Intelligent medicine dispensing robot based on visual recognition technology

Abstract

The present invention provides an intelligent dispensing robot based on visual recognition technology, wherein the intelligent dispensing robot includes: a medicine box attribute information acquisition subsystem, which is used to obtain the medicine box attribute information of the first target medicine box; an automatic medicine loading subsystem, which is used to automatically replenish the first target medicine box based on the medicine box attribute information based on the visual recognition technology; a medicine dispensing demand acquisition subsystem, which is used to obtain the medicine dispensing demand of the target patient; a medicine dispensing plan determination subsystem, which is used to determine the medicine dispensing plan based on the medicine dispensing demand and the medicine box attribute information; an intelligent medicine dispensing subsystem, which is used to perform intelligent medicine dispensing according to the medicine dispensing plan. The intelligent medicine dispensing robot based on visual recognition technology of the present invention introduces visual recognition technology and automatically fills the first target medicine box according to the medicine box attribute information. In addition, the medicine dispensing plan is determined according to the medicine dispensing demand and the medicine box attribute information, and the medicine is dispensed based on the medicine dispensing plan, without the need for manual participation in the medicine dispensing process, thereby avoiding the situation of medicine dispensing errors, and is more intelligent.

Description

Technical Field

[0001] The present invention relates to the field of industrial vision technology, and in particular to an intelligent medicine dispensing robot based on vision recognition technology. Background Art

[0002] At present, the population base of elderly people suffering from chronic diseases is large, and most of these elderly people need to take multiple drugs for long-term treatment. Modern medical technology is advanced, with rich varieties of drugs, detailed classifications, and complex usage. However, due to the decline of physical functions and obvious decline in memory of the elderly, taking and dispensing medicine every day has become a matter that is easy to forget or make mistakes. As a common medical auxiliary product, medicine boxes are inseparable from the disease care effect of the elderly. Traditional medicine boxes have pain points in product shape and function, such as low degree of automation, low degree of intelligence, and lack of humanized design, which obviously cannot meet the increasingly complex medical care needs of the elderly.

[0003] The invention patent with application number: CN201911044150.1 discloses a desktop dispensing robot and a dispensing method thereof, wherein the desktop dispensing robot includes a cabinet, which is provided with a control panel and an operation window for operators to use, and a support plate is provided in the cabinet, through which the space in the cabinet is separated to form an upper and lower space, and the upper space of the cabinet is provided with a syringe transmission module, a rotating manipulator module, a needle cap collection module, an infusion bag transmission module, a vial transmission module, a cap opening and disinfection module, a vial rotation module, a laser scanning module, an empty vial push module, an oscillation dissolution module, and a syringe needle separation and collection module; the support plate is also provided with a needle cap recovery port, a syringe needle recovery port and a vial recovery port, and a needle cap recovery barrel, a syringe needle recovery barrel and a vial recovery barrel are provided in the lower space of the cabinet. The above-mentioned desktop dispensing robot and its dispensing method can effectively reduce the labor intensity of operators, improve dispensing efficiency, and reduce the error rate.

[0004] At that time, the above-mentioned existing technology still required operators to assist in dispensing medicines at the operating window, which resulted in dispensing errors and was not smart enough.

[0005] In view of this, there is an urgent need for intelligent dispensing robots based on visual recognition technology to at least solve the above-mentioned deficiencies. Summary of the invention

[0006] One of the purposes of the present invention is to provide an intelligent medicine dispensing robot based on visual recognition technology, which introduces visual recognition technology and automatically fills a first target medicine box according to the attribute information of the medicine box. In addition, the medicine dispensing requirements of the target patient are obtained, and the medicine dispensing plan is determined according to the medicine dispensing requirements and the attribute information of the medicine box. Then, intelligent medicine dispensing is performed based on the medicine dispensing plan, without the need for manual participation in the medicine dispensing process, thereby avoiding the situation of medicine dispensing errors and being more intelligent.

[0007] The intelligent medicine dispensing robot based on visual recognition technology provided by the embodiment of the present invention includes:

[0008] A medicine box attribute information acquisition subsystem, used to acquire medicine box attribute information of a first target medicine box;

[0009] An automatic medicine loading subsystem, used to automatically replenish the first target medicine box based on the visual recognition technology and the medicine box attribute information;

[0010] The medication demand acquisition subsystem is used to obtain the medication demand of target patients;

[0011] The medicine dispensing plan determination subsystem is used to determine the medicine dispensing plan according to the medicine dispensing requirements and the medicine box attribute information;

[0012] The intelligent medicine dispensing subsystem is used to perform intelligent medicine dispensing according to the medicine collection plan.

[0013] Preferably, the medicine box attribute information acquisition subsystem includes:

[0014] An authorized personnel information acquisition module is used to acquire authorized personnel information, where the authorized personnel information is: information of personnel who are allowed to set the medicine box attribute information of the first target medicine box;

[0015] A setting personnel information acquisition module, used to acquire setting personnel information who sets the medicine box attribute information of the first target medicine box;

[0016] A target setting information determination module is used to determine the target setting information of the target setting person according to the authority person information and the setting person information;

[0017] The medicine box attribute information determination module is used to determine the medicine box attribute information according to the target setting information.

