 2231 is fixedly sleeved on the second connecting rod 221, and when the second driving gear 2232 rotates, the second main gear 2231 is driven to rotate, and the second connecting rod 221 also rotates, so as to drive the baffle 224 at the end of the second connecting rod 221 to rotate on the vertical surface.
Referring to fig. 3 and 5, the flapper 224 is in the initial position in fig. 3, the flapper 224 is parallel with respect to the horizontal plane, and in fig. 5 the flapper 224 is rotated in a vertical plane by the second rotating assembly 22 by a certain angle (e.g., 90 degrees) to dump the trash on the flapper into the corresponding trash storage bin 12.
In order to guide the garbage to the baffle 224, as shown in fig. 2, a guide passage 13 is provided between the baffle 224 and the garbage input port 11, and the guide passage 13 guides the garbage input from the garbage input port 11 to the baffle 224. In an alternative scheme, the position of the guide channel 13 is relatively unchanged, and after the garbage on the barrier 224 is dumped into the corresponding garbage storage bin 12, the barrier 224 is reset under the action of the first rotating assembly 21, so that the guide channel 13 corresponds to the barrier 224.
In this embodiment, the type of the garbage can be detected by a sensor, and specifically, a plurality of sensors are disposed on the barrier 224, and the sensors are used for detecting the type of the garbage, so that the first rotating assembly 21 drives the barrier 224 to move to the corresponding garbage storage box 12. The sensor comprises a conductor detection sensor, a sensor body and a sensor body, wherein the conductor detection sensor is used for detecting the conductivity of an object to be detected so as to determine whether the garbage is metal or not; the sensor also comprises a stress sensor for detecting the elastic force and the hardness of the object to be detected so as to determine that the garbage is glass, paper or plastic. In an optional scheme, an image acquisition device can be further arranged on the dustbin body 1, and the image acquisition device acquires image signals, determines the type of rubbish and automatically classifies and manages the rubbish.
Further, an infrared sensor is provided at the top of each bin 12 to detect whether the bin 12 is full.
The scheme related to the embodiment is completed based on the cooperation of a mechanical structure and an electrical structure, the automatic garbage sorting device further comprises a processor, the processor is respectively connected with the sensors, the first motor 2133 and the second motor 2233, and the processor is used for respectively sending control signals to the first motor 2133 and the second motor 2233 according to the detection results of the sensors, so as to drive the baffle 224 to move, and pour the garbage into the corresponding garbage storage box 12, which is specifically described in the text in embodiment 2.
Be provided with the display screen on the rubbish automatic sorting device, the display screen with the treater is connected, the display screen is used for showing the serial number and the position of dustbin body 1 to and show whether a certain rubbish storage box 12 is fully loaded. In actual use, through infrared sensor or visual signal detection, judge that certain rubbish storage box 12 is full load the back, show on the display screen to indicate the user, and will correspond the full load signal of box and send appointed administrator through wireless signal, indicate the administrator to clear up corresponding box.
In a preferred embodiment, the automatic garbage sorting device is provided with a charging plate, the charging plate is a solar charging plate or a wind energy charging plate, and the charging plate is used for providing a power supply for the automatic garbage sorting device, so that natural resources such as solar energy or wind energy can be fully utilized to generate electricity, and the purpose of environmental protection is achieved. Wherein the charging plate may be connected with the battery unit to charge the battery unit.
Regarding the determination of the position of the flap 224 with respect to different waste storage bins 12, the initial position of the flap 224 may be preset such that the flap 224 corresponds to the position of one of the waste storage bins 12, and in the subsequent process, the corresponding relationship between the flap 224 and the different waste storage bins 12 may be determined according to the rotation angle of the flap 224 on the horizontal plane.
Example 2:
referring to fig. 7, the automatic garbage sorting device according to the foregoing embodiment provides an intelligent garbage classification method based on machine learning, where the intelligent garbage classification method includes the following steps:
step 101: and detecting the garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted.
The automatic garbage sorting device of the embodiment is mainly applied to sorting recyclable garbage, wherein the recyclable garbage includes metal, glass, paper and plastic, that is, the types of garbage to be sorted include metal garbage and non-metal garbage, and the non-metal garbage includes glass garbage, paper garbage and plastic garbage.
