CN211544683U

CN211544683U - Intelligent classification response garbage bin based on vision technique

Info

Publication number: CN211544683U
Application number: CN201922455525.5U
Authority: CN
Inventors: 谢糖; 王昕�
Original assignee: Individual
Current assignee: Individual
Priority date: 2019-12-31
Filing date: 2019-12-31
Publication date: 2020-09-22
Anticipated expiration: 2029-12-31

Abstract

The utility model provides an intelligent classification responds to garbage bin based on vision technique, include: the garbage sorting device comprises a barrel body, a sorting platform arranged below a garbage input port of the barrel body, a garbage sorting channel arranged below the sorting platform and a sorting garbage barrel arranged at the bottom of the barrel body below the garbage sorting channel; a camera is arranged above the sorting platform; the middle shaft of the sorting platform is connected with a first speed reduction control motor; the garbage classification channel is formed by three partition plate structures, wherein the upper parts of the left partition plate structure and the right partition plate structure are movable partition plates, rotating shafts at the lower parts of the movable partition plates are respectively connected with a second speed reduction control motor and a third speed reduction control motor, and the middle partition plate is a fixed partition plate; the camera, the first speed reduction control motor and the second speed reduction control motor are respectively connected with the controller. The garbage sorting system has the advantages that through the specific structural design of the sorting platform and the garbage sorting channel, garbage throwing after image recognition and judgment is achieved, and high accuracy and working stability are guaranteed.

Description

Intelligent classification response garbage bin based on vision technique

Technical Field

The utility model relates to a garbage bin, waste classification field especially relate to an intelligent classification response garbage bin based on vision technique.

Background

According to the current general garbage classification rule in China, garbage is generally classified into four categories: kitchen waste, recoverable rubbish, harmful rubbish, other rubbish, current rubbish classification work is all realized through artifical basically. Because the rule of garbage classification is relatively complicated, the problems of low efficiency, inaccurate judgment and the like occur in manual classification, and a large amount of social resources are occupied.

Automatic identification and classification are carried out through intelligent equipment, and the method is a possible effective means for solving the problems, and is a Chinese patent: CN 201710558594-garbage classification and recycling method, garbage classification device, garbage classification and recycling system, and other prior art propose an image recognition method for garbage classification through image recognition, however, how to construct a feasible and easy-to-use garbage can device through the image recognition method is still blank at present; chinese patent CN 201510963930-trash can, trash classification method and system of trash can, although a system and device for performing trash classification based on image recognition are proposed, the technical details of how to realize classification by making clear the trash can structure are still ambiguous.

SUMMERY OF THE UTILITY MODEL

To the defect that prior art exists and not enough, the utility model discloses specifically adopt following technical scheme:

the utility model provides an intelligent classification response garbage bin based on vision technique which characterized in that includes: the garbage sorting device comprises a barrel body, a sorting platform arranged below a garbage input port of the barrel body, a garbage sorting channel arranged below the sorting platform and a sorting garbage barrel arranged at the bottom of the barrel body below the garbage sorting channel; a camera is arranged above the sorting platform; the middle shaft of the sorting platform is connected with a first speed reduction control motor; the garbage classification channel is composed of three partition plate structures, wherein the part, close to the upper part, of the left partition plate structure and the right partition plate structure is a movable partition plate, a rotating shaft at the lower part of the left movable partition plate is connected with a second speed reduction control motor, a rotating shaft at the lower part of the right movable partition plate is connected with a third speed reduction control motor, and the middle partition plate is a fixed partition plate clamped in a clamping groove in the inner side of the barrel body; the camera, the first speed reduction control motor and the second speed reduction control motor are respectively connected with the controller.

Preferably, a barrel cover and an infrared sensor which are connected with a fourth motor are arranged on the garbage throwing port.

Preferably, a display screen connected with the controller is arranged on the side of the barrel body.

Preferably, a vibration sensor connected with a controller is installed on the sorting platform.

Preferably, the side of the sorting platform is provided with a level detection sensor connected with a controller.

Preferably, the tail ends of the rotating shafts of the left movable partition plate and the right movable partition plate are respectively provided with an extension rod; two sides of the extension rod are respectively provided with a limit switch, and the limit switches are connected with the controller; a limiting part is arranged above the middle partition plate of the garbage classification channel.

The utility model discloses and preferred scheme provides one set of perfect intelligent classification garbage bin scheme based on image recognition technique, and it has realized that image recognition judges rubbish after and puts in through the concrete structural design of letter sorting platform and waste classification passageway, has guaranteed very high accuracy and job stabilization nature.

