Intelligent resource environment-friendly recycling box and method

Abstract

The application provides an intelligent resource environmental protection recycling box and method. Image frame sequences captured by a depth camera arranged on the upper part of the front face of the recycling box and real-time weight data output by a weight sensor installed at the bottom of the classification bin are collected during the whole process of the throwing behavior; then, based on the image frame sequences, the throwing operation is divided into the approaching stage, the aligning stage and the releasing stage by a behavior recognition algorithm, and the movement trajectory of the garbage bag is tracked. In the approaching stage, the garbage bag features are identified to obtain the garbage category prediction result and generate preliminary prompt information; in the aligning stage, the position deviation is detected and the voice prompt is triggered to adjust the position; after the releasing stage is completed, the garbage falling condition is confirmed according to the weight data, and the corresponding prompt is broadcast. Finally, through consistency comparison, classification error correction and score incentive are realized. The above scheme can realize real-time prompt, error correction and result confirmation of the whole process of garbage throwing.

Description

Technical Field

[0001] This application relates to the field of intelligent resource and environmentally friendly recycling bins, and more specifically, to an intelligent resource and environmentally friendly recycling bin and method. Background Technology

[0002] With the acceleration of urbanization and the improvement of environmental awareness, smart resource recycling bins, as an important infrastructure for waste sorting and recycling, are widely deployed in communities, public places and other areas to guide residents to correctly dispose of recyclable waste, hazardous waste and other types of waste.

[0003] Traditional waste recycling bins primarily rely on static classification labels (such as text and color tags) or simple voice prompts to guide users in disposing of their waste. However, these methods cannot monitor user behavior in real time, identify waste categories, or correct disposal errors. Especially when users approach the bin with their trash bags, the lack of dynamic guidance mechanisms leads to frequent disposal failures or misclassifications. Furthermore, traditional recycling bins lack disposal result verification functions, failing to confirm whether waste has correctly landed in the corresponding compartment, further exacerbating the inefficiency and human error in the recycling process. Therefore, how to achieve end-to-end guidance, real-time error correction, and result confirmation through intelligent means has become an urgent problem to be solved to improve the practicality and environmental effectiveness of smart recycling bins. Summary of the Invention

[0004] This application provides an intelligent resource recycling bin and method that can provide real-time prompts, error corrections, and result confirmation throughout the entire waste disposal process.

[0005] Firstly, this application provides a waste disposal prompt method for intelligent resource recycling bins, including: Throughout the entire process of collecting waste disposal, the system collects image frame sequences captured by at least one depth camera located on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments, which correspond one-to-one with each sorting and disposal port. Based on the image frame sequence, the behavior recognition algorithm divides the disposal operation into an approach phase, an alignment phase, and a release phase, starting with the user entering a preset approach area, the first node being the hand entering a preset alignment area above the disposal port, the second node being the user triggering a release action, and the end point being the separation of the garbage bag from the user's hand. The movement trajectory of the garbage bag is tracked throughout the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin. During the approach phase, the characteristics of the garbage bags are identified to obtain their garbage category prediction results, and preliminary prompts are generated to guide the user to the corresponding sorting and disposal port; During the alignment phase, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position. After the release phase is completed and a preset weight stabilization period has elapsed, the system confirms whether the waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin. If the waste has fallen into the bin, a successful disposal confirmation voice message is played. If no valid weight change is detected, a disposal anomaly prompt is triggered.

[0006] In some embodiments, the image frame sequence of the entire delivery process specifically includes: A depth camera is installed in the center of the upper front of the recycling bin. The field of view of the depth camera completely covers all sorting and disposal ports on the bin, as well as the 3m×2m user access and operation area in front of the bin. The depth camera continuously acquires a sequence of depth images including the user's body, hands, and garbage bags, along with a synchronized sequence of color images. It also simultaneously acquires real-time weight data from the corresponding weight sensors in each sorting bin, thus achieving spatiotemporal alignment of images and weight data.

