A verification platform omnidirectional recognition device

Abstract

This invention discloses a full-angle recognition device for a verification platform, comprising a part for collecting expiration date and batch number information and a part for collecting drug appearance information features. The part for collecting expiration date and batch number information includes one grayscale machine vision camera, one set of infrared photoelectric triggers, and five strip auxiliary light sources. The part for collecting drug appearance information features includes one color machine vision camera, two auxiliary light strips, a warning light strip, and a transparent drug verification platform. This invention employs a combination of color and grayscale machine vision cameras, assisted by light sources, to achieve effective image acquisition of drugs and rapid matching of multi-point information. It can simultaneously and quickly complete the verification of drug type, expiration date, and batch number. Furthermore, the high-performance processor significantly improves the device's operating speed and ensures its reliability.

Description

Technical Field

[0001] This invention relates to the field of drug verification technology, and in particular to a verification platform omnidirectional recognition device. Background Technology

[0002] A drug verification platform is a device used at hospital pharmacy dispensing windows to verify the consistency between prescription and actual dispensing medications, and to return the prescriptions and dispensing products. During operation, the device needs to quickly identify both the type of drug and its expiration date and batch number. In most cases, drug type verification requires text and images from the front, back, or top and bottom sides of the drug to assist the verification process. However, the expiration date and batch number are usually printed on the left and right sides of the drug box, not the front, back, or top and bottom sides. Using conventional machine vision methods for drug type and expiration date / batch number identification requires dividing the verification process into two parts: drug appearance verification and expiration date / batch number verification. This increases the verification time and reduces efficiency. When the device is running at the dispensing window, a slow speed can lead to delayed dispensing, overcrowding at the dispensing window, and unnecessary problems.

[0003] To address this issue, we propose a full-angle recognition device for a verification platform. Summary of the Invention

[0004] The purpose of this invention is to solve the problems existing in the prior art, which require two parts to identify the type, expiration date and batch number of medicines, namely, to verify the appearance of the medicine and verify the expiration date and batch number. This results in long verification time and low efficiency, which in turn leads to untimely medicine distribution and accumulation of queues. Therefore, this invention proposes a verification platform with an all-angle recognition device.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A verification platform with full-angle recognition device includes a part for collecting expiration date and batch number information and a part for collecting drug appearance information features. The part for collecting expiration date and batch number information includes a gray machine vision camera, a set of infrared photoelectric triggers, and five strip auxiliary light sources. The part for collecting drug appearance information features includes a color machine vision camera, two auxiliary light strips, a prompt light strip, and a transparent drug verification platform.

[0007] As a further development in this invention, the gray machine vision camera in the part that collects expiration date and batch number information is mainly responsible for image acquisition of the feature surface of the drug's expiration date and batch number.

[0008] As a further development of this invention, four of the five strip-shaped auxiliary light sources are distributed in the four directions of the gray machine vision camera, which are used to provide normal illumination and flashing light for generating stamped characters. The last of the five strip-shaped auxiliary light sources is above the other four strip-shaped auxiliary light sources for auxiliary illumination, and can be turned on and off adaptively according to the operating conditions of the equipment to compensate for brightness.

[0009] As a further development in this invention, the infrared photoelectric trigger serves as a trigger marker for the start of the verification process.

[0010] As a further aspect of this invention, the color machine vision can output both color and grayscale images. When the camera is started, it performs adaptive parameter optimization on the camera, optimizing the exposure time and white balance to ensure that the image features during image acquisition are closer to the real features of the verified drug.

[0011] As a further feature of this invention, the indicator light strip is a feedback signal light, which provides timely feedback on the review process, prompts the user on the progress of the drug review, and has three colors: blue, green and red.

[0012] As a further feature of this invention, the transparent verification platform is the part where medicines are placed for verification. It is made of high-transmittance glass to ensure the accuracy of image acquisition.

[0013] As a further aspect of this invention, the transparent verification platform includes drug model generation, which is one of the core functions of the drug verification platform. It mainly relies on the drug verification platform to collect drug information, process the collected images to extract the drug model, and save the drug model.

[0014] As a further step in this invention, the drug model generation includes the following steps: S1, drug information query; S2, collection of basic drug appearance information; S3, collection of drug appearance features; S4, collection of drug character information; S5, saving the model.

