A cigarette management method, electronic device, medium and product applied to a heating smoking set

Abstract

The application discloses a cigarette management method applied to a heating smoking set, an electronic device, a medium and a product. The method comprises the following steps: obtaining image data of a cigarette inserted into a storage bin of a target heating smoking set, wherein the image data comprises one or more of a tobacco segment end face image, a tobacco segment side wall image, a filter segment end face image and a filter segment side wall image of the cigarette in the storage bin; processing the obtained image data through a multi-task image recognition module to obtain an image recognition result of the cigarette in the storage bin, wherein the image recognition result comprises one or more of a cigarette type recognition result, a cigarette use state recognition result and a cigarette insertion depth recognition result; determining a corresponding cigarette management strategy based on one or more of the cigarette type recognition result, the cigarette use state recognition result and the cigarette insertion depth recognition result in combination with a preset cigarette management strategy library, and managing and controlling the cigarette based on the cigarette management strategy. The accuracy and stability of temperature control of the heating smoking set are improved.

Description

Technical Field

[0001] This invention relates to the technical field of smoking device control, and more particularly to a method, electronic device, medium, and product for managing cigarettes in heated smoking devices. Background Technology

[0002] With the continuous development of new tobacco technologies, heated tobacco products (HTMS) have gradually become the mainstream alternative to traditional cigarettes and are widely used in the consumer market due to their advantages of low-temperature tobacco heating, flameless combustion, low release of harmful substances, and high safety. Existing consumer-grade HTMS use preset fixed heating curves to heat and atomize cigarettes to meet user smoking needs. However, the intelligence level of existing consumer-grade HTMS devices is generally low, lacking targeted cigarette status recognition and intelligent control mechanisms. The devices cannot recognize the cigarette's usage status, easily leading to users repeatedly inserting used cigarettes, causing dry burning and producing a burnt odor. This not only ruins the smoking experience but also poses a safety hazard of overheating and damage. Furthermore, the devices lack precise cigarette insertion depth detection, and improper user insertion and removal can easily lead to deviations in insertion depth, resulting in uneven heating and poor aerosol release stability. Additionally, most HTMS devices on the market use a single fixed heating curve, which cannot adapt to different specifications and types of cigarette materials, requiring users to manually switch and adjust parameters, making operation cumbersome and lacking adaptability. Therefore, the existing control of heated smoke appliances is insufficient to meet the needs of precise management of cigarettes, which seriously affects user experience and equipment safety. Summary of the Invention

[0003] This invention provides a method, electronic device, medium, and product for managing cigarettes in heated smoking appliances, in order to solve the problem that the control of heated smoking appliances is difficult to meet the needs of precise management of cigarettes.

[0004] According to one aspect of the present invention, a method for managing cigarettes applied to heated smoking appliances is provided, comprising: Acquire image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette inside the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. Based on one or more of the images of the end face of the tobacco section, the side wall of the tobacco section, the end face of the filter section, and the side wall of the filter section, the images are processed by a multi-task image recognition module to obtain the image recognition results of the cigarettes in the storage. The image recognition results include one or more of the following: cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. Based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, and in conjunction with a preset cigarette management strategy library, a corresponding cigarette management strategy is determined, and cigarette control is carried out based on the cigarette management strategy.

[0005] Optionally, acquiring image data of the cigarette insertion chamber of the target heated smoking device includes: simultaneously acquiring multiple frames of original images of the cigarette inside the chamber using image acquisition devices positioned at at least two different angles on the inner wall of the cigarette insertion chamber. These multiple frames of original images include one or more of the following: original images of the tobacco section end face, original images of the tobacco section side wall, original images of the filter section end face, and original images of the filter section side wall. A preset sharpness determination algorithm is used to calculate the sharpness evaluation results corresponding to each of the multiple frames of original images. Based on the sharpness evaluation results and preset image filtering conditions, various images of the multiple frames of original images are sorted and filtered to obtain image data of the cigarette insertion chamber. The preset sharpness determination algorithm includes the Laplace variance algorithm.

[0006] Optionally, the multi-task image recognition module includes a shared convolutional feature extraction backbone network and a multi-task recognition network. Based on one or more of the following images—the end face image of the tobacco shreds, the side wall image of the tobacco shreds, the end face image of the filter tip, and the side wall image of the filter tip—the multi-task image recognition module processes the images to obtain the image recognition results of the cigarettes in the storage compartment. This includes: inputting one or more of the following images into the shared convolutional feature extraction backbone network for unified feature extraction to obtain a globally shared feature extraction result. The globally shared feature extraction result includes tobacco color distribution features, cigarette outer wall texture features, cigarette brand identification features, tobacco shreds end face deformation and color change features, filter tip color distribution features, filter tip surface texture features, filter tip outline edge features, and cigarette side wall reference mark position features. The various features in the globally shared feature extraction result are then input into the multi-task recognition network, which processes them separately through parallel task branches and outputs the image recognition results in parallel.

[0007] Optionally, the multi-task recognition network includes a cigarette type recognition network, a cigarette usage status recognition network, and a cigarette insertion depth recognition network. Features from the globally shared feature extraction results are input into the multi-task recognition network, which processes them separately through parallel task branches and outputs image recognition results in parallel. These include: if tobacco color distribution features, cigarette outer wall texture features, and cigarette brand identification features are extracted, these features are input into the cigarette type recognition network and processed in conjunction with a pre-set cigarette type feature library to obtain the cigarette type recognition result. The cigarette type recognition result includes an identifier indicating whether the cigarette is registered and cigarette brand information; if filter color distribution features, filter surface features, etc., are extracted... Texture features, filter tip outline edge features, and tobacco segment end-face deformation and color change features are input into the cigarette usage status recognition network. These features are then processed in conjunction with a preset cigarette usage status feature template to obtain the cigarette usage status recognition result, which includes both used and unused states. If the cigarette sidewall reference mark position features are extracted, these features are input into the cigarette insertion depth recognition network. This network is then processed in conjunction with a preset reference size for the cigarette insertion chamber to obtain the cigarette insertion depth recognition result, which includes the deviation value of the cigarette from the preset standard insertion depth.