Drawings

The invention will be described in further detail with reference to the following drawings and detailed description:

FIG. 1 is an exploded view of the overall structure of the embodiment of the present invention;

FIG. 2 is a schematic diagram of an image recognition structure according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of a sorting platform according to an embodiment of the present invention;

FIG. 4 is a schematic structural diagram of a garbage classification passage according to an embodiment of the present invention;

FIG. 5 is a schematic structural view of a garbage input port according to an embodiment of the present invention;

fig. 6 is a schematic view of the overall structure of the embodiment of the present invention;

fig. 7 is a schematic view of the working flow of the embodiment of the present invention;

in the figure: 1-a sorting platform; 2-left fixed baffle; 3-a left movable partition plate; 4-a middle partition plate; 5-right movable partition board; 6-right fixed baffle; 7-sorting the garbage cans; 8-a camera; 9-barrel cover; 10-a display screen; 11-a first speed reduction control motor; 12-a second reduction control motor; 13-a third speed reduction control motor; 14-a limit stop; 15-an extension rod; 16-limit switches; 17-a fourth motor; 18-barrel head; 19-midbody.

Detailed Description

In order to make the features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail as follows:

as shown in fig. 1 to 6, each functional module of the solution of the present embodiment is disposed on the barrel body. As shown in fig. 1, in this embodiment, the main structure of the bin body can be divided into an upper bin body with a bin cover 9, a middle bracket with a camera 8 and a display screen 10, and a lower bin body with a garbage sorting passage and a sorting garbage bin 7.

As shown in fig. 1-6, wherein the middle bracket is installed above the sorting garbage can 7 and at the side of the garbage sorting passage; the letter sorting platform 1 of axis connection first speed reduction control motor 11 also fixes through the middle part support to lie in between camera 8 and the categorised passageway of rubbish in vertical direction, first speed reduction control motor 11 control letter sorting platform 1 swings left, right or resets to the horizontality.

The side part of the garbage classification channel is also limited by the middle support, and the specific classification function is realized by a left partition plate structure, a middle partition plate 4 structure and a right partition plate structure, wherein the three partition plate structures realize the separation of 4 channels, the left partition plate structure is divided into a left fixed partition plate 2 below and a left movable partition plate 3 above, and the right partition plate structure is divided into a right fixed partition plate 6 below and a right movable partition plate 5 above; in this embodiment, the length of the movable partition plate is required to be greater than the distance between the three partition plate structures, so that when the movable partition plate swings to the limit position, the corresponding channel can be closed, and a slope is formed at the same time, so that garbage can be accurately thrown into the target channel. The rotating shafts of the left movable partition plate 3 and the right movable partition plate 5 are respectively arranged at one end, namely the lower end, close to the left fixed partition plate 2 and the right fixed partition plate 6 and are respectively driven by a second speed reduction control motor 12 and a third speed reduction control motor 13 to swing, so that the selection control of a channel for finally throwing the garbage is completed; the middle clapboard 4 is a fixed clapboard clamped in a clamping groove at the inner side of the barrel body. As shown in fig. 4, in order to make the matching between the garbage sorting channel and the sorting platform 1 smoother, in the embodiment, a limiting member 14 is disposed below the sorting platform 1, and the limiting member 14 also has a function of limiting the rotation of the sorting platform 1, so as to prevent the rotation from exceeding the range of the control logic due to the rotation of more than 90 °. As shown in fig. 2 and 4, in this embodiment, in order to realize the limitation of the turnover of the movable partition board, the rotating shaft ends of the left movable partition board 3 and the right movable partition board 5 are respectively provided with an extension rod 15; limit switch 16 is installed respectively to the both sides of extension rod 15, and wherein the limit switch 16 of one side ensures that the activity baffle is not beyond the scope to the swing of incline direction, and the limit switch 16 of opposite side then sets up for vertical direction for can accurately get back to initial vertical state when guaranteeing the activity baffle return, consequently, when left activity baffle 3 or right activity baffle 5 swung or the return under the control of motor, extension rod 15 can trigger the limit switch 16 of one side wherein, and the swing stops this moment.

In order to improve the performance of the sorting platform 1, in the embodiment, a vibration sensor is mounted on the sorting platform 1 and used for detecting the throwing of the garbage, when the garbage is thrown to the sorting platform 1, the vibration sensor is triggered, the vibration sensor does not designate a special preferred model, and a conventional vibration sensor with a proper size and cost on the market is selected; meanwhile, a horizontal detection sensor is further arranged on the middle support of the side portion of the sorting platform 1 and used for detecting whether the sorting platform 1 restores to a horizontal state after garbage is thrown into the garbage classification channel. In the present embodiment, the level detection sensor may employ an existing hall sensor-based level detection circuit.

In terms of a control system, the present embodiment adopts a single chip microcomputer system or a PLC as a controller to implement the above operation logic, and the above sensors such as the level detection sensor and the hall sensor, the driving devices such as the first deceleration control motor 11, the second deceleration control motor 12, and the third deceleration control motor 13, the image recognition related devices such as the display screen 10 and the camera 8, and the limit switch 16 are electrically connected to the controller respectively.

