Transmission bottle quality detection method of pneumatic logistics transmission system

Abstract

The invention provides a transmission bottle quality detection method for a pneumatic logistics transmission system, and the method comprises the steps: detecting an event that a transmission bottle is placed in a transmission station in real time, obtaining at least one object image of an object loaded in the transmission bottle through an image sensor when the event that the transmission bottle is placed in the transmission station is detected, and transmitting the at least one object image to a server; performing three-dimensional modeling on the article loaded in the transmission bottle by using the article image to generate a three-dimensional model of the article loaded in the transmission bottle, and calculating the mass of the article loaded in the transmission bottle based on the three-dimensional model, and according to the mass of the articles loaded in the conveying bottle, the comprehensive mass of the conveying bottle loaded with the articles is calculated, the conveying speed and the conveying time of the whole conveying process of the conveying bottle can be finely controlled, and the problem of vibration noise generated in the early stage of movement of the conveying bottle is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of pneumatic transmission, and in particular to a method for detecting the quality of a transmission bottle of a pneumatic tube logistics transmission system. Background Art

[0002] Intra-hospital material transportation is an important and very labor-intensive logistics link in hospitals. The frequent material demands of various departments in the hospital result in a large amount of medical materials, including medicines, medical consumables, etc., being frequently transported between departments. For a long time, intra-hospital material transportation in hospitals, that is, material transportation between departments, has mostly relied on manual transportation, which has taken up a lot of time and energy of medical staff, and is prone to errors in the rush, resulting in low efficiency. The introduction of the pneumatic tube logistics transmission system has solved the above problems well. The pneumatic tube logistics transmission system puts items into transmission bottles and transmits them from one transmission station to another through transmission pipelines. It only needs to set up transmission stations in each department to quickly transport medical materials between departments. The transmission bottle in the pneumatic tube logistics transmission system is equivalent to a car on the highway. It is designed as a sturdy container that is impact-resistant and rupture-resistant. It serves as a medical material transmission vehicle and moves at high speed in the transmission pipeline of the pneumatic tube logistics transmission system. The power for the transport bottle to move in the transmission pipeline is provided by the positive / negative air pressure generated by the variable frequency fan of the pneumatic tube transmission system in the transmission pipeline. By adjusting the air pressure applied by the variable frequency fan in the transmission pipeline, the movement speed of the transport bottle can be adjusted in multiple levels.

[0003] Hospitals require a wide variety of medical supplies to be transported within the hospital, many of which vary significantly in mass between gaseous and liquid forms. Once these medical supplies are placed in transport bottles, the varying mass leads to varying transfer speeds and times even at the same air pressure. This hinders the precise management of medical supply transport tasks and also impacts the stability of the transport process, leading to severe vibration and noise issues. While load cells and other mass-measuring devices can be installed at the transport bottle's location at the transport station, their accuracy depends on the stability of the object being placed. Therefore, accurate mass measurement requires a period of stability after the transport bottle is placed, resulting in a delay before the transport action can begin. Chinese patent CN113340386A proposes a method for measuring the quality of a transport bottle in a pneumatic tube transport system. This method employs two photoelectric detection sensors spaced apart in the transport pipe to detect the time difference between the two photoelectric detection points, and then uses this time difference to infer the mass of the transport bottle. Although this method for measuring the mass of a loaded transport bottle is simple and effective, and does not produce hysteresis, it requires the transport bottle to be measured while it is in motion within the transport pipe. The driving air pressure in the early stages of the transport bottle's movement can only be set based on empirical values, making it impossible to achieve precise control of the transport speed and time of the transport bottle throughout the entire transport process, nor can it address the vibration and noise issues generated in the early stages of the transport bottle's movement. Summary of the Invention

[0004] Based on the above problems, the present invention proposes a method for detecting the quality of a transport bottle in a pneumatic tube transport system. The method can achieve precise control of the transmission speed and transmission time of the transport bottle throughout the entire transmission process, and solve the vibration and noise problem generated in the early stages of the movement of the dynamic transport bottle.

