Intelligent garbage classification bin based on voice control and visual processing

Abstract

The utility model relates to and discloses an intelligent garbage classification can based on voice control and visual processing, which comprises a can body and four garbage cans, one side of the can body is provided with a garbage throwing port, a cross-shaped connecting frame is fixedly installed below the can body, a garbage identification table is installed on the cross-shaped connecting frame through a two-degree-of-freedom steering engine holder, and the garbage identification table is provided with a voice control module. A camera is arranged right above the garbage identification table; according to the intelligent garbage classification can based on voice control and visual processing, firstly, garbage images are collected through a camera, recognition and classification are conducted through the convolutional neural network image recognition technology, and an output shaft of a front side wall steering engine drives the tail end of a supporting rod to rotate downwards so that garbage can be thrown into a garbage recognition table below the tail end of a shovel plate; and according to the garbage recognition result, the two-degree-of-freedom steering engine holder can do two-degree-of-freedom movement in the horizontal direction and the vertical direction, garbage on the garbage recognition table is dumped into the corresponding garbage can, and compared with manual classification, garbage classification is more automatic, intelligent and convenient.

Description

Technical Field

[0001] The utility model relates to the technical field of garbage classification, in particular to an intelligent garbage classification barrel based on voice control and visual processing. Background Art

[0002] In recent years, in order to increase garbage treatment capacity and utilization rate, my country has classified domestic waste into categories, and through classified collection, utilization and disposal, achieved garbage reduction and resource utilization, thereby improving our living environment.

[0003] At present, the garbage in the community is mainly sorted by manual picking. The workers who pick and sort the garbage spend a long time in a place full of garbage and are easily harmed by harmful substances in the garbage. In view of this, this paper proposes a garbage sorting bin for automatic garbage sorting. Utility Model Content

[0004] The purpose of the utility model is to solve the problem that garbage is mainly sorted by manual picking in the prior art, and to propose an intelligent garbage sorting bin based on voice control and visual processing.

[0005] In order to achieve the above-mentioned purpose, the utility model adopts the following technical solutions:

[0006] An intelligent garbage sorting bin based on voice control and visual processing, comprising:

[0007] A barrel body, a garbage delivery port is arranged on one side of the barrel body, a cross connecting frame is fixedly installed at the bottom of the barrel body, a garbage identification platform is installed on the cross connecting frame via a two-degree-of-freedom steering gear platform, the cross connecting frame is fixedly connected to the two-degree-of-freedom steering gear platform, an output rotating frame in the two-degree-of-freedom steering gear platform is fixedly connected to the garbage identification platform, a camera is arranged directly above the garbage identification platform, and the camera is fixedly installed on the inner top surface of the barrel body;

[0008] Four trash cans are respectively arranged under the four notches of the cross connecting frame.

[0009] Preferably, a cross steel frame is fixedly installed below the cross connecting frame, a disc base is fixedly installed below the cross steel frame, and the four trash cans are respectively arranged in four notches of the cross steel frame on the disc base.

[0010] Preferably, an arc-shaped plate is fixedly installed below the garbage delivery port, a shovel plate is rotatably installed on one side of the arc-shaped plate, a fixed plate is fixedly installed below the arc-shaped plate, a front side wall servo is fixedly installed on one side of the fixed plate, the output shaft of the front side wall servo is fixedly connected to a support rod through a coupling, and the end of the support rod and the lower bottom surface of the shovel plate are in contact with each other.

[0011] Preferably, two DC drive motors are fixedly mounted on the lower bottom surface of the disc base, the output shafts of the DC drive motors are fixedly connected to wheels via couplings, and two universal wheels are also fixedly mounted on the lower bottom surface of the disc base.

[0012] Preferably, a power distribution box is fixedly installed on the top of the barrel body, a solar panel is fixedly installed on one side of the power distribution box, and ultraviolet disinfection light strips are fixedly installed on the inner walls of the four notches of the cross connecting frame.