In this embodiment, the detection signals matched with the garbage to be sorted may be acquired by using a plurality of sensors and/or an image acquisition device to acquire image signals, so as to determine the type of the garbage to be sorted.
The following explains the acquisition of detection signals matching the garbage to be sorted based on different detection modes respectively.
The first method is as follows:
with reference to fig. 8, in an optional scheme, the to-be-sorted garbage is detected by the conductor detection sensor, and an electrical detection signal is obtained; judging whether the electrical detection signal exceeds a first preset threshold value or not; if the electrical property detection signal exceeds the first preset threshold value, the garbage to be sorted is metal garbage; and if the electrical property detection signal does not exceed the first preset threshold value, the garbage to be sorted is nonmetal garbage.
Still be provided with stress sensor on the rubbish automatic sorting device, if electrical property detected signal does not exceed first predetermined threshold value, then wait to sort still to include after rubbish is non-metallic waste: detecting the garbage to be sorted through the stress sensor to obtain a hardness detection signal; judging whether the hardness detection signal exceeds a second preset threshold value or not; if the hardness detection signal exceeds the second preset threshold value, the garbage to be sorted is glass garbage; if the hardness detection signal does not exceed the second preset threshold, judging whether the hardness detection signal exceeds a third preset threshold, wherein the third preset threshold is smaller than the second preset threshold; if the hardness detection signal exceeds the third preset threshold value, the garbage to be sorted is paper garbage; and if the hardness detection signal does not exceed the third preset threshold value, the garbage to be sorted is plastic garbage.
The first preset threshold, the second preset threshold and the third preset threshold may be determined according to actual conditions, for example, by collecting a large number of samples, the first preset threshold, the second preset threshold and the third preset threshold are set to ensure accuracy.
The conductor detection sensor can be a sensor based on a coil, and obtains an electrical detection signal of garbage to be sorted through electromagnetic induction, wherein the electrical detection signal can be a current detection signal or a voltage detection signal, and a first preset threshold is a preset current threshold or a preset voltage threshold.
In this embodiment, the garbage is detected by the various sensors, the corresponding electrical detection signal and hardness detection signal are obtained, and the type of the garbage is determined according to the electrical detection signal and hardness detection signal, so as to identify the metal garbage, the glass garbage, the paper garbage and the plastic garbage.
The second method comprises the following steps:
be provided with treater and image acquisition device on the rubbish automatic sorting device, the treater with image acquisition device connects, the neural network model is built to treating to sort rubbish in the treater, treat to sort rubbish and detect, acquire with treat to sort rubbish assorted detected signal, in order to confirm treat the type of sorting rubbish includes: acquiring a first detection image of the garbage to be sorted through an image acquisition device; inputting the first detection image into the neural network model, and processing the first detection image through the neural network model to determine the type of the garbage to be sorted. In the embodiment, the type of the garbage is determined by means of image recognition.
In a preferred embodiment, after the first detection image is acquired, a second detection image of the garbage to be sorted is continuously acquired through the image acquisition device, wherein the image acquisition angles of the first detection image and the second detection image are different. Compared with the method of identifying by using one detection image independently, the method of identifying the garbage to be sorted by using a plurality of detection images with different shooting angles can extract more characteristic information, so that the identification accuracy is improved. In this embodiment, a neural network model is used to process the first detection image and the second detection image respectively, and the type of the garbage to be sorted is determined by combining the detection result of the first detection image and the detection result of the second detection image.
In an optional scheme, after the first detection image is obtained, the first detection image is processed through the neural network model, when the first detection image contains less characteristic information, the neural network model cannot accurately judge the type of the garbage to be sorted, if output information which cannot be judged is fed back by the neural network model is received, the processor outputs a third control signal, the first rotating assembly is controlled to move according to the third control signal so as to drive the garbage to be sorted to rotate on a horizontal plane, after the garbage to be sorted rotates to a target position, an image acquisition device is used for obtaining a second detection image, the second detection image is input to the neural network model, the second detection image is processed through the neural network model, and the detection result of the first detection image and the detection result of the second detection image are combined, judging and determining the type of the garbage to be sorted.
The second detection image is selectively acquired in the mode, so that the identification accuracy can be improved, and meanwhile, the efficiency can be improved.
