Garbage classification treatment device

Abstract

The utility model relates to a garbage classification treatment device. The utility model relates to the technical field of garbage classification, and the device comprises a top cover which is provided with a garbage feeding port; the box body is arranged below the top cover, and the box body is used for supporting the top cover; the classified garbage cans are arranged at the bottom of the box body; the classification module is used for receiving to-be-classified garbage entering from the feeding port and sequentially throwing the garbage into the corresponding classification garbage cans according to the types of the garbage; and the control module is electrically connected with the classification module, and the control module is used for identifying the garbage to obtain the type of the garbage and controlling the classification module to sequentially throw the garbage into the corresponding classification garbage cans according to the type of the garbage. According to the garbage classification device, sequential classification work of various kinds of garbage thrown at a time is achieved, and the garbage classification accuracy and efficiency are greatly improved.

Description

Technical Field

[0001] The utility model relates to the technical field of garbage processing, in particular to a garbage classification processing device. Background Art

[0002] In recent years, with the continuous growth of population and rapid economic development, domestic waste has also continued to increase. Correct waste classification and recycling can help reduce environmental pollution, improve the utilization rate of renewable resources, and reduce energy loss. At present, most domestic waste treatment plants still rely on manual sorting to sort waste, which is not only not conducive to the physical and mental health of relevant practitioners, but also has low sorting efficiency and cannot handle the large amount of waste generated in modern social life in a timely manner. Garbage that is not cleaned up in time will not only affect the cleanliness and beauty of the environment, but also easily breed bacteria, produce harmful gases, and endanger human health. Therefore, it is particularly important to develop an automated, efficient, and accurate waste sorting technology.

[0003] Among the current garbage classification methods, some require manual labor to put the garbage into the garbage classification device one by one for classification, resulting in low garbage classification efficiency; another part can put in multiple types of garbage at the same time and use traditional visual methods to identify the image information of the garbage through machine vision technology. However, traditional visual methods are limited by the types of data sets and the scale of models, and their ability to classify garbage in actual situations is poor. They fail to classify garbage efficiently and accurately. If multiple types of garbage are identified and classified at the same time, mistakes are likely to occur, reducing the accuracy of garbage classification. Utility Model Content

[0004] Therefore, the technical problem to be solved by the present invention is to overcome the problems of low efficiency and low accuracy of garbage classification detection in the prior art. In order to solve the above technical problems, the present invention provides a garbage classification processing device, which can sequentially identify and classify multiple garbage input at one time, and realize efficient and accurate classification processing of multiple quantities of garbage.

[0005] In order to achieve the above-mentioned purpose, the main technical solutions adopted by the utility model include: a top cover, on which a garbage feeding port is arranged; a box body, which is arranged under the top cover, and the box body is used to support the top cover; a classified garbage bin, which is arranged at the bottom of the box body; a classification module, which is arranged in the box body and is between the top cover and the classified garbage bin; the classification module is used to receive the garbage to be classified entering from the feeding port, and put the garbage into the corresponding classified garbage bin in sequence according to the type of the garbage; a control module, which is electrically connected to the classification module, and the control module is used to identify the garbage to obtain the type of the garbage, and control the classification module to put the garbage into the corresponding classified garbage bin in sequence according to the type of the garbage.

[0006] In one embodiment of the utility model, the classification module includes: a conveying component, which is arranged below the feeding port; the conveying component is used to receive the garbage and convey the garbage in sequence according to the order of delivery; a sorting component, which is arranged between the conveying component and the classification garbage bin; the sorting component is used to receive the garbage conveyed by the conveying component in sequence, and put the garbage into the corresponding classification garbage bin according to the type of the garbage.

[0007] In one embodiment of the utility model, the conveying component includes: a first conveying component, a second conveying component and a second matrix optical fiber sensor; the first conveying component is arranged below the feeding port, the first conveying component is used to receive the garbage entering from the feeding port, and sequentially convey the garbage to the second conveying component; the second conveying component is arranged below the first conveying component and is located between the first conveying component and the sorting component, the second conveying component is used to sequentially receive the garbage conveyed by the first conveying component, and sequentially convey the garbage to the sorting component; the second matrix optical fiber sensor is arranged in the box body and is located between the first conveying component and the second conveying component; the second matrix optical fiber sensor is used to detect whether there is garbage passing between the first conveying component and the second conveying component; wherein, when only the second matrix optical fiber sensor detects the garbage, the first conveying component stops conveying.

[0008] In one embodiment of the utility model, the first conveying component includes: a primary conveyor belt, which is located below the feeding port and is horizontally arranged in the box body; a first preset distance is arranged between the end of the primary conveyor belt away from the feeding port and the inner wall of the box body, and the first preset distance is used to pass the garbage; a first matrix optical fiber sensor, which is arranged at the end of the primary conveyor belt away from the feeding port; the first matrix optical fiber sensor is used to detect whether the garbage is transmitted to the end of the primary conveyor belt away from the feeding port; wherein, when the first matrix optical fiber sensor does not detect the garbage, the conveying direction of the primary conveyor belt is from the end close to the feeding port to the end away from the feeding port; when only the first matrix optical fiber sensor detects the garbage, the primary conveyor belt reduces the conveying speed; when the first matrix optical fiber sensor and the second matrix optical fiber sensor detect the garbage at the same time, the conveying direction of the primary conveyor belt is changed from the end away from the feeding port to the end close to the feeding port until the first matrix optical fiber sensor fails to detect the garbage.