[0005] In view of this, the present invention proposes a method for detecting the quality of a transport bottle of a pneumatic tube transport system, comprising:

[0006] Real-time detection of the event of a carrier bottle being placed in a transfer station;

[0007] When an event of a transport bottle being placed in a transport station is detected, at least one image of the items loaded in the transport bottle is acquired by an image sensor;

[0008] Performing three-dimensional modeling of the article loaded in the transport bottle using the article image to generate a three-dimensional model of the article loaded in the transport bottle;

[0009] Calculating the mass of the item loaded in the transport bottle based on the three-dimensional model;

[0010] The total mass of the transport bottle after loading the items is calculated based on the mass of the items loaded in the transport bottle.

[0011] Furthermore, the step of using the item image to perform three-dimensional modeling on the item loaded in the transport bottle to generate a three-dimensional model of the item loaded in the transport bottle specifically includes:

[0012] Inputting the item image into a pre-trained medical material recognition model for item recognition, wherein the medical material recognition model is a deep learning model for identifying the type of medical materials that is trained using photos of medical materials as training samples;

[0013] Dividing the items loaded in the transport bottle into several components of different materials according to the item identification result;

[0014] Generating a component model corresponding to each component, having the same shape and size as the corresponding part of the item loaded in the transport bottle;

[0015] A three-dimensional model of the items loaded in the transport bottle is constructed using the component models corresponding to the components.

[0016] Furthermore, the step of calculating the mass of the items loaded in the transport bottle based on the three-dimensional model specifically includes:

[0017] Matching material data corresponding to each component from a pre-built component database, wherein the material data includes the density of the corresponding material;

[0018] Calculating the volume of each component based on the shape and size of the model components;

[0019] The mass of each component is calculated based on the density and volume of each component;

[0020] The component mass of each component is then determined as the mass of the contents of the carrier.

[0021] Furthermore, the step of calculating the mass of the items loaded in the transport bottle based on the three-dimensional model further includes:

[0022] Determining whether the components constituting the article loaded in the transport bottle include a standard component, wherein the standard component includes a multi-material component composed of multiple components of a single material;

[0023] Identifying specification parameters of the standard component through the image of the object;

[0024] Reading the component quality of the standard component from the component database according to the specification parameters;

[0025] The mass of the item loaded in the transport bottle is obtained by adding the mass of the standard component and the calculated mass of the non-standard component.

[0026] Furthermore, before the step of detecting in real time the event of the transport bottle being placed in the transport station, the method further includes:

[0027] Receive item information input by a user regarding an item to be loaded into the transport bottle for transport, the item information including one or more of the name, item type, item composition, item volume, or mass of the item to be loaded;

[0028] The item information of the item to be loaded is written into the NFC tag carried by the transport bottle used to load the item to be loaded through an NFC reader / writer.

[0029] Furthermore, the item information includes the item name and / or item type information of the item to be loaded. After the step of detecting in real time that the transport bottle is placed in the transport station, the process further includes:

[0030] When an event of a transport bottle being placed in a transport station is detected, the item information in the NFC tag carried by the transport bottle is read by an NFC reader;

[0031] After inputting the item image into a pre-trained medical material recognition model for item recognition, the method further includes:

[0032] Determining whether a first item type identified by the medical supply identification model is consistent with a second item type read by the NFC reader;

[0033] When the first item type and the second item type are inconsistent, comparing the appearance similarity between the first item type and the second item type;

[0034] Whether to continue to perform subsequent quality inspection steps is determined based on the appearance similarity between the first item type and the second item type.

[0035] Furthermore, the step of comparing the appearance similarity between the first item type and the second item type specifically includes:

[0036] determining an image of the article loaded in the transport bottle obtained by an image sensor as a first image corresponding to the first article type;

[0037] Reading a representative image of the second item type from a medical supplies database as a second image corresponding to the second item type;

[0038] The first image and the second image are input into a pre-trained medical material similarity comparison model for comparison to obtain the appearance similarity between the first item type and the second item type.

[0039] Furthermore, determining whether to continue to perform subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes:

[0040] When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the first item type identified by the medical supply identification model does not match the item loaded in the transport bottle;

[0041] determining an image of the object loaded in the transport bottle acquired by an image sensor as a candidate training sample associated with the second object type;

[0042] The second object type is determined to be the object type of the object loaded in the transport bottle and subsequent quality inspection steps are performed.