In an actual application scenario, after the second detection image is acquired, the to-be-sorted garbage can be driven to return to the initial position through the first rotating assembly, or the to-be-sorted garbage can be continuously kept at the target position, after the type of the to-be-sorted garbage is determined, a first control signal is generated according to the target position and a path between the to-be-sorted garbage storage boxes, and then the to-be-sorted garbage is driven to move to the upper side of the corresponding garbage storage boxes under the control of the first control signal.
In an actual application scene, objects can be classified through classification strategies such as a support vector machine, a naive Bayes classifier, logistic regression and a decision tree, and the classification strategies can obtain more accurate classification results for different scenes and some public test data sets. In recent years, with the progress of deep learning means, researchers have proposed networks such as vgnet, ResNet, and SeNet, and have obtained good results in a large number of classification tasks for standard data sets. However, the foregoing network model still has some limitations in practical application, and on one hand, the foregoing method often has better performance on a public data set, but a larger deviation often exists on measured data; on the other hand, in order to achieve higher accuracy, these network models are often more complex and generally have better effects on high-performance servers, but in practical applications, a small device based on an embedded system or a microprocessor is often required to be designed, and such models are difficult to implement on the microprocessor due to large occupied memory and hard disk space.
In order to solve the foregoing problems, the present embodiment determines the type of garbage by using a lightweight neural network model, and has the characteristics of fast calculation speed, low complexity, and small memory space occupied by the model, so that the method is conveniently and quickly implemented on a miniaturized low-cost microprocessor, and the deployment cost is saved. The following specifically describes a process of constructing a neural network model and a structure of the neural network model.
Firstly, acquiring a plurality of types of garbage images, enriching a training set, performing normalization operation on each garbage image, and normalizing each garbage image into an RGB image with a preset size, wherein the corresponding number of samples of the garbage image is not specifically limited, for example, the number of samples of the garbage image may be 6000 or other values. The predetermined size may be set according to practical situations, for example, the predetermined size is a tensor 224 × 3, where 224 × 224 represents the length and width of the picture, and 3 represents R, G, B three channels.
Sequentially inputting the garbage images into the neural network model, and processing the garbage images through the neural network model to obtain classification results corresponding to the garbage images; judging whether the classification result is consistent with the actual category of the garbage image; if the neural network model is consistent with the neural network model, constructing the neural network model in the processor; if not, adjusting weights of all layers in the neural network model according to a gradient descent method, re-training the garbage image according to the adjusted neural network model until a classification result output by the neural network model is consistent with the actual class of the garbage image, and constructing the final neural network model in the processor.
As shown in fig. 9, the neural network model includes: an input layer, a convolutional layer, a plurality of separable convolutional layers, a pooling layer, a full-link layer, a flexibility maximizing layer, and an output layer; the input layer is used for receiving the first detection image; the convolution layer is used for right first detection image carries out the convolution filtration for the first time, obtains initial rubbish characteristic, separable convolution layer is used for right initial rubbish characteristic carries out multiple convolution operation, obtains deep level rubbish characteristic, the pooling layer is used for sampling rubbish characteristic, it is right that the complete linkage layer is used for rubbish characteristic is purified, flexible maximization layer is used for carrying out normalization process to rubbish characteristic, the output layer is used for exporting the type that rubbish corresponds.
In the present embodiment, the depth separable convolutional layer is added, and the conventional convolutional layer is reduced, so as to reduce the complexity of the model. The activation functions of the convolutional layer and the depth separable convolutional layer both use a ReLU (reduced linear unit, abbreviated as ReLU) activation function, and the step size, the convolutional kernel size and the number of convolutional kernels of each convolutional layer can be freely set. The full connection layer is not provided with an activation function and is directly connected with the maximum flexible layer.
In the training process, adjusting parameters of the neural network model, optimizing a loss function, and generating a final neural network model when the cost loss function is reduced to an ideal degree and the training reaches the required maximum iteration times; in the training process of the neural network, a loss function is established by using cross entropy, and the specific formula is as follows:
where n denotes the number of samples and i denotes the ith sample. y iscFor variables labeled as garbage types, the actual type representing the current sample is c, pcThe output of the neural network model is represented as the predicted probability of the current sample being considered by the neural network model as class c.
The number of the separable convolution layers is not particularly limited, and may be 13, for example. Also, a pooling layer may be provided between each two separable convolutional layers to prevent over-fitting.