[0015] As a further step in this invention, the drug verification process is as follows: Step 1: Scan the drug model and prescription information in the drug model; Step 2: Collect the image features of the drug placed on the drug verification platform; Step 3: Verify the drug information. If a verification error occurs, proceed to Step 4. If there is no error, proceed to Step 5; Step 4: Re-verify; Step 5: Determine whether the verification is complete; Step 6: Drug verification is complete.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0017] This invention mainly relies on machine vision technology, which involves acquiring images of production materials through image acquisition devices, analyzing and processing the acquired images (such as image blurring, smoothing, binarization, etc., and extracting image features), and using the extracted image features to build a model of the production materials. When needed in production, the established visual model can be called to match the production materials. The implementation of machine vision technology usually depends on components such as machine vision cameras, light sources, and controllers.

[0018] This device uses a combination of color machine vision cameras and grayscale machine vision cameras, along with a light source and a special positioning design, to achieve 360-degree image acquisition. At the same time, it is equipped with a high-performance processor, which greatly improves the device's operating speed and ensures its reliability.

[0019] In this invention, by combining machine vision technology with a drug model, the collected multi-point information is quickly matched, enabling the simultaneous and rapid verification of drug type, expiration date, and batch number.

[0020] Specifically, the drug verification platform can perform 360-degree panoramic batch number and expiration date identification. It proposes a unified modeling method for different drugs and uses the device and supporting modeling methods to achieve one-time verification of drug appearance, expiration date and batch number. This shortens the drug verification process and improves the efficiency of drug verification, so that drug distribution is not affected, thereby preventing the accumulation of people at the drug distribution window and causing unnecessary trouble. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of a method for generating a drug model in a full-angle recognition device for a verification platform proposed in this invention;

[0022] Figure 2 This is a flowchart illustrating the workflow of a drug verification platform in a full-angle identification device for verification platforms proposed in this invention. Detailed Implementation

[0023] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0024] In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0025] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installed," "equipped with," "sleeved / connected," "connected," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0026] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0027] A verification platform omnidirectional recognition device includes three parts: device hardware, drug model generation, and device usage method. The device hardware includes a part for collecting expiration date and batch number information and a part for collecting drug appearance information features. Further, the part for collecting expiration date and batch number information includes one gray machine vision camera, one set of infrared photoelectric triggers, and five strip auxiliary light sources. The part for collecting drug appearance information features includes one color machine vision camera, two auxiliary light strips, a prompt light strip, and a transparent drug verification platform. Furthermore, both the gray machine vision camera and the color machine vision camera have at least 5 million pixels.

[0028] The following is a detailed explanation of the above:

[0029] The main function of the expiration date and batch number information collection section is to use a combination of camera and light to collect the expiration date and batch number information on the side of the drug's appearance that contains the information. The expiration date and batch number information that can be collected includes expiration date and batch number with ordinary printed font, dot matrix character, and stamped character.

[0030] Among them, the gray machine vision camera in the part that collects the expiration date and batch number information is mainly responsible for the image acquisition of the feature surface of the drug's expiration date and batch number. When the camera is started, it will be initialized and the camera parameters (such as exposure time, brightness, response time) will be adaptively optimized to ensure that the acquired images have good recognition during the image acquisition process. At the same time, during the image acquisition process, image detection will be performed adaptively and the acquired images will be cropped to improve the information concentration of the acquired images without affecting the effectiveness of the acquisition.

[0031] Four of the five strip auxiliary light sources are distributed in the four directions of the gray machine vision camera (up, down, left, right) and have two functions: one is to provide normal illumination, and the other is to provide flashing light (flash) for the generation of stamped characters. By acquiring images by alternately lighting them in the four directions (up, down, left, right), and then synthesizing the images, a high-contrast stamped image can be obtained based on the image shadow features in each direction, thus completing the acquisition of the stamped image.

[0032] In addition to the four strip auxiliary light sources distributed in the top, bottom, left, and right directions of the gray machine vision camera, the remaining strip auxiliary light source is used above the other four strip auxiliary light sources for auxiliary lighting. It can be turned on and off adaptively according to the equipment's operating conditions to compensate for brightness.

[0033] The infrared photoelectric trigger serves as the start of the verification process. This trigger replaces the manual clicking of buttons or the use of operation switches in the software program, thereby improving the automation level of the equipment and increasing work efficiency.

[0034] The drug appearance information feature collection section mainly involves collecting information features of the drug's appearance, including text, graphics, and colors;

[0035] The color machine vision system can output both color and grayscale images. Upon camera startup, it adaptively optimizes camera parameters, including exposure time and white balance, to ensure that image features closely approximate the actual characteristics of the drug being verified. When acquiring a standard frontal image of the drug, a color image can be output to provide a more intuitive visual experience. For ordinary boxed drugs, considering the need for efficiency and accuracy in drug verification, grayscale images are prioritized and output as the appearance feature image when acquiring images of the drug's appearance features. For drugs with multiple specifications, since their appearances are basically similar or their information is completely identical, with only some color differences, a color image can be output, using the image color as an important reference for drug verification.