[0007] In some embodiments, tracking the movement trajectory of garbage bags specifically includes: Depth information and contour features of the user's hand region are extracted from the depth image frame sequence. Based on the depth threshold and instance segmentation model, accurate segmentation of the area where the hand is stuck to the garbage bag is completed. In the segmented garbage bag area, a deep learning target tracker is used to continuously associate the spatial position of the garbage bag in consecutive image frames from the start of the approach phase to the complete process of the garbage bag falling into the target sorting bin, thereby generating the complete motion trajectory of the garbage bag.

[0008] In some embodiments, during the approach phase, identifying the characteristics of the garbage bag to obtain its garbage category prediction specifically includes: Detect whether there are pre-printed classification identification codes and classification color markings on the surface of the garbage bag from the image frame sequence; If a valid classification identifier or classification color identifier is detected, the waste category prediction result is obtained through decoding or color matching; If no valid classification identifier is detected, or the recognition confidence level is lower than the preset threshold, a voice or screen prompt will be triggered before the user enters the alignment stage, guiding the user to manually select the waste category through the human-computer interaction interface of the recycling bin, and the result of the user's manual selection will be used as the waste category prediction result; if the user does not complete the manual selection and directly enters the alignment stage, the waste category prediction result of this disposal will be marked as pending confirmation, and subsequent verification will be completed through classification bin category matching after disposal is completed.

[0009] In some embodiments, during the alignment phase, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position. Specifically, this includes: Around each sorting and disposal port, an area-array infrared detection array is deployed to form a two-dimensional position detection field covering the preset effective disposal area of ​​the corresponding disposal port; or the spatial coordinates of the garbage bag and the target sorting and disposal port are obtained in real time through the depth camera. During the alignment phase, the relative positional deviation between the center of the garbage bag and the center of the effective disposal area of ​​the target sorting and disposal port is calculated in real time using the area array infrared detection array or depth camera. If the relative position deviation continues to exceed the preset deviation threshold and the duration exceeds the preset stability judgment duration, the user will be guided to move their hand in the specified direction via voice prompts before the user performs the release action, and the relative position status of the garbage bag and the effective disposal area will be displayed in real time on the screen of the bin.

[0010] In some embodiments, after the release phase is completed and a preset weight stabilization period has elapsed, confirming whether the waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin specifically includes: Each sorting and disposal port corresponds to an independent sealed sorting bin, and each sorting bin is independently equipped with a weight sensor with matching weighing accuracy at the bottom. Based on the completion time of the release phase, after waiting for a preset weight stabilization period of 300ms-600ms, the stable weight value output by the corresponding classification bin weight sensor is obtained, and the difference between it and the baseline weight value collected within 100ms before the release phase is calculated as the actual weight change. If the actual weight change exceeds the preset minimum weight threshold, the waste is determined to have successfully fallen into the corresponding sorting bin; if it does not exceed the preset minimum weight threshold, the disposal is determined to be abnormal, triggering a check for blockage at the disposal port and recording the abnormal event simultaneously.

[0011] In some embodiments, the method further includes: Once it is determined that the waste has successfully fallen into the corresponding sorting bin, the preset standard waste category corresponding to the sorting bin is obtained; The waste category prediction results obtained in the approach phase are compared with the preset standard waste category corresponding to the sorting bin for consistency. If the two match, a confirmation voice message will be played to confirm successful waste disposal, and the corresponding environmental protection points will be added to the user's account. If they do not match, an error message will be played to indicate a waste disposal error, the user's incorrect disposal event will be recorded, no environmental protection points will be added, and waste sorting guidance information will be pushed to the user's linked terminal.

[0012] Secondly, this application provides an intelligent resource recycling bin, which includes a waste disposal prompting unit, the waste disposal prompting unit specifically including: The data acquisition module is used to collect image frame sequences captured by at least one depth camera set on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, throughout the entire waste disposal process, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments, which are set up one by one with each sorting and disposal port. The processing module is used to divide the disposal operation into an approach phase, an alignment phase, and a release phase based on the image frame sequence and through a behavior recognition algorithm. The phases are: the user entering a preset approach area as the starting point; the hand entering a preset alignment area above the disposal port as the first node; the user triggering a release action as the second node; and the garbage bag separating from the user's hand as the ending point. The module also tracks the movement trajectory of the garbage bag throughout the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin. The prompting module is also used to identify the characteristics of the garbage bag during the approach phase to obtain its garbage category prediction result, and generate preliminary prompt information to guide the user to the corresponding sorting and disposal port; The prompting module is also used to trigger a voice prompt to guide the user to adjust the disposal position if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port during the alignment stage. The prompting module is also used to confirm whether the garbage has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin after the release phase is completed and a preset weight stabilization period has elapsed; if it has fallen into the bin, a successful disposal confirmation voice will be played; if no effective weight change is detected, a disposal anomaly prompt will be triggered.