[0009] In one embodiment of the utility model, the second conveying component includes: a secondary conveyor belt, which is located below the primary conveyor belt, horizontally arranged in the box body, and arranged at the end of the primary conveyor belt away from the feeding port; a second preset distance is set between the end of the secondary conveyor belt close to the primary conveyor belt and the inner wall of the box body, and the second preset distance is used to pass the garbage; the secondary conveyor belt is used to receive the garbage conveyed by the primary conveyor belt, and convey the garbage to the sorting component in sequence; a third matrix optical fiber sensor, which is located at the end of the secondary conveyor belt close to the primary conveyor belt; the third matrix optical fiber sensor is used to detect whether the garbage is conveyed to the end of the secondary conveyor belt close to the primary conveyor belt; wherein, when the third matrix optical fiber sensor does not detect the garbage, the conveying direction of the secondary conveyor belt is from the end away from the primary conveyor belt to the end close to the primary conveyor belt; when the third matrix optical fiber sensor detects the garbage, the secondary conveyor belt reduces the conveying speed until the third matrix optical fiber sensor fails to detect the garbage.

[0010] In one embodiment of the utility model, the sorting component includes: a tray, which is located below the secondary conveyor belt and is arranged at one end of the secondary conveyor belt close to the primary conveyor belt; the tray is used to receive the garbage conveyed by the secondary conveyor belt; a servo, which is located between the tray and the classified trash can and is rotatably connected to the tray; the servo is used to drive the tray to put the garbage into the classified trash can; a fourth matrix fiber optic sensor, which is located in the box and is arranged between the tray and the secondary conveyor belt; the fourth matrix fiber optic sensor is used to detect whether the garbage passes between the secondary conveyor belt and the tray; wherein, when the fourth matrix fiber optic sensor is located in the box, the fourth matrix fiber optic sensor is used to detect whether the garbage passes between the secondary conveyor belt and the tray; When the four-matrix fiber optic sensor detects the garbage, the secondary conveyor belt stops conveying; when the third matrix fiber optic sensor and the fourth matrix fiber optic sensor detect the garbage at the same time, the secondary conveyor belt changes the conveying direction from one end close to the primary conveyor belt to one end far away from the primary conveyor belt until the third matrix fiber optic sensor fails to detect the garbage; a camera is arranged in the box; the camera is used to take an image of the garbage in the tray; wherein the control module is used to identify the type of the garbage according to the image, and control the sorting component to put the garbage into the corresponding classified garbage bin according to the type of the garbage.

[0011] In one embodiment of the utility model, a full load detection sensor is provided in the classified trash can; the full load detection sensor is used to detect whether the volume ratio of the garbage in the classified trash can to the classified trash can exceeds a preset ratio; wherein, when the full load detection sensor detects that the volume ratio of the garbage in the classified trash can to the classified trash can exceeds a preset ratio, the full load detection sensor sends a full load signal to the control module, and the control module issues a full load warning; a garbage compression device is also provided in the classified trash can, and the garbage compression device is used to compress the garbage in the classified trash can; the garbage compression device also includes a piercing rebound device, and the piercing rebound device is used to pierce the garbage.

[0012] The above technical solution of the utility model has the following beneficial effects compared with the prior art:

[0013] The garbage classification and processing device described in the utility model can classify multiple different types of garbage thrown in at one time in sequence through the classification module. Through the cooperation of the classification module and the control module, the function of accurately identifying and accurately classifying a large number of garbage in sequence is realized, avoiding the possibility of wrong classification when there are multiple types of garbage, improving the accuracy of garbage classification and sorting efficiency. In addition, the garbage detection model trained by the target data set collected independently in this application has an extremely high recognition accuracy, thereby greatly improving the efficiency and accuracy of garbage classification. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to make the content of the utility model easier to understand, the utility model is further described in detail according to the specific embodiments of the utility model in combination with the accompanying drawings, wherein

[0015] Figure 1 This is a schematic diagram of the overall structure of a garbage classification and processing device provided in a preferred embodiment of the utility model;

[0016] Figure 2 for Figure 1 A schematic diagram of the structure of the conveying assembly of the garbage sorting and processing device;

[0017] Figure 3 for Figure 1 A schematic diagram of the structure of the sorting component of the garbage sorting and processing device;

[0018] Figure 4 for Figure 1 A schematic structural diagram of a classified garbage bin of the garbage classification processing device;

[0019] Figure 5 for Figure 1 A schematic structural diagram of a garbage compression device of the garbage classification and processing device;

[0020] Figure 6 for Figure 1 A schematic structural diagram of the piercing rebound device of the garbage classification and processing device;

[0021] Figure 7 for Figure 1 A schematic diagram of the structure of the high-torque telescopic motor of the garbage sorting and processing device;

[0022] Figure 8 for Figure 1 A schematic side view of the structure of the garbage classification and processing device;

[0023] Fig. 9 for Figure 1 A schematic diagram of the structure of the single-chip microcomputer control module of the garbage classification and processing device;

[0024] Fig.10 for Figure 1 A specific overall workflow diagram of the control module of the garbage classification and processing device;

[0025] Fig.11 for Figure 1 A flow chart of establishing a detection model for the garbage classification and processing device.

[0026] Explanation of the reference numerals in the specification: 1. Top cover; 11. Feeding port; 12. Visualization module; 2. Box; 3. Conveying assembly; 311. Primary conveyor belt; 312. First matrix fiber optic sensor; 313. Horizontal bracket; 32. Second matrix fiber optic sensor; 331. Secondary conveyor belt; 332. Third matrix fiber optic sensor; 333. Fixed bracket; 4. Sorting assembly; 41. Pallet; 42. Servo; 43. Fourth matrix fiber optic sensor; 44. Support frame; 45. Camera; 5. Classification trash can; 51. Full load detection sensor; 52. Trash compression device; 521. Piercing rebound device; 53. High-torque telescopic motor. Specific embodiments

[0027] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments so that those skilled in the art can better understand the present invention and implement it, but the embodiments are not intended to limit the present invention.

