A garbage collection and metering system based on RFID technology

Abstract

The present invention relates to the field of intelligent sanitation technology, and in particular to a waste collection and metering system based on RFID technology, comprising an electronic tag and a card reader. The electronic tag is attached to a user's waste bag and contains pre-stored tag information regarding the bag's capacity. The card reader is attached to a trash can at a waste collection site. When a user places a waste bag into the trash can, the card reader reads the tag information stored in the electronic tag. A data platform is used to calculate and analyze the amount of waste at the waste collection site based on the capacity information in the tags corresponding to all the waste bags. The present invention can significantly improve the accuracy and efficiency of waste collection and metering, achieving waste measurement and statistics.

Description

Technical Field

[0001] The present invention relates to the field of intelligent sanitation technology, and in particular to a garbage collection and metering system based on RFID technology. Background Art

[0002] With rising global environmental awareness and the growing popularity of resource recycling, waste sorting, as a key measure to reduce environmental pollution and improve resource recycling rates, has been widely implemented in many developed countries and has achieved remarkable results. However, in the actual implementation of waste sorting, there is still an uneven distribution of resident participation, and some residents still lack awareness of sorting. Furthermore, the management of sorted waste still relies primarily on manual labor, especially the measurement of sorted waste, which is still mainly based on manual operations and record-keeping. This results in slow progress and low operational efficiency. Summary of the Invention

[0003] (1) Technical issues to be resolved

[0004] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a garbage collection and metering system based on RFID technology, which solves the technical problem that the existing garbage metering work mainly relies on manual labor and has low operational efficiency.

[0005] (2) Technical solution

[0006] In order to achieve the above objectives, the main technical solutions adopted by the present invention include:

[0007] The embodiment of the present invention provides a garbage collection and metering system based on RFID technology, comprising: an electronic tag and a card reader;

[0008] The electronic tag is set on the garbage bag used by the user, and the electronic tag pre-stores tag information about the capacity information of the garbage bag;

[0009] The card reader is set on the trash can at the garbage recycling site. When the user puts the garbage bag into the trash can, the card reader is used to read the tag information stored in the electronic tag on the garbage bag;

[0010] The data platform is used to count the amount of garbage at the garbage recycling site and perform data analysis based on the capacity information in the label information corresponding to all garbage bags.

[0011] Optionally, the electronic tag is a passive electronic tag, and the card reader is a radio frequency reader. The radio frequency reader activates the passive electronic tag by transmitting a first radio frequency signal, so that the passive electronic tag transmits a second radio frequency signal containing tag information.

[0012] Optionally, the card reader includes:

[0013] A controller module, configured to generate an initial first radio frequency signal;

[0014] A radio frequency circuit, configured to modulate an initial first radio frequency signal;

[0015] an antenna, configured to transmit the modulated first radio frequency signal;

[0016] as well as,

[0017] When the card reader is used to receive the second radio frequency signal,

[0018] The antenna is further configured to receive a second radio frequency signal;

[0019] The radio frequency circuit is further configured to demodulate the second radio frequency signal;

[0020] The controller module is further configured to receive the demodulated second radio frequency signal.

[0021] Optionally, the label information includes: the volume capacity of the garbage bag and the type of garbage it is used to hold.

[0022] Optionally, the metering system further includes:

[0023] An image acquisition device is provided on the top of the trash can and is used to obtain a first image of the trash can when the card reader reads the electronic tag of a newly-added trash bag;

[0024] An image analysis device is used to identify a newly thrown garbage bag in the first image based on the first image; and, based on the image of the newly thrown garbage bag, identify the fullness of the newly thrown garbage bag; and predict the predicted volume of the newly thrown garbage bag based on the fullness of the newly thrown garbage bag and the garbage type in the label information.