[0018] Preferably, the automatic charge subsystem comprises:

[0019] A captured image acquisition module, used to acquire a captured image of the medicine to be replenished;

[0020] The module for determining identification information of the medicine to be replenished is used to determine the identification information of the medicine to be replenished according to the captured image based on the visual recognition technology. The identification information of the medicine to be replenished includes: identifying the packaging and identifying the type of medicine;

[0021] A packaging disassembly plan determination module is used to determine the packaging disassembly plan of the medicine to be replenished according to the identified packaging;

[0022] A medicine box input plan determination module is used to determine the medicine box input plan for the medicine to be replenished based on the identified medicine type and medicine box attribute information;

[0023] The automatic replenishment module is used to automatically replenish the first target medicine box according to the packaging disassembly plan and the medicine box input plan.

[0024] Preferably, the module for determining identification information of the medicine to be replenished includes:

[0025] A text information determination submodule, for determining text information on the outer packaging of the medicine to be replenished according to the captured image based on visual recognition technology;

[0026] A first text semantics determination submodule, used to determine the first text semantics according to the text information;

[0027] A second text semantics determination submodule, used to determine the second text semantics according to the first text semantics, the second text semantics including: drug name semantics and drug packaging semantics;

[0028] A submodule for determining identification information of the drug to be replenished, used for determining identification information of the drug to be replenished according to the semantics of the second text;

[0029] Among them, the text information determination submodule includes:

[0030] An outer packaging contour information determination unit, used to determine the outer packaging contour information according to the captured image;

[0031] A text anchor point determination unit, used to determine the text anchor point on the outer packaging according to the outer packaging contour information and the preset packaging specification;

[0032] A local captured image determining unit, used to determine the local captured image according to the text anchor point;

[0033] The text information determination unit is used to determine the text information based on the character recognition technology and the local captured image.

[0034] Preferably, the medication demand acquisition subsystem includes:

[0035] A medication taking method acquisition module, used to acquire at least one medication taking method input by a target patient;

[0036] A medication dispensing demand determination module is used to determine medication dispensing demand based on medication administration methods;

[0037] Among them, the module for determining the dispensing demand includes:

[0038] The medication information determination submodule is used to determine the expected medication time point and the medication type and medication dosage corresponding to the expected medication time point according to the medication administration method;

[0039] The current time acquisition submodule is used to obtain the current time;

[0040] The medication demand determination submodule is used to take the medication type and medication amount corresponding to the estimated medication time point reached at the current time as the medication demand if the current time reaches any estimated medication time point.

[0041] Preferably, the medication plan determination subsystem includes:

[0042] A first target drug determination module, used to determine a first target drug in a first target medicine box according to the medicine box attribute information;

[0043] A second target medicine box determination module, used to determine a second target medicine box in the first target medicine box according to the medication type in the medication dispensing requirement and the first target medicine;

[0044] A calibration information acquisition module, used to acquire calibration information of a second target medicine box;

[0045] The medication plan determination module is used to determine the medication plan based on the medication dosage and calibration information.

[0046] Preferably, the intelligent medicine dispensing subsystem comprises:

[0047] A calibration medicine box rotation module, used to control the second target medicine box to rotate to below a preset medicine suction tube according to the calibration information;

[0048] The intelligent medicine dispensing module is used to control the medicine aspirating tube to absorb a dosage of the second target medicine from the second target medicine box when the second target medicine box rotates to below the medicine aspirating tube.

[0049] Preferably, the calibration medicine box rotation module comprises:

[0050] A first medicine box number acquisition submodule, used to acquire a first medicine box number of a first target medicine box that is currently closest to a preset medicine aspiration tube;

[0051] A second medicine box number obtaining submodule, used for obtaining a second medicine box number of a second target medicine box that has not been subjected to the aspiration operation;

[0052] A first target difference calculation submodule, used to calculate a first target difference between the first medicine box number and the second medicine box number;

[0053] A second target difference calculation submodule, used for determining a second target difference according to the maximum number of the first target medicine box and the first target difference;

[0054] A comparison result determination submodule, used to determine a comparison result between the first target difference and the second target difference, and associate it with the second target medicine box;

[0055] A third target medicine box control submodule, used to determine the third target medicine box according to the comparison result, and control the third target medicine box to rotate to below the medicine suction tube according to the comparison result of the third target medicine box;

[0056] The loop submodule is used to repeatedly execute the first medicine box number acquisition submodule, the second medicine box number acquisition submodule, the first target difference calculation submodule, the second target difference calculation submodule, the comparison result determination submodule and the third target medicine box control submodule until all the second target medicine boxes that have not undergone the absorption operation have completed the absorption operation.

[0057] Preferably, the intelligent medication dispensing module comprises:

[0058] A captured information determination submodule, used to determine the captured information of the second target drug according to the calibration information;

[0059] A grasping feature determination submodule is used to determine grasping features according to grasping information;

[0060] The medicine dispensing test record acquisition submodule is used to obtain the medicine dispensing test record;

[0061] The medicine dispensing model training submodule is used to train the medicine dispensing model according to the medicine dispensing test records;

[0062] The medicine grabbing strategy acquisition submodule is used to input the grabbing features into the medicine grabbing model to obtain the medicine grabbing strategy of the medicine suction tube. The medicine grabbing strategy includes: the suction nozzle of the medicine suction tube and the suction force of the medicine suction tube;

[0063] The medicine picking submodule is used to control the medicine aspirating tube to suck the second target medicine in the second target medicine box according to the medicine picking strategy.