[0028] Reference Figures 1 to 8As shown, a preferred embodiment of the utility model provides a garbage sorting and processing device, hereinafter referred to as "this device", which includes a top cover 1, a box body 2, a classification garbage bin 5 and a classification module; wherein, the top cover 1 is arranged on the top of the box body 2, the classification garbage bin 5 is arranged at the bottom of the box body 2, a garbage feeding port 11 is arranged on the top cover 1, and the classification module is arranged in the box body 2 and is located between the top cover 1 and the classification garbage bin 5. The classification module is used to receive the garbage to be classified entering from the feeding port 11, and put the garbage into the corresponding classification garbage bin 5 according to the type of garbage; in order to make this device fully automated, the present application is provided with a control module, which is electrically connected to the classification module, and the control module is used to identify the type of garbage obtained by the garbage, and control the classification module to put the garbage into the corresponding classification garbage bin 5 according to the type of garbage. In this embodiment, refer to Fig. 9 As shown, the control module adopts a single-chip microcomputer control module, which ensures the efficiency and accuracy of the entire garbage classification process; and according to China's current garbage classification and the "Domestic Waste Classification Mark" issued by the Ministry of Housing and Urban-Rural Development in 2019, this embodiment sets the number of classified garbage bins 5 to 4, namely, recyclable garbage bins, kitchen waste garbage bins, other garbage bins and hazardous waste bins.

[0029] In order to make the device more practical in actual use, in a specific embodiment, the device is provided with a visualization module 12, and the visualization module 12 is arranged on the top cover 1; the visualization module 12 includes a 5.8G radar sensing module and a display screen, and the 5.8G radar sensing module uses the Doppler effect to realize real-time monitoring of pedestrians. The output end of the 5.8G radar sensing module is electrically connected to the single-chip control module. When someone approaches, the 5.8G radar sensing module will send a high level to the single-chip control module, and then control the display screen to jump out of the video playback, and the display screen enters the garbage classification interface. At this time, the device enters the working state of garbage classification. When no one passes by, the display screen plays the garbage classification promotion video in a loop. At this time, the device is in a non-working standby state.

[0030] The classification module includes a transmission component 3 and a sorting component 4; the transmission component 3 is arranged below the feeding port 11, and the transmission component is used to receive the garbage to be classified entering from the feeding port 11, and transmit the garbage to the sorting component 4 in sequence; the sorting component 4 is located between the transmission component 3 and the classification garbage bin 5, and the sorting component 4 is used to receive the garbage transmitted by the transmission component in sequence, and put the garbage into the classification garbage bin 5 in sequence according to the type of garbage, thereby realizing the sequential and accurate classification of multiple garbage.

[0031] Specifically, the conveying component 3 includes a first conveying component, a second conveying component and a second matrix optical fiber sensor 32; the first conveying component is arranged below the feeding port 11, and the garbage to be classified enters the first conveying component from the feeding port 11, and the first conveying component conveys the garbage to the second conveying component in sequence; the second conveying component is located below the first conveying component, and the second conveying component receives the garbage conveyed by the first conveying component in sequence, and then conveys it to the sorting component 4 in sequence; the second matrix optical fiber sensor 32 is arranged between the first conveying component and the second conveying component. In this embodiment, the second matrix optical fiber sensor 32 is arranged on the inner wall of the box body 2, and the second matrix optical fiber sensor 32 is used to detect whether there is garbage passing between the first conveying component and the second conveying component. If there is garbage passing, it means that there is garbage being conveyed from the first conveying component to the second conveying component. At this time, in order to avoid multiple garbage falling from the first conveying component to the second conveying component at the same time, when only the second matrix optical fiber sensor 32 detects garbage, the first conveying component stops working and no longer conveys.

[0032] Among them, the first conveying component includes a first-stage conveyor and a first matrix optical fiber sensor 312; the first-stage conveyor belt 31 is horizontally arranged below the feeding port 11. In this embodiment, a horizontal bracket 313 is arranged in the box body 2, and the first-stage conveyor belt is arranged on the horizontal bracket 313. In particular, in order to make the garbage fall from the first-stage conveyor belt 31 to the second conveying member, the first-stage conveyor belt 31 is far away from the feeding port 11, that is, one end close to the second conveying component, and a certain distance is set between the first-stage conveyor belt 31 and the inner wall of the box body 2. This distance can make the garbage fall smoothly from the first-stage conveyor belt 31 to the second conveying component; in order to detect the position of the garbage conveyed in the first-stage conveyor belt 31, the present application is in the first-stage conveyor The first matrix optical fiber sensor 312 is provided at the end of the belt 31 away from the feeding port 11. The first matrix optical fiber sensor 312 is used to detect whether the garbage is transmitted to the end of the primary conveyor belt 31 away from the feeding port 11; when the first matrix optical fiber sensor 312 does not detect the garbage, the primary conveyor belt 31 operates normally, and the transmission direction at this time is from the end close to the feeding port 11 to the end away from the feeding port 11. In this embodiment, the transmission speed of the primary conveyor belt 31 is 100rpm when it operates normally; when the first matrix optical fiber sensor 312 detects the garbage, the first matrix optical fiber sensor 312 transmits the detection signal to the control module, and the control module controls The primary conveyor belt 31 slows down, but does not change the conveying direction of the primary conveyor belt 31. In this embodiment, the conveying speed of the primary conveyor belt 31 is slowed down to 25rpm. When the first matrix optical fiber sensor 312 detects a signal, it means that the garbage has approached the edge of the primary conveyor belt 31. The purpose of slowing down the speed at this time is to avoid multiple garbage from falling onto the second conveying component at the same time and affecting the accuracy of subsequent garbage classification. The garbage on the primary conveyor belt 31 away from the feeding port 11 continues to be conveyed by the primary conveyor belt 31 until it falls onto the second conveying component. At this time, the second matrix optical fiber sensor 32 located between the first conveying component and the second conveying component will capture the garbage. The second matrix fiber optic sensor 32 transmits the detection signal to the control module, and the control module controls the primary conveyor belt 31 to stop running. If the first matrix fiber optic sensor 312 can still detect that the garbage is within the detection range at this time, the control module will receive the detection signal of the first matrix fiber optic sensor 312 and the detection signal of the second matrix fiber optic sensor 32 at the same time, and the control module will control the transmission direction of the primary conveyor belt 31 to change to the opposite direction, that is, the transmission direction is changed from the end away from the feeding port 11 to the end close to the feeding port 11, until the first matrix fiber optic sensor 312 no longer detects the garbage, so as to prevent the garbage from continuing to fall into the second transmission component.