[0025] Optionally, the image analysis device includes:

[0026] a first image recognition module configured to perform a subtraction between the first image and a second image of the trash can obtained when the trash bag was last placed in the trash can, and to define an area where the difference is greater than a first preset value as a region of interest corresponding to the newly placed trash bag in the first image;

[0027] The second image recognition module is used to predict the fullness of a newly-added garbage bag in the region of interest;

[0028] The volume prediction module is used to determine the predicted volume of the newly-thrown garbage bag based on the volume capacity and fullness of the newly-thrown garbage bag.

[0029] Optionally, the image analysis device further includes:

[0030] An image processing module is configured to perform a subtraction between the first image and the second image to obtain a difference image, and to assign zero to the pixel values of the region of interest in the difference image to obtain a feature image;

[0031] a third image recognition module, configured to predict the predicted density of the newly-thrown garbage bag based on the feature image and the garbage type of the newly-thrown garbage bag;

[0032] The weight prediction module is used to determine the predicted weight of the newly-thrown garbage bag based on the predicted density and predicted volume of the newly-thrown garbage bag.

[0033] Optionally, the image analysis device further includes:

[0034] a fourth image recognition module, configured to identify the type of garbage in a newly-added garbage bag based on the region of interest in the first image;

[0035] A judgment module is used to judge whether the types of garbage in the newly-added garbage bag only include the types of garbage currently contained in the garbage bin. If so, it is determined that the garbage classification is correct.

[0036] Among them, the second image recognition module, the third image recognition module and the fourth image recognition module are respectively: deep learning models with appropriate model parameters obtained through a pre-training process.

[0037] Optionally, the data platform includes:

[0038] The sequence processing module is used to obtain the total amount of garbage in all garbage bins at the garbage collection site, and calculate the total amount of garbage at each preset sampling period since the last emptying of the garbage collection site to form a time series of the total amount of garbage;

[0039] A time prediction model for predicting, based on the time series, a future time point at which the total amount of garbage at the garbage collection site reaches a preset value;

[0040] A scheduling module, configured to determine a time to schedule a garbage truck to a garbage collection site based on the future time point;

[0041] The time prediction model is a machine learning model with appropriate model parameters obtained through a pre-training process.

[0042] Optionally, the data platform further includes:

[0043] The data preprocessing module is used to record the total amount of new garbage at the garbage recycling site based on the predicted volume and predicted weight of the newly-added garbage bag predicted by the image analysis device each time a new garbage bag is added.

[0044] (3) Beneficial effects

[0045] The garbage collection and metering system proposed in an embodiment of the present invention includes an electronic tag and a card reader; the electronic tag is set on the garbage bag used by the user, and the electronic tag pre-stores tag information about the capacity information of the garbage bag; the card reader is set on the garbage bin of the garbage recycling site, and when the user puts the garbage bag into the garbage bin, the card reader is used to read the tag information stored in the electronic tag on the garbage bag; the data platform is used to count the garbage volume of the garbage recycling site and perform data analysis based on the capacity information in the tag information corresponding to all garbage bags.

[0046] Based on the electronic tags attached to the garbage bags and the card reader installed on the trash can, each time a garbage bag is placed in the trash can, the card reader can automatically identify the garbage bag and count the number of garbage bags, significantly improving the accuracy and efficiency of garbage collection and measurement, and achieving garbage measurement and statistics. Furthermore, the metering system provided by the present invention can provide real-time monitoring and alarm functions for garbage collection sites, helping managers optimize garbage management processes. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a schematic diagram of the architecture of a garbage collection and metering system based on RFID technology provided in an embodiment;

[0048] Figure 2 A schematic structural diagram of a card reader provided in an embodiment;

[0049] Figure 3 Schematic diagram of the architecture of the image acquisition device and the image analysis device in the embodiment;

[0050] Figure 4 Schematic diagram of the data platform architecture in the embodiment. DETAILED DESCRIPTION

[0051] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation methods in conjunction with the accompanying drawings.

[0052] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0053] Example 1

[0054] like Figure 1As shown, this embodiment provides a garbage collection and metering system based on RFID technology, including: an electronic tag and a card reader.

