A smart sanitation waste sorting monitoring and feedback system
By using an intelligent sanitation waste sorting monitoring and feedback system, which integrates image recognition and weight data, accurate waste sorting and disposal can be achieved. This solves the waste sorting problem caused by residents' failure to sort their waste, improves sorting accuracy, and reduces processing costs.
Patent Information
- Application Number
- CN202511056125.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-30
- Publication Date
- 2026-11-17
- Estimated Expiration
- 2045-07-30
AI Technical Summary
Currently, some residents in certain communities do not sort their garbage when disposing of it, which increases the difficulty of garbage sorting, wastes manpower and resources, and causes environmental pollution.
The system employs an intelligent sanitation waste sorting monitoring and feedback system, which includes waste bins with electric lids, information collection devices, softness collection devices, waste detection modules inside the bins, data analysis modules, and control modules. Through image recognition and weight data fusion, it achieves accurate waste sorting and disposal.
It improves the accuracy of waste sorting, reduces mixed waste, lowers the need for manual sorting, reduces back-end processing costs, and adapts to changes in waste characteristics in different regions and seasons through its self-evolution capabilities.
Smart Images

Figure CN120698103B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of waste treatment technology, and in particular to an intelligent sanitation waste sorting monitoring and feedback system. Background Technology
[0002] Waste sorting is the foundation for the operation of waste disposal facilities. Implementing household waste sorting can effectively improve the urban and rural environment and promote resource recycling. Based on the scientific and rational sorting of household waste, a corresponding supporting system for household waste sorting should be developed, and terminal disposal facilities should be improved to ensure seamless integration of sorting, collection, recycling, utilization, and treatment facilities.
[0003] Chinese Patent Publication No. CN105772403A discloses a waste sorting and detection system, including a belt conveyor, a waste claw, a camera device, and a computer processing system. The belt conveyor is used to transport waste bags containing waste information. Each waste bag is labeled with waste type information and an identification code containing the waste owner's information. The waste claw, controlled by a drive mechanism, places the waste bag onto the belt conveyor. The camera device collects the waste type information and identification code of the waste bags on the belt conveyor and feeds this data back to the computer processing system. The computer processing system calculates the waste disposal fee based on the waste type, quantity, and price of the waste bags and provides this information back to the waste owner. This waste sorting and detection system automates waste sorting, detection, and fee management, significantly improving waste disposal efficiency and bringing social benefits such as a cleaner and more comfortable living environment.
[0004] It is evident that the existing technology has the following problems: at present, some residents in some communities do not sort their garbage in advance, or they put sorted garbage into other types of garbage bins, which increases the difficulty of subsequent garbage sorting, wastes a lot of manpower and resources, and causes environmental pollution. Summary of the Invention
[0005] To address this issue, the present invention provides an intelligent sanitation waste sorting monitoring and feedback system to overcome the problem in the existing technology that, at present, some residents in some communities do not sort their waste in advance, or put sorted waste into other types of waste bins, which increases the difficulty of subsequent waste sorting and wastes a lot of manpower and resources.
[0006] To achieve the above objectives, the present invention provides an intelligent sanitation waste sorting monitoring and feedback system, comprising, A trash can for holding trash, equipped with a controllable electric lid; including a trash can assembly, which consists of several different types of trash cans placed side by side, each trash can being equipped with an individual switch that can be controlled to open and close. Garbage bag storage bins are used to hold garbage sorting bags when sorting garbage; Information collection device, used to collect information on disposed waste and user input; A softness collection device is installed at the lower part of the information collection device to collect the softness of the garbage bag. The bin-in-bin waste detection module is used to detect the proportion of corresponding waste types in each type of waste bin to determine the accuracy of waste sorting; The data analysis module is connected to the information acquisition device and the garbage detection module in the bin, respectively. It determines the type of garbage based on the garbage weight information and garbage image information to obtain a first determination result. It compares the first determination result with the input information to obtain a garbage classification comparison result. Based on the garbage classification comparison result, it determines whether garbage needs to be separated, or adjusts the weight of garbage weight information and garbage image information to the first determination result. The waste judgment module is used to determine whether the analysis results of the data analysis module are correct based on the data from the waste detection module in the bin. A control module, which is connected to the garbage bin assembly, controls the opening of the electric lid of the garbage bin assembly based on the analysis data from the data analysis module; The waste information includes waste weight information and waste image information; the input information is the type of waste to be disposed of actively selected by the user.