[0033] The second conveying component includes a secondary conveyor belt 331 and a third matrix optical fiber sensor 332; the secondary conveyor belt 331 is horizontally arranged below the primary conveyor belt 31. In the present embodiment, a fixed bracket 333 is arranged below the horizontal bracket 313 in the box body 2, and the secondary conveyor belt 331 is fixed in the fixed bracket 333; similarly, in order to make the garbage fall from the secondary conveyor belt 331 to the sorting component 4, a certain distance is also arranged between the end of the secondary conveyor belt 331 close to the primary conveyor belt 31 and the inner wall of the box body 2, and the distance can make the garbage fall smoothly from the secondary conveyor belt 331 to the sorting component 4; in order to detect the position of the garbage conveyed in the secondary conveyor belt 331, the present application is provided with a third matrix optical fiber sensor 332 at one end of the secondary conveyor belt 331 close to the primary conveyor belt 31, and the third matrix optical fiber sensor 332 is used To detect whether the garbage is transmitted to the end of the primary conveyor belt 31 close to the secondary conveyor belt 331; when the garbage falls from the primary conveyor belt 31 to the secondary conveyor belt 331 but is not captured by the third matrix optical fiber sensor, the transmission direction of the secondary conveyor belt 331 is from the direction away from the primary conveyor belt 31 to the direction close to the primary conveyor belt 31. In this embodiment, the transmission speed of the secondary transmission is 60rpm at this time; when the garbage is transmitted to the detection range of the third matrix optical fiber sensor 332, it means that the garbage is about to fall from the secondary conveyor belt 331 to the sorting component 4. In order to prevent multiple garbage from falling at the same time, when the control module obtains the detection signal of the third matrix optical fiber sensor 332, it controls the secondary conveyor belt 331 to reduce the speed. In this embodiment, the speed is reduced to 25rmp until garbage falls to the sorting component 4.

[0034] The sorting component 4 includes a tray 41, a steering gear 42, a fourth matrix optical fiber sensor 43 and a camera 45; the tray 41 is rotatably connected to the steering gear 42, and the combination of the tray 41 and the steering gear 42 is arranged between the classified garbage bin 5 and the secondary conveyor belt 331, and is located at one end of the secondary conveyor belt 331 close to the primary conveyor belt 31. In this embodiment, the steering gear 42 is arranged on the inner wall of the bottom of the box body 2 through the support frame 44; the tray 41 is used to receive garbage dropped from the secondary conveyor belt 331; the camera 45 is used to capture the image of the garbage on the tray 41 and transmit the garbage image to the control module, the control module obtains the garbage image data, and identifies the type of the garbage based on the garbage image data. In this embodiment, in order to obtain the most accurate image data, the camera 45 is arranged directly above the tray 41; the steering gear 42 is used to drive the tray 41 to put the garbage into the corresponding in the classified trash can 5; the fourth matrix fiber optic sensor 43 is arranged between the secondary conveyor belt 331 and the tray 41. In the present embodiment, the fourth matrix fiber optic sensor is arranged on the inner wall of the box body 2; the fourth matrix fiber optic sensor 43 is used to detect whether there is garbage falling from the secondary conveyor belt 331 to the tray 41. When the fourth matrix fiber optic sensor captures garbage falling, it transmits the detection signal to the control module, and the control module controls the secondary conveyor belt 331 to stop running immediately; if the control module obtains the detection signals of the third matrix fiber optic sensor 332 and the fourth matrix fiber optic sensor 43 at the same time, it means that there is still garbage that may fall from the secondary conveyor belt 331 to the tray 41. At this time, the control module controls the secondary conveyor belt 331 to change the transmission direction, and the transmission direction is changed to the direction away from the primary conveyor belt 31 until the third matrix fiber optic sensor 332 no longer detects garbage.

[0035] The present application provides a garbage compression device 52 driven by a large torque telescopic motor 53 in the recyclable garbage bin. In the present embodiment, the large torque telescopic motor 53 is provided at the bottom of one of the classified garbage bins 5, and the garbage compression device 52 is used to compress the recyclable garbage. The garbage compression device 52 also includes a piercing rebound device 521, which is used to pierce the sealed bottles and cans to facilitate exhaust and compression; the present application provides a full load detection sensor 51 in the classified garbage bin 5, which is used to detect whether the volume ratio of the garbage in the classified garbage bin 5 to the classified garbage bin 5 exceeds a preset ratio; in the present embodiment, a full load detection sensor 51 is provided at the position of 3 / 4 of the height of each garbage bin. When the stacking height of the garbage exceeds the position of 3 / 4 of the classified garbage bin 5, the distance measured by the full load detection sensor 51 changes, and the full load signal is sent to the single-chip control module, which controls the display screen to issue a full load reminder and display the full load information.