In an actual application scene, the size of the improved neural network model is only 30MB, the neural network model is constructed on a raspberry device with a 1G internal memory and an external 8G memory card, and the type of input garbage can be accurately detected in real time.
The third method comprises the following steps:
in another optional scheme, in order to improve the accuracy of detection, a first mode and a second mode can be combined, and the type of the garbage to be sorted is judged and determined by fusing the detection result of the sensor and the detection result of image acquisition. For example, if the detection result of the sensor is the same as the detection result of the image acquisition, the type of the garbage to be sorted can be directly determined; if the detection result of the sensor is different from the detection result of the image acquisition, the detection image is sent to the server, the type of the garbage to be sorted is determined by the server according to the detection image, and the automatic garbage sorting device acquires the detection result of the server to determine the type of the garbage to be sorted.
Meanwhile, if any one of the detection result of the sensor and the detection result of the image acquisition is the same as the detection result of the server, the weight occupied by the detection result of the sensor and the detection result of the image acquisition is corrected so as to be used as the final detection result according to the detection result of the side with the larger weight; and if any detection result of the sensor and the detection result of the image acquisition is different from the detection result of the server, correcting the threshold value corresponding to the sensor and the parameter corresponding to the neural network model so as to improve the respective detection results.
In another alternative, with reference to fig. 10, after the garbage to be sorted enters the garbage bin body, the characteristics of the garbage to be sorted are obtained through a sensor; the method comprises the steps of obtaining a detection image of garbage to be sorted through an image acquisition device, then extracting features of the garbage to be sorted through a neural network model, fusing the features obtained through a sensor and the features obtained through the neural network model, and determining the type of the garbage to be sorted. In this embodiment, the garbage to be sorted is detected from different dimensions, so that the detection accuracy can be improved.
Meanwhile, the characteristics of the garbage to be sorted are added into the training set to optimize the labels, enrich training data and further improve the accuracy of the neural network model.
Step 102: generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to a plurality of garbage storage boxes; and controlling the second rotating assembly to move according to the second control signal so as to drive the garbage to be sorted to rotate on a vertical surface, and pouring the garbage to be sorted into a corresponding garbage storage box.
In this embodiment, the garbage storage boxes corresponding to the garbage to be sorted are determined according to the type of the garbage to be sorted, a motion track is determined according to a path between the current position of the garbage to be sorted and the garbage storage boxes, and then a first control signal is generated according to the motion track, where the first control signal includes a rotation direction of a motor, a rotation speed of the motor, and a rotation time of the motor, so that the first control signal drives the corresponding motor to rotate to drive the first rotating assembly to move, thereby switching the positions of the garbage to be sorted relative to the plurality of garbage storage boxes, and moving the garbage to be sorted to the upper side of the corresponding garbage storage boxes.
And after the garbage to be sorted moves to the upper part of the corresponding garbage storage box, controlling the second rotating assembly to move according to the second control signal so as to drive the garbage to be sorted to rotate on the vertical surface and dump the garbage to be sorted into the corresponding garbage storage box, wherein the second control signal comprises the rotating direction of the motor, the rotating speed of the motor and the rotating time of the motor.
In an actual application scenario, there are situations that the sorting operation of the garbage input before is not completed, and the garbage input occurs, and at this time, there may be two situations: (1) the baffle is not corresponding to the guide channel, and the garbage which is put into later can be directly put into the garbage storage box corresponding to the guide channel. Influencing the sorting of the garbage; (2) the baffle does not correspond with the guide channel, but the garbage thrown into the baffle last time is still on the baffle, and the garbage thrown into the baffle later can fall on the baffle, if the garbage thrown into the baffle last time is different from the garbage thrown into the baffle last time, the automatic garbage sorting device can not carry out the operation next step, and the sorting of the garbage is influenced.
In order to solve the aforementioned problem, in a preferred embodiment, a movable plate is provided on the guide passage, and the movable plate is used for opening or closing the opening of the guide passage, and for example, the position of the movable plate can be switched by a cylinder or a solenoid valve.