[0064] The intelligent medicine dispensing robot based on visual recognition technology provided by the embodiment of the present invention also includes:

[0065] Reminder subsystem, used to remind target patients to take medicine in time;

[0066] Among them, the reminder subsystem includes:

[0067] The sensing information acquisition module is used to obtain the sensing information of the medicine dispensing box after the intelligent medicine dispensing is completed;

[0068] The medication collection determination module is used to determine whether the target patient has completed medication collection based on the sensing information;

[0069] A remote binding information acquisition module is used to obtain remote binding information if the target patient has not completed medication collection within the target time after the intelligent medication dispensing is completed;

[0070] A first reminder information determination module, used to determine the first reminder information according to the remote binding information and send it to the remote binding person;

[0071] An environmental information acquisition module, used to acquire environmental information;

[0072] A reminder person identification module is used to try to identify the reminder person based on environmental information;

[0073] The second reminder information broadcasting module is used to broadcast the second reminder information based on the preset voice broadcasting rules if the recognition attempt is successful.

[0074] The intelligent medicine dispensing method based on visual recognition technology provided by the embodiment of the present invention includes:

[0075] Step 1: Obtaining medicine box attribute information of the first target medicine box;

[0076] Step 2: Based on visual recognition technology, automatically replenish the first target medicine box according to the medicine box attribute information;

[0077] Step 3: Obtain the medication requirements of target patients;

[0078] Step 4: Determine the medication plan based on medication requirements and medicine box attribute information;

[0079] Step 5: Dispense medicine intelligently according to the medication plan.

[0080] The beneficial effects of the present invention are:

[0081] The present invention introduces visual recognition technology and automatically fills the first target medicine box according to the medicine box attribute information. In addition, the target patient's medication needs are obtained, and the medication plan is determined according to the medication needs and the medicine box attribute information. Then, intelligent medication is dispensed based on the medication plan, without the need for manual participation in the medication dispensing process, thus avoiding medication dispensing errors and being more intelligent.

[0082] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in the present application documents.

[0083] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0085] Figure 1 A schematic diagram of an intelligent dispensing robot based on visual recognition technology in an embodiment of the present invention;

[0086] Figure 2Schematic diagram of an intelligent medication dispensing method based on visual recognition technology in an embodiment of the present invention. DETAILED DESCRIPTION

[0087] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0088] The embodiment of the present invention provides an intelligent medicine dispensing robot based on visual recognition technology, such as Figure 1 As shown, including:

[0089] The medicine box attribute information acquisition subsystem 1 is used to acquire the medicine box attribute information of the first target medicine box; wherein the first target medicine box is: a medicine box set on the intelligent medicine dispensing robot for storing medicines to be dispensed; the medicine box attribute information is: medicine box number and medicine box contents;

[0090] The automatic medicine loading subsystem 2 is used to automatically replenish the first target medicine box based on the visual recognition technology and the medicine box attribute information; wherein, when automatically replenishing the first target medicine box, the medicine to be replenished and the medicine box attribute information are identified based on the visual recognition technology, and the mechanical arm is used to put the medicine to be replenished into the corresponding first target medicine box;

[0091] The medication demand acquisition subsystem 3 is used to acquire the medication demand of the target patient; wherein the medication demand is: what kind of medicine needs to be dispensed at the current moment, and how much of each medicine is dispensed;

[0092] The medicine taking plan determination subsystem 4 is used to determine the medicine taking plan according to the medicine dispensing requirements and the medicine box attribute information; wherein the medicine taking plan is: when to take a certain amount of medicine from which first target medicine box;

[0093] The intelligent medicine dispensing subsystem 5 is used to perform intelligent medicine dispensing according to the medicine dispensing plan, wherein intelligent medicine dispensing refers to accurately sucking the required amount of medicine using a medicine suction tube connected to a vacuum pump.

[0094] The working principle and beneficial effects of the above technical solution are:

[0095] This application introduces visual recognition technology to automatically fill the first target medicine box according to the medicine box attribute information. In addition, the target patient's medication needs are obtained, and the medication plan is determined based on the medication needs and the medicine box attribute information. Then, intelligent medication is dispensed based on the medication plan, without the need for manual participation in the medication dispensing process, avoiding medication dispensing errors, and being more intelligent.

[0096] In one embodiment, the medicine box attribute information acquisition subsystem includes:

[0097] The authorized personnel information acquisition module is used to acquire the authorized personnel information, where the authorized personnel information is: the personnel information that is allowed to set the medicine box attribute information of the first target medicine box; wherein the personnel information is: the personnel's unique identity information;

[0098] The setting personnel information acquisition module is used to acquire the setting personnel information for setting the medicine box attribute information of the first target medicine box; wherein, setting the medicine box attribute information of the first target medicine box includes: setting what kind of medicine is placed in the first target medicine box;

[0099] The target setting information determination module is used to determine the target setting information of the target setting person according to the authorized person information and the setting person information; wherein the target setting person is: a person who has the setting authority to set the medicine box attribute information of the first target medicine box determined by authority matching based on the setting person information and the authorized person information; the target setting information is: the medicine box attribute information setting operation of the target setting person;

[0100] The medicine box attribute information determination module is used to determine the medicine box attribute information according to the target setting information.

[0101] The working principle and beneficial effects of the above technical solution are:

[0102] This application introduces authorized personnel information and setting personnel information, determines the medicine box attribute information setting operation of the target setting personnel based on the authorized personnel information and the setting personnel information, and determines the medicine box attribute information based on the target setting information, thereby improving the suitability of obtaining the medicine box attribute information.