[0043] Furthermore, determining whether to continue to perform subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes:

[0044] When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the item information in the NFC tag carried by the transport bottle does not match the item loaded in the transport bottle;

[0045] A reminder message is issued indicating that the NFC tag information is incorrect or the loaded item is incorrect, and the transmission instruction is terminated. Subsequent quality inspection steps are no longer performed.

[0046] Furthermore, the item information includes component specification information of the item to be loaded. After the step of dividing the items loaded in the transport bottle into several components of different materials according to the item identification result, the following steps are further included:

[0047] Determine a component having component specification information in the item information obtained by reading the NFC reader as a target component;

[0048] determining whether the component specification information of the target component read by the NFC reader includes the component quality of the target component;

[0049] When the component specification information of the target component read by the NFC reader does not include the component quality of the target component, determining whether the component specification information of the target component read by the NFC reader includes a parameter that can be converted to obtain the component quality of the target component;

[0050] When the component specification information of the target component read by the NFC reader does not include parameters that can be converted to obtain the component mass of the target component, a step of calculating the component mass of the target component according to the density and volume of the target component is performed.

[0051] The present invention proposes a method for detecting the quality of a transport bottle in a pneumatic tube transport system. The method detects in real time the event of a transport bottle being placed in a transport station. When the event of the transport bottle being placed in the transport station is detected, an image sensor is used to obtain at least one image of the items loaded in the transport bottle. The items loaded in the transport bottle are stereo-modeled using the image of the items to generate a stereo model of the items loaded in the transport bottle. The mass of the items loaded in the transport bottle is calculated based on the stereo model. The total mass of the transport bottle after loading the items is calculated based on the mass of the items loaded in the transport bottle. This method can achieve refined control of the transmission speed and transmission time of the entire transport process of the transport bottle and solve the problem of vibration and noise generated in the early stage of the movement of the dynamic transport bottle. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 The present invention provides a flowchart of a method for detecting the quality of a transport bottle of a pneumatic tube logistics transport system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0055] In the description of the present invention, the term "plurality" refers to two or more. Unless otherwise specified, the terms "upper" and "lower" are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific manner. Therefore, they should not be construed as limiting the present invention. The terms "connected," "mounted," and "fixed," etc., should be interpreted broadly. For example, "connected" can refer to fixed, removable, or integral connections; directly or indirectly through an intermediary. A person of ordinary skill in the art will understand the specific meanings of these terms in the present invention based on the specific circumstances. Furthermore, the terms "first," "second," etc., etc., are used for descriptive purposes only and should not be construed to indicate or imply relative importance or to implicitly specify the number of the technical features indicated. Therefore, a feature designated "first," "second," etc., may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0056] Throughout this specification, terms such as "one embodiment," "some implementations," and "specific examples" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0057] A method for detecting the quality of a transport bottle of a pneumatic tube logistics transport system according to some embodiments of the present invention will be described below with reference to the accompanying drawings.

[0058] like Figure 1 As shown, the present invention proposes a method for detecting the quality of a transport bottle of a pneumatic tube transport system, comprising:

[0059] Real-time detection of the event of a carrier bottle being placed in a transfer station;

[0060] When an event of a transport bottle being placed in a transport station is detected, at least one image of the items loaded in the transport bottle is acquired by an image sensor;

[0061] Performing three-dimensional modeling of the article loaded in the transport bottle using the article image to generate a three-dimensional model of the article loaded in the transport bottle;

[0062] Calculating the mass of the item loaded in the transport bottle based on the three-dimensional model;

[0063] The total mass of the transport bottle after loading the items is calculated based on the mass of the items loaded in the transport bottle.

[0064] In the technical solutions of some embodiments of the present invention, an infrared sensor is provided in the transfer station. When a transport bottle is placed in the transfer station, an event in which the infrared signal is blocked is determined as an event in which the transport bottle is placed in the transfer station.