[0013] Thirdly, this application provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described method for prompting waste disposal for intelligent resource and environmentally friendly recycling bins.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for prompting waste disposal in an intelligent resource recycling bin.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: In the process of providing waste disposal prompts in the intelligent resource recycling bin: First, a depth camera installed on the upper front of the bin and weight sensors installed at the bottom of each sorting compartment simultaneously collect image frame sequences and real-time weight data of the entire disposal process, achieving spatiotemporal alignment of multi-source data and obtaining dynamic perception basic data for the entire disposal process; Second, based on the collected image frame sequences, a behavior recognition algorithm accurately divides the disposal operation into the approach stage, alignment stage, and release stage, tracking the movement trajectory of the waste bag throughout the process, achieving precise staged control of the entire disposal behavior chain; Third, a tiered system is set for different stages of the entire disposal process. The real-time prompting and error correction mechanism identifies garbage bag characteristics to predict garbage category and guides users to the corresponding disposal slot during the approach phase. During the alignment phase, it detects position deviations in real time and guides users to adjust. During the release phase, it confirms the garbage entry result through weight data. This solves the core pain points of traditional recycling bins, which lack dynamic guidance, pre-emptive error correction, and disposal result verification. Finally, by comparing the consistency between the disposed garbage category and the standard category of the target bin, it completes the closed-loop verification of classification compliance. Combined with environmental protection points incentives and classification guidance information push, it achieves long-term guidance for users to standardize disposal behavior. It can fully realize real-time prompting, dynamic error correction, and result confirmation throughout the entire garbage disposal process. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a waste disposal prompt method for a smart resource recycling bin, according to some embodiments of this application; Figure 2 This is a schematic diagram of the structure of a waste disposal prompting unit according to some embodiments of this application; Figure 3 This is a schematic diagram of the structure of a computer device that implements a waste disposal prompt method for a smart resource recycling bin, according to some embodiments of this application. Detailed Implementation

[0017] To better understand the above technical solutions, a detailed description of the technical solutions will be provided below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the following description of the embodiments is for illustration and explanation of this application and is not intended to limit this application. Where specific technologies or conditions are not specified in the embodiments, they are performed in accordance with the technologies or conditions described in the literature in the art or in accordance with the product instructions. Reagents or instruments used in the embodiments of this application whose manufacturers are not specified are all conventional products that can be obtained commercially.

[0018] refer to Figure 1 The figure is an exemplary flowchart of a waste disposal prompt method for a smart resource recycling bin, according to some embodiments of this application. The method mainly includes the following steps: In step 101, during the entire waste disposal process, the system collects a sequence of image frames captured by at least one depth camera located on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments, which correspond one-to-one with each sorting and disposal port.

[0019] A depth camera positioned at the top front of the recycling bin captures visual information of the user's disposal behavior, covering all sorting slots and the area in front of the bin to ensure complete monitoring of the entire process from the user approaching to releasing the waste. A weight sensor verifies the physical changes in the disposal result. In some embodiments, the acquisition of image frame sequences of the entire disposal process can be achieved in the following manner: A depth camera is installed in the center of the upper front of the recycling bin. The field of view of the depth camera completely covers all sorting and disposal ports on the bin, as well as the user approach and operation area in front of the bin within a preset range, such as 3 meters × 2 meters. This is only an example and is not intended to limit the specific scope of this application. The depth camera continuously acquires a sequence of depth images including the user's body, hands, and garbage bags, along with a synchronized sequence of color images. It also simultaneously acquires real-time weight data from the corresponding weight sensors in each sorting bin, thereby achieving spatiotemporal alignment of images and weight data.