[0103] In one embodiment, the automatic charge subsystem comprises:

[0104] The captured image acquisition module is used to acquire the captured image of the medicine to be replenished; wherein the medicine to be replenished is: the medicine to be stored in the intelligent dispensing robot; the captured image is: the captured photo of the medicine to be replenished;

[0105] The module for determining identification information of the drugs to be replenished is used to determine the identification information of the drugs to be replenished according to the captured images based on the visual recognition technology. The identification information of the drugs to be replenished includes: identifying the packaging and identifying the type of drugs. Among them, identifying the packaging is: how the drugs to be replenished are packaged, such as: plastic bottles and cans, and for example: bar packaging; identifying the type of drugs is: the identification result of the drug type of the drugs to be replenished;

[0106] The module for determining the packaging disassembly plan is used to determine the packaging disassembly plan of the medicine to be replenished according to the identified packaging; wherein the packaging disassembly plan is: the disassembly plan corresponding to the identified packaging, such as: unscrewing the lid of the medicine bottle, or: puncturing the aluminum packaging;

[0107] The medicine box input plan determination module is used to determine the medicine box input plan of the medicine to be replenished according to the identified medicine type and medicine box attribute information; wherein the medicine box input plan is: into which first target medicine box the medicine is input;

[0108] The automatic replenishment module is used to automatically replenish the first target medicine box according to the packaging disassembly plan and the medicine box input plan.

[0109] The working principle and beneficial effects of the above technical solution are:

[0110] At present, tablets are usually put into medicine boxes by manual assembly. However, the manual loading process is cumbersome and easily causes drug contamination. Therefore, this application introduces the captured image of the medicine to be replenished, and determines the identification packaging and identification type of the medicine to be replenished based on visual recognition technology. According to the identification packaging, the packaging disassembly plan is determined, and then the medicine box input plan is determined based on the identification type of medicine and medicine box attribute information. The packaging of the medicine to be replenished is disassembled based on the packaging disassembly plan, and then based on the medicine box input plan, the unpacked medicine to be replenished is added to the corresponding first target medicine box, which is more hygienic and convenient.

[0111] In one embodiment, the module for determining identification information of the medicine to be replenished includes:

[0112] The text information determination submodule is used to determine the text information on the outer packaging of the medicine to be replenished based on the captured image based on the visual recognition technology; wherein the text information is: the identification characters on the outer packaging of the medicine to be replenished;

[0113] A first text semantics determination submodule is used to determine the first text semantics according to the text information; wherein the first text semantics is: the meaning represented by the text information;

[0114] The second text semantics determination submodule is used to determine the second text semantics according to the first text semantics, wherein the second text semantics includes: drug name semantics and drug packaging semantics; wherein the drug name semantics is: the meaning of the drug name represented by the text information; and the drug packaging semantics is: the meaning of the drug packaging represented by the text information;

[0115] A submodule for determining identification information of the drug to be replenished, used for determining identification information of the drug to be replenished according to the semantics of the second text;

[0116] Among them, the text information determination submodule includes:

[0117] The outer packaging contour information determining unit is used to determine the outer packaging contour information according to the captured image; wherein the outer packaging contour information is: the outer packaging three-dimensional scanning information;

[0118] The text anchor point determination unit is used to determine the text anchor point on the outer packaging according to the outer packaging contour information and the preset packaging specification; wherein the preset packaging specification is: the packaging standard of the drug; the text anchor point is: the standard marking position of the text on the outer packaging contour information;

[0119] A partial captured image determining unit is used to determine a partial captured image according to a text anchor point; wherein the partial captured image is: a partial image corresponding to the text anchor point in the captured image;

[0120] The text information determination unit is used to determine the text information based on the character recognition technology and the local captured image.

[0121] The working principle and beneficial effects of the above technical solution are:

[0122] This application determines the outer packaging contour information of the drug to be supplemented, determines the text anchor point on the outer packaging based on the outer packaging contour information and the preset packaging specifications, extracts the local captured image corresponding to the text anchor point, and determines the text information based on the character recognition technology and the local captured image. The meaning represented by the text information is obtained, and the semantics of the drug name and drug packaging therein are determined, and the semantics of the drug name and drug packaging are used as the identification information of the drug to be supplemented, thereby improving the efficiency and accuracy of obtaining the identification information of the drug to be supplemented.

[0123] In one embodiment, the medication requirement acquisition subsystem includes:

[0124] The drug taking method acquisition module is used to acquire at least one drug taking method input by the target patient; wherein the drug taking method is: what kind of drug to take and how to take it, for example: drug A, one tablet in the morning and one tablet in the evening, two tablets each;

[0125] A medication dispensing demand determination module is used to determine medication dispensing demand based on medication administration methods;

[0126] Among them, the module for determining the dispensing demand includes:

[0127] The medication information determination submodule is used to determine the expected medication time point and the medication type and medication dosage corresponding to the expected medication time point according to the medication administration method;

[0128] The current time acquisition submodule is used to obtain the current time;

[0129] The medication demand determination submodule is used to take the medication type and medication amount corresponding to the estimated medication time point reached at the current time as the medication demand if the current time reaches any estimated medication time point.

[0130] The working principle and beneficial effects of the above technical solution are:

[0131] This application obtains the medication taking method input by the target patient, and determines the immediate medication dispensing needs based on the medication taking method, so that the acquisition of medication dispensing needs is more reasonable.