[0007] Furthermore, the information collection device includes, The support platform is used to place the waste to be tested; A camera, mounted on top of the support platform, is used to capture images of the waste placed on the support platform; An interactive device, located in the middle of the testing platform, is used to facilitate interaction between the system and the user. The interactive device includes an input keyboard and a display screen. The input keyboard is used to collect information input by the user, and the display screen is used to display feedback information.
[0008] Furthermore, the information acquisition device identifies the garbage image formed by the garbage image information, extracts the feature elements of the garbage to obtain the actual feature elements, and the data analysis module compares the actual feature elements with the standard feature elements to obtain the feature element matching degree. Based on the feature element matching degree, the garbage type corresponding to the garbage image is determined. The featured elements include the number of prominent locations and the softness of the waste.
[0009] Furthermore, the data analysis module obtains the actual waste classification matching degree based on the overall waste type information and the input information, compares the actual waste classification matching degree with the standard waste classification matching degree to obtain the waste classification matching degree comparison result, and determines whether the waste needs to be sorted or whether to control the opening of the electric bin lid based on the waste classification matching degree comparison result.
[0010] Furthermore, the data analysis module includes, The first data analysis unit is used to make an initial determination of the material of the waste based on the number of prominent points in the waste image; The second data analysis unit is used to correct the initial judgment results based on the softness of the waste; The third data analysis unit is used to determine whether to make a second determination of the type of waste based on the comparison between the actual waste weight and the waste weight within the standard waste weight range.
[0011] Furthermore, the waste determination module determines whether the waste should be disposed of or whether the standard softness index range should be adjusted based on the secondary determination result of the third data analysis unit.
[0012] Furthermore, the control module controls the electric box cover. If the actual weight of the waste is less than the minimum standard weight of the waste, the corresponding standard softness index range will be expanded based on the difference between the minimum standard weight of the waste and the actual weight of the waste. If the actual weight of the waste is within the standard waste weight range, the waste is determined to be Class I waste, and the control module controls the electric lid to open the electric lid for waste disposal. If the actual weight of the waste is greater than the maximum standard weight of the waste, the corresponding standard softness index range will be reduced based on the difference between the actual weight of the waste and the maximum standard weight of the waste.
[0013] Furthermore, the bin-in-bin waste detection module performs secondary waste data collection on the waste inside the waste bin, generates bin-in-bin waste data, and determines the type of waste inside the bin based on the bin-in-bin waste data.
[0014] Furthermore, the judgment module compares the types of waste in the bin with the secondary judgment result to obtain the accuracy of waste classification in the bin. Based on the comparison result of the accuracy of waste classification in the bin with the accuracy of waste classification within the standard waste classification accuracy range, it determines whether to adjust the weight weight of waste weight information on the overall weight of waste, or to increase consumption points for users.
[0015] Furthermore, the determination module makes its judgment based on the comparison results of the accuracy of garbage classification. If the accuracy of waste sorting in the bin is less than the minimum value of the standard waste sorting accuracy range, then the weight is increased based on the difference between the minimum value of the standard waste sorting accuracy range and the accuracy of waste sorting in the bin. If the accuracy of waste sorting in the bin is within the standard waste sorting accuracy range, the original weight remains unchanged. If the accuracy of waste sorting in the bin exceeds the maximum value of the standard waste sorting accuracy range, the user will receive additional points.
[0016] Compared to existing technologies, this system achieves precise waste sorting by using waste bins with electrically operated lids, preventing the mixing of different types of waste. Each bin operates independently, opening only when the corresponding type of waste is detected, improving the spatial resolution of waste softness detection and adapting to the detection needs of waste with different shapes. Standardized detection contact surfaces ensure consistent pressure testing of waste of different volumes, and real-time monitoring of waste composition within the bins provides data support for classification accuracy assessment and system optimization. By fusing image recognition and weight data, the system calculates the proportion of each type of waste, compares it with corresponding standard ranges to generate an accuracy score, which is used to adjust model weights or user incentives. By comparing user input information with actual detection data, the system automatically adjusts the weight ratio of weight and image information.