[0013] Preferably, a voice control chip and a voice announcer are fixedly mounted on the periphery of the barrel body, and a lighting strip is fixedly mounted on the inner top surface of the barrel body.

[0014] Compared with the prior art, the utility model provides an intelligent garbage sorting bin based on voice control and visual processing, which has the following beneficial effects:

[0015] The intelligent garbage sorting bin based on voice control and visual processing divides four garbage bins into recyclable garbage bin, hazardous garbage bin, dry garbage bin and wet garbage bin respectively, then cooperates with the camera to collect images of garbage, and uses convolutional neural network image recognition technology to perform image recognition and classification. After the recognition is completed, the output shaft of the front side wall servo drives the end of the support rod to rotate downward, releases the support of the support rod to the shovel plate, and the end of the shovel plate rotates freely downward to put the garbage onto the garbage identification table below the end of the shovel plate. According to the result of garbage identification, the two-degree-of-freedom servo gimbal can make two-degree-of-freedom movements in the horizontal and vertical directions to dump the garbage on the garbage identification table into the corresponding garbage bin, thereby greatly improving the efficiency of garbage sorting, while also reducing the cost of garbage transportation, improving the convenience of garbage placement at night, reducing environmental pollution caused by garbage, and providing visualized garbage data for garbage sorting workers. Compared with traditional manual sorting, garbage sorting is more automated, intelligent and convenient. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the structure of an intelligent garbage sorting bin based on voice control and visual processing.

[0017] Figure 2 A partial cross-sectional view of an intelligent garbage sorting bin based on voice control and visual processing.

[0018] Figure 3 Schematic diagram of the connection between the front wall servo and the shovel in the intelligent garbage sorting bin based on voice control and visual processing Figure 1 .

[0019] Figure 4 Schematic diagram of the connection between the front wall servo and the shovel in the intelligent garbage sorting bin based on voice control and visual processing Figure 2 .

[0020] Figure 5 This is a schematic diagram of the structure of the garbage recognition station in the intelligent garbage sorting bin based on voice control and visual processing.

[0021] Figure 6 Schematic diagram of convolutional neural network image recognition technology for smart garbage sorting bins based on voice control and visual processing.

[0022] Figure 7 Flowchart of convolutional neural network deep learning for intelligent garbage sorting bins based on voice control and visual processing.

[0023] Figure 8 Schematic diagram of the garbage type identification process of the intelligent garbage sorting bin based on voice control and visual processing.

[0024] Fig. 9 Schematic diagram of the garbage sorting process of the intelligent garbage sorting bin based on voice control and visual processing.

[0025] Fig.10 Schematic diagram of the capacity detection process of intelligent garbage sorting bins based on voice control and visual processing.

[0026] Fig.11 Flowchart for voice-controlled movement and lighting of smart waste sorting bins based on voice control and vision processing.

[0027] Fig.12 Schematic diagram of the cloud data upload process for smart garbage sorting bins based on voice control and visual processing.

[0028] In the figure: 1. barrel; 2. trash can; 3. wheels; 4. DC drive motor; 5. universal wheel; 6. voice control chip; 7. voice announcer; 8. distribution box; 9. solar panel; 10. touch screen; 11. garbage inlet; 12. arc plate; 13. disc base; 14. cross steel frame; 15. cross connecting frame; 16. ultraviolet disinfection light strip; 17. mounting block; 18. garbage identification table; 19. two-degree-of-freedom servo gimbal; 20. front side wall servo; 21. support rod; 22. camera; 23. lighting light strip; 24. shovel plate; 25. fixing plate; 26. aggregation trough; 27. diversion port. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments.

[0030] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation on the present invention.