[0132] In one embodiment, the medication plan determination subsystem includes:

[0133] A first target drug determination module is used to determine the first target drug in the first target drug box according to the drug box attribute information; wherein the first target drug is: the drug stored in the first target drug box;

[0134] The second target medicine box determination module is used to determine the second target medicine box in the first target medicine box according to the medication type in the medication dispensing requirement and the first target medicine; wherein the second target medicine box is: the medicine box corresponding to the medication dispensing requirement;

[0135] The calibration information acquisition module is used to acquire the calibration information of the second target medicine box; wherein the calibration information is: the medicine box attribute information of the second target medicine box, including: the medicine box number, the current position and the information of the contents of the medicine box, etc.;

[0136] The medication plan determination module is used to determine the medication plan based on the medication dosage and calibration information.

[0137] The working principle and beneficial effects of the above technical solution are:

[0138] The present application determines the first target drug in the first target medicine box, and determines the medicine box corresponding to the medicine dispensing requirement, that is, the second target medicine box, according to the type of medicine in the dispensing requirement and the first target drug. Obtain the medicine box number and current position of the second target medicine box, determine the medicine dispensing plan according to the dosage and calibration information, and obtain the medicine dispensing plan more appropriately.

[0139] In one embodiment, the intelligent medication dispensing subsystem includes:

[0140] The calibration medicine box rotation module is used to control the second target medicine box to rotate to the bottom of the preset medicine suction tube according to the calibration information; wherein, the second target medicine box is controlled to rotate to the bottom of the preset medicine suction tube by starting the turntable stepper motor on which the first target medicine box is placed to rotate forward, driving the medicine box rack to rotate, and sending the second target medicine box fixed on the rack to the bottom of the medicine suction tube; the medicine suction tube is connected to a vacuum pump;

[0141] The intelligent medicine dispensing module is used to control the medicine aspirating tube to absorb the second target medicine in the second target medicine box when the second target medicine box rotates to the bottom of the medicine aspirating tube. The second target medicine is the medicine in the second target medicine box.

[0142] The working principle and beneficial effects of the above technical solution are:

[0143] The present application determines the relative position of the second target medicine box in the medicine box rack based on the calibration information, controls the second target medicine box to rotate to below a preset medicine suction tube, and then controls the medicine suction tube to suck the second target medicine dose from the second target medicine box, thereby improving the accuracy of medicine dispensing.

[0144] In one embodiment, calibrating the medicine box rotation module includes:

[0145] The first medicine box number acquisition submodule is used to acquire the first medicine box number of the first target medicine box that is currently closest to the preset medicine aspiration tube; wherein the first medicine box number is: the medicine box number of the first target medicine box that is currently closest to the preset medicine aspiration tube, for example: 3;

[0146] The second medicine box number acquisition submodule is used to acquire the second medicine box number of the second target medicine box that has not been subjected to the suction operation; wherein the suction operation is: a medicine taking operation corresponding to the current medicine taking demand; the second medicine box number is: the medicine box number of the second target medicine box that has not been subjected to the suction operation, such as: 2, 5 and 8;

[0147] A first target difference calculation submodule is used to calculate a first target difference between the first medicine box number and the second medicine box number; wherein the first target difference is: a result obtained by subtracting the first medicine box number from the second medicine box number, such as: +1, -2 and -5;

[0148] The second target difference calculation submodule is used to determine the second target difference according to the maximum number of the first target medicine boxes and the first target difference; wherein the maximum number is determined according to the total number of the first target medicine boxes, for example: if the total number of the first target medicine boxes is 8, then the maximum number is 8; the second target difference is: if the first target difference is a positive number, then the second target difference is the maximum number minus the first target difference; if the first target difference is not a positive number, then the second target difference is the maximum number plus the first target difference;

[0149] A comparison result determination submodule is used to determine the comparison result of the first target difference and the second target difference, and associate it with the second target medicine box; wherein the comparison result is: the first target difference or the second target difference with the smaller absolute value among the first target difference and the second target difference;

[0150] The third target medicine box control submodule is used to determine the third target medicine box according to the comparison result, and control the third target medicine box to rotate to the bottom of the medicine suction tube according to the comparison result of the third target medicine box; wherein, when determining the third target medicine box according to the comparison result, the second target medicine box with the smallest absolute value of the comparison result is used as the third target medicine box;

[0151] The loop submodule is used to repeatedly execute the first medicine box number acquisition submodule, the second medicine box number acquisition submodule, the first target difference calculation submodule, the second target difference calculation submodule, the comparison result determination submodule and the third target medicine box control submodule until all the second target medicine boxes that have not undergone the absorption operation have completed the absorption operation.

[0152] The working principle and beneficial effects of the above technical solution are:

[0153] The present application determines the medicine box number of the first target medicine box that is currently closest to the preset medicine suction tube, and at the same time, obtains the medicine box number of the second target medicine box that has not been sucked, and determines the second target difference according to the maximum number of the first target medicine box and the first target difference. The first target difference or the second target difference with the smaller absolute value among the first target difference and the second target difference is obtained as the comparison result. The smaller the absolute value of the comparison result, the closer the corresponding second target medicine box is to the medicine suction tube. The second target medicine box with the smallest absolute value of the comparison result is used as the third target medicine box. At the same time, the comparison result is analyzed, and the rotation direction of the turntable stepper motor control is determined according to the sign of the comparison result, and the rotation angle of the turntable stepper motor control is determined according to the absolute value of the comparison result. The turntable stepper motor controls the minimum rotation variable, thereby improving the dispensing speed.