In the embodiment, the position of the baffle is obtained, and whether the position of the baffle corresponds to the position of the guide channel is judged; if the position of the baffle plate does not correspond to the position of the guide channel, closing the opening of the guide channel; if the position of the baffle corresponds to the position of the guide channel, judging whether the baffle is provided with the garbage to be sorted; if so, closing the opening of the guide channel; if not, the opening of the guide channel is opened.
Different from the prior art, the intelligent garbage classification method of the embodiment includes: detecting the garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted; generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to the plurality of garbage storage boxes; and controlling the second rotating assembly to move according to a second control signal so as to drive the garbage to be sorted to rotate on the vertical surface, and pouring the garbage to be sorted into the corresponding garbage storage box. The intelligent garbage classification method can automatically identify and classify garbage to solve the confusion of people about garbage putting, and can reduce the difficulty and workload of manual sorting, save the labor cost and time cost and improve the recovery and utilization efficiency of resources.
Example 3:
referring to fig. 11, fig. 11 is a schematic structural diagram of an automatic garbage sorting device according to an embodiment of the present invention. The automatic sorting apparatus for garbage of the present embodiment includes one or more processors 41 and a memory 42. In fig. 11, one processor 41 is taken as an example.
The processor 41 and the memory 42 may be connected by a bus or other means, and fig. 11 illustrates the connection by a bus as an example.
The memory 42, which is a non-volatile computer-readable storage medium based on the intelligent garbage classification method, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the intelligent garbage classification method based on machine learning in embodiment 2 and corresponding program instructions. The processor 41 implements the functions of the intelligent machine-learning-based garbage classification method of embodiment 2 by executing various functional applications and data processing of the intelligent machine-learning-based garbage classification method by executing nonvolatile software programs, instructions, and modules stored in the memory 42.
The memory 42 may include, among other things, high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 42 may optionally include memory located remotely from processor 41, which may be connected to processor 41 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
Please refer to fig. 7 to fig. 10 and the related text description for an intelligent garbage classification method based on machine learning, which will not be described again.
It should be noted that, for the information interaction, execution process and other contents between the modules and units in the apparatus and system, the specific contents may refer to the description in the embodiment of the method of the present invention because the same concept is used as the embodiment of the processing method of the present invention, and are not described herein again.
Those of ordinary skill in the art will appreciate that all or part of the steps of the various methods of the embodiments may be implemented by associated hardware as instructed by a program, which may be stored on a computer-readable storage medium, which may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like.
It will be understood by those skilled in the art that the foregoing is only a preferred embodiment of the present invention, and is not intended to limit the invention, and that any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (10)
1. An intelligent garbage classification method based on machine learning is characterized in that the intelligent garbage classification method is applied to an automatic garbage sorting device, the automatic garbage sorting device comprises a garbage can body, a first rotating assembly and a second rotating assembly, a plurality of garbage storage boxes are arranged in the garbage can body, and the intelligent garbage classification method comprises the following steps:
detecting garbage to be sorted, and acquiring a detection signal matched with the garbage to be sorted so as to determine the type of the garbage to be sorted;
generating a first control signal and a second control signal based on the type of the garbage to be sorted, and controlling the first rotating assembly to move according to the first control signal so as to drive the garbage to be sorted to rotate on a horizontal plane and switch the positions of the garbage to be sorted relative to a plurality of garbage storage boxes; and controlling the second rotating assembly to move according to the second control signal so as to drive the garbage to be sorted to rotate on a vertical surface, and pouring the garbage to be sorted into a corresponding garbage storage box.
2. The intelligent garbage classification method according to claim 1, wherein the automatic garbage sorting device is provided with a conductor detection sensor, and the step of detecting the garbage to be sorted and acquiring a detection signal matched with the garbage to be sorted to determine the type of the garbage to be sorted comprises the steps of:
detecting the garbage to be sorted through the conductor detection sensor to obtain an electrical property detection signal;
judging whether the electrical detection signal exceeds a first preset threshold value or not;
if the electrical property detection signal exceeds the first preset threshold value, the garbage to be sorted is metal garbage;
and if the electrical property detection signal does not exceed the first preset threshold value, the garbage to be sorted is nonmetal garbage.