[0008] Optionally, based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, and in combination with a preset cigarette management strategy library, a corresponding cigarette management strategy is determined. This includes: retrieving the preset cigarette management strategy library; matching one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results with various control strategies in the preset cigarette management strategy library to obtain matching results; and selecting one or more strategies with the highest matching degree from the matching results to combine them into a cigarette management strategy.

[0009] Optionally, the control strategies in the preset cigarette management strategy library include: if the cigarette registration status in the cigarette type identification result is unregistered, a cigarette registration error message is generated to remind the user to complete the cigarette information registration; if the cigarette registration status in the cigarette type identification result is registered and the cigarette usage status identification result is unused, the appropriate cigarette heating parameters are matched based on the cigarette brand information in the cigarette type identification result, a cigarette heating command is generated based on the cigarette heating parameters, and the cigarette heating operation is controlled based on the cigarette heating command; if the cigarette usage status identification result is used, an audible and visual warning is triggered, and the cigarette heating function is prohibited; if the deviation value in the cigarette insertion depth identification result is greater than the preset allowable deviation range, an insertion depth adjustment prompt message is generated based on the deviation value to remind the user to adjust the cigarette insertion position to within the standard range.

[0010] Optionally, the method also includes: when sufficient cigarette usage feedback data has been collected, or when the recognition accuracy of the multi-task image recognition module drops to a set threshold, obtaining the currently effective cigarette management strategy and the corresponding actual cigarette usage feedback data, using the cigarette management strategy and the actual cigarette usage feedback data as incremental training samples, and sending them back to the multi-task image recognition module for iterative optimization training to obtain the optimized multi-task image recognition module.

[0011] According to another aspect of the present invention, a cigarette management device for use in heated smoking appliances is provided, comprising: The image data acquisition module is used to acquire image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette in the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. The image recognition result determination module is used to process one or more of the following images based on the end face image of the tobacco section, the side wall image of the tobacco section, the end face image of the filter section, and the side wall image of the filter section through the multi-task image recognition module to obtain the image recognition result of the cigarettes in the storage. The image recognition result includes one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result. The cigarette management module is used to determine the corresponding cigarette management strategy based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, combined with a preset cigarette management strategy library, and to manage cigarettes based on the cigarette management strategy.

[0012] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and A memory that is communicatively connected to at least one processor; wherein, The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the cigarette management method for a heated smoking appliance according to any embodiment of the present invention.

[0013] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement a cigarette management method for a heated smoking appliance according to any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement a cigarette management method for a heated smoking appliance according to any embodiment of the present invention.

[0015] The technical solution of this invention acquires image data of a cigarette inserted into the chamber of a target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. This achieves omnidirectional image acquisition from the end face and side wall, ensuring complete information coverage and avoiding feature loss due to a single viewpoint, significantly improving the reliability of subsequent identification. Based on one or more of the images of the end face of the tobacco section, the image of the side wall of the tobacco section, the image of the end face of the filter section, and the image of the side wall of the filter section, a multi-task image recognition module processes the data to obtain the image recognition result of the cigarette in the chamber. The image recognition result includes a cigarette type identification result and a cigarette usage status identification result. The system simultaneously completes multiple recognition tasks by analyzing one or more of the results from the cigarette insertion depth identification. This eliminates the need for sequential calculations by different modules, simplifying the system architecture, reducing computational power consumption, shortening recognition time, and enhancing real-time performance. Based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, and combined with a preset cigarette management strategy library, the system determines the corresponding cigarette management strategy. Cigarette control is then implemented based on this strategy. This allows for decision-making based on multi-dimensional identification results combined with a standardized strategy library, ensuring unified judgment logic and standardized control rules. This enables intelligent, refined, and automated control, improving equipment operational stability and operational safety. This solution ensures comprehensive recognition information through multi-angle image acquisition. Multi-task recognition can output multiple key data points at once, obtaining cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. It is highly efficient and highly integrated. Based on the multi-dimensional recognition results and a standardized strategy library, it can manage and control the cigarettes with accurate judgment and rapid response. It can realize intelligent and refined management of cigarettes, avoiding the trouble of manual adjustment by users and the problems of uneven heating and low efficiency caused by insertion depth deviation, thus improving the standardization of cigarette use and the user experience.

[0016] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a cigarette management method applied to heated smoking appliances provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of a cigarette management method for heated smoking appliances provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of a cigarette management device for heated smoking appliances provided in Embodiment 3 of the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device that implements the cigarette management method for heated smoking appliances according to embodiments of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] Example 1 Figure 1 This is a flowchart of a cigarette management method for heated smoking appliances provided in Embodiment 1 of the present invention. This embodiment is applicable to situations involving cigarette management in heated smoking appliances. The method can be executed by a cigarette management device for heated smoking appliances, which can be implemented in hardware and / or software. This cigarette management device can be configured in electronic devices such as the smart chip and controller of the heated smoking appliance. Figure 1 As shown, the method includes: S110. Obtain image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette in the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section.

[0022] The target heated tobacco device refers to a specialized device used to heat a cigarette for smoking, and can be any heated tobacco device currently in use by any target user. The image data is a collection of multi-angle visual images acquired by an image acquisition device within the cigarette insertion chamber, including but not limited to images of the tobacco section end face, tobacco section sidewall, filter section end face, and filter section sidewall. The tobacco section end face image is an image of the top cross-section of the tobacco portion of the cigarette; the tobacco section sidewall image is an image of the outer cylindrical surface of the tobacco section; the filter section end face image is an image of the cross-section of the end of the cigarette filter; and the filter section sidewall image is an image of the outer cylindrical surface of the filter section. These four types of images comprehensively present the appearance characteristics of the cigarette from different locations and perspectives.