[0154] In one embodiment, the intelligent medication dispensing module comprises:

[0155] The captured information determination submodule is used to determine the captured information of the second target drug according to the calibration information; wherein the captured information is: information related to the captured second target drug, such as: the existence form of the second target drug, including: capsules, tablets and glass jars;

[0156] The grab feature determination submodule is used to determine the grab features according to the grab information; wherein the grab feature is: a characteristic representation of the grab information, such as: the grab information type and the corresponding information content (for example: the form of the drug: capsule, weight: 90 mg, etc.);

[0157] The medicine dispensing test record acquisition submodule is used to obtain the medicine dispensing test record; wherein the medicine dispensing test record is: a record of the process of manually setting the suction nozzle and the suction pressure of the medicine suction tube to dispensing medicine, and the medicine dispensing test record only selects the record of successful medicine dispensing;

[0158] The medicine dispensing model training submodule is used to train the medicine dispensing model according to the medicine dispensing test records; wherein the medicine dispensing model is: an AI model that learns the dispensing logic of the intelligent dispensing robot;

[0159] The medicine grabbing strategy acquisition submodule is used to input the grabbing features into the medicine grabbing model to obtain the medicine grabbing strategy of the medicine suction tube. The medicine grabbing strategy includes: the suction nozzle of the medicine suction tube and the suction force of the medicine suction tube;

[0160] The medicine picking submodule is used to control the medicine suction tube to suck the second target medicine in the second target medicine box according to the medicine picking strategy. The medicine suction tube can be preset with a variety of suction nozzles. After the suction nozzle of the medicine suction tube of the medicine picking strategy is determined, the suction nozzle of the corresponding shape is selected to suck the medicine.

[0161] The working principle and beneficial effects of the above technical solution are:

[0162] The present application determines the grabbing information of the second target drug, and characterizes the grabbing information to obtain the grabbing features. According to the successful grabbing test records of the drug grabbing results, the drug grabbing model is trained, the grabbing features are input into the drug grabbing model, the appropriate suction nozzle of the drug suction tube and the suction force of the drug suction tube are determined, and according to the drug grabbing strategy, the drug suction tube is controlled to absorb the second target drug in the second target medicine box, avoiding the situation of falling or failing to suck up the drug when sucking it, which is more appropriate.

[0163] The embodiment of the present invention provides an intelligent medicine dispensing robot based on visual recognition technology, further comprising:

[0164] Reminder subsystem, used to remind target patients to take medicine in time;

[0165] Among them, the reminder subsystem includes:

[0166] The sensing information acquisition module is used to obtain the sensing information of the medicine dispensing box after the intelligent medicine dispensing is completed; wherein the sensing information is: the relevant information of whether the medicine in the medicine dispensing box has been taken out, such as: the gravity sensing information of the medicine dispensing box;

[0167] The medication collection determination module is used to determine whether the target patient has completed medication collection based on the sensing information;

[0168] The remote binding information acquisition module is used to obtain remote binding information if the target patient has not completed the medication within the target time after the intelligent medication dispensing is completed; wherein the target time is preset manually; the remote binding information is: the remote terminal bound to the intelligent medication dispensing robot;

[0169] A first reminder information determination module is used to determine the first reminder information according to the remote binding information and send it to the remote binding personnel; wherein the first reminder information is: information on the remote terminal sent by the intelligent dispensing robot to the remote binding personnel, such as: "It is detected that the medicines dispensed by the intelligent dispensing robot associated with you have not been taken away";

[0170] An environmental information acquisition module is used to acquire environmental information; wherein the environmental information is: three-dimensional scanning information within a preset range around the intelligent dispensing machine;

[0171] A reminder person identification module is used to try to identify the reminder person according to the environmental information; wherein the reminder person is: the person corresponding to the human body model identified in the three-dimensional scanning information;

[0172] The second reminder information broadcast module is used to broadcast the second reminder information based on the preset voice broadcast rules if the recognition attempt is successful. The preset voice broadcast rules are manually pre-set; the second reminder information is: voice broadcast information, such as: "The medicine has not been taken out, please take it out in time."

[0173] The working principle and beneficial effects of the above technical solution are:

[0174] After the medicine dispensing box completes the dispensing, the target patient may not necessarily take the medicine in time, and there may be cases where the patient forgets to take the medicine. Therefore, this application obtains the sensing information of the medicine dispensing box after the intelligent medicine dispensing is completed, and determines whether the target patient has completed taking the medicine based on the sensing information. If the target patient has not completed taking the medicine within the target time after the intelligent medicine dispensing is completed, a first reminder message is sent to the remote terminal bound to the intelligent medicine dispensing robot. The remote bound person can be the target patient himself or a relative of the target patient, which is convenient for timely reminders. In addition, three-dimensional scanning information within a preset range around the intelligent medicine dispensing machine is introduced. When the reminder person is identified, a voice broadcast rule is introduced to broadcast the second reminder message, which improves the comprehensiveness of the reminder.

[0175] The embodiment of the present invention provides an intelligent prescription method based on visual recognition technology, such as Figure 2 As shown, including:

[0176] Step 1: Obtaining medicine box attribute information of the first target medicine box;

[0177] Step 2: Based on visual recognition technology, automatically replenish the first target medicine box according to the medicine box attribute information;

[0178] Step 3: Obtain the medication requirements of target patients;

[0179] Step 4: Determine the medication plan based on medication requirements and medicine box attribute information;

[0180] Step 5: Dispense medicine intelligently according to the medication plan.