3. The intelligent garbage classification method according to claim 2, wherein a stress sensor is further disposed on the automatic garbage sorting device, and if the electrical detection signal does not exceed the first preset threshold, the method further comprises the following steps after the garbage to be sorted is non-metal garbage:
detecting the garbage to be sorted through the stress sensor to obtain a hardness detection signal;
judging whether the hardness detection signal exceeds a second preset threshold value or not;
and if the hardness detection signal exceeds the second preset threshold value, the garbage to be sorted is glass garbage.
4. The intelligent garbage classification method according to claim 3, further comprising:
if the hardness detection signal does not exceed the second preset threshold, judging whether the hardness detection signal exceeds a third preset threshold, wherein the third preset threshold is smaller than the second preset threshold;
if the hardness detection signal exceeds the third preset threshold value, the garbage to be sorted is paper garbage;
and if the hardness detection signal does not exceed the third preset threshold value, the garbage to be sorted is plastic garbage.
5. The intelligent garbage classification method according to claim 1, wherein a processor and an image acquisition device are arranged on the automatic garbage sorting device, the processor is connected with the image acquisition device, a neural network model is built in the processor, the garbage to be sorted is detected, and the obtaining of the detection signal matched with the garbage to be sorted to determine the type of the garbage to be sorted comprises:
acquiring a first detection image of the garbage to be sorted through an image acquisition device;
inputting the first detection image into the neural network model, and processing the first detection image through the neural network model to determine the type of the garbage to be sorted.
6. The intelligent garbage classification method according to claim 5, wherein the neural network model construction process comprises:
acquiring various types of garbage images;
sequentially inputting the garbage images into the neural network model, and processing the garbage images through the neural network model to obtain classification results corresponding to the garbage images;
judging whether the classification result is consistent with the actual category of the garbage image;
if the neural network model is consistent with the neural network model, constructing the neural network model in the processor;
if not, adjusting weights of all layers in the neural network model according to a gradient descent method, re-training the garbage image according to the adjusted neural network model until a classification result output by the neural network model is consistent with the actual class of the garbage image, and constructing the final neural network model in the processor.
7. The intelligent garbage classification method of claim 5, wherein the neural network model comprises: an input layer, a convolutional layer, a plurality of separable convolutional layers, a pooling layer, a full-link layer, a flexibility maximizing layer, and an output layer;
the input layer is used for receiving first detection image, the convolution layer is used for right first detection image carries out the convolution filtration for the first time, obtains initial rubbish characteristic, separable convolution layer is used for right initial rubbish characteristic carries out multiple convolution operation, obtains deep level rubbish characteristic, the pooling layer is used for sampling rubbish characteristic, it is right that the full articulamentum is used for rubbish characteristic is purified, flexible maximize layer is used for carrying out normalization processing to rubbish characteristic, the output layer is used for exporting the type that rubbish corresponds.
8. The intelligent garbage classification method according to claim 5, further comprising:
after the first detection image is obtained, continuously obtaining a second detection image of the garbage to be sorted through the image acquisition device, wherein the first detection image and the second detection image have different image obtaining angles;
inputting the second detection image into the neural network model, and processing the second detection image through the neural network model;
and judging and determining the type of the garbage to be sorted according to the detection result of the first detection image and the detection result of the second detection image.
9. The intelligent garbage classification method according to any one of claims 1 to 8, wherein the automatic garbage sorting device comprises a baffle plate and a guide channel, the baffle plate is used for bearing the garbage to be sorted, the guide channel is used for guiding the garbage to be sorted onto the baffle plate, and the intelligent garbage classification method further comprises the following steps:
acquiring the position of a baffle, and judging whether the position of the baffle corresponds to the position of the guide channel;
if the position of the baffle plate does not correspond to the position of the guide channel, closing the opening of the guide channel;
if the position of the baffle corresponds to the position of the guide channel, judging whether the baffle is provided with the garbage to be sorted;
if so, closing the opening of the guide channel;
if not, the opening of the guide channel is opened.