[0023] Specifically, the image acquisition device captures images of the cigarette inserted into the chamber of the heated smoking device, and obtains one or more of the following images: the end face of the tobacco section, the side wall of the tobacco section, the end face of the filter section, and the side wall of the filter section, capturing the appearance features of the cigarette from multiple perspectives.

[0024] In this embodiment, by using multi-angle, full-position image acquisition, the details of the cigarette can be fully preserved, making up for the lack of information from a single shooting perspective, providing sufficient and reliable raw data for subsequent image recognition, and effectively reducing recognition errors caused by feature omissions.

[0025] Optionally, acquiring image data of the cigarette insertion chamber of the target heated smoking device includes: simultaneously acquiring multiple frames of original images of the cigarette inside the chamber using image acquisition devices positioned at at least two different angles on the inner wall of the cigarette insertion chamber. The multiple frames of original images include one or more of the following: original images of the tobacco section end face, original images of the tobacco section sidewall, original images of the filter section end face, and original images of the filter section sidewall. A preset sharpness determination algorithm is used to calculate the sharpness evaluation results corresponding to each of the multiple frames of original images. Based on the sharpness evaluation results and preset image filtering conditions, various images of the multiple frames of original images are sorted and filtered to obtain image data of the cigarette insertion chamber. The preset sharpness determination algorithm includes the Laplace variance algorithm.

[0026] The preset sharpness determination algorithm is an image quality evaluation rule based on the Laplacian variance algorithm, used to quantitatively judge the sharpness of images. The sharpness evaluation result is a numerical value characterizing the image sharpness. The value directly reflects the sharpness of a single frame image; the higher the value, the richer the image details and the clearer the picture, providing a quantitative basis for subsequent image selection. The preset sharpness determination algorithm can be called to calculate the sharpness evaluation results corresponding to multiple original images. The preset image selection conditions refer to the pre-set conditions for image selection based on image sharpness, used to select valid image data. It can be set to determine the image with the highest sharpness evaluation result as the image data for subsequent image processing, or the image with image sharpness meeting the preset sharpness threshold can be set as the valid image. For example, the image with the highest sharpness evaluation result among multiple original images of the tobacco section end face, multiple original images of the tobacco section side wall, multiple original images of the filter section end face, and multiple original images of the filter section side wall can be summarized into the image data of the cigarette insertion chamber of the target heated tobacco device.

[0027] Specifically, multiple frames of original images of the cigarettes inside the insertion chamber are simultaneously captured using image acquisition devices at at least two different angles on the inner wall of the chamber. These images can cover different parts of the cigarette, including the end face and side wall of the tobacco section, as well as the end face and side wall of the filter section. A pre-defined sharpness determination algorithm, the Laplace variance algorithm, is then used to calculate the sharpness evaluation result for each frame of the original image. Finally, this evaluation result is combined with pre-defined image filtering conditions to sort and filter all the original images. Images that meet the pre-defined image filtering conditions are identified as the final usable image data for the cigarette insertion chamber. For example, if multiple images among the multiple frames of original images meet the pre-defined sharpness threshold in the pre-defined image filtering conditions, the image with the highest sharpness evaluation result is identified as the final usable image data for the cigarette insertion chamber.

[0028] For example, a miniature CMOS camera (e.g., pixel density ≥ 3 million, supporting macro photography) and a ring-shaped LED fill light are installed at the cigarette insertion port of the heated tobacco device. This camera is used to capture one or more of the following images when the cigarette is inserted: the original image of the tobacco end face, the original image of the tobacco sidewall, the original image of the filter end face, and the original image of the filter sidewall. The camera's optical axis forms a 15-30 degree angle with the cigarette axis to ensure that the end face and part of the sidewall images are captured simultaneously.

[0029] In this embodiment, simultaneous shooting using multi-view devices can fully capture the features of each part of the cigarette. Combined with the Laplace variance algorithm, the image clarity is accurately evaluated and the best image is selected, effectively avoiding problems such as blurring and ghosting, ensuring the quality of the original image, providing a highly reliable data source for subsequent image recognition, and improving the overall recognition accuracy.

[0030] S120. Based on one or more of the following images: end face image of tobacco section, side wall image of tobacco section, end face image of filter section, and side wall image of filter section, the image is processed by a multi-task image recognition module to obtain the image recognition result of the cigarette in the compartment. The image recognition result includes one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result.

[0031] The multi-task image recognition module can be understood as a network unit integrating feature extraction and multi-class judgment functions. It is a functional unit capable of simultaneously performing multiple recognition analyses on multiple image data sets. It can extract image features in parallel and complete various judgments. It may include a shared convolutional feature extraction backbone network and a multi-task recognition network. The shared convolutional feature extraction backbone network is used to extract features from the image data, obtaining the feature extraction results. The multi-task recognition network is used to perform multi-task recognition processing on the feature extraction results, obtaining the image recognition results. The image recognition results can be specifically understood as the judgment information output by the multi-task image recognition module, including but not limited to cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results.

[0032] Specifically, one or more of the following images are input into the multi-task image recognition module: the end face image of the tobacco section, the side wall image of the tobacco section, the end face image of the filter section, and the side wall image of the filter section. The module then extracts the image features and analyzes them simultaneously, and outputs one or more of the following three types of recognition results: cigarette type, usage status, and insertion depth.

[0033] In this embodiment, by integrating multi-dimensional image information, the identification basis is more sufficient, and multiple detections are completed in parallel by a single module without splitting the processing flow. This reduces computing power consumption, improves computing efficiency, and ensures the accuracy and real-time performance of various identification results.