[0036] In actual operation, the entire workflow is as follows: the 5.8G radar sensing module uses the Doppler effect to detect pedestrians in real time and determine whether to throw in garbage: when no one is approaching, the display screen on the top cover 1 will play the garbage classification promotion video in a loop, and the device is in standby mode; when someone is approaching, the 5.8G radar sensing module will send a high level to the single-chip control module, and the control module will control the display screen to jump out of the video playback and enter the working state. When garbage is thrown in, it is assumed in this embodiment that three garbages are thrown in at one time, namely A, B, and C. In the working state, the primary conveyor belt 31 idles at a constant speed of 100rpm, and the secondary conveyor belt 331 does not move; garbage falls from the feeding port 11 to the primary conveyor belt 31, and is driven by the primary conveyor belt 31 to move toward one end close to the secondary conveyor belt 331 until it blocks the first matrix optical fiber sensor 312. After the first matrix optical fiber sensor 312 detects the garbage, it transmits the signal to the control module. After receiving the signal, the control module controls the speed of the primary conveyor belt 31 to slow down to 25rpm, so that the garbage moves slowly. When garbage falls from the primary conveyor belt 31 to the secondary conveyor belt 331, the primary conveyor belt 31 moves to the secondary conveyor belt 331. When the conveyor belt 331 is in motion, it will be captured by the second matrix optical fiber sensor 32 between the primary conveyor belt 31 and the secondary conveyor belt 331. At this time, the primary conveyor belt 31 stops running. If the first matrix optical fiber sensor 312 still detects that there is still garbage in the detection range, the control module will simultaneously receive the detection signals of the first matrix optical fiber sensor 312 and the second matrix optical fiber sensor 32, and the control module will control the primary conveyor belt 31 to run toward the end close to the feeding port 11 until the first matrix optical fiber sensor 312 no longer detects objects, so as to prevent subsequent garbage from continuing to fall. When garbage falls onto the secondary conveyor belt 331, even if there are multiple garbage falling at the same time due to objective factors, they will be dispersed due to the impact and transported one by one by the secondary conveyor belt 331 to the tray 41 for detection, which will not affect this workflow; the secondary conveyor belt 331 runs at a speed of 60rpm in the direction close to the primary conveyor belt 31 until the garbage blocks the third matrix fiber optic sensor 332, and the third matrix fiber optic sensor 332 transmits the detection signal to the control module, and the control module controls the secondary conveyor belt 331 to move slowly at a speed of 25rmp until garbage falls into the tray 41; after the fourth matrix fiber optic sensor 43 captures the falling signal of the garbage, it transmits the detection signal to the control module, and the control module controls the secondary conveyor belt 331 to stop running immediately. If the third matrix fiber optic sensor 332 detects that there are still objects in the detection range at this time, the control module will receive the detection signals of the third matrix fiber optic sensor 332 and the fourth matrix fiber optic sensor 43 at the same time, and the control module controls the secondary conveyor belt 331 to run in the direction away from the primary conveyor belt 31 until the third matrix fiber optic transmitter no longer detects the signal. At this point, the device has completed the task of sorting out a single piece of garbage from multiple pieces of garbage thrown in at the same time.After that, the conveying component 3 is on standby, waiting for the garbage to be put into the corresponding classified garbage bin 5. After the garbage is put into the corresponding classified garbage bin 5, the secondary conveyor belt 331 moves at a speed of 60rpm in the direction close to the primary conveyor belt 31 for a certain period of time. This time is the time it takes for the secondary conveyor belt 331 to move one circle at a speed of 60rpm. This operation is to verify the number of garbage dropped by the primary conveyor belt 31 last time. If only one garbage is dropped, the garbage will be classified at this time, and the secondary conveyor belt 331 is empty. The movement of the primary conveyor belt 31 should be continued to classify the subsequent garbage; if x (x>1) garbage is dropped at a time, it can be considered that the secondary conveyor belt 331 only classifies one garbage, and there are x-1 garbage left, and the secondary conveyor belt 331 needs to be continued. The transmission process is verified by whether the third matrix optical fiber sensor 332 recognizes garbage during the movement of the secondary conveyor belt 331. If the third matrix optical fiber sensor 332 does not detect garbage, it proves that only one piece of garbage fell from the primary conveyor belt 31 last time, and the secondary conveyor belt 331 is empty, then the secondary conveyor belt 331 stops, and the device returns to the initial state; otherwise, it proves that multiple pieces of garbage fell from the primary conveyor belt 31 last time, and there is still garbage on the secondary conveyor belt 331. The secondary conveyor belt 331 will execute according to the above logic cycle until the third matrix optical fiber sensor 332 does not detect the garbage signal in a certain cycle, and the program returns to the initial state. After the garbage falls into the tray 41, the camera 45 takes an image of the garbage, the control module obtains the image data and identifies the type of garbage based on the image data, and controls the servo 42 to put the garbage on the tray 41 into the corresponding classified garbage bin 5, so as to realize the sequential classification of garbage. The whole process is efficient and accurate.

[0037] refer to Fig.10 , Fig.11 As shown, the control module of the utility model is configured to have the above functions, specifically including:

[0038] S1, obtaining image data of the garbage to be classified, preprocessing the image data to obtain processed image data;

[0039] Among them, the preprocessing methods mainly include: scaling, brightness normalization, removing redundant parts, and edge filling. Scaling mainly scales the input image data to 640x640 pixels to meet the input requirements of the network structure. Since in the actual machine, the part captured by the camera 45 is the steering wheel and its surrounding area, in the step of removing redundant parts, the surrounding area is removed and filled with a solid color to prevent false recognition results generated by the surrounding area. In actual application scenarios, different garbage input may cause inconsistent exposure of the camera 45, which in turn leads to different brightness of the input photos. In brightness normalization, the brightness of the input image data is unified to a pre-set value. These preprocessing methods can effectively improve the robustness of the model after the image data is input into the model.