10. An automatic sorting device for garbage is characterized by comprising at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor and programmed to perform the intelligent garbage classification method of any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910755986.6A CN110589282A (en) | 2019-08-16 | 2019-08-16 | Intelligent garbage classification method based on machine learning and automatic garbage sorting device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910755986.6A CN110589282A (en) | 2019-08-16 | 2019-08-16 | Intelligent garbage classification method based on machine learning and automatic garbage sorting device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110589282A true CN110589282A (en) | 2019-12-20 |
Family
ID=68854469
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910755986.6A Pending CN110589282A (en) | 2019-08-16 | 2019-08-16 | Intelligent garbage classification method based on machine learning and automatic garbage sorting device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110589282A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027639A (en) * | 2019-12-24 | 2020-04-17 | 北华大学 | Cloud-based garbage sorting system and method, information processing terminal and storage medium |
| CN111186656A (en) * | 2020-01-10 | 2020-05-22 | 上海电力大学 | Target garbage classification method and intelligent garbage can |
| CN111573037A (en) * | 2020-05-16 | 2020-08-25 | 邵丽丹 | Intelligent garbage can convenient for garbage classification and recycling for residential community |
| CN111992516A (en) * | 2020-08-19 | 2020-11-27 | 淮北速达医疗转运服务有限公司 | Intelligent garbage classification system |
| CN112660640A (en) * | 2020-12-09 | 2021-04-16 | 西安文理学院 | Garbage classification device and method |
| CN113173349A (en) * | 2021-05-06 | 2021-07-27 | 上海市格致中学 | Garbage classification system and method |
| CN113213016A (en) * | 2021-05-31 | 2021-08-06 | 南京工业大学 | Garbage recognition and classification algorithm based on ViT, and device and control method thereof |
| CN113401543A (en) * | 2021-06-17 | 2021-09-17 | 广东机电职业技术学院 | Garbage classification device and classification method based on unmanned trolley |
| CN113731832A (en) * | 2021-11-04 | 2021-12-03 | 南京信息工程大学 | Garbage sorting treatment method and system for garbage transfer station |
| CN114239676A (en) * | 2021-11-08 | 2022-03-25 | 苏州纳故环保科技有限公司 | Image classification method based on improved ResNet network |
| CN117522388A (en) * | 2023-11-08 | 2024-02-06 | 永昊环境科技(集团)有限公司 | Intelligent sanitation processing method for urban environment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107857032A (en) * | 2017-12-15 | 2018-03-30 | 柳州市鼎福通用机械厂 | A kind of city intelligent dustbin and its control method |
| CN108341184A (en) * | 2018-03-01 | 2018-07-31 | 安徽省星灵信息科技有限公司 | A kind of intelligent sorting dustbin |
| CN208683621U (en) * | 2018-08-30 | 2019-04-02 | 德江县民族中医院 | A kind of medical Environmental-protection garbage collection device taken out convenient for rubbish |
| CN109625680A (en) * | 2019-01-28 | 2019-04-16 | 中国地质大学（武汉） | A kind of intelligent recognition classification recycling dustbin |
| CN208813938U (en) * | 2018-08-29 | 2019-05-03 | 广州大学 | A kind of garbage classification device |
| CN110002141A (en) * | 2019-04-29 | 2019-07-12 | 肇庆市忠良环境卫生管理有限公司 | A kind of Intelligent refuse classification and recyclable device |
| CN110116415A (en) * | 2019-06-12 | 2019-08-13 | 中北大学 | A kind of Bottle & Can class rubbish identification sorting machine people based on deep learning |
-
2019
- 2019-08-16 CN CN201910755986.6A patent/CN110589282A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107857032A (en) * | 2017-12-15 | 2018-03-30 | 柳州市鼎福通用机械厂 | A kind of city intelligent dustbin and its control method |
| CN108341184A (en) * | 2018-03-01 | 2018-07-31 | 安徽省星灵信息科技有限公司 | A kind of intelligent sorting dustbin |
| CN208813938U (en) * | 2018-08-29 | 2019-05-03 | 广州大学 | A kind of garbage classification device |
| CN208683621U (en) * | 2018-08-30 | 2019-04-02 | 德江县民族中医院 | A kind of medical Environmental-protection garbage collection device taken out convenient for rubbish |
| CN109625680A (en) * | 2019-01-28 | 2019-04-16 | 中国地质大学（武汉） | A kind of intelligent recognition classification recycling dustbin |
| CN110002141A (en) * | 2019-04-29 | 2019-07-12 | 肇庆市忠良环境卫生管理有限公司 | A kind of Intelligent refuse classification and recyclable device |