[0034] Optionally, the multi-task image recognition module includes a shared convolutional feature extraction backbone network and a multi-task recognition network. Based on one or more of the following images—the end face image of the tobacco shreds, the side wall image of the tobacco shreds, the end face image of the filter tip, and the side wall image of the filter tip—the multi-task image recognition module processes the images to obtain the image recognition results of the cigarettes in the storage compartment. This includes: inputting one or more of the following images into the shared convolutional feature extraction backbone network for unified feature extraction to obtain a globally shared feature extraction result. The globally shared feature extraction result includes tobacco color distribution features, cigarette outer wall texture features, cigarette brand identification features, tobacco shreds end face deformation and color change features, filter tip color distribution features, filter tip surface texture features, filter tip outline edge features, and cigarette side wall reference mark position features. The various features in the globally shared feature extraction result are then input into the multi-task recognition network, which processes them separately through parallel task branches and outputs the image recognition results in parallel.

[0035] The shared convolutional feature extraction backbone network is a fundamental network built upon a convolutional neural network (CNN) model. It's used to extract visual features from multiple images uniformly, focusing on uncovering commonalities and details. The multi-task recognition network, based on its parallel branches, utilizes the extracted features to complete different recognition tasks. It can be constructed using a lightweight multi-task network (MobileNetV3). The globally shared feature extraction result can be understood as a set of various visual features output by the backbone network. This serves as core data for subsequent recognition, encompassing information such as cigarette color, texture, contour, deformation, markings, and reference markers, providing comprehensive data support for subsequent recognition tasks.

[0036] Specifically, one or more of the following images are input into a recognition module composed of a shared convolutional feature extraction backbone network and a multi-task recognition network: the end face image of the tobacco shreds, the side wall image of the tobacco shreds, the end face image of the filter tip, and the side wall image of the filter tip. The shared convolutional feature extraction backbone network extracts multiple types of globally shared features, including but not limited to the color, texture, contour, deformation, identification, and reference marker features of the tobacco shreds and filter tip. The extracted features are then input into the multi-task recognition network, which performs calculations separately using parallel task branches and outputs the recognition results synchronously.

[0037] In this embodiment, by sharing backbone network reuse features, redundant calculations are avoided, significantly reducing computing power consumption and runtime latency; comprehensive and rich feature dimensions ensure sufficient basis for identification and judgment, and the parallel branch processing mode can efficiently complete multiple identification tasks, taking into account both overall operating efficiency and identification accuracy.

[0038] Optionally, the multi-task recognition network includes a cigarette type recognition network, a cigarette usage status recognition network, and a cigarette insertion depth recognition network. Features from the globally shared feature extraction results are input into the multi-task recognition network, which processes them separately through parallel task branches and outputs image recognition results in parallel. These include: if tobacco color distribution features, cigarette outer wall texture features, and cigarette brand identification features are extracted, these features are input into the cigarette type recognition network and processed in conjunction with a pre-set cigarette type feature library to obtain the cigarette type recognition result. The cigarette type recognition result includes an identifier indicating whether the cigarette is registered and cigarette brand information; if filter color distribution features, filter surface features, etc., are extracted... Texture features, filter tip outline edge features, and tobacco segment end-face deformation and color change features are input into the cigarette usage status recognition network. These features are then processed in conjunction with a preset cigarette usage status feature template to obtain the cigarette usage status recognition result, which includes both used and unused states. If the cigarette sidewall reference mark position features are extracted, these features are input into the cigarette insertion depth recognition network. This network is then processed in conjunction with a preset reference size for the cigarette insertion chamber to obtain the cigarette insertion depth recognition result, which includes the deviation value of the cigarette from the preset standard insertion depth.

[0039] Specifically, the globally shared features are input into the three corresponding branch networks of the multi-task recognition network for parallel processing, and the corresponding features are matched with the corresponding sub-networks to carry out the recognition work. If the tobacco color, outer wall texture, and brand logo features are extracted, they are input into the cigarette type recognition network, and the cigarette registration status and brand information are determined by comparing them with the preset feature library. If the filter color, texture, outline, and tobacco end face deformation and color features are extracted, they are input into the cigarette usage status recognition network, and the status template is matched to distinguish whether the cigarette has been used or not. If the side wall reference mark position features are extracted, they are input into the insertion depth recognition network, and the insertion depth deviation value is calculated by combining it with the chamber reference size. Finally, one or more recognition results are output. For example, the tobacco color distribution features, cigarette outer wall texture features, and cigarette brand identification features are input into the cigarette type recognition network, and similarity matching is performed in combination with a preset cigarette type feature library. The cigarette type recognition result is obtained when the matching threshold is not less than 90%. The filter color distribution features, filter surface texture features, and tobacco segment end face deformation and color change features are input into the cigarette usage status recognition network, and comparison is performed in combination with a preset cigarette usage status feature template. The confidence level is output according to the Sigmoid activation function, and the judgment threshold is not less than 95%, to obtain the cigarette usage status recognition result. The cigarette side wall reference mark position features are input into the cigarette insertion depth recognition network, and the pixel-to-physical distance mapping relationship obtained by the pre-completed camera calibration is used for calculation. The insertion depth judgment tolerance is ±0.5mm, to obtain the cigarette insertion depth recognition result.

[0040] In this embodiment, corresponding features are accurately assigned according to task characteristics, and judgment is made by combining a dedicated reference library and templates, resulting in highly targeted and accurate identification; each branch operates in parallel without interfering with each other, resulting in fast processing speed, and the clearly defined network structure also facilitates individual optimization and maintenance in the later stages.

[0041] S130. Based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, and in conjunction with a preset cigarette management strategy library, determine the corresponding cigarette management strategy, and control cigarettes based on the cigarette management strategy.

[0042] The preset cigarette management strategy library is a collection of pre-stored control rules and judgment logic, integrating handling solutions corresponding to different cigarette types, usage states, and insertion depths. The cigarette management strategy is a specific execution rule retrieved from the strategy library based on the identification results, used to guide the equipment to complete corresponding cigarette control actions. Optionally, the construction steps of the preset cigarette management strategy library include: sorting out various combinations of cigarette types, usage states, and insertion depths; setting judgment conditions, trigger actions, and execution parameters for each condition; classifying and storing all rules and completing logic verification; and completing the strategy library construction after testing and optimization in actual scenarios. The determination steps of the cigarette management strategy include: retrieving the preset cigarette management strategy library; matching at least one of the cigarette type identification results, cigarette usage state identification results, and cigarette insertion depth identification results with the control strategies in the library; selecting the single or multiple strategies with the highest matching degree and combining them to obtain the final cigarette management strategy.