[0036] The two auxiliary light strips primarily serve to provide supplementary lighting for illumination.

[0037] The indicator light strip is a feedback signal light, providing timely feedback on the review process and indicating the progress of the medication review to the user. It has three colors: blue, green, and red. Details are as follows:

[0038] The blue color indicates that when the teacher at the window scans the prescription, the device enters the medication verification state, and the light strip displays blue until the prescription verification enters the pass state (including manual pass and automatic pass of device verification) or the error state, at which point its display color changes to green or red.

[0039] The indicator light is green when the prescription is scanned. From the start of the prescription review process to the completion of the review, if there are no instances where the actual quantity of drugs dispensed exceeds the quantity that should be dispensed, the expiration date or batch number of the drugs has expired, or any drugs in the prescription are not identified or have expiration dates or batch numbers that are not identified, the prescription review is successful and the indicator light will be green.

[0040] A red indicator light indicates a verification error. From the start of the prescription verification process until the verification of all medications in the prescription is completed, if any of the following occurs: the actual quantity of medications dispensed exceeds the quantity that should have been dispensed, the expiration date or batch number of the medications has expired, or any medications in the prescription were not identified, or the expiration date or batch number of any medications in the prescription were not identified, the medication verification error will be indicated by a red indicator light.

[0041] The transparent verification platform is where medicines are placed for verification. It is made of high-transparency glass to ensure the accuracy of image acquisition.

[0042] The transparent verification platform includes drug model generation, which is one of the core functions of the drug verification platform. It mainly relies on the platform to collect drug information, process the collected images to extract the drug model, and save the drug model. Drug model generation involves the following steps (refer to...). Figure 1 ):

[0043] S1. Drug information query: Query the drug database to find drugs for which a drug model needs to be built;

[0044] S2. Basic appearance information of drugs: Collect the overall appearance of drugs in the drug maintenance interface of the drug verification platform, select the appearance outline of the drug in the image captured by the camera, and learn the appearance of the drug.

[0045] S3. Drug appearance feature collection: Collect drug appearance features in the drug maintenance interface of the drug verification platform, select drug appearance features in the images captured by the camera, and learn the drug appearance features.

[0046] S4. Drug character information collection: Collect drug character features in the drug maintenance interface of the drug verification platform, select drug character appearance features in the image captured by the camera, extract the drug expiration date and batch number, and learn and save the drug expiration date and batch number information.

[0047] S5. Save the model: First, input the basic information of the drug, then collect the appearance features of the drug through the image acquisition unit, then collect the expiration date and batch number features of the drug through the image acquisition unit, and finally save the drug model to the local machine and the database.

[0048] The drug verification platform is an important tool for drug verification. During its operation, the platform matches real-time images with the models it loads to verify the verification results. The drug verification workflow is as follows (refer to...). Figure 2 ):

[0049] Step 1: Scan the drug model and prescription information in the drug model; scan the prescription and load the drug model and prescription information of the prescription in the review from the drug model library.

[0050] The second step is to collect image features of the medicines placed on the medicine verification platform. The medicines are placed on the working platform of the medicine verification platform, and the image features of the medicines are collected in real time. The verification is triggered by an infrared trigger, and the light strip on the bottom of the device indicates that it is in operation.

[0051] The third step is to verify the drug information. The side camera and the bottom camera collect the drug's expiration date, batch number, and appearance characteristics, respectively. The real-time characteristics of the collected drugs are matched with the information in the drug database. The results are analyzed based on the matching degree and whether the matching is successful to determine whether the drug placed has passed the verification. The verification results are then returned and statistically analyzed. If a verification error occurs, proceed to the fourth step. If there is no error, proceed to the fifth step.

[0052] Step 4: Re-verification. When errors occur during drug verification, such as the quantity of drugs dispensed exceeding the quantity required in the prescription, the expiration date or batch number of the drug being expired, the drug's appearance characteristics not being identified, or the expiration date or batch number information not being identified, the indicator light will turn red, indicating that the drug verification is incorrect and needs to be re-verified. In this case, manual confirmation can be used to force the verification to pass.