[0020] It should be noted that the aforementioned spatiotemporal alignment refers to synchronizing the timestamps of image frames with the timestamps of weight data to ensure data consistency on the timeline, for example, through NTP (Network Time Protocol) or hardware synchronization signals. In specific implementations, depth cameras can use RGB-D sensors, such as Intel RealSense or Kinect, to ensure depth accuracy within 10cm to support precise position detection, which will not be elaborated further here.

[0021] In step 102, based on the image frame sequence, the behavior recognition algorithm divides the disposal operation into an approach phase (i.e., from the start point to the first node), an alignment phase (i.e., from the first node to the second node), and a release phase (i.e., from the second node to the end point), with the user entering the preset approach area as the starting point, the hand entering the preset alignment area above the disposal port as the first node, the user triggering the release action as the second node, and the separation of the garbage bag from the user's hand as the end point; and tracks the movement trajectory of the garbage bag during the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin.

[0022] The behavior recognition algorithm can employ a deep learning-based model, such as YOLO combined with LSTM sequence analysis, to detect user posture and action nodes. The preset approach area is a rectangular area of ​​3 meters × 2 meters in front of the bin for user access and operation, and the alignment area is a cubic space of 20cm × 20cm above the disposal opening. Separation of the garbage bag from the hand can be achieved through depth change and contour break detection.

[0023] In some embodiments, tracking the movement trajectory of garbage bags can be achieved in the following ways: Depth information and contour features of the user's hand region are extracted from the depth image frame sequence. Based on the depth threshold and instance segmentation model, accurate segmentation of the area where the hand is stuck to the garbage bag is completed. In the segmented garbage bag area, a deep learning target tracker is used to continuously associate the spatial position of the garbage bag in consecutive image frames from the start of the approach phase to the complete process of the garbage bag falling into the target sorting bin, thereby generating the complete motion trajectory of the garbage bag.

[0024] In practical implementation, the instance segmentation model can be Mask R-CNN, and the depth threshold can be set to the hand depth ±5cm to distinguish adhesion; the target tracker can use SORT or DeepSORT algorithm to ensure trajectory smoothness and robustness. In this application, the complete motion trajectory of the garbage bag can be represented as a three-dimensional coordinate sequence for subsequent deviation calculation and category prediction.

[0025] In step 103, during the approach phase, the characteristics of the garbage bag are identified to obtain its garbage category prediction result, and preliminary prompts are generated to guide the user to the corresponding sorting and disposal port.

[0026] The characteristics of the garbage bag include surface markings, color, and shape. In some embodiments, during the approach phase, identifying the characteristics of the garbage bag to obtain its garbage category prediction result can be achieved in the following ways: Detect whether there are pre-printed classification identification codes and classification color markings on the surface of the garbage bag from the image frame sequence; If a valid classification identifier or classification color identifier is detected, the waste category prediction result is obtained through decoding or color matching; If no valid classification identifier is detected or the recognition confidence level is lower than the preset threshold, a voice or screen prompt will be triggered before the user enters the alignment stage, guiding the user to manually select the waste category through the human-computer interaction interface of the recycling bin, and the result of the user's manual selection will be used as the waste category prediction result.

[0027] In practice, the classification identification code can be a QR code or a barcode, and the color code, for example, green for kitchen waste; the confidence threshold can be set to 0.8, and the initial prompts can be broadcast through the voice module, such as the voice module broadcasting "Please go to the recyclable waste disposal port" or guiding via LED indicator lights. This is only an example and is not intended to limit the specific scope of this application.

[0028] In step 104, during the alignment stage, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position.