[0055] The electronic tag is set on the garbage bag used by the user, and the electronic tag pre-stores tag information about the capacity information of the garbage bag.

[0056] The card reader is arranged on a trash can at a trash recycling site. When a user puts a trash bag into the trash can, the card reader is used to read the tag information stored in the electronic tag on the trash bag.

[0057] The data platform is used to count the amount of garbage at the garbage recycling site and perform data analysis based on the capacity information in the label information corresponding to all garbage bags.

[0058] To more precisely count the number of garbage bags or the amount of garbage for each classification, the label information may include: the garbage bag number, capacity information, and the type of garbage it contains. Specifically, the garbage types may include recyclable garbage, hazardous waste, kitchen waste, and other garbage. This label information may also be distinguished by the different digits in the garbage bag number. For example, the last digit of the number being 0, 1, 2, or 3 represents dry garbage, wet garbage, hazardous waste, and recyclable garbage, respectively. The capacity information of the garbage bag may be volumetric or weight capacity. This capacity information may also be distinguished by the different digits in the garbage bag number. For example, the first digit of the number may represent the volumetric capacity. For example, 1 represents a garbage bag with a volumetric capacity of 1L, 2 represents a garbage bag with a volumetric capacity of 2L, and 5 represents a garbage bag with a volumetric capacity of 5L. Furthermore, to facilitate user differentiation between garbage bags for different types of garbage, four colors of garbage bags may be provided: blue, red, green, and black, corresponding to recyclable garbage, hazardous waste, kitchen waste, and other garbage.

[0059] Based on the electronic tags attached to the garbage bags and the card readers installed on the trash cans, each time a garbage bag is placed in the trash can, the card reader can automatically identify the type of garbage contained in the garbage bag and count the number of garbage bags, significantly improving the accuracy and efficiency of garbage collection and measurement, and achieving the measurement and statistics of classified garbage. Furthermore, the metering system provided by the present invention can provide real-time monitoring and alarm functions for garbage collection sites, helping managers optimize garbage management processes.

[0060] In a preferred embodiment of this embodiment, the electronic tag is a passive electronic tag, and the reader is a radio frequency reader. The radio frequency reader activates the passive electronic tag by transmitting a first radio frequency signal, causing the passive electronic tag to emit a second radio frequency signal containing the tag information. The passive electronic tag comprises a coil and a chip. The coil converts electrical energy from the received first radio frequency signal to activate the chip within the electronic tag. The electronic tag then transmits the tag information pre-stored in the chip via the second radio frequency signal. Passive electronic tags do not require internal batteries and are activated by the radio frequency signal from the radio frequency reader, making them suitable for large-scale, low-cost applications.

[0061] Specifically, each electronic tag is pre-assigned a unique identification number to identify the uniqueness of each garbage bag, ensuring that the information of different garbage bags in the metering system will not be confused or repeated. The electronic tag has the characteristics of being waterproof, dustproof and high temperature resistant, ensuring that it can still work stably in a variety of complex environments of garbage disposal. At the same time, the electronic tag should meet the specific working frequency band and reading distance requirements. Preferably, the electronic tag can use a high-frequency RFID tag that can be accurately identified within a range of 1 meter. The electronic tag is fixed to the side of the garbage bag. During installation, use a strong adhesive or heat sealing technology to firmly attach the electronic tag to the garbage bag to prevent it from falling off or being damaged during use, and ensure that the electronic tag always remains readable in each link of garbage bag placement, transportation and processing, so as to facilitate automatic identification in the subsequent garbage disposal process and reduce manual work.

[0062] Specifically, if Figure 2 As shown, the card reader includes a controller module, a radio frequency circuit and an antenna that are communicatively connected in sequence, as follows:

[0063] The controller module is configured to generate an initial first radio frequency signal.

[0064] The radio frequency circuit is used to modulate the initial first radio frequency signal.

[0065] The antenna is used to transmit the modulated first radio frequency signal.