[0031] See attached Figure 1-12 , an intelligent garbage sorting bin based on voice control and visual processing, comprising a bin body 1, a garbage inlet 11 for putting garbage is arranged on one side of the bin body 1, an arc plate 12 is fixedly installed below the garbage inlet 11, a 16×10 cm touch screen 10 is arranged above the garbage inlet 11, the touch screen 10 is fixedly connected to the bin body 1, a shovel plate 24 is rotatably connected to the lower right side of the arc plate 12 through two hinges, a collection groove 26 for centrally placing garbage is arranged on the upper end surface of the shovel plate 24, a fixing plate 25 is fixedly connected below the arc plate 12, the bin body 1 is specifically made of acrylic material, a camera 22 is fixedly installed on the inner top surface of the bin body 1, a lighting strip 23 is arranged on the outer ring of the camera 22, and the lighting strip 23 is fixedly installed on the inner top surface of the bin body 1;

[0032] A front side wall servo 20 is fixedly installed on one side of the fixed plate 25, and a support rod 21 is fixedly installed on the output shaft of the front side wall servo 20 through a coupling. The end of the support rod 21 is in contact with the lower end surface of the shovel plate 24. After the front side wall servo 20 is energized, the output shaft of the front side wall servo 20 can drive the end of the support rod 21 to rotate in a vertical plane, thereby changing the support height of the end of the support rod 21. After the end of the support rod 21 rotates downward, the end of the shovel plate 24 is automatically rotated downward under the pressure of its own weight and the gravity of the load-bearing garbage above it. After the shovel plate 24 rotates downward and tilts, it is used to dump garbage. After the shovel plate 24 rotates upward and keeps level with the curved plate 12, it is used to carry garbage.

[0033] In this embodiment, a power distribution box 8 is fixedly installed above the barrel body 1, and a voice announcer 7 and a voice control chip 6 are also fixedly installed on the barrel body 1. The power distribution box 8 is fixedly installed with a main control board, a 32-way servo drive board, an STM32F103C8T6 chip, an STM32F429BIT6 chip, an ESP8266 chip, a 5v / 3.3V voltage regulator chip, a 12V lithium battery, an ESP8266WIFI module and a single-chip microcomputer, wherein the 32-way servo drive board It can be used for eight servos to operate. The 32-channel servo driver board is connected to the STM32F103C8T6 chip and the main control board through wires. The ESP8266 chip is connected to the STM32F429BIT6 chip through wires. The camera 22, the front side wall servo 20 and the voice announcer 7 are also connected to the 32-channel servo driver board, the STM32F103C8T6 chip, the 5v / 3.3V voltage regulator chip and the 12V lithium battery in the power distribution box 8 through wires.

[0034] In this embodiment, a cross connecting frame 15 is coaxially fixedly installed on the lower bottom surface of the barrel body 1, and a two-degree-of-freedom servo platform 19 is fixedly installed above the cross connecting frame 15. The two-degree-of-freedom servo platform 19 is connected to the 32-way servo drive board, STM32F103C8T6 chip, 5v / 3.3V voltage regulator chip and 12V lithium battery in the distribution box 8 through wires. The output rotating frame in the two-degree-of-freedom servo platform 19 is fixedly installed with a garbage identification platform 18. The garbage identification platform 18 is disc-shaped as a whole. The center part of the upper end surface of the garbage identification platform 18 is concave downward, which is used to concentrate the garbage on the center part of the garbage identification platform 18, and the outer periphery of the concave part of the center of the garbage identification platform 18 is connected to the outside with four guide ports 27 for guiding the flow of garbage. The four guide ports 27 of the garbage identification platform 18 correspond to the four notches of the cross connecting frame 15 respectively.

[0035] The two-degree-of-freedom servo gimbal 19 can perform two-degree-of-freedom motion in the horizontal and vertical directions, and has four preset states. The four states dump the garbage from the garbage identification station 18 into the four gaps corresponding to the cross connecting frame 15. When the garbage is thrown into the gathering groove 26 of the shovel plate 24 in the barrel body 1 from the garbage delivery port 11, the camera 22 first collects the image of the garbage and sends the image data to the main control board, and the convolutional neural network image recognition technology is used for image recognition.