[0065] In the technical solutions of some other embodiments of the present invention, an NFC reader is provided in the transmission station, and an NFC tag is provided on the transmission bottle. When a transmission bottle carrying an NFC tag is placed in the transmission station, the event in which the NFC reader reads the information in the NFC tag is determined as the event in which the transmission bottle is placed in the transmission station. More specifically, a tray or base for placing the transmission bottle is provided in the transmission station, and the NFC reader in the transmission station is provided at a position on the surface of the tray or base for placing the transmission bottle. Preferably, a limiting structure for constraining the placement position of the transmission bottle is provided on the tray or base in the transmission station. When the transmission bottle is placed in the transmission station, under the constraint of the limiting structure, its NFC tag is exactly located at the position where the NFC reader is located, so that when the transmission bottle is placed in the transmission station, the NFC reader can quickly sense and read the data in the NFC tag.

[0066] Furthermore, the transfer station is equipped with an image sensor, such as a camera, for capturing images of items loaded within the transport bottle. The outer shell of the transport bottle includes a transparent outer shell, at least partially allowing visibility of the internal loading space. Preferably, the outer shell of the transport bottle is a transparent annular outer shell, and the transfer station is equipped with multiple image sensors at different angles to capture images of the items within the transport bottle from multiple angles through the transparent annular outer shell.

[0067] In some other embodiments of the present invention, the transfer station is provided with only one image sensor for acquiring images of items loaded in the transport bottle, and the transfer station is further provided with a rotation mechanism for rotating the transport bottle. After the transport bottle is placed in the transfer station, the rotation mechanism drives the transport bottle to rotate relative to the image sensor within the transfer station, so that the image sensor can capture images of the items loaded in the transport bottle from multiple angles.

[0068] Furthermore, the step of using the item image to perform three-dimensional modeling on the item loaded in the transport bottle to generate a three-dimensional model of the item loaded in the transport bottle specifically includes:

[0069] Inputting the item image into a pre-trained medical material recognition model for item recognition, wherein the medical material recognition model is a deep learning model for identifying the type of medical materials that is trained using photos of medical materials as training samples;

[0070] Dividing the items loaded in the transport bottle into several components of different materials according to the item identification result;

[0071] Generating a component model corresponding to each component, having the same shape and size as the corresponding part of the item loaded in the transport bottle;

[0072] A three-dimensional model of the items loaded in the transport bottle is constructed using the component models corresponding to the components.

[0073] The medical supply recognition model is a specialized model trained using deep learning technology for medical supply recognition. Before implementing the transport bottle quality inspection method, a large number of photos of medical supplies are collected and classified to obtain sample data sets of medical supplies corresponding to different classification labels. These sample data sets are used for training to obtain the medical supply recognition model.

[0074] In the step of classifying the item in the transport bottle into several components of different materials based on the item identification results, if the item in the transport bottle is entirely made of the same material, no classification is performed. For example, if the item in the transport bottle is a test report, it is entirely made of paper and no component classification is performed. However, if the item in the transport bottle is a medicine bottle containing liquid medicine, the bottle cap, bottle body, and the liquid medicine in the bottle are all made of different materials, and the item is classified into three components: the bottle cap, bottle body, and liquid medicine.

[0075] Furthermore, the step of calculating the mass of the items loaded in the transport bottle based on the three-dimensional model specifically includes:

[0076] Matching material data corresponding to each component from a pre-built component database, wherein the material data includes the density of the corresponding material;

[0077] Calculating the volume of each component based on the shape and size of the model components;

[0078] The mass of each component is calculated based on the density and volume of each component;

[0079] The component mass of each component is then determined as the mass of the contents of the carrier.

[0080] The component database is a pre-built database containing data on the components and materials of various medical supplies. The database contains the correspondence between various medical supplies and their components. Therefore, when classifying the items in the transport bottle into components made of different materials based on the item identification results, after identifying the type of medical supply to which the items in the transport bottle belong, the database can be used to query the components that make up the medical supply.

[0081] Furthermore, the step of calculating the mass of the items loaded in the transport bottle based on the three-dimensional model further includes:

[0082] Determining whether the components constituting the article loaded in the transport bottle include a standard component, wherein the standard component includes a multi-material component composed of multiple components of a single material;

[0083] Identifying specification parameters of the standard component through the image of the object;

[0084] Reading the component quality of the standard component from the component database according to the specification parameters;

[0085] The mass of the item loaded in the transport bottle is obtained by adding the mass of the standard component and the calculated mass of the non-standard component.