[0043] Specifically, the identification results of one or more of the cigarette type, usage status, and insertion depth are matched with the rules in the preset cigarette management strategy library to determine the exclusive cigarette management strategy, and the corresponding control actions are strictly executed in accordance with the strategy.

[0044] In this embodiment, identification information can be flexibly combined to adapt to different scenarios. The standardized strategy library makes the control logic uniform and standardized, with fast response speed, enabling automated and refined control, and ensuring the safe operation and compliance of equipment.

[0045] Based on the above embodiments, the method further includes: when sufficient cigarette usage feedback data has been collected, or when the recognition accuracy of the multi-task image recognition module drops to a set threshold, obtaining the currently effective cigarette management strategy and the corresponding actual cigarette usage feedback data, using the cigarette management strategy and the actual cigarette usage feedback data as incremental training samples, and sending them back to the multi-task image recognition module for iterative optimization training to obtain the optimized multi-task image recognition module.

[0046] Among them, the sufficient cigarette usage feedback data refers to real-world scenario data collected by the system over a long period, including the usage status and control execution results of cigarettes that meet the preset quantity standards. The set threshold is a pre-defined critical value for recognition accuracy, used to determine whether to trigger the model optimization process. For example, it can be set that when the actual recognition accuracy of the module is lower than this value, the model optimization process can be triggered to optimize the multi-task image recognition module.

[0047] Specifically, when sufficient feedback data on cigarette usage has been collected, or when the recognition accuracy of the multi-task image recognition module drops to a preset threshold, the system extracts the currently effective cigarette management strategy and the corresponding actual cigarette usage feedback data, and uses them as incremental training samples to send back to the recognition module for iterative training, ultimately completing module optimization.

[0048] In this embodiment, by setting a model optimization mechanism, real-world scene data can be supplemented in real time, model defects can be continuously corrected, recognition accuracy can be effectively improved and stabilized, the module can be continuously adapted to actual use scenarios, and training costs can be reduced according to the incremental training mode, thereby improving the adaptive capability and long-term operational reliability of the entire system.

[0049] The technical solution of this embodiment acquires one or more of the following image data: the end face and side wall of the tobacco section, filter section, and other four types of images of the cigarette inserted into the heating device's compartment. These images are then input into a multi-task image recognition module, which simultaneously identifies the cigarette type, usage status, and insertion depth. Finally, it combines one or more recognition results to retrieve a preset cigarette management strategy library, matches and executes corresponding control operations. This solution ensures comprehensive recognition information through multi-angle image acquisition. Multi-task recognition can output multiple key data points at once, obtaining cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. It is highly efficient and integrated. Control is achieved by combining multi-dimensional recognition results with a standardized strategy library, resulting in accurate judgment and rapid response. This enables intelligent and refined cigarette management, avoiding the inconvenience of manual adjustment and the problems of uneven heating and low efficiency caused by insertion depth deviations, thus improving the standardization of cigarette device use and the user experience.

[0050] Example 2 Figure 2 This is a flowchart of a cigarette management method applied to heated smoking appliances provided in Embodiment 2 of the present invention. The method in this embodiment is a further optimization of the method in the above embodiments. Optionally, a preset cigarette management strategy library is retrieved, and one or more of the following are matched with various control strategies in the preset cigarette management strategy library based on cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results to obtain matching results; from the matching results, one or more strategies with the highest matching degree are selected and combined to form a cigarette management strategy. For example... Figure 2 As shown, the method includes: S210. Obtain image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette in the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section.

[0051] S220. Based on one or more of the following images: end face image of tobacco section, side wall image of tobacco section, end face image of filter section, and side wall image of filter section, the image is processed by a multi-task image recognition module to obtain the image recognition result of the cigarette in the compartment. The image recognition result includes one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result.

[0052] S230. Retrieve the preset cigarette management strategy library, and match it with various control strategies in the preset cigarette management strategy library based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results to obtain the matching results.

[0053] Specifically, the system calls a preset cigarette management strategy library, compares one or more identification results (cigarette type, usage status, insertion depth) with various control strategies in the library one by one, and finally outputs the corresponding matching result. For example, it can calculate the similarity between one or more identification results (cigarette type, usage status, insertion depth) and each record in various control strategies in the library. If there is a record with a similarity higher than a preset similarity threshold, the matching result is considered successful; otherwise, the matching result is considered unsuccessful.

[0054] In this embodiment, logical comparison is completed based on a standardized strategy library. The matching rules are clear, the execution speed is fast, and multiple types of identification information can be flexibly combined for judgment, adapting to different use scenarios and ensuring that the matching results are objective, consistent and reliable.

[0055] Optionally, the control strategies in the preset cigarette management strategy library include: if the cigarette registration status in the cigarette type identification result is unregistered, a cigarette registration error message is generated to remind the user to complete the cigarette information registration; if the cigarette registration status in the cigarette type identification result is registered and the cigarette usage status identification result is unused, the appropriate cigarette heating parameters are matched based on the cigarette brand information in the cigarette type identification result, a cigarette heating command is generated based on the cigarette heating parameters, and the cigarette heating operation is controlled based on the cigarette heating command; if the cigarette usage status identification result is used, an audible and visual warning is triggered, and the cigarette heating function is prohibited; if the deviation value in the cigarette insertion depth identification result is greater than the preset allowable deviation range, an insertion depth adjustment prompt message is generated based on the deviation value to remind the user to adjust the cigarette insertion position to within the standard range.