[0036] In the present embodiment, a cross steel frame 14 is fixedly installed below the cross connecting frame 15, and a disc base 13 is fixedly installed below the cross steel frame 14. Four trash cans 2 are pushed and pulled on the disc base 13. The cross-section of the trash can 2 is a quarter circle. The four trash cans 2 are mainly divided into recyclable trash cans, hazardous trash cans, dry trash cans, and wet trash cans. Grips are provided on the outer sides of the trash cans 2. The four trash cans 2 are respectively pushed and pulled and embedded in the four notches of the cross steel frame 14, and the four trash cans 2 are respectively arranged directly below the four notches of the cross connecting frame 15, that is, the four notches of the cross steel frame 14 and the four notches of the cross connecting frame 15 are sequentially connected to each other.

[0037] In this embodiment, ultraviolet disinfection light strips 16 are fixedly installed on the inner walls of the four notches of the cross connecting frame 15, that is, the four ultraviolet disinfection light strips 16 are respectively arranged directly above the four garbage cans 2 in the cross steel frame 14, and mounting blocks 17 are fixedly installed at the four notches of the cross connecting frame 15, and ultrasonic ranging sensors and temperature and humidity sensors are fixedly installed on the mounting blocks 17, that is, four ultrasonic ranging sensors and four temperature and humidity sensors are respectively arranged directly above the four garbage cans 2, the ultrasonic ranging sensor, the temperature and humidity sensor, the touch display screen 10, the ultraviolet disinfection light strips 16 are connected to the main control board, the 32-way servo drive board, the STM32F103C8T6 chip, the 5v / 3.3V voltage regulator chip, the 12V lithium battery and the STM32F429BIT6 chip through wires, the ultrasonic ranging sensor is mainly used to detect the height of the garbage accumulation in the garbage can 2 directly below the main control board, and the temperature and humidity sensor.

[0038] In this embodiment, a solar panel 9 is fixedly installed on the upper end of the distribution box 8. The solar panel 9 is interconnected with a 12V lithium battery through a wire. The solar panel is a rectangular structure of 35cm×24cm, and the variable tilt angle range is 5°-60°. The solar panel 9 is mainly used to convert solar energy into electrical energy, and the 12V lithium battery stores electricity. Two mutually symmetrical DC drive motors 4 are fixedly installed on the lower bottom surface of the disc base 13. The output shafts of the two DC drive motors 4 are respectively fixedly connected to wheels 3 through couplings. Two universal wheels 5 are also fixedly installed on the lower bottom surface of the disc base 13. The two universal wheels 5 and the two wheels 3 are respectively arranged at the four corners below the disc base 13, and are used to support the disc base 13 and the objects above it;

[0039] The two DC drive motors 4 are respectively connected through the main control board, 32-way servo drive board, STM32F103C8T6 chip, 5v / 3.3V voltage regulator chip, 12V lithium battery and STM32F429BIT6 chip, and then connected to the voice control chip 6, and connected with the dual-way motor drive board through wires. The acrylic barrel body 1 and the trash can 2 are both 5mm thick acrylic plates, the shovel plate 24, the distribution box 8, and the garbage identification table 18 are all 2mm thick acrylic plates. The acrylic barrel body 1 has a radius of 20cm as a whole, and the support rod 21 is a cylindrical structure of 5cm×5cm×10cm. The trash can 2 is a fan column with a side length of 18.5cm, an arc length of 13.5cm, and a height of 28cm. The disc base 13 is a circular structure of 50cm×50cm, and the garbage identification table is a disc with a diameter of 26cm.

[0040] The utility model

[0041] 1) Identify the type of garbage; the garbage disposal personnel place the garbage on the shovel board 24 or drop it onto the garbage identification table 18 through the garbage disposal port 11, which is illuminated by the lighting strip 23, and then cooperates with the camera 22 to collect images of the garbage and send the image data to the main control board, and uses the convolutional neural network image recognition technology to perform image recognition and classification on the collected images;