[0066] When the card reader is used to receive the second radio frequency signal,

[0067] The antenna is further configured to receive a second radio frequency signal;

[0068] The radio frequency circuit is further configured to demodulate the second radio frequency signal;

[0069] The controller module is further configured to receive the demodulated second radio frequency signal.

[0070] In addition, the metering system provided in this embodiment may also include a communication module and a power module. The communication module is connected to all card readers within the garbage station, receiving tag information read by the card readers and forwarding it to the data platform. The power module is electrically connected to the controller module and the RF circuit to provide power. The power module may also have intelligent power management capabilities, issuing an alarm when the battery is low or automatically entering low-power mode, ensuring continuous and stable operation of the device in various environments.

[0071] In one specific application, the antenna transmits the first RF signal and receives the second RF signal, serving as a key component for data communication between the reader and the electronic tag. The antenna can be mounted on the top or side of the trash can, covering the area where the trash bag is placed. This ensures that the electronic tag is quickly activated and its information is read when the trash bag is dropped in. The RF circuit modulates and transmits the first RF signal and demodulates the second RF signal returned by the electronic tag. The controller module is the core of the entire reader, coordinating the antenna and RF circuit, processing acquired tag information, and interacting with the communication module. The controller module has built-in data processing capabilities, enabling filtering, verification, and organization of read tag information. The controller module can communicate directly with the communication module via a network cable or wirelessly via the RF circuit. The communication module transmits tag information read by the reader to the data platform in real time. The communication module supports Wi-Fi, LoRa, NB-IoT, and other communication methods, adapting to network conditions in various scenarios and building a localized Internet of Things. Through the communication module, the data platform can remotely monitor and manage waste placement and sorting data.

[0072] More specifically, the controller module filters, verifies, and organizes tag information as follows:

[0073] Filtration process:

[0074] When a garbage bag is put into the trash bag multiple times, it may cause the card reader to read the electronic tag repeatedly. In this embodiment, when duplicate tag information is recognized, only the tag information read for the first time is retained as valid information. Next, in the case where multiple trash cans for carrying different types of garbage may be set side by side, it is easy for the card reader of trash can A to read the electronic tag in trash can B. By setting a minimum signal strength threshold, it is possible to eliminate misreading caused by electronic tags in other trash cans that are relatively far away interfering with the signal of the card reader set on the current trash can. Only when the signal strength received by the card reader exceeds the set minimum signal strength threshold will the card reader consider this to be a valid read event. Then, a time window can be defined, and continuous reading of the same tag within this time window is regarded as one operation, avoiding frequent reading of the same electronic tag in a short period of time. This time window can be the average time interval between two adjacent emptying of the garbage collection station.

[0075] Verification processing:

[0076] RFID tags support cyclic redundancy checks (CRCs). Card readers can perform CRC checks after receiving tag information to ensure data transmission integrity. Furthermore, further verification can be performed based on business logic, such as checking whether the garbage type in the garbage bag is legal, whether the capacity is appropriate, and whether the production date of the garbage bag is within its validity period. Verifying the legality of the garbage type in the garbage bag involves parsing the garbage type specified in the tag and then matching it with the garbage type in the current trash can. If there is a mismatch, the garbage bag has been placed in the wrong trash can.

[0077] Finishing:

[0078] The read tag information is organized in a unified standard format to facilitate subsequent data processing and analysis by the data platform. This includes but is not limited to ID numbers, classification codes, capacity units, date formats, etc. In addition, the tag information can be associated with other relevant data such as the location and delivery time of the garbage collection site to form a complete record to track the flow path of the garbage bag and provide richer contextual information for data analysis by the data platform. Finally, for tag information that fails the above inspection or has incorrect data format, an exception handling mechanism can be set up, such as recording error logs and sending alerts to notify administrators.

[0079] Example 2

[0080] In actual use, garbage bags placed by users may be overfilled, exceeding their designed volume capacity, or underfilled, failing to reach their designed volume capacity. In this case, simply relying on a card reader to read the tag information in the electronic tag may result in the capacity information obtained being inconsistent with the actual amount of garbage. To address this issue, the metering system provided in this embodiment, in addition to the electronic tag and card reader provided in the first embodiment, further includes:

[0081] The image acquisition device is arranged on the top of the trash can and is used to obtain a first image inside the trash can when the card reader reads the electronic tag of a newly-added trash bag.