[0029] In some embodiments, during the alignment phase, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position. Specifically, this can be achieved in the following ways: Around each sorting and disposal port, an area-array infrared detection array can be deployed to form a two-dimensional position detection field covering the preset effective disposal area of ​​the corresponding disposal port; or a depth camera can be used to obtain the spatial coordinates of the garbage bag and the target sorting and disposal port in real time. During the alignment phase, the relative positional deviation between the center of the garbage bag and the center of the effective disposal area of ​​the target sorting and disposal port is calculated in real time using the area array infrared detection array or depth camera. If the relative position deviation continues to exceed the preset deviation threshold and the duration exceeds the preset stability judgment duration, the user will be guided to move their hand in the specified direction via voice prompts before the user performs the release action, and the relative position status of the garbage bag and the effective disposal area will be displayed in real time on the screen of the bin.

[0030] The deviation threshold can be set to 5cm, the stability judgment time can be 300ms, and the voice prompt can be, for example, "Please move 10cm to the left." The screen display can be indicated by AR overlay arrows. This is only an example and is not intended to limit the specific scope of this application.

[0031] In step 105, after the release phase is completed and a preset weight stabilization period has elapsed, the system confirms whether the waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin. If the waste has fallen into the bin, a successful disposal confirmation voice message is played. If no effective weight change is detected, a disposal anomaly prompt is triggered.

[0032] In some embodiments, after the release phase is completed and a preset weight stabilization period has elapsed, the system confirms whether waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin. Specifically, this can be achieved in the following manner: Each sorting and disposal port corresponds to an independent sealed sorting bin, and each sorting bin is independently equipped with a weight sensor with matching weighing accuracy at the bottom. Based on the completion time of the release phase, after waiting for a preset weight stabilization period of 300ms-600ms, the stable weight value output by the corresponding classification bin weight sensor is obtained, and the difference between it and the baseline weight value collected within 100ms before the release phase is calculated as the actual weight change. If the actual weight change exceeds the preset minimum weight threshold, the waste is determined to have successfully fallen into the corresponding sorting bin; if it does not exceed the preset minimum weight threshold, the disposal is determined to be abnormal, triggering a check for blockage at the disposal port and recording the abnormal event simultaneously.

[0033] The minimum weight threshold can be set to 10g. Anomaly prompts may include a voice message "Delivery failed, please check the delivery port" and log entries for maintenance.

[0034] In the above embodiments, the physical verification of waste disposal results is completed through independent closed-loop weight acquisition and weight difference comparison under precise timing control, accurately determining whether the waste has successfully fallen into the corresponding sorting bin. The hardware foundation is an independent weighing unit, isolating interference. Each sorting disposal port corresponds to an independent sealed sorting bin, and a weight sensor with matching accuracy is installed separately at the bottom of the bin. Its core function is to ensure that the weight data of each bin is completely independent and is not affected by the disposal in other bins or the external environment, ensuring that the detected weight change comes only from the waste disposal in this bin, providing an interference-free hardware prerequisite for accurate judgment.

[0035] In addition, the above embodiments, through timing control, can avoid impact fluctuations and obtain the true weight, eliminating the weight value distortion caused by the impact force of the garbage bag falling and the shaking inside the container, and setting two levels of precise timing control: Before deployment: Take the weight of the pod within 100ms before the release action is completed as the baseline weight (i.e. the original net weight of the pod before deployment). After placement: Starting from the moment the release action is completed, wait for 300ms-600ms for the weight to stabilize. This is suitable for the height range of 0.5m-1.2m from the placement port of a regular recycling bin to the bottom of the bin. Ensure that the waste has fallen and the weight value has stabilized before reading the stable weight value, thus completely filtering out numerical errors caused by the impact of falling.

[0036] The core judgment in this application is based on weight difference comparison. The actual weight change is calculated by subtracting the baseline weight from the stable weight value. This difference is the actual weight of the garbage disposed of and serves as the sole criterion for judgment. If the difference exceeds the preset minimum weight threshold (10g): it indicates that there is a valid weight increase in the bin, and it is determined that the garbage has successfully fallen into the corresponding sorting bin; If the difference does not reach the threshold, it means there is no effective weight change in the bin, and the garbage has not fallen into the bin normally (mostly due to blockage at the disposal port, disposal error, etc.), and it is directly judged as an abnormal disposal.