[0181] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. Intelligent dispensing robot based on visual recognition technology, characterized by: include: A medicine box attribute information acquisition subsystem, used to acquire medicine box attribute information of a first target medicine box; An automatic medicine loading subsystem, used to automatically replenish the first target medicine box based on the visual recognition technology and the medicine box attribute information; The medication demand acquisition subsystem is used to obtain the medication demand of target patients; The medicine dispensing plan determination subsystem is used to determine the medicine dispensing plan according to the medicine dispensing requirements and the medicine box attribute information; Intelligent medicine dispensing subsystem, used to intelligently dispense medicine according to the medicine dispensing plan; Automatic charging subsystem, including: A captured image acquisition module, used to acquire a captured image of the medicine to be replenished; The module for determining identification information of the medicine to be replenished is used to determine the identification information of the medicine to be replenished according to the captured image based on the visual recognition technology. The identification information of the medicine to be replenished includes: identifying the packaging and identifying the type of medicine; A packaging disassembly plan determination module is used to determine the packaging disassembly plan of the medicine to be replenished according to the identified packaging; A medicine box input plan determination module is used to determine the medicine box input plan for the medicine to be replenished based on the identified medicine type and medicine box attribute information; An automatic replenishment module, used to automatically replenish the first target medicine box according to the packaging disassembly plan and the medicine box input plan; The module for determining identification information of drugs to be replenished includes: A text information determination submodule, for determining text information on the outer packaging of the medicine to be replenished according to the captured image based on visual recognition technology; A first text semantics determination submodule, used to determine the first text semantics according to the text information; A second text semantics determination submodule, used to determine the second text semantics according to the first text semantics, the second text semantics including: drug name semantics and drug packaging semantics; A submodule for determining identification information of the drug to be replenished, used for determining identification information of the drug to be replenished according to the semantics of the second text; Among them, the text information determination submodule includes: An outer packaging contour information determination unit, used to determine the outer packaging contour information according to the captured image; A text anchor point determination unit, used to determine the text anchor point on the outer packaging according to the outer packaging contour information and the preset packaging specification; A local captured image determining unit, used to determine the local captured image according to the text anchor point; A text information determination unit, used to determine text information based on character recognition technology and a local captured image; Intelligent drug dispensing subsystem, including: A calibration medicine box rotation module, used to control the second target medicine box to rotate to below a preset medicine suction tube according to the calibration information; An intelligent medicine dispensing module, used for controlling the medicine aspirating tube to absorb a dose of the second target medicine from the second target medicine box when the second target medicine box rotates to below the medicine aspirating tube; Calibration cartridge rotation module, including: A first medicine box number acquisition submodule, used to acquire a first medicine box number of a first target medicine box that is currently closest to a preset medicine aspiration tube; A second medicine box number obtaining submodule, used for obtaining a second medicine box number of a second target medicine box that has not been subjected to the aspiration operation; A first target difference calculation submodule, used to calculate a first target difference between the first medicine box number and the second medicine box number; A second target difference calculation submodule, used for determining a second target difference according to the maximum number of the first target medicine box and the first target difference; A comparison result determination submodule, used to determine a comparison result between the first target difference and the second target difference, and associate it with the second target medicine box; A third target medicine box control submodule, used to determine the third target medicine box according to the comparison result, and control the third target medicine box to rotate to below the medicine suction tube according to the comparison result of the third target medicine box; a loop submodule, for repeatedly executing the first medicine box number acquisition submodule, the second medicine box number acquisition submodule, the first target difference calculation submodule, the second target difference calculation submodule, the comparison result determination submodule, and the third target medicine box control submodule until all the second target medicine boxes that have not been subjected to the absorption operation have completed the absorption operation; Intelligent medication dispensing module, including: A captured information determination submodule, used to determine the captured information of the second target drug according to the calibration information; A grasping feature determination submodule is used to determine grasping features according to grasping information; The medicine dispensing test record acquisition submodule is used to obtain the medicine dispensing test record; The medicine dispensing model training submodule is used to train the medicine dispensing model according to the medicine dispensing test records; The medicine grabbing strategy acquisition submodule is used to input the grabbing features into the medicine grabbing model to obtain the medicine grabbing strategy of the medicine suction tube. The medicine grabbing strategy includes: the suction nozzle of the medicine suction tube and the suction force of the medicine suction tube; The medicine picking submodule is used to control the medicine aspirating tube to suck the second target medicine in the second target medicine box according to the medicine picking strategy.

2. The intelligent drug dispensing robot based on visual recognition technology as claimed in claim 1, characterized in that: The medicine box attribute information acquisition subsystem includes: An authorized personnel information acquisition module is used to acquire authorized personnel information, where the authorized personnel information is: information of personnel who are allowed to set the medicine box attribute information of the first target medicine box; A setting personnel information acquisition module, used to acquire setting personnel information who sets the medicine box attribute information of the first target medicine box; A target setting information determination module is used to determine the target setting information of the target setting person according to the authority person information and the setting person information; The medicine box attribute information determination module is used to determine the medicine box attribute information according to the target setting information.

3. The intelligent drug dispensing robot based on visual recognition technology as claimed in claim 1, characterized in that: The medication demand acquisition subsystem includes: A medication taking method acquisition module, used to acquire at least one medication taking method input by a target patient; A medication dispensing demand determination module is used to determine medication dispensing demand based on medication administration methods; Among them, the module for determining the dispensing demand includes: The medication information determination submodule is used to determine the expected medication time point and the medication type and medication dosage corresponding to the expected medication time point according to the medication administration method; The current time acquisition submodule is used to obtain the current time; The medication demand determination submodule is used to take the medication type and medication amount corresponding to the estimated medication time point reached at the current time as the medication demand if the current time reaches any estimated medication time point.