[0082] Specifically, the image acquisition device can be a miniature camera, which is arranged on the inner side of the top of the trash can lid. In order to ensure the brightness of the shooting environment, it can also be equipped with an LED fill light.

[0083] An image analysis device is used to identify a newly thrown garbage bag in the first image based on the first image; and, based on the image of the newly thrown garbage bag, identify the fullness of the newly thrown garbage bag; and predict the predicted volume of the newly thrown garbage bag based on the fullness of the newly thrown garbage bag and the garbage type in the label information.

[0084] Specifically, the image analysis devices may be multiple, one for each trash can, and located at the same location as the card reader. Alternatively, to save costs, each waste collection station may be equipped with one image analysis device, configured to receive and analyze the first image captured by the image acquisition devices installed in all trash cans at that station.

[0085] In a specific embodiment, the image analysis device includes: a first image recognition module, a second image recognition module and a volume prediction module.

[0086] The first image recognition module is used to subtract the first image from the second image of the trash can obtained when the trash bag was last thrown in, and use the area where the difference is greater than a first preset value as the area of interest corresponding to the newly thrown trash bag in the first image.

[0087] The second image recognition module is used to predict the fullness of a newly-thrown garbage bag in the region of interest.

[0088] The volume prediction module is used to determine the predicted volume of the newly-added garbage bag based on the volume capacity and fullness of the newly-added garbage bag. Specifically, the predicted volume is the product of the volume capacity and fullness of the newly-added garbage bag.

[0089] More preferably, in order to simultaneously obtain weight information of a newly-added garbage bag based on the first image, the image analysis device further includes: an image processing module, a third image recognition module, and a weight prediction module.

[0090] An image processing module is configured to perform a subtraction between the first image and the second image to obtain a difference image, and to assign zero to the pixel values of the region of interest in the difference image to obtain a feature image;

[0091] a third image recognition module, configured to predict the predicted density of the newly-thrown garbage bag based on the feature image and the garbage type of the newly-thrown garbage bag;

[0092] The weight prediction module is used to determine the predicted weight of the newly-added garbage bag based on the predicted density and predicted volume of the newly-added garbage bag. Specifically, the predicted density and the predicted volume are multiplied, and the product obtained is the predicted weight.

[0093] It should be noted that the density of a bag of garbage is proportional to its weight. When a garbage bag is placed on top of other garbage bags in a trash can, its density is positively correlated with the slump of the other garbage bags. That is, the greater the density and weight of the garbage, the more likely it is that its weight will cause the other garbage bags to collapse and deform. Therefore, this embodiment uses an image processing module to extract the deformation features of the other garbage bags before and after a new garbage bag is added. It then uses a third image recognition module to predict the density of the garbage bag based on the deformation information of the other garbage bags contained in the feature image. Furthermore, since different types of garbage experience different stress deformation patterns, taking the garbage type into consideration can more accurately predict the garbage density.

[0094] More preferably, in order to further improve the accuracy of the predicted weight output by the weight prediction module, the predicted weight may be compensated based on the following formula.

[0095] Where W is the predicted weight after compensation, is the product of predicted density and predicted volume, is the predicted density, V is the predicted volume, is the volume compensation coefficient corresponding to different types of garbage, is the density compensation coefficient corresponding to different garbage types. and are empirical constants whose value range is real numbers, and different types of garbage correspond to and different.

[0096] and It can be measured experimentally. Specifically, for a type of garbage, the laboratory randomly puts various types of garbage of this type into garbage bags and measures their true weights. Then, the bags are randomly placed in a garbage bin to obtain the first image and the second image. The above image analysis device is used to obtain the predicted density, predicted volume and predicted weight of the garbage bag. The true weight of the garbage bag is equivalent to the predicted weight after compensation using the above formula. After repeated experiments to obtain multiple sets of data, nonlinear fitting is used based on the multiple sets of data to calculate the weight of the garbage bag in the above formula. and Solve it and get and The optimal value of .