[0037] It should be noted that the above embodiments use closed-loop processing. That is, if the delivery is successful based on the weight determination, the result is confirmed. If an anomaly is detected, a voice prompt is immediately triggered to prompt the user to check for a problem with the delivery port. At the same time, the abnormal event log is recorded for subsequent equipment maintenance, which can form a complete weight verification closed loop. This will not be elaborated further here.

[0038] It should be noted that the three stages of approaching, aligning, and releasing in the aforementioned embodiments only complete the basic functions of "guiding the disposal location and confirming the physical entry of waste into the bin," but cannot identify the classification error problem of "waste being disposed of in the bin, but the category does not match the bin location standard." Therefore, in some embodiments, the method may further include: Once it is determined that the waste has successfully fallen into the corresponding sorting bin, the preset standard waste category corresponding to the sorting bin is obtained; The waste category prediction results obtained in the approach phase are compared with the preset standard waste category corresponding to the sorting bin for consistency. If the two match, a confirmation voice message will be played to confirm successful waste disposal, and corresponding environmental protection points will be added to the user's account. If they do not match, an error message will be played to indicate a waste disposal error, the user's incorrect disposal event will be recorded, environmental protection points will not be added, and waste sorting guidance information will be pushed to the user's bound terminal. The points can be added by the user logging in by scanning a QR code, and the push can be done through a WeChat mini program or APP, which will not be elaborated here.

[0039] It should be noted that by adding a consistency comparison step in the above embodiments, the criteria for successful delivery are upgraded from physical warehousing to compliant warehousing based on category matching. This completes the entire closed loop from user approach, delivery guidance, location correction, warehousing confirmation to category compliance verification, filling the technical gap that traditional solutions cannot achieve real-time identification and immediate correction of category errors after delivery.

[0040] In another aspect, in some embodiments, this application provides an intelligent resource recycling bin, which includes a waste disposal prompt unit, as referenced. Figure 2 The figure is a schematic diagram of the structure of a waste disposal prompting unit according to some embodiments of this application. The waste disposal prompting unit 200 includes: a collection module 201, a processing module 202, and a prompting module 203, which are described below: The acquisition module 201 in this application is mainly used to collect image frame sequences captured by at least one depth camera set on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, during the entire process of garbage disposal, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments that correspond one-to-one with each sorting and disposal port. Processing module 202 in this application is mainly used to divide the disposal operation into an approach phase (from the start point to the first node), an alignment phase (from the first node to the second node), and a release phase (from the second node to the end point) based on the image frame sequence and through a behavior recognition algorithm, with the user entering a preset approach area as the starting point, the hand entering a preset alignment area above the disposal port as the first node, the user triggering a release action as the second node, and the separation of the garbage bag from the user's hand as the end point; and to track the movement trajectory of the garbage bag during the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin; The prompt module 203 in this application is mainly used to identify the characteristics of the garbage bag during the approach stage to obtain its garbage category prediction result, and generate preliminary prompt information to guide the user to the corresponding classification and disposal port; In a specific implementation, the prompting module 203 is also used to trigger a voice prompt to guide the user to adjust the disposal position if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port during the alignment stage. In addition, the prompting module 203 is also used to confirm whether the garbage has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin after the release phase is completed and a preset weight stabilization period has elapsed; if it has fallen into the bin, a voice message confirming successful disposal is played; if no effective weight change is detected, a disposal anomaly prompt is triggered.

[0041] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, the processor being configured to acquire the code and execute the above-described method for prompting waste disposal for intelligent resource and environmentally friendly recycling bins.

[0042] In some embodiments, reference Figure 3 This figure is a schematic diagram of the structure of a computer device implementing a waste disposal prompt method for a smart resource recycling bin, according to some embodiments of this application. The method in the above embodiments can be... Figure 3 The computer device shown is used to implement this, and the computer device 300 includes at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304.

[0043] The processor 301 may be a general-purpose central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more devices used to control the execution of the waste disposal prompting method for the smart resource recycling bin in this application.

[0044] The communication bus 302 may include a path for transmitting information between the aforementioned components.

[0045] The memory 303 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or it may be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), a magnetic disk or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but is not limited thereto. The memory 303 may exist independently and be connected to the processor 301 via a communication bus 302. The memory 303 may also be integrated with the processor 301.