[0040] S2, establish a garbage detection model; specifically includes:

[0041] S21, establish an initial YOLOv5-serial model; the YOLOv5 model includes: a Backbone module, a Neck module, a Head module and a serial communication module; wherein the serial communication module is used to receive the output data of the Head module and transmit the output data for communication;

[0042] Among them, YOLOv5 (You Only Look Once version 5) is the latest version of the target detection algorithm YOLO (You Only Look Once). YOLOv5 is a target detection model used to identify and locate objects in videos or images;

[0043] The Backbone module uses the CSP-Darknet53 (feature extraction network) architecture, which is based on the Darknet53 (convolutional neural) network and applies the Cross Stage Partial (CSP) strategy to reduce the number of model parameters and enhance feature extraction efficiency; the CSP strategy effectively reduces redundant gradient information by segmenting the base feature map and merging it in the cross-stage hierarchy, thereby improving the real-time detection speed of the model. Among them, the Backbone module includes: convolution layer (Conv), batch normalization (BN) and SiLU activation function; the convolution layer (Conv) is the core of image processing in deep learning, especially in computer vision tasks. It extracts features from the input image through convolution operations, which involves a small, learnable filter (or convolution kernel) sliding on the input image to calculate the dot product between the filter and the local area of ​​the image. Each convolution kernel is responsible for extracting a specific feature in the image, such as edges, colors or textures. By stacking multiple convolution layers, the network is able to learn feature hierarchies from simple to complex. Batch Normalization (BN) is a technique used to speed up the training of deep networks. It also helps reduce sensitivity to initialization weights and helps prevent overfitting. It makes the network's learning process more stable by normalizing the layer inputs to a mean of 0 and a standard deviation of 1. The BN layer is usually placed after the convolutional layer and before the activation function. It also introduces two learnable parameters, the scale factor and the shift, to maintain the expressiveness of the network. This treatment can reduce the so-called internal covariate shift problem, which is the change in the distribution of network layer inputs during training.

[0044] The Neck module uses SPPF (Spatial Pyramid Pooling Fast) and improved PANet (Path Aggregation Network). SPPF significantly improves the receptive field without reducing the network speed by aggregating the input information and outputting a fixed-length output. The improvement of PANet introduces BottleNeckCSP (bottleneck cross-stage part) into its structure, which further optimizes the information flow and facilitates accurate pixel-level positioning. The Neck module includes Feature Pyramid Networks (FPN) and Path Aggregation Network (PAN). The Head module is used to locate and classify objects in the YOLOv5-based object detection system;

[0045] The head module consists of multiple submodules that work together to generate accurate object detection results. The head module includes three convolutional layers for predicting bounding box positions, scores, and object categories. In addition, this application uses SiLU (Sigmoid Linear Unit) and Sigmoid functions in terms of activation functions. SiLU is applied to the convolution operation of the hidden layer, while Sigmoid is used for the output layer. The SiLU activation function (Sigmoid Linear Unit) is an activation function used to introduce nonlinearity, so that the deep learning network can learn complex feature representations. The SiLU function is expressed as:

[0046] f(x) = x·σ(x);

[0047] Among them, σ(x) is the Sigmoid function, expressed as: f(x) is the output of the SiLU function for input x.

[0048] The main advantage of the SiLU function is its smooth non-monotonic nature, which allows negative values ​​to pass through, in contrast to the traditional ReLU (Rectified Linear Unit) activation function, which outputs all negative values ​​as zero. This property enables the SiLU activation function to provide better performance than ReLU in some cases, as it allows the network to capture information about negative input values, thereby improving the learning dynamics and the overall performance of the model.

[0049] The serial module is designed to output the output of the Head module through the hardware serial port. The module receives the detection result data from the Head module and converts it into a data frame in a specific format for transmission through a standard serial communication interface. This application has made a series of improvements based on YOLOv5, including replacing the Focus layer to reduce parameters and calculations, and eliminating grid sensitivity issues, thereby improving the bounding box detection capability. The serial module is designed to send information to the serial port of the hardware device running the model in order to achieve the purpose of sending classification results to the main control. First, the serial module obtains the recognition information obtained by the head module from the output layer of the model, including object identification, bounding box coordinates, category labels, confidence scores, etc. Then, the module encodes this information into a decimal data frame, converts it to hexadecimal, and outputs it to the serial port. The data frame structure of the module is as follows:

[0050] 1. Start bit: The data frame starts with a specific start bit, which is used to mark the beginning of the data frame. Here it is 1.

[0051] 2. Data payload:

[0052] a) Object Identifier: A unique identifier for each detected object. It is a 16-bit standard UUID.

[0053] b) Bounding box coordinates: Contains the bounding box coordinates of each detected object. Here, it is x coordinate, y coordinate, width, height. The above data are all 8-bit floating point numbers.

[0054] c) Category label: The category label of the detected object. It is a 4-bit integer.

[0055] d) Confidence score: The confidence score of the detection result. It is an 8-bit floating point number.

[0056] 3. Check bit: Check bit or checksum used to verify the integrity and correctness of the data frame during transmission. This data frame format uses CRC16 check.

[0057] 4. End bit: A line break character, marking the end of the data frame.