[0017] Furthermore, a fixed detection position is provided for the waste, ensuring consistency in camera shooting angle and pressure detection for softness. This avoids image distortion or pressure detection errors caused by waste misalignment, improving image recognition accuracy. The platform's size is adaptable to common garbage bags, meeting specific scenario requirements. Combined with image recognition algorithms, it effectively improves the accuracy of material identification for recyclables, especially for regularly shaped waste (such as cardboard boxes and aluminum cans). Comparison between user input information and actual detection results generates classification error data, which is used to adjust model weights. The cooperation between the platform and the camera standardizes the position and angle of waste detection, providing a high-quality data foundation for AI recognition.
[0018] Furthermore, by using dual threshold determination of the length and angle of prominent points, the "shape analysis" in traditional image recognition is transformed into quantifiable geometric features, solving the problem of distinguishing between soft and hard materials. Deeply integrating visual features (prominent points) and physical features (softness) forms a dual "visual-tactile" detection dimension, improving the coverage of common household waste sorting scenarios. Automatically adjusting feature parameters based on historical data enables the system to have self-learning capabilities, quickly adapting to changes in waste characteristics across different regions and seasons. This feature extraction and analysis method, through improved hardware precision and algorithmic innovation, significantly enhances the intelligence level of the waste sorting system, providing key technical support for achieving a closed-loop chain of "accurate identification - automatic sorting - efficient processing."
[0019] Furthermore, the matching accuracy determination combines real-time data with historical experience to avoid misjudgments or overly strict standards caused by a "one-size-fits-all" approach. From user declaration, system detection, lid control to sorting feedback, a complete behavioral guidance chain is formed, achieving continuous improvement in classification accuracy. By reducing mixed waste and lowering the need for manual sorting, backend processing costs are significantly reduced, while extending equipment life (due to reduced damage to the lid from hard foreign objects).
[0020] Furthermore, a three-dimensional classification system based on "highlighting key points (visual perception), softness (tactile sensation), and weight (density)" is constructed to overcome the limitations of single-dimensional classification and achieve three-dimensional recognition of material, shape, and density. Weight-driven adjustments to the softness range enable the system to intelligently accommodate "lightweight / high-density" variants of waste. Data accumulation throughout the entire process, from initial assessment and correction to secondary assessment, supports continuous iteration of the AI model, resulting in exponential growth in classification capabilities over time. Through precise inter-module collaboration and data-driven design, this system not only significantly improves waste classification accuracy but also constructs a self-evolving intelligent classification system, providing a feasible technological paradigm for intelligent urban sanitation.
[0021] Furthermore, by combining weight deviation with softness compliance rate, misjudgments caused by single weight anomalies are avoided. The weight ratio of weight, softness, and image is adjusted in real time based on the secondary detection results, enabling the system to have self-evolution capabilities.
[0022] Furthermore, for mixed waste with abnormal weight, increasing the weight factor improves the system's sensitivity to identifying "weight meets standards but composition is abnormal," reducing the false positive rate. Points can be redeemed for garbage bags or community services, creating a positive cycle of "accurate disposal - earning rewards." Through a non-linear mapping formula between accuracy and weight, precise correction of classification errors is achieved, avoiding system oscillations caused by "one-size-fits-all" adjustments. The points mechanism not only increases participation but also feeds back user behavior data to optimize the model, forming a closed loop of "human classification - system learning - more accurate identification." The standard accuracy range and weight adjustment strategy dynamically evolve with historical data and processing needs, ensuring high robustness of the system across different scenarios. Attached Figure Description
[0023] Figure 1 This is a flowchart of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 2 This is a flowchart illustrating the judgment process of the data analysis module of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 3 This is a flowchart of the waste disposal determination process of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 4 This is a flowchart illustrating the weight changes of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment. Detailed Implementation
[0024] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.
[0025] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.
[0026] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.
[0027] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.
[0028] Please see Figure 1-4 As shown, Figure 1 This is a flowchart of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 2 This is a flowchart illustrating the judgment process of the data analysis module of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 3 This is a flowchart of the waste disposal determination process of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment; Figure 4 This is a flowchart illustrating the weight changes of the intelligent sanitation waste sorting monitoring and feedback system described in this embodiment.