4. The intelligent drug dispensing robot based on visual recognition technology as claimed in claim 1, characterized in that: The medication plan determination subsystem includes: A first target drug determination module is used to determine the first target drug in the first target drug box according to the drug box attribute information; wherein the first target drug is: the drug stored in the first target drug box; The second target medicine box determination module is used to determine the second target medicine box in the first target medicine box according to the medication type in the medication dispensing requirement and the first target medicine; wherein the second target medicine box is: the medicine box corresponding to the medication dispensing requirement; A calibration information acquisition module, used to acquire calibration information of a second target medicine box; The medication plan determination module is used to determine the medication plan based on the medication dosage and calibration information.

5. The intelligent drug dispensing robot based on visual recognition technology as claimed in claim 1, characterized in that: Also includes: Reminder subsystem, used to remind target patients to take medicine in time; Among them, the reminder subsystem includes: The sensing information acquisition module is used to obtain the sensing information of the medicine dispensing box after the intelligent medicine dispensing is completed; The medication collection determination module is used to determine whether the target patient has completed medication collection based on the sensing information; A remote binding information acquisition module is used to obtain remote binding information if the target patient has not completed medication collection within the target time after the intelligent medication dispensing is completed; A first reminder information determination module, used to determine the first reminder information according to the remote binding information and send it to the remote binding person; An environmental information acquisition module, used to acquire environmental information; A reminder person identification module is used to try to identify the reminder person based on environmental information; The second reminder information broadcasting module is used to broadcast the second reminder information based on the preset voice broadcasting rules if the recognition attempt is successful.

6. An intelligent medication dispensing method based on visual recognition technology, characterized in that: include: Step 1: Obtaining medicine box attribute information of the first target medicine box; Step 2: Based on visual recognition technology, automatically replenish the first target medicine box according to the medicine box attribute information; Step 3: Obtain the medication requirements of target patients; Step 4: Determine the medication plan based on medication requirements and medicine box attribute information; Step 5: Intelligently dispense medicine according to the medication plan; Step 2: Based on visual recognition technology, automatically replenish the first target medicine box according to the medicine box attribute information, including: Acquire a photographed image of the medicine to be replenished; Based on visual recognition technology, the identification information of the medicine to be replenished is determined according to the captured image. The identification information of the medicine to be replenished includes: identifying the packaging and identifying the type of medicine; Determine the packaging disassembly plan for the medicine to be replenished based on the identified packaging; Determine the medicine box input plan for the medicine to be replenished based on the identified drug type and medicine box attribute information; Automatically replenish the first target medicine box according to the packaging disassembly plan and medicine box input plan; Based on visual recognition technology, the identification information of the medicine to be replenished is determined according to the captured image, including: Based on visual recognition technology, the text information on the outer packaging of the medicine to be replenished is determined according to the captured image; Determine the first text semantics according to the text information; Determine the second text semantics according to the first text semantics, the second text semantics including: drug name semantics and drug packaging semantics; Determine identification information of the drug to be replenished according to the semantics of the second text; Among them, based on visual recognition technology, the text information on the outer packaging of the medicine to be replenished is determined according to the captured image, including: Determine the outer packaging contour information according to the captured image; Determine the text anchor point on the outer packaging according to the outer packaging contour information and the preset packaging specifications; According to the text anchor point, determine the local captured image; Determine text information based on character recognition technology and local captured images; Step 5: Perform intelligent medication dispensing according to the medication plan, including: According to the calibration information, the second target medicine box is controlled to rotate to below the preset medicine suction tube; When the second target medicine box rotates to below the medicine suction tube, the medicine suction tube is controlled to suck the second target medicine in the second target medicine box; According to the calibration information, the second target medicine box is controlled to rotate to below the preset medicine suction tube, including: Obtaining the first medicine box number of the first target medicine box that is currently closest to the preset medicine aspiration tube; Acquire a second medicine box number of a second target medicine box that has not been subjected to the aspiration operation; Calculate a first target difference between the first medicine box number and the second medicine box number; Determining a second target difference value according to the maximum number of the first target medicine box and the first target difference value; Determine a comparison result of the first target difference and the second target difference, and associate it with the second target medicine box; Determine a third target medicine box according to the comparison result, and control the third target medicine box to rotate to below the medicine suction tube according to the comparison result of the third target medicine box; Repeat the first medicine box number acquisition submodule, the second medicine box number acquisition submodule, the first target difference calculation submodule, the second target difference calculation submodule, the comparison result determination submodule and the third target medicine box control submodule until all the second target medicine boxes that have not been sucked have completed the sucking operation; When the second target medicine box rotates to the bottom of the medicine suction tube, controlling the medicine suction tube to suck the second target medicine in the second target medicine box includes: Determine the capture information of the second target drug according to the calibration information; Determine the crawling features based on the crawling information; Obtain medication test records; Train the medicine dispensing model based on the medicine dispensing test records; The grabbing features are input into the drug grabbing model to obtain the drug grabbing strategy of the drug suction tube, which includes: the suction mouth of the drug suction tube and the suction force of the drug suction tube; According to the medicine picking strategy, the medicine suction tube is controlled to suck the second target medicine in the second target medicine box.

Images