[0058] In terms of loss functions, this application adopts binary cross entropy (BCE) and complete cross union (CIoU) loss calculation methods, which are used to calculate the loss of detection category and objectness and position loss respectively. Through this method, this application effectively improves the accuracy and efficiency of the model in identifying garbage. Among them, the binary cross entropy (Binary Cross-Entropy, BCE) loss is a loss function for binary classification tasks. In target detection, it is used to calculate the difference between the predicted object category or object existence and the true label. Given the predicted value p (the probability of predicting the positive class) and the true label y (0 or 1), the BCE loss calculation formula is:

[0059] LBCE(p,y)=-[ylog(p)+(1-y)log(1-p)];

[0060] Complete intersection-union (Complete-IoU) loss is a bounding box regression loss function for object position estimation in target detection. It not only considers the overlapping area between the predicted box and the true box, but also includes the center point distance and aspect ratio between them, providing more accurate gradient information. The calculation formula of CIoU loss is:

[0061]

[0062] Among them, IoU is the intersection over union ratio between the predicted box and the true box; ρ(b,b^gt) is the Euclidean distance between the center b of the predicted box and the center b^gt of the true box; c is the diagonal length of the minimum closed area containing the predicted box and the true box; v measures the consistency of the aspect ratio and is used to penalize shape deviation; α is a trade-off coefficient used to balance the overlapping area and shape difference.

[0063] S22, obtaining a target data set, and optimizing the target data set using data enhancement technology to obtain an optimized data set; wherein the data enhancement technology method includes: random image rotation, random scaling, random color jitter, perspective transformation, and random cropping;

[0064] The target dataset used in this application includes four common categories and 24 subcategories of garbage in life, and collects data and annotates their locations. The target dataset includes 5126 training set images and 570 test set images. The target dataset is shown in Table 1.

[0065] Table 1

[0066]

[0067]

[0068] This application performs several data enhancements on the training set and test set images taken. These data enhancement methods include Rotation (random image rotation), Scaling (random scaling), Color Jittering (random color jittering), Perspective Transformation (perspective transformation), and Random Cropping (random cropping). These data enhancement methods take into account the characteristics of garbage image data in garbage classification and the actual practical scenarios in the actual implementation process, and can effectively improve the recognition rate of the model trained by the data set. Among them, since garbage may face different directions during garbage classification, it is necessary to apply Rotation in the data set to improve reliability; when applying this method, a random angle is selected for the input image, and then the new position of each pixel in the image after rotation is calculated according to the selected rotation angle. For blank areas that may appear after rotation, pure white is filled as a supplement; since the size of garbage in the image may be different during the actual classification process, the Scaling method needs to be used to improve the generalization of the model; when applying this method, a scaling factor is first randomly generated between [0.8,1.2], and then the image is scaled using the selected scaling factor. The scaling process uses a bilinear interpolation method. Finally, the image is adjusted back to 640*640 pixels. Since the camera 45 may generate noise during the actual classification process, possible stains may cause reflective or blurred images, and changes in lighting conditions may cause inconsistencies in the contrast, saturation, brightness, etc. of the image, this method needs to be used to improve the robustness of the model. The application of this method is divided into several independent image processing steps: 1. Add simulated black and white and color noise to the image. The noise is implemented using a Gaussian noise generator; 2. Randomly select several stain patterns, including circular spots (simulating water droplets or dust), irregular stripes (simulating fingerprints), and thick straight lines (simulating scratches) to generate the coordinate range to be processed. Then, apply the stain effect to the image to be processed. This step uses random non-uniform Gaussian blur, random brightness increase and decrease, superimpose a gray-black irregular piece and perform semi-transparent processing on it to simulate as many stain effects as possible. Since the actual camera 45 is not necessarily installed at a 90-degree vertical angle to the classified garbage, a suitable perspective transformation of the data set can fully improve the adaptability of the model in different deployment environments. This step first generates a random trapezoid as the quadrilateral after perspective transformation, then uses the projection perspective transformation function (getPerspectiveTransform) of perspective transformation (opencv) to obtain the perspective transformation matrix, and finally uses the perspective transformation function (warpPerspective) to apply the transformation to the image. Due to the actual characteristics of garbage, most garbage will be missing, so it is necessary to add random cropped blocks to the image to increase the model's ability to recognize incomplete targets.This step randomly generates several 10-50 pixel cropped square coordinate areas in the image, and then fills these areas with random solid colors to cover the target features. The data augmentation method can effectively improve the quality of the data set and greatly enhance the model's ability to cope with various situations.

[0069] S23, training the initial YOLOv5-serial model by minimizing the loss function according to the optimized data set, so that the initial YOLOv5-serial model converges, and using the converged initial YOLOv5-serial model as a garbage detection model.

[0070] S3, inputting the processed image data into the garbage detection model to obtain the garbage type;

[0071] S4, controlling the sorting component 4 to put the garbage into the corresponding classified garbage bin 5 according to the type of garbage;

[0072] S5, if a full load signal is obtained when the volume ratio of the garbage in the classified garbage bin 5 to the classified garbage bin 5 exceeds a preset ratio, a full load warning is issued.

[0073] In this embodiment, the parameters of each device are shown in Table 2, and the designed parameters are shown in Table 3.

[0074] Table 2

[0075]

[0076] Table 3

[0077] parameter value Epoch 200 batch-size 32 weights Yolov5s.pt Imgsize 640 optimizer SGD workers 8

[0078] In this implementation, the accuracy of garbage classification is close to 100%, which greatly increases the accuracy of garbage classification.

[0079] In the description of the present utility model, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the present utility model, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0080] In the present invention, unless otherwise clearly specified and limited, the terms "install", "connect", "connect", "fix" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0081] In the present utility model, unless otherwise clearly specified and limited, when a first feature is “on” or “below” a second feature, it may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, when a first feature is “above”, “above” or “above” a second feature, it may be that the first feature is directly above or obliquely above the second feature, or it may simply mean that the first feature is higher in level than the second feature. When a first feature is “below”, “below” or “below” a second feature, it may be that the first feature is directly below or obliquely below the second feature, or it may simply mean that the first feature is lower in level than the second feature.