[0086] Standard components are single-material or multi-material components pre-configured in the component database. The database stores their component quality data. As components of frequently transported items, these components typically have fixed specifications and quality. For example, medicine bottles of various sizes are commonly used containers for pharmaceuticals, and their specifications and quality are relatively fixed. Defining a standard component for the bottle body and cap as a whole avoids the need for mass calculation during each transport of medical supplies.

[0087] The non-standard components are components that do not have unified specification parameters and are difficult to unify into standard specification parameters. Their quality needs to be calculated each time medical supplies are transported.

[0088] Furthermore, before the step of detecting in real time the event of the transport bottle being placed in the transport station, the method further includes:

[0089] Receive item information input by a user regarding an item to be loaded into the transport bottle for transport, the item information including one or more of the name, item type, item composition, item volume, or mass of the item to be loaded;

[0090] The item information of the item to be loaded is written into the NFC tag carried by the transport bottle used to load the item to be loaded through an NFC reader / writer.

[0091] Specifically, the user enters the item information of the item to be loaded through a terminal device such as a personal computer or a tablet computer, or reads the QR code on the item to be loaded through a barcode reader to obtain the item information of the item to be loaded.

[0092] In the technical solution of the above embodiment, the carrier bottle is provided with an NFC tag. The NFC reader / writer is communicatively connected to a terminal device such as a personal computer or tablet computer to obtain the item information of the item to be loaded from the terminal device and then write the information to the NFC tag carried by the carrier bottle used to load the item to be loaded.

[0093] Furthermore, the item information includes the item name and / or item type information of the item to be loaded. After the step of detecting in real time that the transport bottle is placed in the transport station, the process further includes:

[0094] When an event of a transport bottle being placed in a transport station is detected, the item information in the NFC tag carried by the transport bottle is read by an NFC reader;

[0095] After inputting the item image into a pre-trained medical material recognition model for item recognition, the method further includes:

[0096] Determining whether a first item type identified by the medical supply identification model is consistent with a second item type read by the NFC reader;

[0097] When the first item type and the second item type are inconsistent, comparing the appearance similarity between the first item type and the second item type;

[0098] Whether to continue to perform subsequent quality inspection steps is determined based on the appearance similarity between the first item type and the second item type.

[0099] Furthermore, the NFC tag carried on the transport bottle also records the specifications and / or quality information of the transport bottle itself. The NFC reader set up in the transmission station can query the weight of the empty transport bottle by reading the specifications in the NFC tag, or directly read the weight of the empty transport bottle.

[0100] Furthermore, when the item information contains the item name of the item to be loaded but does not contain its item type information, the item type of the item to be loaded is determined by querying the correspondence between the item name and item type preset in the medical supplies database, and then a step of determining whether the first item type obtained by item identification is consistent with the second item type read by the NFC reader is performed.

[0101] Furthermore, the so-called subsequent quality inspection steps include:

[0102] When an event of a transport bottle being placed in a transport station is detected, at least one image of the items loaded in the transport bottle is acquired by an image sensor;

[0103] Performing three-dimensional modeling of the article loaded in the transport bottle using the article image to generate a three-dimensional model of the article loaded in the transport bottle;

[0104] Calculating the mass of the item loaded in the transport bottle based on the three-dimensional model;

[0105] The total mass of the transport bottle after loading the items is calculated based on the mass of the items loaded in the transport bottle.

[0106] Furthermore, after the step of determining whether the first item type obtained by item identification is consistent with the second item type obtained by reading the NFC reader, the method further includes:

[0107] When the first article type is the same as the second article type, the subsequent quality inspection steps are directly performed.

[0108] Furthermore, the step of comparing the appearance similarity between the first item type and the second item type specifically includes:

[0109] determining an image of the article loaded in the transport bottle obtained by an image sensor as a first image corresponding to the first article type;

[0110] Reading a representative image of the second item type from a medical supplies database as a second image corresponding to the second item type;

[0111] The first image and the second image are input into a pre-trained medical material similarity comparison model for comparison to obtain the appearance similarity between the first item type and the second item type.