[0056] Specifically, the preset cigarette management strategy library contains multiple judgment rules. The system performs differentiated control based on the rules in the preset cigarette management strategy library. Specifically: when the cigarette type identification result shows that the cigarette registration status is not registered, a registration error prompt is immediately generated to guide the user to complete the information entry; when the cigarette type identification result shows that the cigarette registration status is registered and the cigarette usage status identification result is unused, the corresponding heating parameters are matched with the brand information and an instruction is issued to automatically execute the heating operation; when the cigarette usage status identification result shows that the cigarette is used, an audible and visual warning is activated and the heating function is locked; when the cigarette insertion depth deviation is detected to be greater than the preset allowable deviation range, a prompt message is generated and pushed to the user, reminding the user to adjust the cigarette insertion position to within the standard range.

[0057] In this embodiment, the control rules in the preset cigarette management strategy library cover various usage scenarios. The logic is clear and the classification is well-defined, which can realize automated judgment and handling in all scenarios. It can not only standardize the use of equipment and avoid ineffective operations and safety hazards, but also accurately adjust the operating parameters according to the characteristics of cigarettes, thereby improving the user experience and the safety of equipment operation.

[0058] S240. Select one or more strategies with the highest matching degree from the matching results and combine them to form a cigarette management strategy.

[0059] Specifically, the system selects one or more control strategies with the highest matching degree from the strategy matching results and integrates them into the final cigarette management strategy. For example, if it identifies that a cigarette is registered, in use, and has an excessive insertion depth deviation, the system selects the two strategies with the highest matching degree from the matching results: audible and visual warning, prohibition of heating, and insertion position adjustment prompt. It combines these two strategies to form a final cigarette management strategy that simultaneously executes warnings, locks heating, and provides insertion position reminders.

[0060] In this embodiment, the method prioritizes the rule with the best adaptability and can also flexibly combine multiple strategies to work together, adapt to complex working conditions, ensure that the control plan fits the actual situation, and make the handling logic more comprehensive and the execution effect more accurate.

[0061] S250. Control cigarettes based on cigarette management strategies.

[0062] The technical solution of this embodiment collects four types of image data: the end face and side wall of the tobacco section, the filter section, and the tobacco insertion chamber of the heated smoking device. These images are then input into a multi-task image recognition module to calculate three types of recognition results: cigarette type, usage status, and insertion depth. Subsequently, a preset cigarette management strategy library is retrieved, and the recognition results are matched one by one with the control rules in the library. The single or multiple strategies with the highest matching degree are selected and combined into a final management strategy. Finally, the entire process of cigarette management is completed based on this strategy. This solution ensures comprehensive recognition information through multi-angle image acquisition. Multi-task recognition can output multiple key data points at once, obtaining cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. It is highly efficient and highly integrated. Management is based on a standardized strategy library combined with multi-dimensional recognition results. The strategy matching and combination modes are flexible, ensuring not only accurate recognition and rapid response but also adaptability to different usage scenarios for differentiated management. This avoids the inconvenience of manual adjustment by users and the problems of uneven heating and low efficiency caused by insertion depth deviation, improving the intelligent and refined management of cigarettes and enhancing the standardization and user experience of smoking devices.

[0063] Example 3 Figure 3 This is a schematic diagram of a cigarette management device applied to a heated smoking appliance, provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: The image data acquisition module 310 is used to acquire image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette in the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. The image recognition result determination module 320 is used to process one or more of the following images based on the end face image of the tobacco section, the side wall image of the tobacco section, the end face image of the filter section, and the side wall image of the filter section through the multi-task image recognition module to obtain the image recognition result of the cigarette in the compartment. The image recognition result includes one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result. The cigarette management module 330 is used to determine the corresponding cigarette management strategy based on one or more of the cigarette type identification result, cigarette usage status identification result, and cigarette insertion depth identification result, combined with a preset cigarette management strategy library, and to manage cigarettes based on the cigarette management strategy.

[0064] The technical solution of this embodiment involves acquiring image data of the cigarette inserted into the chamber of the target heated smoking device through an image data acquisition module. The image data includes one or more of the following: an image of the end face of the tobacco section, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. An image recognition result determination module processes one or more of the following images through a multi-task image recognition module to obtain the image recognition result of the cigarette in the chamber. The image recognition result includes one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result. A cigarette management module determines the corresponding cigarette management strategy based on one or more of the following: cigarette type recognition result, cigarette usage status recognition result, and cigarette insertion depth recognition result, combined with a preset cigarette management strategy library, and performs cigarette management based on the cigarette management strategy. This solution ensures comprehensive recognition information through multi-angle image acquisition. Multi-task recognition can output multiple key data points at once, obtaining cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. It is highly efficient and highly integrated. Based on the multi-dimensional recognition results and a standardized strategy library, it can manage and control the cigarettes with accurate judgment and rapid response. It can realize intelligent and refined management of cigarettes, avoiding the trouble of manual adjustment by users and the problems of uneven heating and low efficiency caused by insertion depth deviation, thus improving the standardization of cigarette use and the user experience.

[0065] Based on the above embodiments, optionally, the image data acquisition module 310 is specifically used to simultaneously acquire multiple frames of original images of the cigarette inside the chamber through image acquisition devices set at at least two different angles on the inner wall of the cigarette insertion chamber. The multiple frames of original images include one or more of the original images of the tobacco section end face, the tobacco section side wall, the filter section end face, and the filter section side wall. A preset sharpness determination algorithm is used to calculate the sharpness evaluation results corresponding to the multiple frames of original images. Based on the sharpness evaluation results and preset image filtering conditions, the various images of the multiple frames of original images are sorted and filtered to obtain the image data of the cigarette insertion chamber. The preset sharpness determination algorithm includes the Laplace variance algorithm.