[0053] Step 5: Determine if the verification is complete. If no verification errors were found with the medications dispensed (including the quantity of medications dispensed being less than or equal to the quantity required in the prescription, the medications not being expired, and the appearance and batch / expiration information of the medications being collected correctly), then the prescription needs to be determined as complete. If there are other medications in the prescription that require verification, the verification process continues. If the verification of all medications in the prescription has been completed, proceed directly to the completion stage.

[0054] Step 6: Once the medication verification is complete, the indicator light will turn green, indicating that the medication verification is finished.

[0055] This device is equipped with a set of matching drug models. A drug model is a formalized description of the appearance, batch number, and expiration date of a drug. The description includes the drug's name, batch number, and expiration date text information, multi-faceted images of the drug's appearance, and batch number and expiration date images. This description does not correspond to a specific type of drug, but rather is an abstract, unified description of the common characteristics of all drugs. Through this description, all actual drugs are reconstructed, achieving a unified, simplified, and accurate generalization of various types of drugs. The drug models required by this device include information such as drug appearance image features, drug expiration date and batch number image features, and drug character features.

[0056] In existing drug verification devices, the verification process is usually divided into two parts or only the appearance of the drug is verified, and the expiration date and batch number of the drug are obtained by other methods (such as by system data push). These methods usually have slow drug verification speed and may not be able to guarantee the real-time nature of drug verification.

[0057] Existing drug verification devices generally only use drug appearance modeling, without modeling drug expiration date and batch number, or simply do not verify drug batch number. When drugs need to be verified, the only way is to "borrow" data from the hospital's HIS system to verify the expiration date and batch number. However, during the operation of this device, through the cooperation of machine vision technology and drug model, the collected multi-point information is quickly matched, and the function of verifying drug type, expiration date and batch number can be completed simultaneously and quickly.

[0058] In addition, the technology of this invention has been implemented in the drug verification platform. After on-site environmental testing and experiments under various conditions, it can be used to model various drugs and use drug models on the drug verification platform. Moreover, it can accurately provide feedback on whether the drug verification results are correct during use.

[0059] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

[0060] In the description of this invention, it should be understood that the terms indicating orientation or positional relationship are based on the orientation or positional relationship shown in the drawings and are only for the convenience of describing the invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention.

Claims

1. A verification platform omnidirectional recognition device, characterized in that, include: This includes the collection of expiration date and batch number information, as well as the collection of drug appearance information and characteristics. The expiration date and batch number information acquisition section includes one grayscale machine vision camera, one set of infrared through-beam triggers, and five bar-shaped auxiliary light sources. Four of the bar-shaped auxiliary light sources are respectively arranged above, below, left, and right of the grayscale machine vision camera, and are configured to alternately flash to illuminate the side of the drug containing the expiration date and batch number information from different directions. The grayscale machine vision camera acquires corresponding images during each illumination. The device performs image synthesis processing on the images acquired from each direction to obtain a high-contrast stamped image for identifying the stamped expiration date and batch number based on the shadow characteristics of each image. The fifth bar-shaped auxiliary light source is positioned above the four bar-shaped auxiliary light sources for auxiliary illumination and adaptively turns on or off according to the device's operating status for brightness compensation. The infrared through-beam triggers serve as a trigger indicator to initiate the verification process, triggering the grayscale machine vision camera to perform image acquisition. The expiration date and batch number information collection section can collect and identify expiration date and batch numbers in ordinary printed fonts, dot matrix characters, and stamped characters. The drug appearance information feature acquisition section includes one color machine vision camera, two auxiliary light strips, a prompt light strip, and a transparent drug verification platform.

2. The verification platform omnidirectional recognition device according to claim 1, characterized in that, The color machine vision system can output both color and grayscale images.

3. The verification platform omnidirectional recognition device according to claim 1, characterized in that, The indicator light strip is a feedback signal light and has three colors.

4. The verification platform omnidirectional recognition device according to claim 1, characterized in that, The transparent drug verification platform includes drug model generation.

5. The verification platform omnidirectional recognition device according to claim 4, characterized in that, The drug model generation includes the following steps: S1. Drug Information Inquiry; S2. Collection of basic appearance information of drugs; S3. Collection of drug appearance characteristics; S4. Drug character information collection; S5. Save the model.

6. The verification platform omnidirectional recognition device according to claim 5, characterized in that, The drug verification process is as follows: Step 1: Scan the drug model and prescription information in the drug model; The second step is to collect image features of the drugs placed on the drug verification platform. The third step is to verify the drug information. If an error is found, proceed to the fourth step; otherwise, proceed to the fifth step. Step 4: Re-verify; Step 5: Determine whether the review is complete; Step 6: Drug verification completed.

Images