[0029] This embodiment provides an intelligent sanitation waste sorting monitoring and feedback system, including, A trash can for holding trash, equipped with a controllable electric lid; including a trash can assembly, which consists of several different types of trash cans placed side by side, each trash can being equipped with an individual switch that can be controlled to open and close. Garbage bag storage bins are used to hold garbage sorting bags when sorting garbage; Information collection device, used to collect information on disposed waste and user input; A softness collection device, located at the lower part of the information collection device, is used to collect the softness of the garbage bag. It includes two sets of opposing vertically arranged combined upright plates. Each combined upright plate is equipped with a displacement sensor and several pressure sensors. The displacement sensor is used to collect the distance the garbage bag retracts when pressure is applied to the garbage bag, and the pressure sensor is used to collect the pressure value of the garbage bag rebounding when pressure is applied to the garbage bag. The bin-in-bin waste detection module determines the accuracy of waste sorting by detecting the proportion of corresponding waste types in each type of waste bin. The data analysis module is connected to the information acquisition device and the garbage detection module in the bin, respectively. It determines the type of garbage based on the garbage weight information and garbage image information to obtain a first determination result. It compares the first determination result with the input information to obtain a garbage classification comparison result. Based on the garbage classification comparison result, it determines whether garbage needs to be separated, or adjusts the weight of garbage weight information and garbage image information to the first determination result. The waste judgment module is used to determine whether the analysis results of the data analysis module are correct based on the data from the waste detection module in the bin. A control module, which is connected to the garbage bin assembly, controls the opening of the electric lid of the garbage bin assembly based on the analysis data from the data analysis module; The waste information includes waste weight information and waste image information; the input information is the type of waste to be disposed of actively selected by the user.
[0030] The combined vertical plate is a 20cm×30cm rectangular plane, and the pressure sensors are arranged in a 6×8 array (48 in total), with a distribution density of 1 sensor per 12.5cm. 2 .
[0031] Precise waste sorting is achieved by using waste bins with electrically operated lids, preventing the mixing of different types of waste. Each bin has an independent switch, opening only when the corresponding type of waste is detected, improving the spatial resolution of waste softness detection and adapting to the detection needs of waste with different shapes. Standardized detection contact surfaces ensure consistent pressure testing of waste of different volumes, and real-time monitoring of waste composition within the bins provides data support for classification accuracy assessment and system optimization. By fusing image recognition and weight data, the proportion of each type of waste is calculated, and an accuracy score is generated by comparing it with corresponding standard ranges, which is used to adjust model weights or user incentives. The weight ratio of weight and image information is automatically adjusted by comparing user input information with actual detection data.
[0032] Specifically, the information collection device includes, The support platform is used to place the waste to be tested; A camera, which is mounted on top of the support platform, is used to capture image information of the garbage placed on the support platform; An interactive device, located in the middle of the testing platform, is used to facilitate interaction between the system and the user. The interactive device includes an input keyboard and a display screen. The input keyboard is used to collect information input by the user, and the display screen is used to display feedback information.
[0033] The system provides a fixed detection location for waste, ensuring consistency in camera shooting angles and pressure detection for softness. This avoids image distortion or pressure detection errors caused by waste misalignment, improving image recognition accuracy. The platform's size is adaptable to common waste bags, meeting specific scenario requirements. Combined with image recognition algorithms, it effectively improves the accuracy of material identification for recyclables, especially for regularly shaped waste (such as cardboard boxes and aluminum cans). Comparison between user input information and actual detection results generates classification error data, which is used to adjust model weights. The cooperation between the platform and the camera standardizes the location and angle of waste detection, providing a high-quality data foundation for AI recognition.
[0034] Specifically, the information acquisition device identifies the garbage image formed by the garbage image information, extracts the feature elements of the garbage to obtain the actual feature elements, the data analysis module compares the actual feature elements with the standard feature elements to obtain the feature element matching degree, and determines the type of garbage corresponding to the garbage image based on the feature element matching degree. The featured elements include the number of prominent locations and the softness of the waste.
[0035] In this embodiment, the user places the garbage bag on the carrying platform. At this time, the camera takes a picture of the garbage bag to obtain a garbage image. The camera uses a high-resolution camera. The data analysis module analyzes the number of protruding points on the garbage image. The protruding points are determined based on the length and angle of the protruding points. A protruding point is defined as one whose length is greater than a preset value of the maximum length of the garbage image and whose angle is less than a preset angle. The protruding points are calculated. For example, in this embodiment, the maximum length of the detected garbage image is 50cm, the preset length value is set to 0.2, and the preset angle is set to 120°. The preset length value and the preset angle are determined based on historical data. Protruding points with a length greater than 10cm are marked. The angle of the marked protruding points is detected. If the angle of the marked protruding point is less than the preset angle, the point is determined to be a protruding point and recorded as one protruding point.