[0042] The neural network includes an input layer, a convolution layer, a pooling layer, a fully connected layer, and an output layer. Compared with general neural networks, convolutional neural networks have a network structure in image understanding. Convolutional neural networks can better adapt to the structure of images and perform feature extraction and classification at the same time, so that feature extraction is helpful for feature classification and weight sharing can reduce the training parameters of the network, so that the neural network structure will become simpler and more adaptable. The main control board first sends the received image data to the input layer of the convolutional neural network, and then the input layer normalizes the image data before transmitting the image data to the convolution layer. Then the convolution layer extracts features from the image through six convolution kernels with different parameters, thereby obtaining six different feature maps, and finally transmitting the data of the feature maps to the pooling layer;

[0043] Since the image passes through the convolution layer, the amount of image data will increase, but in order to reduce the amount of data, it is pooled and the feature map is divided into small 3×3 areas, so that the maximum value in the small area is retained and the remaining values ​​are discarded. However, in order to improve the accuracy of image recognition and reduce the amount of image data, we perform two convolution and pooling operations on the convolutional neural network. After two convolution and pooling operations, the two-dimensional data of all feature maps will be converted into one-dimensional data, and then enter the fully connected layer. Through the neurons between the fully connected layer and the output layer, the image data is further processed. The output layer outputs a total of four results "1000", "0100", "0010" and "0001", which correspond to "dry garbage", "wet garbage", "recyclable garbage" and "hazardous garbage", respectively, and "dry garbage", "wet garbage", "recyclable garbage" and "hazardous garbage" correspond to the four gaps of the cross connecting frame 15, respectively.

[0044] 2) The garbage enters the corresponding garbage bin; after the above-mentioned garbage identification is completed, the output shaft of the front side wall servo 20 first drives the end of the support rod 21 to rotate downward, and after the shovel plate 24 releases the support of the support rod 21, the end of the shovel plate 24 freely rotates downward to put the garbage onto the garbage identification table 18 below the end of the shovel plate 24, and according to the result of the garbage identification, the two-degree-of-freedom servo pan-tilt 19 can make two-degree-of-freedom movements in the horizontal and vertical directions to dump the garbage on the garbage identification table 18 into the four gaps of the corresponding cross-connecting frame 15, and then pour it into the corresponding garbage bin 2;

[0045] The specific process is as follows: the single-chip microcomputer represents the four results output by the output layer with the numbers "1", "2", "3" and "4", and sends the numbers corresponding to the identified garbage types to the 32-way servo drive board through the serial port. When the 32-way servo drive board does not receive the data sent by the single-chip microcomputer, it will control the servo to remain in the initial position, that is, the front inner wall shovel plate 24 is in the open state, and the garbage identification platform 18 remains in the horizontal state;

[0046] When the 32-way servo driver board receives a signal from the single-chip microcomputer, the 32-way servo driver board will execute different programs according to different signals. If the data received by the 32-way servo driver board is "dry garbage", the front inner wall baffle is in an upward open state, and the garbage identification platform 18 is driven by the two-degree-of-freedom servo gimbal 19 to tilt 45° to the left, and the garbage enters the "dry garbage bin"; if the data received by the 32-way servo driver board is "wet garbage", the garbage identification platform 18 tilts 45° to the right, and the garbage enters the "wet garbage bin"; if the data received by the 32-way servo driver board is "recyclable garbage", the front inner wall baffle is in an upward open state, and the angle of the servo on the front side of the box becomes 90°, the garbage identification platform 18 tilts forward 45°, and the garbage enters the "recyclable garbage bin"; if the data received by the 32-way servo driver board is "hazardous garbage", the garbage identification platform 18 tilts backward 45°, and the garbage enters the "hazardous garbage bin".

[0047] 3) Identification of trash can capacity; four ultrasonic distance measuring sensors are used above the four trash cans 2 to monitor the distance between the trash in the four trash cans 2 and the ultrasonic distance measuring sensors in real time, and send the data to the single-chip microcomputer. The single-chip microcomputer calculates the received data and compares it with the actual height of the trash can 2. If the garbage storage volume exceeds 90%, the single-chip microcomputer will send the compared data to the background management center to notify the management personnel to clean the trash can 2 in time and turn on the ultraviolet disinfection light strip 16 for disinfection. At the same time, the capacity of each trash can 2 will be displayed on the front touch display screen 10.