[0066] Optionally, the multi-task image recognition module includes a shared convolutional feature extraction backbone network and a multi-task recognition network; the image recognition result determination module 320 is specifically used to input one or more of the following images—the end face image of the tobacco shreds, the side wall image of the tobacco shreds, the end face image of the filter tip, and the side wall image of the filter tip—to the shared convolutional feature extraction backbone network for unified feature extraction, obtaining a globally shared feature extraction result. This globally shared feature extraction result includes tobacco color distribution features, cigarette outer wall texture features, cigarette brand identification features, tobacco shreds end face deformation and color change features, filter tip color distribution features, filter tip surface texture features, filter tip outline edge features, and cigarette side wall reference mark position features. The various features in the globally shared feature extraction result are input to the multi-task recognition network, which processes them separately through parallel task branches and outputs the image recognition result in parallel.

[0067] Optionally, the image recognition result determination module 320 is specifically used to: if the extracted tobacco color distribution features, cigarette outer wall texture features, and cigarette brand identification features are obtained, input the tobacco color distribution features, cigarette outer wall texture features, and cigarette brand identification features into the cigarette type recognition network, process them in conjunction with a preset cigarette type feature library, and obtain the cigarette type recognition result, which includes an identifier indicating whether the cigarette is registered and cigarette brand information; if the extracted filter color distribution features, filter surface texture features, filter outline edge features, and tobacco segment end face deformation and color change features are obtained, input the filter color distribution features, filter ... outer wall texture features, and tobacco brand identification features are obtained, input the filter outer wall texture features, and cigarette brand identification features into the cigarette type recognition network, process them in conjunction with a preset cigarette type feature library, and obtain the cigarette type recognition result, which includes an identifier indicating whether the cigarette is registered and cigarette brand information. The surface texture features of the mouthpiece, the outline and edge features of the filter tip, and the deformation and color change features of the tobacco segment end face are input into the cigarette usage status recognition network. The network is then processed in conjunction with a preset cigarette usage status feature template to obtain the cigarette usage status recognition result, which includes the used state and the unused state. If the reference mark position features of the cigarette side wall are extracted, these features are input into the cigarette insertion depth recognition network. The network is then processed in conjunction with the preset reference size of the cigarette insertion chamber to obtain the cigarette insertion depth recognition result, which includes the deviation value of the cigarette from the preset standard insertion depth.

[0068] Optionally, the cigarette management module 330 is specifically used to retrieve a preset cigarette management strategy library, and match one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results with various management strategies in the preset cigarette management strategy library to obtain matching results; select one or more strategies with the highest matching degree from the matching results and combine them to form a cigarette management strategy.

[0069] Optionally, the control strategies in the preset cigarette management strategy library include: if the cigarette registration status in the cigarette type identification result is unregistered, a cigarette registration error message is generated to remind the user to complete the cigarette information registration; if the cigarette registration status in the cigarette type identification result is registered and the cigarette usage status identification result is unused, the appropriate cigarette heating parameters are matched based on the cigarette brand information in the cigarette type identification result, a cigarette heating command is generated based on the cigarette heating parameters, and the cigarette heating operation is controlled based on the cigarette heating command; if the cigarette usage status identification result is used, an audible and visual warning is triggered, and the cigarette heating function is prohibited; if the deviation value in the cigarette insertion depth identification result is greater than the preset allowable deviation range, an insertion depth adjustment prompt message is generated based on the deviation value to remind the user to adjust the cigarette insertion position to within the standard range.

[0070] Optionally, the device is also used to acquire the currently effective cigarette management strategy and the corresponding actual cigarette usage feedback data when sufficient cigarette usage feedback data has been collected, or when the recognition accuracy of the multi-task image recognition module drops to a set threshold. The cigarette management strategy and the actual cigarette usage feedback data are then used as incremental training samples and sent back to the multi-task image recognition module for iterative optimization training to obtain the optimized multi-task image recognition module.

[0071] The cigarette management device for heated smoking appliances provided in this embodiment of the invention can execute the cigarette management method for heated smoking appliances provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0072] Example 4 Figure 4 This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0073] like Figure 4As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded into the RAM 13 from storage unit 18. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0074] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0075] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as cigarette management methods applied to heated smoking appliances.

[0076] In some embodiments, the cigarette management method applied to a heated smoking appliance may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the cigarette management method applied to a heated smoking appliance described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the cigarette management method applied to a heated smoking appliance by any other suitable means (e.g., by means of firmware).

[0077] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0078] Computer programs for implementing the present invention's method for managing cigarettes in heated smoking appliances can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0079] Example 5 Embodiment 5 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a cigarette management method applied to a heated smoking appliance, the method comprising: Acquire image data of the cigarette inserted into the chamber of the target heated smoking device. The image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette inside the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section. Based on one or more of the images of the end face of the tobacco section, the side wall of the tobacco section, the end face of the filter section, and the side wall of the filter section, the images are processed by a multi-task image recognition module to obtain the image recognition results of the cigarettes in the storage. The image recognition results include one or more of the following: cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. Based on one or more of the cigarette type identification results, cigarette usage status identification results, and cigarette insertion depth identification results, and in conjunction with a preset cigarette management strategy library, a corresponding cigarette management strategy is determined, and cigarette control is carried out based on the cigarette management strategy.

[0080] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0081] To provide interaction with an object, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the object; and a keyboard and pointing device (e.g., a mouse or trackball) through which the object provides input to the electronic device. Other types of devices can also be used to provide interaction with the object; for example, feedback provided to the object can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the object can be received in any form (including sound input, voice input, or tactile input).

[0082] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., a computer with a graphical user interface or web browser through which an item can interact with the implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0083] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0084] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0085] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements a cigarette management method for heated tobacco appliances according to any embodiment of the invention.