[0036] By employing dual threshold determination based on the length and angle of prominent points, the "shape analysis" in traditional image recognition is transformed into quantifiable geometric features, solving the challenge of distinguishing between soft and hard materials. Deeply fusing visual features (prominent points) and physical features (softness) creates a dual "visual-tactile" detection dimension, improving coverage in common household waste sorting scenarios. Automatically adjusting feature parameters based on historical data enables the system to self-learn and quickly adapt to changes in waste characteristics across different regions and seasons. This feature extraction and analysis method, through improved hardware precision and algorithmic innovation, significantly enhances the intelligence level of the waste sorting system, providing crucial technical support for achieving a closed-loop process of "accurate identification - automatic sorting - efficient processing."
[0037] Specifically, the data analysis module obtains the actual waste classification matching degree based on the overall waste type information and the input information, compares the actual waste classification matching degree with the standard waste classification matching degree to obtain the waste classification matching degree comparison result, and determines whether the waste needs to be sorted or whether to control the opening of the electric bin lid based on the waste classification matching degree comparison result.
[0038] The matching accuracy assessment combines real-time data with historical experience to avoid misjudgments or overly strict standards caused by a "one-size-fits-all" approach. From user declaration, system detection, lid control to sorting feedback, a complete behavioral guidance chain is formed, continuously improving sorting accuracy. By reducing mixed waste and minimizing manual sorting needs, backend processing costs are significantly reduced, while equipment lifespan is extended (due to reduced damage to lids from hard foreign objects).
[0039] Specifically, the data analysis module includes, The first data analysis unit is used to make an initial determination of the material of the waste based on the number of prominent points in the waste image; The second data analysis unit is used to correct the initial judgment results based on the softness of the waste; The third data analysis unit is used to determine whether to make a second determination of the type of waste based on the comparison between the actual waste weight and the waste weight within the standard waste weight range.
[0040] Specifically, the waste determination module determines whether the waste should be disposed of or whether the standard softness index range should be adjusted based on the secondary determination result of the third data analysis unit.
[0041] Specifically, the control module controls the electric box cover. If the actual weight of the waste is less than the minimum standard weight of the waste, the corresponding standard softness index range will be expanded based on the difference between the minimum standard weight of the waste and the actual weight of the waste. If the actual weight of the waste is within the standard waste weight range, the waste is determined to be Class I waste, and the control module controls the electric lid to open the electric lid for waste disposal. If the actual weight of the waste is greater than the maximum standard weight of the waste, the corresponding standard softness index range will be reduced based on the difference between the actual weight of the waste and the maximum standard weight of the waste.
[0042] The first data analysis unit is used to make an initial determination of the material of the waste based on the number of prominent points in the waste image. For example, if the number of protruding points is greater than 10, the garbage bag is initially determined to contain Class I garbage, which includes metal and glass. Setting the threshold for the number of protruding points of Class I garbage to 10 can effectively distinguish metal and glass from other materials. For daily life in the community, most garbage bags contain household waste, so the types of garbage can be distinguished well. If the number of protruding points is between 7 and 10, the garbage bag is initially determined to be Class II waste, which includes plastics and textiles. In daily life, residents have a high demand for bottled liquids, so the threshold for the number of protruding points for plastics and textiles is set between 7 and 10. If the number of protruding points is between 4 and 6, the garbage bag is initially determined to contain three types of garbage, including paper and wood. Since the contents of the three types of garbage do not significantly affect the number of protruding points, the threshold for the number of protruding points is set between 4 and 6. If the number of protruding points is between 0 and 3, the garbage bag is initially determined to contain four types of garbage, including kitchen waste and household waste. For these four types of garbage, since most of the items are relatively soft and cannot support the garbage bag, thus creating too many protruding points, the number of protruding points is set between 0 and 3. The second data analysis unit is used to correct the initial judgment results based on the softness of the waste; For cases where the initial determination result is any type of waste, the control module controls the softness sensing device to compress the waste bag. The pressure sensors on the softness sensing device