As shown in fig. 1 and 5, in this embodiment, the upper barrel body covers the outer side of the middle support, the garbage input opening at the top of the upper barrel body is composed of a barrel head 18 and a middle body 19, a closed cavity is formed between the barrel head 18 and the middle body 19, a circuit board containing an infrared sensing device is arranged at the front end inside the cavity, a fourth motor 17 is arranged at the rear end inside the cavity, and the fourth motor 17 controls the automatic opening and closing of the barrel cover 9. The structure of the part can directly adopt similar Chinese patents: 201420122968.7, the prior art protocol.

In the present embodiment, the first deceleration-controlled motor 11, the second deceleration-controlled motor 12, the third deceleration-controlled motor 13, and the fourth motor 17 may be stepping motors, or may be ordinary motors with a decelerator.

As shown in fig. 7, the device provided in this embodiment is combined with a scheme of performing garbage classification through image recognition in the prior art, so as to achieve the object of the present invention. The software program of image recognition can be directly arranged in the controller, and the image processing and the garbage type recognition are realized through the operation of the microprocessor; considering the constraints of cost, data storage capacity and operation speed, the embodiment may also adopt the scheme provided by chinese patent CN201510963930, the work of image recognition and judgment is placed in a cloud server or a cloud platform, the controller may be wirelessly connected with the cloud computing platform through the internet of things communication modules such as 5G, and only the uploading of the image acquired by the camera 8 and the downloading of the recognition result need to be completed, and the recognition result is displayed through the display screen 10.

If the result of cloud platform discernment rubbish is "rubbish from cooking", then after the controller received information, left side activity baffle 3 was to the left bank swing, extension rod 15 stopped when touching slant limit switch 16, wait left side activity baffle swing back that targets in place left, sorting platform 1 was to the left bank swing, then rubbish can fall into in the middle of "rubbish from cooking" classification dustbin automatically under the action of gravity, after the action execution finishes, sorting platform 1 and left activity baffle 3 are controlled again to the controller and are succeeded the return, extension rod 15 touches and stops when vertical to limit switch 16. Similarly, when the recognition result is "recoverable article", the action is not executed to left movable partition plate 3, and sorting platform 1 swings to the left in an inclined manner, so that garbage can be ensured to automatically fall into the sorting garbage can with "recoverable article".

As a preferred scheme provided by this embodiment, the image recognition software specifically adopted by the method can also be trained and generated by the following detection method based on visual fine-grained image processing, and the method includes the following steps:

1. constructing a fine-grained neural network based on vision;

2. training the constructed neural network, comprising:

a. preprocessing the image, such as edge detection, filtering operation, binarization, image scaling, normalization and the like;

b. extracting features, and further reducing data dimensions such as lbp features, hog features and the like for the preprocessed image;

c. and (4) classifying or fitting by adopting a machine learning method, such as svm, bp, logistic regression and the like.

3. And detecting the trained vision-based fine-grained neural network.

The present invention is not limited to the above-mentioned preferred embodiments, and other various intelligent classification sensing trash cans based on visual technology can be obtained by anyone who can enlighten the present invention.

Claims

1. The utility model provides an intelligent classification response garbage bin based on vision technique which characterized in that includes: the garbage sorting device comprises a barrel body, a sorting platform arranged below a garbage input port of the barrel body, a garbage sorting channel arranged below the sorting platform and a sorting garbage barrel arranged at the bottom of the barrel body below the garbage sorting channel; a camera is arranged above the sorting platform; the middle shaft of the sorting platform is connected with a first speed reduction control motor; the garbage classification channel is composed of three partition plate structures, wherein the part, close to the upper part, of the left partition plate structure and the right partition plate structure is a movable partition plate, a rotating shaft at the lower part of the left movable partition plate is connected with a second speed reduction control motor, a rotating shaft at the lower part of the right movable partition plate is connected with a third speed reduction control motor, and the middle partition plate is a fixed partition plate clamped in a clamping groove in the inner side of the barrel body; the camera, the first speed reduction control motor and the second speed reduction control motor are respectively connected with the controller.

2. The vision technology-based intelligent classification induction trash can of claim 1, wherein: and a barrel cover and an infrared sensor which are connected with a fourth motor are arranged on the garbage throwing port.

3. The vision technology-based intelligent classification induction trash can of claim 1, wherein: the lateral part of the barrel body is provided with a display screen connected with the controller.

4. The vision technology-based intelligent classification induction trash can of claim 1, wherein: and the sorting platform is provided with a vibration sensor connected with the controller.

5. The vision technology-based intelligent classification induction trash can of claim 1, wherein: and a horizontal detection sensor connected with the controller is arranged on the side part of the sorting platform.

6. The vision technology-based intelligent classification induction trash can of claim 1, wherein: the tail ends of the rotating shafts of the left movable partition plate and the right movable partition plate are respectively provided with an extension rod; two sides of the extension rod are respectively provided with a limit switch, and the limit switches are connected with the controller; a limiting part is arranged above the middle partition plate of the garbage classification channel.