| CN110116415A (en) * | 2019-06-12 | 2019-08-13 | 中北大学 | A kind of Bottle & Can class rubbish identification sorting machine people based on deep learning |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111027639A (en) * | 2019-12-24 | 2020-04-17 | 北华大学 | Cloud-based garbage sorting system and method, information processing terminal and storage medium |
| CN111186656A (en) * | 2020-01-10 | 2020-05-22 | 上海电力大学 | Target garbage classification method and intelligent garbage can |
| CN111573037B (en) * | 2020-05-16 | 2022-02-25 | 邵丽丹 | Intelligent garbage can convenient for garbage classification and recycling for residential community |
| CN111573037A (en) * | 2020-05-16 | 2020-08-25 | 邵丽丹 | Intelligent garbage can convenient for garbage classification and recycling for residential community |
| CN111992516A (en) * | 2020-08-19 | 2020-11-27 | 淮北速达医疗转运服务有限公司 | Intelligent garbage classification system |
| CN112660640A (en) * | 2020-12-09 | 2021-04-16 | 西安文理学院 | Garbage classification device and method |
| CN113173349A (en) * | 2021-05-06 | 2021-07-27 | 上海市格致中学 | Garbage classification system and method |
| CN113213016A (en) * | 2021-05-31 | 2021-08-06 | 南京工业大学 | Garbage recognition and classification algorithm based on ViT, and device and control method thereof |
| CN113401543A (en) * | 2021-06-17 | 2021-09-17 | 广东机电职业技术学院 | Garbage classification device and classification method based on unmanned trolley |
| CN113731832A (en) * | 2021-11-04 | 2021-12-03 | 南京信息工程大学 | Garbage sorting treatment method and system for garbage transfer station |
| CN113731832B (en) * | 2021-11-04 | 2022-02-15 | 南京信息工程大学 | Garbage sorting treatment method and system for garbage transfer station |
| CN114239676A (en) * | 2021-11-08 | 2022-03-25 | 苏州纳故环保科技有限公司 | Image classification method based on improved ResNet network |
| CN117522388A (en) * | 2023-11-08 | 2024-02-06 | 永昊环境科技(集团)有限公司 | Intelligent sanitation processing method for urban environment |
| CN117522388B (en) * | 2023-11-08 | 2024-04-12 | 永昊环境科技(集团)有限公司 | Intelligent sanitation processing method for urban environment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110589282A (en) | Intelligent garbage classification method based on machine learning and automatic garbage sorting device | |
| CN110884791A (en) | Vision garbage classification system and classification method based on TensorFlow | |
| CN109928107A (en) | A kind of automatic classification system | |
| CN105642645B (en) | Cities and towns composting dry type Automated fine method for separating and its equipment | |
| CN110937280B (en) | Audio-visual combination based intelligent garbage classification and recovery method and terminal | |
| CN203782622U (en) | Intelligent garbage-collecting-sorting educating robot | |
| CN110355188B (en) | Smart garbage classification in-situ reduction recycling system based on multiple land utilization | |
| CN109250353A (en) | A kind of dustbin and storage medium with image identification function | |
| CN110466911A (en) | Automatic sorting garbage bin and classification method | |
| CN103318582A (en) | Automatic classification trash can | |
| CN109018777A (en) | A kind of refuse classification method | |
| CN109250351A (en) | A kind of garbage classification device | |
| Salem et al. | A critical review of existing and emerging technologies and systems to optimize solid waste management for feedstocks and energy conversion | |
| CN113319007A (en) | Household garbage sorting method based on machine vision recognition | |
| CN113859803A (en) | Intelligent identification trash can and intelligent identification method thereof | |
| Ji et al. | Automatic sorting of low-value recyclable waste: a comparative experimental study | |
| CN210417814U (en) | Automatic garbage sorting device | |
| CN111137602B (en) | Garbage classification method and garbage classification system based on artificial intelligence | |
| CN116510901A (en) | Intelligent household garbage sorting system | |
| Kapadia et al. | Dry waste segregation using seamless integration of deep learning and industrial machine vision | |
| CN215708916U (en) | Intelligent recognition garbage can | |
| CN112620165B (en) | Garbage classification method | |
| CN217576626U (en) | Classification garbage can | |
| Sanathkumar et al. | Smart Waste Segregation | |
| CN211443620U (en) | Garbage can for classified detection of garbage |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191220 |
|
| RJ01 | Rejection of invention patent application after publication |