are evenly distributed on two combined vertical plates. At this point, the softness sensing device determines the compression pressure value based on the initial determination result and compresses the waste bag a preset distance. The compression contact area is a constant. The actual softness index corresponding to any type of waste is then calculated by dividing the compression displacement distance of the waste bag by the pressure value. The actual softness index is compared with the standard softness index for that type of waste to obtain a softness index comparison result. Based on this softness index comparison result, it is determined whether the initial determination result needs to be corrected. For example, this embodiment uses a type of waste as an example. The standard softness index of Class I waste is set to 0.5 mm / N. The displacement distance of Class I waste is 1.8 mm and the pressure value is 10 N. The actual softness index of Class I waste is 0.8. Since the actual softness index of Class I waste is not in the standard softness index range [0.4, 0.6], it is determined that it is not Class I waste. The standard softness index range is set as [0.7, 0.9] for Class II waste, [1.0, 1.5] for Class III waste, and [1.6, 2.0] for Class IV waste. The third data analysis unit analyzes the actual weight of a type of waste, comparing it with a standard waste weight range. This standard waste weight range is determined based on historical data showing the actual proportion of each type of waste. If the actual weight of the waste is less than the minimum standard weight of the waste, the standard softness index range of the waste type will be expanded based on the difference between the minimum standard weight of the waste and the actual weight of the waste. If the actual weight of the waste is within the standard waste weight range, the waste is classified as Class 1 waste and disposed of accordingly. If the actual weight of the waste is greater than the maximum standard weight of the waste, the standard softness index range of a certain type of waste will be reduced based on the difference between the actual weight of the waste and the maximum standard weight of the waste. In this embodiment, the standard weight range for a type of waste is set as [5kg, 10kg], and the adjustment amount for the standard softness index range corresponding to the weight difference is set to 0.1. For example, if the actual weight of a type of waste is 3kg and its actual softness is 0.7, then the adjustment amount for the standard softness index range of type 1 waste is (5-3)×0.1=0.2. The adjusted standard softness index range for type 1 waste is [0.4+0.2, 0.6+0.2], that is, [0.6, 0.8]. Since 0.7 falls into the adjusted standard softness index range, it is determined to be type 1 waste. The actual weight of the waste was 7kg and the actual softness was 0.5. Since the actual weight of the waste was within the standard waste weight range, it was determined that the waste was classified as waste and was disposed of directly. If the actual weight of Class 1 waste is 13kg and the actual softness is 0.3, then the corresponding adjustment amount for the standard softness index range of Class 1 waste is (13-10)×0.1=0.3. The adjusted standard softness index range for Class 1 waste is [0.4-0.3, 0.6-0.3], that is, [0.1, 0.3]. Therefore, it is determined that the bin lid should not be opened and repackaging is required.
[0043] This system constructs a three-dimensional classification system based on "highlighting key points (visual perception), softness (tactile sensation), and weight (density)," breaking through the limitations of single-dimensional classification and achieving three-dimensional recognition of material, shape, and density. Weight-driven adjustments to the softness range enable the system to intelligently accommodate "lightweight / high-density" variants of waste. Data accumulation throughout the entire process, from initial assessment and correction to secondary assessment, supports continuous iteration of the AI model, resulting in exponential growth in classification capabilities over time. Through precise inter-module collaboration and data-driven design, this system not only significantly improves waste classification accuracy but also constructs a self-evolving intelligent classification system, providing a practical technological paradigm for intelligent urban sanitation.
[0044] Specifically, the bin-in-bin waste detection module performs secondary waste data collection on the waste inside the waste bin, generates bin-in-bin waste data, and determines the type of waste inside the bin based on the bin-in-bin waste data.
[0045] The bin-in-garbage detection module performs bin-in-garbage detection based on the visual assessment of the secondary judgment results. The waste in the bin is detected based on the secondary judgment result. If the secondary judgment result indicates that it is Class I waste, the waste detection module in the bin determines the classification accuracy based on physical matching degree and image matching degree. Among them, the physical matching degree is determined based on the actual weight and softness of the waste. For example, if the actual weight of the waste is 12kg, it is outside the standard waste weight range [5, 10kg] corresponding to Class I waste (-20%), and the softness is 0.5 within the range [0.4, 0.6] (+100%), the overall physical matching degree is 80%; the image matching degree is 70% for Class I waste and 30% for non-Class I waste, which is based on the waste storage unit in the waste detection module inside the bin to identify Class I waste.