[0097] In addition, in order to be able to identify whether the garbage in the garbage bag newly put in by the user has been reasonably classified, the image analysis device also includes:

[0098] a fourth image recognition module, configured to identify the type of garbage in a newly-added garbage bag based on the region of interest in the first image;

[0099] The judgment module is used to judge whether the garbage types in the newly-thrown garbage bag only include the garbage types currently contained in the garbage bin. If so, it is judged that the garbage classification is correct; otherwise, it is judged that the garbage classification is wrong.

[0100] Specifically, the second image recognition module, the third image recognition module, and the fourth image recognition module are all deep learning models with adapted model parameters obtained through a pre-training process. More preferably, the deep learning model is a CNN model. The pre-training process of the CNN model adopts existing technology.

[0101] In the training data set of the CNN model corresponding to the second image recognition module, the images used for training can be multiple sampling images taken by a micro camera after the garbage bag is filled with garbage to different fullness and placed in the garbage can at different angles. The annotation information of the sampling image is the fullness of the garbage bag.

[0102] Similarly, in the training dataset for the CNN model corresponding to the third image recognition module, the training images can be multiple sample images taken from different angles inside a trash can, pre-stuffed with garbage of the corresponding type. The garbage within the bag (different garbage within the same garbage type, such as plastic and metal within recyclables) has different densities. The images are then annotated with the density of the garbage bag. The density of the garbage bag can be determined by weighing the bag, then determining its volume using the displacement method, and then dividing the weight by the volume.

[0103] Similarly, in the training data set of the CNN model corresponding to the fourth image recognition module, the images used for training may be multiple sampling images taken by a micro camera, in which random types of garbage are pre-placed in a garbage bag and placed in a trash can at different angles. The labeling information of the sampling images is all the types of garbage in the garbage bag.

[0104] Example 3

[0105] Based on the metering systems provided in Examples 1 and 2, the data platform, using the collected data, can better manage waste collection sites. For example, it can identify issues such as fluctuations in waste sorting accuracy and abnormal waste collection volumes, allowing administrators to address these issues in a targeted manner and improve the accuracy of waste sorting and recycling. Furthermore, the data platform can provide real-time monitoring and alerting capabilities to help managers optimize waste management processes.

[0106] Specifically, the data platform includes:

[0107] The data preprocessing module is used to record the total amount of new garbage at the garbage recycling site based on the predicted volume and predicted weight of the newly-added garbage bag predicted by the image analysis device each time a new garbage bag is added.

[0108] The sequence processing module is used to obtain the total amount of garbage in all garbage bins at the garbage collection site, and calculate the total amount of garbage at each preset sampling period since the last emptying of the garbage collection site to form a time series of the total amount of garbage;

[0109] A time prediction model, for predicting, based on the time series, a future time point at which the total amount of garbage at the garbage site will reach a preset value;

[0110] The scheduling module is used to determine the time to schedule the garbage truck to go to the garbage collection site based on the future time point.

[0111] Specifically, the time prediction model may be: an LSTM model obtained through a pre-training process and having adapted model parameters; or an ARIMA model obtained through a pre-training process and having suitable model parameters.

[0112] In a specific implementation, taking the ARIMA model as a time prediction model as an example, this embodiment provides a method for performing data analysis on a data platform, which is as follows:

[0113] Use ARIMA (AutoRegressive Integrated Moving Average) to predict garbage disposal volume. The ARIMA model is a time series forecasting method that is particularly suitable for data with trends and seasonality.

[0114] The specific steps are as follows:

[0115] S1. Historical data cleaning and preprocessing

[0116] The format of this historical data is: the time it takes for a trash bin to be emptied and for it to reach 10%, 20%, ... 100% of capacity. These ten time periods form a time series. (This could also be based on the total trash capacity of the entire site and the time it takes to reach 0%, 20%, ... 100% of the site capacity.)