[0048] 4) Voice control function; the specific process is as follows: the voice control chip 6 first recognizes the command word. If the forward or move command word is recognized, the voice control chip sends two identical low or high levels to the dual-way motor drive board. If the left or right turn command word is recognized, the voice control chip sends two opposite level signals to the dual-way motor drive board. If the lighting or turning off the light command word, the voice control chip sends a high level signal or a low level signal to the relay.

[0049] 5) Internet of Things function; the specific process is as follows: 4 temperature and humidity sensors and 4 ultrasonic ranging sensors obtain relevant data of the current environment and send them to the single-chip microcomputer, which then sends the data to the ESP8266WIFI module. The ESP8266WIFI module connects to the Alibaba Cloud server through the MQTT protocol and then reports the data. The real-time reported data can be viewed by configuring the user interface.

[0050] 6) Cleaning of the garbage bin; the specific process is as follows: the garbage cleaning personnel only need to pull out the garbage bin 2 to be cleaned, clean it, and then push the garbage bin 2 into the intelligent classification garbage classification bin to complete the cleaning of the garbage bin 2.

[0051] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes within the technical scope disclosed by the present invention according to the technical scheme and the utility model concept of the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An intelligent garbage sorting bin based on voice control and visual processing, characterized in that: include: A barrel body (1), a garbage delivery port (11) is provided on one side of the barrel body (1), a cross connecting frame (15) is fixedly installed below the barrel body (1), a garbage identification platform (18) is installed on the cross connecting frame (15) via a two-degree-of-freedom steering gear platform (19), the cross connecting frame (15) and the two-degree-of-freedom steering gear platform (19) are fixedly connected, an output rotating frame in the two-degree-of-freedom steering gear platform (19) and the garbage identification platform (18) are fixedly connected, a camera (22) is provided directly above the garbage identification platform (18), and the camera (22) is fixedly installed on the inner top surface of the barrel body (1); Four trash cans (2) are respectively arranged below the four notches of the cross connection frame (15).

2. The intelligent garbage classification bin based on voice control and visual processing according to claim 1 is characterized by: A cross steel frame (14) is fixedly installed below the cross connecting frame (15), a disc base (13) is fixedly installed below the cross steel frame (14), and the four garbage cans (2) are respectively arranged in four notches of the cross steel frame (14) on the disc base (13).

3. The intelligent garbage classification bin based on voice control and visual processing according to claim 1 is characterized by: An arc-shaped plate (12) is fixedly installed below the garbage delivery port (11), a shovel plate (24) is rotatably installed on one side of the arc-shaped plate (12), a fixed plate (25) is fixedly installed below the arc-shaped plate (12), a front side wall steering gear (20) is fixedly installed on one side of the fixed plate (25), an output shaft of the front side wall steering gear (20) is fixedly connected to a support rod (21) via a coupling, and the end of the support rod (21) and the lower bottom surface of the shovel plate (24) are in contact with each other.

4. The intelligent garbage classification bin based on voice control and visual processing according to claim 2 is characterized by: Two DC drive motors (4) are fixedly mounted on the lower bottom surface of the disc base (13); the output shaft of the DC drive motor (4) is fixedly connected to the wheel (3) via a coupling; and two universal wheels (5) are also fixedly mounted on the lower bottom surface of the disc base (13).

5. The intelligent garbage classification bin based on voice control and visual processing according to claim 1 is characterized by: A power distribution box (8) is fixedly mounted above the barrel body (1), a solar panel (9) is fixedly mounted on one side of the power distribution box (8), and ultraviolet disinfection light strips (16) are fixedly mounted on the inner walls of the four notches of the cross connecting frame (15).

6. The intelligent garbage classification bin based on voice control and visual processing according to claim 1 is characterized by: A voice control chip (6) and a voice announcer (7) are fixedly mounted on the periphery of the barrel body (1), and a lighting strip (23) is fixedly mounted on the inner top surface of the barrel body (1).

Images