[0046] By combining weight deviation and softness compliance rate, misjudgment caused by single weight abnormality is avoided. The weight ratio of weight, softness and image is adjusted in real time based on the secondary detection results, so that the system has the ability to self-evolve.
[0047] Specifically, the judgment module compares the types of waste in the bin with the secondary judgment result to obtain the accuracy of waste classification in the bin. Based on the comparison result of the waste classification accuracy in the bin with the standard waste classification accuracy range, it determines whether to adjust the weight of waste weight information on the overall weight of waste, or to increase consumption points for users.
[0048] Specifically, the determination module makes its judgment based on the comparison results of the accuracy of garbage classification. If the accuracy of waste sorting in the bin is less than the minimum value of the standard waste sorting accuracy range, then the weight is increased based on the difference between the minimum value of the standard waste sorting accuracy range and the accuracy of waste sorting in the bin. If the accuracy of waste sorting in the bin is within the standard waste sorting accuracy range, the original weight remains unchanged. If the accuracy of waste sorting in the bin exceeds the maximum value of the standard waste sorting accuracy range, the user will receive additional points.
[0049] In this embodiment, a standard waste sorting accuracy range is set [85%, 90%], wherein the standard waste sorting accuracy range is determined based on historical data and waste treatment technology requirements. If the accuracy of waste sorting in the bin is less than the minimum value of the standard waste sorting accuracy range, the increased weight is determined by multiplying the original weight by 1 and adding the product of the difference between the minimum value of the standard waste sorting accuracy range and the accuracy of waste sorting in the bin, divided by the minimum value of the standard waste sorting accuracy range. If the accuracy of waste sorting in the bin is within the standard waste sorting accuracy range, the original weight remains unchanged. If the accuracy of waste sorting in the bin is less than the minimum value of the standard waste sorting accuracy range, then the reduced weight is determined by multiplying the original weight by 1, subtracting the waste sorting accuracy in the bin from the maximum value of the standard waste sorting accuracy range, and dividing by the maximum value of the standard waste sorting accuracy range.
[0050] For mixed waste with abnormal weight, increasing the weight factor improves the system's sensitivity to identifying "weight meets standards but composition is abnormal," reducing the false positive rate. Points can be redeemed for garbage bags or community services, creating a positive cycle of "accurate disposal - earning rewards." A non-linear mapping formula between accuracy and weight allows for precise correction of classification errors, avoiding system oscillations caused by "one-size-fits-all" adjustments. The points mechanism not only increases participation but also uses user behavior data to feed back into model optimization, forming a closed loop of "human classification - system learning - more accurate identification." The standard accuracy range and weight adjustment strategy dynamically evolve with historical data and processing needs, ensuring high robustness across different scenarios.
[0051] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.
[0052] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. An intelligent sanitation waste sorting monitoring and feedback system, characterized in that, include A trash can, used to hold trash, is equipped with a controllable electric lid; This includes a trash can assembly, which consists of several different types of trash cans placed side by side, each of which has an individual switch that can be controlled. Garbage bag storage bins are used to hold garbage sorting bags when sorting garbage; Information collection device, used to collect information on disposed waste and user input; A softness collection device, located at the lower part of the information collection device, is used to collect the softness of garbage bags. It includes two sets of opposing vertically arranged combined upright plates. Each combined upright plate is equipped with a displacement sensor and several pressure sensors. The displacement sensor is used to collect the distance the garbage bag retracts when pressure is applied to the garbage bag, and the pressure sensor is used to collect the pressure value of the garbage bag rebounding when pressure is applied to the garbage bag. The actual softness index is determined by the ratio of distance to pressure value. The bin-in-bin waste detection module determines the accuracy of waste sorting by detecting the proportion of corresponding waste types in each type of waste bin. The data analysis module is connected to the information acquisition device and the garbage detection module in the bin, respectively. It determines the type of garbage based on the garbage weight information and garbage image information to obtain a first determination result. It compares the first determination result with the input information to obtain a garbage classification comparison result. Based on the garbage classification comparison result, it determines whether garbage needs to be separated, or adjusts the weight of garbage weight information and garbage image information to the first determination result. The waste judgment module is used to determine whether the analysis results of the data analysis module are correct based on the data from the waste detection module in the bin. A control module, which is connected to the trash can assembly, controls the opening of the electric lid of the trash can assembly based on the analysis data from the data analysis module; The waste information includes waste weight information and waste image information; the input information is the type of waste to be disposed of actively selected by the user; the data analysis module includes a first data analysis unit, a second data analysis unit, and a third data analysis unit. The first data analysis unit is used to make an initial determination of the waste material based on the number of protruding points in the waste image. The protruding points are determined based on the length and angle of the protruding points. A protruding point is determined when the length of the protruding point is greater than a preset value of the maximum length of the waste image and the angle of the protruding point is less than a preset angle. The second data analysis unit is used to correct the initial determination result based on the comparison result of the actual softness index and the standard softness index range of the corresponding category. The third data analysis unit is used to determine whether to make a secondary determination of the waste type based on the comparison result of the actual waste weight and the waste weight range of the standard waste weight range. The waste determination module determines whether the waste should be disposed of or whether the standard softness index range should be adjusted based on the secondary determination result of the third data analysis unit.