[0117] Noise removal: First, clean the historical data to remove incomplete, duplicate or obviously erroneous data records.

[0118] Standardization: Convert data in different formats into a unified standard format to facilitate subsequent analysis.

[0119] Filling missing values: For some possible missing data, reasonable filling or estimation can be performed based on historical data or other relevant data.

[0120] S2. Determine model parameters

[0121] (1) Stationarity test: Use tests such as ADF (Augmented Dickey-Fuller) to check whether the time series is stationary. If it is not stationary, you may need to perform a difference operation (d parameter) until the series becomes stationary.

[0122] (2) ACF and PACF plots: Draw the autocorrelation function (ACF) and partial autocorrelation function (PACF) plots to help determine the order of AR(p) and MA(q). Generally, a tailing ACF plot and a truncated PACF plot indicate that the AR model is suitable; conversely, a tailing PACF plot and a truncated ACF plot indicate that the MA model is suitable.

[0123] (3) Selecting optimal parameters: Information criteria such as AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) can be used to find the best (p, d, q) combination through grid search or other optimization algorithms.

[0124] S3. Model fitting and validation

[0125] Train Model: Train an ARIMA model using the selected (p, d, q) parameters.

[0126] Residual analysis: Check whether the model residuals are white noise, that is, there is no obvious pattern or trend. If there is significant autocorrelation in the residuals, the model parameters may need to be adjusted.

[0127] Cross-validation: Cross-validate the model using rolling forecasts or hold-out methods to evaluate its predictive performance.

[0128] S4. Prediction and Anomaly Detection

[0129] Use the validated ARIMA model described above to generate forecasts.

[0130] For example, when the garbage collected by a garbage station reaches 90% of its total capacity, it is necessary to consider sending a garbage truck to collect the garbage. The time from the last time the garbage bin was emptied to the time the capacity reached 10%, 20%,... 60% is obtained to form a time series, and the time required for the garbage volume to reach 70%, 80%, and 90% is predicted. The time from the time it reaches 90% is subtracted from the time the garbage station is emptied to the current time point, which is the time from the current time to the time when the garbage reaches 90% of the total capacity. If this time is within the preset range (for example, 4 hours), you can consider sending a garbage truck to collect the garbage.

[0131] In another specific implementation, taking the LSTM model as a time prediction model as an example, this embodiment provides a method for performing data analysis on a data platform.

[0132] The historical data format used for training the LSTM model is the total amount of garbage at a garbage collection site after a preset time period since it was emptied, forming a historical time series about the total amount of garbage. Preferably, the preset time period can be half an hour to two hours.

[0133] The LSTM model is trained using the above historical time series to obtain a trained LSTM model.

[0134] When it is used to predict the total amount of garbage at the current garbage site, the time series of the total amount of garbage at the current garbage site obtained by the sequence processing module is input into the trained LSTM model, and it can then predict the predicted total amount of garbage sequence after every preset time period after the current time point.

[0135] The first future time point in the predicted total garbage volume sequence that exceeds 90% of the total capacity of the garbage site is used as a warning point. If the difference between this warning point and the current time point is less than a preset range (for example, 4 hours), it can be considered to send a garbage truck to collect the garbage.

[0136] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0137] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each process flow and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions.

[0138] It should be noted that, in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims enumerating several means, several of these means may be embodied by one and the same hardware. The use of the words first, second, third etc. is for convenience only and does not indicate any order. These words may be understood as part of the component name.

[0139] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" 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 present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0140] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments after learning the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0141] Obviously, those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations fall within the scope of the claims and their equivalents, the present invention shall also include such modifications and variations.