[0082] In the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the utility model. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0083] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are illustrative and cannot be understood as limitations of the present invention. Ordinary technicians in the field can change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A garbage classification and processing device, characterized in that: include: A top cover, on which a garbage feeding port is provided; A box body, which is arranged under the top cover, and the box body is used to support the top cover; A classified trash can, which is arranged at the bottom of the box; A classification module is arranged in the box body and between the top cover and the classification trash can; the classification module is used to receive the garbage to be classified entering from the feeding port, and put the garbage into the corresponding classification trash can in sequence according to the type of the garbage; A control module, which is electrically connected to the classification module, and is used to identify the garbage to obtain the type of the garbage, and control the classification module to put the garbage into the corresponding classification garbage bin in sequence according to the type of the garbage; The classification module comprises: A conveying assembly is arranged below the feeding port; the conveying assembly is used to receive the garbage and convey the garbage in sequence according to the feeding order; A sorting component is arranged between the conveying component and the classified trash can; the sorting component is used to sequentially receive the garbage conveyed by the conveying component and place the garbage into the corresponding classified trash can according to the type of the garbage.

2. The garbage classification and processing device according to claim 1 is characterized in that: The transmission component comprises: a first transmission component, a second transmission component and a second matrix optical fiber sensor; The first conveying assembly is arranged below the feeding port, and is used to receive garbage entering from the feeding port and sequentially convey the garbage to the second conveying assembly; The second conveying component is arranged below the first conveying component and between the first conveying component and the sorting component, and the second conveying component is used to sequentially receive the garbage conveyed by the first conveying component and sequentially convey the garbage to the sorting component; The second matrix fiber optic sensor is arranged in the box and between the first conveying component and the second conveying component; the second matrix fiber optic sensor is used to detect whether there is garbage passing between the first conveying component and the second conveying component; wherein, when only the second matrix fiber optic sensor detects the garbage, the first conveying component stops conveying.

3. The garbage classification and processing device according to claim 2 is characterized in that: The first transmission component comprises: A primary conveyor belt, which is located below the feeding port and is horizontally arranged in the box; a first preset distance is set between an end of the primary conveyor belt away from the feeding port and the inner wall of the box, and the first preset distance is used for passing the garbage; A first matrix optical fiber sensor is arranged at one end of the primary conveyor belt away from the feeding port; the first matrix optical fiber sensor is used to detect whether the garbage is conveyed to the end of the primary conveyor belt away from the feeding port; Among them, when the first matrix optical fiber sensor fails to detect the garbage, the conveying direction of the primary conveyor belt is from an end close to the feeding port to an end away from the feeding port; when only the first matrix optical fiber sensor detects the garbage, the primary conveyor belt reduces the conveying speed; when the first matrix optical fiber sensor and the second matrix optical fiber sensor detect the garbage at the same time, the conveying direction of the primary conveyor belt is changed from an end away from the feeding port to an end close to the feeding port until the first matrix optical fiber sensor fails to detect the garbage.

4. The garbage classification and processing device according to claim 3 is characterized in that: The second transmission component comprises: A secondary conveyor belt is located below the primary conveyor belt, is horizontally arranged in the box, and is arranged at an end of the primary conveyor belt away from the feeding port; a second preset distance is set between an end of the secondary conveyor belt close to the primary conveyor belt and the inner wall of the box, and the second preset distance is used for passing the garbage; the secondary conveyor belt is used to receive the garbage conveyed by the primary conveyor belt, and convey the garbage to the sorting assembly in sequence; A third matrix optical fiber sensor is located at one end of the secondary conveyor belt close to the primary conveyor belt; the third matrix optical fiber sensor is used to detect whether the garbage is conveyed to the end of the secondary conveyor belt close to the primary conveyor belt; Among them, when the third matrix optical fiber sensor does not detect the garbage, the conveying direction of the secondary conveyor belt is from one end away from the primary conveyor belt to one end close to the primary conveyor belt; when the third matrix optical fiber sensor detects the garbage, the secondary conveyor belt reduces the conveying speed until the third matrix optical fiber sensor fails to detect the garbage.

5. The garbage classification and processing device according to claim 4 is characterized in that: The sorting component comprises: A tray, which is located below the secondary conveyor belt and is arranged at one end of the secondary conveyor belt close to the primary conveyor belt; the tray is used to receive the garbage conveyed by the secondary conveyor belt; A steering gear is located between the tray and the classified trash can and is rotatably connected to the tray; the steering gear is used to drive the tray to put the trash into the classified trash can; a fourth matrix optical fiber sensor, which is located in the box and is arranged between the tray and the secondary conveyor belt; the fourth matrix optical fiber sensor is used to detect whether the garbage passes between the secondary conveyor belt and the tray; wherein, when the fourth matrix optical fiber sensor detects the garbage, the secondary conveyor belt stops conveying; when the third matrix optical fiber sensor and the fourth matrix optical fiber sensor detect the garbage at the same time, the secondary conveyor belt changes the conveying direction from one end close to the primary conveyor belt to one end away from the primary conveyor belt until the third matrix optical fiber sensor fails to detect the garbage; A camera is arranged in the box; the camera is used to capture images of the garbage in the tray; wherein the control module is used to identify the type of the garbage based on the image, and control the sorting component to place the garbage into the corresponding classified garbage bin based on the type of the garbage.

6. The garbage classification and processing device according to claim 1, characterized in that: A full load detection sensor is provided in the classified trash can; the full load detection sensor is used to detect whether the volume ratio of the garbage in the classified trash can to the classified trash can exceeds a preset ratio; wherein, when the full load detection sensor detects that the volume ratio of the garbage in the classified trash can to the classified trash can exceeds a preset ratio, the full load detection sensor sends a full load signal to the control module, and the control module issues a full load warning; The classified trash can is also provided with a garbage compression device, which is used to compress the garbage in the classified trash can; the garbage compression device also includes a piercing rebound device, which is used to pierce the garbage.

Images