[0046] The memory 303 stores program code for executing the scheme of this application, and its execution is controlled by the processor 301. The processor 301 executes the program code stored in the memory 303. The program code may include one or more software modules. In the above embodiment, the determination of the garbage disposal prompt can be implemented by the processor 301 and one or more software modules in the program code in the memory 303.

[0047] Communication interface 304 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0048] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0049] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0050] In addition, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for prompting waste disposal in a smart resource recycling bin.

[0051] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0052] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of the invention. Therefore, if these modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include these modifications and variations.

Claims

1. A method for providing waste disposal prompts for intelligent resource recycling bins, characterized in that, The method includes the following steps: Throughout the entire process of collecting waste disposal, the system collects image frame sequences captured by at least one depth camera located on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments, which correspond one-to-one with each sorting and disposal port. Based on the image frame sequence, the behavior recognition algorithm divides the disposal operation into an approach phase, an alignment phase, and a release phase, starting with the user entering a preset approach area, the first node being the hand entering a preset alignment area above the disposal port, the second node being the user triggering a release action, and the end point being the separation of the garbage bag from the user's hand. The movement trajectory of the garbage bag is tracked throughout the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin. During the approach phase, the characteristics of the garbage bags are identified to obtain their garbage category prediction results, and preliminary prompts are generated to guide the user to the corresponding sorting and disposal port; During the alignment phase, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position. After the release phase is completed and a preset weight stabilization period has elapsed, the system confirms whether the waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin. If the waste has fallen into the bin, a successful disposal confirmation voice message is played. If no valid weight change is detected, a disposal anomaly prompt is triggered.

2. The method according to claim 1, characterized in that, The image frame sequence of the entire acquisition and delivery process specifically includes: A depth camera is installed in the center of the upper front of the recycling bin. The field of view of the depth camera completely covers all sorting and disposal ports on the bin, as well as the 3m×2m user access and operation area in front of the bin. The depth camera continuously acquires a sequence of depth images including the user's body, hands, and garbage bags, along with a synchronized sequence of color images. It also simultaneously acquires real-time weight data from the corresponding weight sensors in each sorting bin, thus achieving spatiotemporal alignment of images and weight data.

3. The method according to claim 2, characterized in that, Tracking the movement of garbage bags specifically includes: Depth information and contour features of the user's hand region are extracted from the depth image frame sequence. Based on the depth threshold and instance segmentation model, accurate segmentation of the area where the hand is stuck to the garbage bag is completed. In the segmented garbage bag area, a deep learning target tracker is used to continuously associate the spatial position of the garbage bag in consecutive image frames from the start of the approach phase to the complete process of the garbage bag falling into the target sorting bin, thereby generating the complete motion trajectory of the garbage bag.

4. The method according to claim 1, characterized in that, In the approach phase, identifying the characteristics of the garbage bags to obtain their garbage category prediction results specifically includes: Detect whether there are pre-printed classification identification codes and classification color markings on the surface of the garbage bag from the image frame sequence; If a valid classification identifier or classification color identifier is detected, the waste category prediction result is obtained through decoding or color matching; If no valid classification identifier is detected, or the recognition confidence level is lower than a preset threshold, a voice or screen prompt will be triggered before the user enters the alignment stage, guiding the user to manually select the waste category through the human-computer interaction interface of the recycling bin, and the result of the user's manual selection will be used as the waste category prediction result; if the user does not complete the manual selection and directly enters the alignment stage, the waste category prediction result for this disposal will be marked as pending confirmation, and subsequent verification will be completed through category matching in the classification bin after disposal is completed; if the user does not complete the manual selection and directly enters the alignment stage, the waste category prediction result for this disposal will be marked as pending confirmation, and subsequent verification will be completed through category matching in the classification bin after disposal is completed.