Claims

1. A garbage collection and metering system based on RFID technology, characterized in that: include: Electronic tags and card readers; The electronic tag is set on the garbage bag used by the user, and the electronic tag pre-stores tag information about the capacity information of the garbage bag; The card reader is set on the trash can at the garbage recycling site. When the user puts the garbage bag into the trash can, the card reader is used to read the tag information stored in the electronic tag on the garbage bag; The data platform is used to count the amount of garbage collected at the recycling station and perform data analysis based on the capacity information in the labels corresponding to all garbage bags; The metering system further comprises: An image acquisition device is provided on the top of the trash can and is used to obtain a first image of the trash can when the card reader reads the electronic tag of a newly-added trash bag; an image analysis device configured to identify a newly-placed garbage bag in the first image based on the first image; and, based on the image of the newly-placed garbage bag, identify the fullness of the newly-placed garbage bag; and predict a predicted volume of the newly-placed garbage bag based on the fullness of the newly-placed garbage bag and the type of garbage in the label information; The image analysis device comprises: a first image recognition module configured to perform a subtraction between the first image and a second image of the trash can obtained when the trash bag was last placed in the trash can, and to define an area where the difference is greater than a first preset value as a region of interest corresponding to the newly placed trash bag in the first image; The second image recognition module is used to predict the fullness of a newly-added garbage bag in the region of interest; A volume prediction module is used to determine the predicted volume of a newly-added garbage bag based on the volume capacity and fullness of the newly-added garbage bag; The image analysis device further comprises: An image processing module is configured to perform a subtraction between the first image and the second image to obtain a difference image, and to assign zero to the pixel values of the region of interest in the difference image to obtain a feature image; a third image recognition module, configured to predict the predicted density of the newly-thrown garbage bag based on the feature image and the garbage type of the newly-thrown garbage bag; A weight prediction module, configured to determine a predicted weight of a newly-added garbage bag based on a predicted density and a predicted volume of the newly-added garbage bag; The data platform includes: The data preprocessing module is used to record the total amount of new garbage at the garbage recycling site based on the predicted volume and predicted weight of the newly-added garbage bag predicted by the image analysis device each time a new garbage bag is added.

2. The metering system according to claim 1, characterized in that The electronic tag is a passive electronic tag, and the card reader is a radio frequency reader. The radio frequency reader activates the passive electronic tag by transmitting a first radio frequency signal, so that the passive electronic tag transmits a second radio frequency signal containing tag information.

3. The metering system according to claim 2, characterized in that The card reader comprises: A controller module, configured to generate an initial first radio frequency signal; A radio frequency circuit, configured to modulate an initial first radio frequency signal; an antenna, configured to transmit the modulated first radio frequency signal; as well as, When the card reader is used to receive the second radio frequency signal, The antenna is further configured to receive a second radio frequency signal; The radio frequency circuit is further configured to demodulate the second radio frequency signal; The controller module is further configured to receive the demodulated second radio frequency signal.

4. The metering system according to claim 1, characterized in that The label information includes: the volume capacity of the garbage bag and the type of garbage it is used to hold.

5. The metering system according to claim 1, characterized in that The image analysis device further comprises: a fourth image recognition module, configured to identify the type of garbage in a newly-added garbage bag based on the region of interest in the first image; A judgment module is used to judge whether the types of garbage in the newly-added garbage bag only include the types of garbage currently contained in the garbage bin. If so, it determines whether the garbage classification is correct. Among them, the second image recognition module, the third image recognition module and the fourth image recognition module are respectively: deep learning models with appropriate model parameters obtained through a pre-training process.

6. The metering system according to any one of claims 1 to 5, characterized in that: The data platform includes: The sequence processing module is used to obtain the total amount of garbage in all garbage bins at the garbage collection site, and calculate the total amount of garbage at each preset sampling period since the last emptying of the garbage collection site to form a time series of the total amount of garbage; A time prediction model for predicting, based on the time series, a future time point at which the total amount of garbage at the garbage collection site reaches a preset value; A scheduling module, configured to determine a time to schedule a garbage truck to a garbage collection site based on the future time point; The time prediction model is a machine learning model with appropriate model parameters obtained through a pre-training process.

Images