2. The intelligent sanitation waste sorting monitoring and feedback system according to claim 1, characterized in that, The information collection device, include, The support platform is used to place the waste to be tested; A camera, mounted on top of the support platform, is used to capture images of the waste placed on the support platform; An interactive device, located in the middle of the testing platform, is used to facilitate interaction between the system and the user. The interactive device includes an input keyboard and a display screen. The input keyboard is used to collect information input by the user, and the display screen is used to display feedback information. 3.The intelligent environmental sanitation garbage classification monitoring and feedback system according to claim 2, characterized in that, The data analysis module obtains the actual waste classification matching degree based on the overall waste type information and the input information. It compares the actual waste classification matching degree with the standard waste classification matching degree to obtain the waste classification matching degree comparison result. Based on the waste classification matching degree comparison result, it determines whether the waste needs to be separated, or whether to control the opening of the electric bin lid.
4. The intelligent environmental sanitation garbage classification monitoring and feedback system according to claim 3, characterized in that, The control module controls the electric box cover. If the actual weight of the waste is less than the minimum standard weight of the waste, the corresponding standard softness index range will be expanded based on the difference between the minimum standard weight of the waste and the actual weight of the waste. If the actual weight of the waste is within the standard waste weight range, the waste is determined to be Class I waste, and the control module controls the electric lid to open the electric lid for waste disposal. If the actual weight of the waste is greater than the maximum standard weight of the waste, the corresponding standard softness index range will be reduced based on the difference between the actual weight of the waste and the maximum standard weight of the waste.
5. The intelligent environmental sanitation garbage classification monitoring and feedback system according to claim 4, characterized in that, The bin-in-garbage detection module performs secondary garbage data collection on the garbage inside the garbage bin, generates bin-in-garbage data, and determines the type of garbage inside the bin based on the bin-in-garbage data. 6.The intelligent environmental sanitation garbage classification monitoring and feedback system according to claim 5, characterized in that, The waste sorting module compares the types of waste in the bin with the secondary sorting results to obtain the accuracy of waste sorting in the bin. Based on the comparison results of the waste sorting accuracy in the bin with the standard waste sorting accuracy range, it determines whether to adjust the weight of waste weight information on the overall weight of waste, or to increase consumption points for users. 7.The intelligent environmental sanitation garbage classification monitoring and feedback system according to claim 6, characterized in that, The waste identification module makes its judgment based on the comparison results of waste classification accuracy. If the accuracy of waste sorting in the bin is less than the minimum value of the standard waste sorting accuracy range, then the weight is increased based on the difference between the minimum value of the standard waste sorting accuracy range and the accuracy of waste sorting in the bin. If the accuracy of waste sorting in the bin is within the standard waste sorting accuracy range, the original weight remains unchanged. If the accuracy of waste sorting in the bin exceeds the maximum value of the standard waste sorting accuracy range, the user will receive additional points.
Citation Information
Patent Citations
Waste classification detecting system
CN105772403A
Garbage collection method and device
CN106927164A
Road garbage can capable of achieving accurate classifying
CN107042970A
Household waste sorting method and system based on Internet big data and image treatment technology
CN110921146A
Garbage classification quality evaluation method and system
CN112749747A