[0112] In the step of determining an image of the item in the transport bottle acquired by the image sensor as the first image corresponding to the first item type, when the image sensor acquires multiple images of the item in the transport bottle, the image that best represents the appearance characteristics of the item in the transport bottle is selected as the first image, or the image that best displays the most appearance details of the item in the transport bottle is selected as the first image.

[0113] In the technical solution of the above embodiment, representative images corresponding to the item type of each medical supply are pre-configured in the medical supply database. When the appearance similarity calculation needs to be performed, the representative images of the corresponding item type are read from the medical supply database for comparison.

[0114] The medical supplies similarity comparison model is a specialized model trained using deep learning technology for comparing the appearance similarity of medical supplies. Before implementing the transport bottle quality inspection method, a large number of photos of medical supplies are collected and classified. Medical supplies of the same type are marked with higher similarity values, and medical supplies of different types are marked with lower similarity values. This obtains sample data sets corresponding to medical supplies of different similarities. These sample data sets are used for training to obtain the medical supplies similarity comparison model.

[0115] In the technical solution of the above embodiment, the similarity values ​​can be obtained by performing pixel-level comparison calculations on photos of similar medical supplies, or by extracting image features, vectorizing the image features, and then performing quantitative calculations to obtain the similarity values.

[0116] Furthermore, the above method also includes:

[0117] Configuring a first similarity coefficient corresponding to medical supplies of the same type and a similarity coefficient corresponding to medical supplies of different types, wherein the first similarity coefficient is greater than the second similarity coefficient;

[0118] Determine two photos of the same type of medical supplies whose similarity needs to be calculated as a first image and a second image respectively;

[0119] extracting first image features from the first image and extracting second image features from the second image;

[0120] Vectorizing the first image feature into a first vector, and vectorizing the second image feature into a second vector;

[0121] calculating a first similarity between the first vector and the second vector;

[0122] The first similarity is multiplied by the first similarity coefficient to obtain a similarity value between the medical supplies corresponding to the first image and the second image.

[0123] Furthermore, the above method also includes:

[0124] Determine the photos of two medical supplies of different types whose similarity needs to be calculated as the third image and the fourth image respectively;

[0125] extracting a third image feature from the third image and extracting a fourth image feature from the fourth image;

[0126] Vectorizing the third image feature into a third vector, and vectorizing the fourth image feature into a fourth vector;

[0127] calculating a second similarity between the third vector and the fourth vector;

[0128] The second similarity is multiplied by the second similarity coefficient to obtain a similarity value between the medical supplies corresponding to the third image and the fourth image.

[0129] Furthermore, determining whether to continue to perform subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes:

[0130] When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the first item type identified by the medical supply identification model does not match the item loaded in the transport bottle;

[0131] determining an image of the object loaded in the transport bottle acquired by an image sensor as a candidate training sample associated with the second object type;

[0132] The second object type is determined to be the object type of the object loaded in the transport bottle and subsequent quality inspection steps are performed.

[0133] Furthermore, after the step of determining the image of the object loaded in the transport bottle obtained by the image sensor as a candidate training sample associated with the second object type, the method further includes:

[0134] Determine whether the data volume of the candidate training sample is greater than a preset training data volume threshold;

[0135] When the data volume of the candidate training samples is greater than a preset training data volume threshold, the medical material recognition model is trained again using the candidate training samples.

[0136] Furthermore, determining whether to continue to perform subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes:

[0137] When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the item information in the NFC tag carried by the transport bottle does not match the item loaded in the transport bottle;

[0138] A reminder message is issued indicating that the NFC tag information is incorrect or the loaded item is incorrect, and the transmission instruction is terminated. Subsequent quality inspection steps are no longer performed.

[0139] In the technical solution of the above-mentioned embodiment, when the appearance similarity between the first and second item types exceeds a preset similarity threshold, it indicates that the item information in the NFC tag carried by the transport bottle does not match the item loaded in the transport bottle. This mismatch can occur in two ways: one is that the item information in the NFC tag is correct and represents the item to be transferred, but an operator error caused the wrong item to be loaded; the other is that the loaded item is correct, but the item information was mistakenly written to the NFC tag as the information of another item. In either case, it may cause an abnormal item transfer or lead to other subsequent problems, necessitating the termination of the transfer instruction and the retry of the transfer after the operator has corrected the information in the item or NFC tag.