[0086] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0087] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for managing cigarettes used in heated smoking appliances, characterized in that, include: Acquire image data of the cigarette inserted into the chamber of the target heated smoking device, wherein the image data includes one or more of the following: an image of the end face of the tobacco section of the cigarette in the chamber, an image of the side wall of the tobacco section, an image of the end face of the filter section, and an image of the side wall of the filter section; Based on one or more of the end face image of the tobacco section, the side wall image of the tobacco section, the end face image of the filter section, and the side wall image of the filter section, the image recognition results of the cigarettes in the compartment are obtained by processing through a multi-task image recognition module. The image recognition results include one or more of the following: cigarette type recognition results, cigarette usage status recognition results, and cigarette insertion depth recognition results. Based on one or more of the cigarette type identification result, cigarette usage status identification result, and cigarette insertion depth identification result, a corresponding cigarette management strategy is determined in conjunction with a preset cigarette management strategy library, and cigarette control is performed based on the cigarette management strategy.

2. The method according to claim 1, characterized in that, The acquisition of image data of the cigarette inserted into the chamber of the target heated smoking device includes: By using image acquisition devices installed at at least two different angles on the inner wall of the cigarette insertion chamber, multiple original images of the cigarette inside the chamber are acquired simultaneously. The multiple original images include one or more of the following: original image of the tobacco section end face, original image of the tobacco section side wall, original image of the filter section end face, and original image of the filter section side wall. A preset sharpness determination algorithm is used to calculate the sharpness evaluation results corresponding to the multiple original images. Based on the sharpness evaluation results and preset image filtering conditions, the various images of the multiple original images are sorted and filtered to obtain the image data of the cigarette insertion chamber. The preset sharpness determination algorithm includes the Laplace variance algorithm.

3. The method according to claim 1, characterized in that, The multi-task image recognition module includes a shared convolutional feature extraction backbone network and a multi-task recognition network; The image recognition result of the cigarettes in the storage compartment is obtained by processing one or more of the following images: the end face image of the tobacco segment, the side wall image of the tobacco segment, the end face image of the filter segment, and the side wall image of the filter segment, through a multi-task image recognition module: One or more of the images of the tobacco shred end face, the tobacco shred sidewall, the filter tip end face, and the filter tip sidewall are input into the shared convolutional feature extraction backbone network for unified feature extraction to obtain a global shared feature extraction result. The global shared feature extraction result includes tobacco color distribution features, cigarette outer wall texture features, cigarette brand identification features, tobacco shred end face deformation and color change features, filter tip color distribution features, filter tip surface texture features, filter tip outline edge features, and cigarette sidewall reference mark position features. Each feature in the globally shared feature extraction result is input into the multi-task recognition network, which processes them separately through parallel task branches and outputs the image recognition result in parallel.

4. The method according to claim 3, characterized in that, The multi-task recognition network includes a cigarette type recognition network, a cigarette usage status recognition network, and a cigarette insertion depth recognition network. The process involves inputting various features from the globally shared feature extraction results into the multi-task recognition network, which then processes them through parallel task branches and outputs the image recognition results in parallel, including: If the tobacco color distribution features, the cigarette outer wall texture features, and the cigarette brand identification features are extracted, then the tobacco color distribution features, the cigarette outer wall texture features, and the cigarette brand identification features are input into the cigarette type recognition network, and processed in conjunction with a preset cigarette type feature library to obtain the cigarette type recognition result. The cigarette type recognition result includes an identifier indicating whether the cigarette is registered and cigarette brand information. If the filter tip color distribution features, filter tip surface texture features, filter tip outline edge features, and tobacco segment end face deformation and color change features are extracted, then these features are input into the cigarette usage status recognition network and processed in conjunction with a preset cigarette usage status feature template to obtain the cigarette usage status recognition result. The cigarette usage status recognition result includes a used state and an unused state. If the position features of the reference mark on the sidewall of the cigarette are extracted, the position features of the reference mark on the sidewall of the cigarette are input into the cigarette insertion depth recognition network, and processed in conjunction with the preset reference size of the cigarette insertion chamber to obtain the cigarette insertion depth recognition result. The cigarette insertion depth recognition result includes the deviation value of the cigarette relative to the preset standard insertion depth.

5. The method according to claim 1, characterized in that, The step of determining a corresponding cigarette management strategy based on one or more of the cigarette type identification result, the cigarette usage status identification result, and the cigarette insertion depth identification result, combined with a preset cigarette management strategy library, includes: The preset cigarette management strategy library is retrieved, and one or more of the cigarette type identification result, cigarette usage status identification result, and cigarette insertion depth identification result are matched with various control strategies in the preset cigarette management strategy library to obtain a matching result. The strategy with the highest matching degree is selected from the matching results and combined to form the cigarette management strategy.

6. The method according to claim 5, characterized in that, The control strategies in the preset cigarette management strategy library include: If the cigarette registration status in the cigarette type identification result is "unregistered", a cigarette registration error message will be generated to remind the user to complete the registration of cigarette information into the database. If the cigarette registration status in the cigarette type identification result is "registered" and the cigarette usage status identification result is "unused", then the appropriate cigarette heating parameters are matched based on the cigarette brand information in the cigarette type identification result, a cigarette heating command is generated based on the cigarette heating parameters, and the cigarette heating operation is controlled based on the cigarette heating command. If the cigarette usage status identification result is "used", then an audible and visual warning operation is triggered, and the cigarette heating function is prohibited from being started. If the deviation value in the cigarette insertion depth recognition result is greater than the preset allowable deviation range, an insertion depth adjustment prompt message is generated based on the deviation value to remind the user to adjust the cigarette insertion position to within the standard range.

7. The method according to claim 1, characterized in that, The method also includes: Once sufficient feedback data on cigarette usage has been collected, or when the recognition accuracy of the multi-task image recognition module drops to a set threshold, the currently effective cigarette management strategy and the corresponding actual cigarette usage feedback data are obtained. The cigarette management strategy and the actual cigarette usage feedback data are used as incremental training samples and fed back to the multi-task image recognition module for iterative optimization training to obtain the optimized multi-task image recognition module.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, which enables the at least one processor to perform the cigarette management method for heated smoking appliances as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the cigarette management method for any one of claims 1-7 applied to a heated smoking appliance.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the cigarette management method for heated smoking appliances according to any one of claims 1-7.

Images