5. The method according to claim 1, characterized in that, During the alignment phase, if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port, a voice prompt is triggered to guide the user to adjust the disposal position. Specifically, this includes: Around each sorting and disposal port, an area-array infrared detection array is deployed to form a two-dimensional position detection field covering the preset effective disposal area of ​​the corresponding disposal port; or the spatial coordinates of the garbage bag and the target sorting and disposal port are obtained in real time through the depth camera. During the alignment phase, the relative positional deviation between the center of the garbage bag and the center of the effective disposal area of ​​the target sorting and disposal port is calculated in real time using the area array infrared detection array or depth camera. If the relative position deviation continues to exceed the preset deviation threshold and the duration exceeds the preset stability judgment duration, the user will be guided to move their hand in the specified direction via voice prompts before the user performs the release action, and the relative position status of the garbage bag and the effective disposal area will be displayed in real time on the screen of the bin.

6. The method according to claim 1, characterized in that, After the release phase is completed and a preset weight stabilization period has elapsed, the system confirms whether the waste has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin. This specifically includes: Each sorting and disposal port corresponds to an independent sealed sorting bin, and each sorting bin is independently equipped with a weight sensor with matching weighing accuracy at the bottom. Based on the completion time of the release phase, after waiting for a preset weight stabilization period of 300ms-600ms, the stable weight value output by the corresponding classification bin weight sensor is obtained, and the difference between it and the baseline weight value collected within 100ms before the release phase is calculated as the actual weight change. If the actual weight change exceeds the preset minimum weight threshold, the waste is determined to have successfully fallen into the corresponding sorting bin; if it does not exceed the preset minimum weight threshold, the disposal is determined to be abnormal, triggering a check for blockage at the disposal port and recording the abnormal event simultaneously.

7. The method according to claim 1, characterized in that, The method further includes: Once it is determined that the waste has successfully fallen into the corresponding sorting bin, the preset standard waste category corresponding to the sorting bin is obtained; The waste category prediction results obtained in the approach phase are compared with the preset standard waste category corresponding to the sorting bin for consistency. If the two match, a confirmation voice message will be played to confirm successful waste disposal, and the corresponding environmental protection points will be added to the user's account. If they do not match, an error message will be played to indicate a waste disposal error, the user's incorrect disposal event will be recorded, no environmental protection points will be added, and waste sorting guidance information will be pushed to the user's linked terminal.

8. A smart resource recycling bin, comprising a waste disposal prompt unit, characterized in that, The waste disposal prompt unit specifically includes: The data acquisition module is used to collect image frame sequences captured by at least one depth camera set on the upper front of the recycling bin, covering all sorting and disposal ports and the area in front of the bin, throughout the entire waste disposal process, as well as real-time weight data output by weight sensors installed at the bottom of the sorting compartments, which are set up one by one with each sorting and disposal port. The processing module is used to divide the disposal operation into an approach phase, an alignment phase, and a release phase based on the image frame sequence and through a behavior recognition algorithm. The phases are: the user entering a preset approach area as the starting point; the hand entering a preset alignment area above the disposal port as the first node; the user triggering a release action as the second node; and the garbage bag separating from the user's hand as the ending point. The module also tracks the movement trajectory of the garbage bag throughout the entire process from the start of the approach phase to the garbage bag falling into the target sorting bin. The prompting module is also used to identify the characteristics of the garbage bag during the approach phase to obtain its garbage category prediction result, and generate preliminary prompt information to guide the user to the corresponding sorting and disposal port; The prompting module is also used to trigger a voice prompt to guide the user to adjust the disposal position if a positional deviation is detected between the garbage bag and the preset effective disposal area of ​​the target sorting and disposal port during the alignment stage. The prompting module is also used to confirm whether the garbage has fallen into the corresponding sorting bin based on the real-time weight data output by the weight sensor of the corresponding sorting bin after the release phase is completed and a preset weight stabilization period has elapsed; if it has fallen into the bin, a successful disposal confirmation voice will be played; if no effective weight change is detected, a disposal anomaly prompt will be triggered.

9. A computer device, characterized in that, The computer device includes a memory and a processor. The memory is used to store computer programs, and the processor is used to call and run the computer programs from the memory, causing the computer device to perform the waste disposal reminder method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions or code that, when executed on a computer, cause the computer to implement the garbage disposal prompt method as described in any one of claims 1 to 7.

Images