[0140] Furthermore, the item information includes component specification information of the item to be loaded. After the step of dividing the items loaded in the transport bottle into several components of different materials according to the item identification result, the following steps are further included:

[0141] Determine a component having component specification information in the item information obtained by reading the NFC reader as a target component;

[0142] determining whether the component specification information of the target component read by the NFC reader includes the component quality of the target component;

[0143] When the component specification information of the target component read by the NFC reader does not include the component quality of the target component, determining whether the component specification information of the target component read by the NFC reader includes a parameter that can be converted to obtain the component quality of the target component;

[0144] When the component specification information of the target component read by the NFC reader does not include parameters that can be converted to obtain the component mass of the target component, a step of calculating the component mass of the target component according to the density and volume of the target component is performed.

[0145] It should be noted that the component specification information in the item information does not necessarily include the component mass, but may be presented in the form of other parameters of the component that can reflect its mass. For example, liquid medicines will be marked with the medicine name and liquid capacity in their item information. The density can be queried through its name, and then its mass can be obtained by multiplying it with its capacity.

[0146] Furthermore, after the step of determining whether the component specification information of the target component read by the NFC reader includes parameters that can be converted to obtain the component quality of the target component, the method further includes:

[0147] When the component specification information of the target component read by the NFC reader includes the component mass of the target component, or the component specification information of the target component read by the NFC reader includes parameters that can be converted to the component mass of the target component, the total mass of the items loaded in the transport bottle is calculated directly using the component mass of the target component included in the component specification information of the target component read by the NFC reader, or the component mass of the target component converted from the parameters that can be converted to the component mass of the target component, without performing the step of calculating the component mass of the target component based on its volume and density.

[0148] Furthermore, when the target component is a standard component and the component specification information of the target component read by the NFC reader includes the component mass of the target component, or the component specification information of the target component read by the NFC reader includes parameters that can be converted to the component mass of the target component, the component specification information of the target component read by the NFC reader includes the component mass of the target component, or the component mass of the target component converted from the component specification information of the target component read by the NFC reader includes the parameters that can be converted to the component mass of the target component is directly used to calculate the total mass of the items loaded in the transport bottle, rather than using the component mass of the standard component read from the component database according to the specification parameters to calculate the total mass of the items loaded in the transport bottle, so as to avoid incorrect identification of the specification parameters of the standard component during item identification, resulting in the component mass read from the component database being an incorrect component mass.

[0149] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0150] While embodiments of the present invention have been described above, these embodiments do not exhaustively describe all details and do not limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the above description. These embodiments are selected and described in detail in this specification in order to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better utilize the present invention and its modifications. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for detecting the quality of a transport bottle of a pneumatic tube transport system, characterized in that: include: Real-time detection of the event of a carrier bottle being placed in a transfer station; When an event of a transport bottle being placed in a transport station is detected, at least one image of the items loaded in the transport bottle is acquired by an image sensor; Performing three-dimensional modeling of the article loaded in the transport bottle using the article image to generate a three-dimensional model of the article loaded in the transport bottle; Calculating the mass of the item loaded in the transport bottle based on the three-dimensional model; The total mass of the transport bottle after loading the items is calculated based on the mass of the items loaded in the transport bottle.

2. The method for detecting the quality of a transport bottle of a pneumatic tube transport system according to claim 1, characterized in that: The steps of using the image of the object to perform three-dimensional modeling on the object in the transport bottle to generate a three-dimensional model of the object in the transport bottle specifically include: Inputting the item image into a pre-trained medical material recognition model for item recognition, wherein the medical material recognition model is a deep learning model for identifying the type of medical materials that is trained using photos of medical materials as training samples; Dividing the items loaded in the transport bottle into several components of different materials according to the item identification result; Generating a component model corresponding to each component, having the same shape and size as the corresponding part of the item loaded in the transport bottle; A three-dimensional model of the items loaded in the transport bottle is constructed using the component models corresponding to the components.

3. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 2, characterized in that: The step of calculating the mass of the item loaded in the transport bottle based on the three-dimensional model specifically includes: Matching material data corresponding to each component from a pre-built component database, wherein the material data includes the density of the corresponding material; Calculating the volume of each component based on the shape and size of the model components; The mass of each component is calculated based on the density and volume of each component; The component mass of each component is then determined as the mass of the contents of the carrier.

4. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 3, characterized in that: The step of calculating the mass of the item loaded in the transport bottle based on the three-dimensional model further includes: Determining whether the components constituting the article loaded in the transport bottle include a standard component, wherein the standard component includes a multi-material component composed of multiple components of a single material; Identifying specification parameters of the standard component through the image of the object; Reading the component quality of the standard component from the component database according to the specification parameters; The mass of the item loaded in the transport bottle is obtained by adding the mass of the standard component and the calculated mass of the non-standard component.

5. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 2, characterized in that: Prior to the step of detecting in real time the event of a carrier being placed in a transfer station, the following steps may also be included: Receive item information input by a user regarding an item to be loaded into the transport bottle for transport, the item information including one or more of the name, item type, item composition, item volume, or mass of the item to be loaded; The item information of the item to be loaded is written into the NFC tag carried by the transport bottle used to load the item to be loaded through an NFC reader / writer.

6. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 5, characterized in that: The item information includes the name and / or type of the item to be loaded. After the step of detecting in real time that the transport bottle is placed in the transport station, the method further includes: When an event of a transport bottle being placed in a transport station is detected, the item information in the NFC tag carried by the transport bottle is read by an NFC reader; After inputting the item image into a pre-trained medical material recognition model for item recognition, the method further includes: Determining whether a first item type identified by the medical supply identification model is consistent with a second item type read by the NFC reader; When the first item type and the second item type are inconsistent, comparing the appearance similarity between the first item type and the second item type; Whether to continue to perform subsequent quality inspection steps is determined based on the appearance similarity between the first item type and the second item type.

7. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 6, characterized in that: The step of comparing the appearance similarity between the first item type and the second item type specifically includes: determining an image of the article loaded in the transport bottle obtained by an image sensor as a first image corresponding to the first article type; Reading a representative image of the second item type from a medical supplies database as a second image corresponding to the second item type; The first image and the second image are input into a pre-trained medical material similarity comparison model for comparison to obtain the appearance similarity between the first item type and the second item type.

8. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 6, characterized in that: Determining whether to continue performing subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes: When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the first item type identified by the medical supply identification model does not match the item loaded in the transport bottle; determining an image of the object loaded in the transport bottle acquired by an image sensor as a candidate training sample associated with the second object type; The second object type is determined to be the object type of the object loaded in the transport bottle and subsequent quality inspection steps are performed.

9. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 6, characterized in that: Determining whether to continue performing subsequent quality inspection steps based on the appearance similarity between the first item type and the second item type specifically includes: When the appearance similarity between the first item type and the second item type is greater than a preset similarity threshold, determining that the item information in the NFC tag carried by the transport bottle does not match the item loaded in the transport bottle; A reminder message is issued indicating that the NFC tag information is incorrect or the loaded item is incorrect, and the transmission instruction is terminated. Subsequent quality inspection steps are no longer performed.

10. The method for detecting the quality of a transport bottle of a pneumatic tube logistics transmission system according to claim 5, characterized in that: The item information includes component specification information of the item to be loaded. After the step of dividing the item loaded in the transport bottle into several components of different materials according to the item identification result, the method further includes: Determine a component having component specification information in the item information obtained by reading the NFC reader as a target component; determining whether the component specification information of the target component read by the NFC reader includes the component quality of the target component; When the component specification information of the target component read by the NFC reader does not include the component quality of the target component, determining whether the component specification information of the target component read by the NFC reader includes a parameter that can be converted to obtain the component quality of the target component; When the component specification information of the target component read by the NFC reader does not include parameters that can be converted to obtain the component mass of the target component, a step of calculating the component mass of the target component according to the density and volume of the target component is performed.

Images