WO2021010277A1 - Drug identification device, drug identification method, and program - Google Patents

Drug identification device, drug identification method, and program Download PDF

Info

Publication number
WO2021010277A1
WO2021010277A1 PCT/JP2020/026850 JP2020026850W WO2021010277A1 WO 2021010277 A1 WO2021010277 A1 WO 2021010277A1 JP 2020026850 W JP2020026850 W JP 2020026850W WO 2021010277 A1 WO2021010277 A1 WO 2021010277A1
Authority
WO
WIPO (PCT)
Prior art keywords
drug
registered image
recognition
similarity
image
Prior art date
Application number
PCT/JP2020/026850
Other languages
French (fr)
Japanese (ja)
Inventor
横内　康治
Original Assignee
富士フイルム富山化学株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 富士フイルム富山化学株式会社 filed Critical 富士フイルム富山化学株式会社
Priority to JP2021533015A priority Critical patent/JP7280361B2/en
Publication of WO2021010277A1 publication Critical patent/WO2021010277A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G16H20/10—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance relating to drugs or medications, e.g. for ensuring correct administration to patients

Definitions

  • the present invention relates to a drug identification device, a drug identification method and a program, and particularly to a drug identification device, a drug identification method and a program for identifying a drug having a similar drug.
  • Patent Document 1 for the purpose of preventing erroneous determination of drugs having similar shapes, colors, and sizes, a technique for inspecting whether or not a drug is dispensed according to a prescription using a photographed image of the drug.
  • a drug image registered image
  • a drug similar to the drug is acquired from a drug database and dispensed with the acquired drug image. It is collated with the photographed image of the drug.
  • the present invention has been made in view of such circumstances, and an object of the present invention is a drug identification device and a drug capable of accurately and efficiently identifying drugs having similar drugs. To provide identification methods and programs.
  • the drug identification device for achieving the above object is a drug identification device that identifies a first drug when the first drug has a similar second drug.
  • a first registered image which is an image of the first drug
  • a second registered image which is an image of the second drug when the second drug is present, are stored in relation to the first registered image.
  • the image input unit in which the target image of the drug to be identified is input the drug area extraction unit that extracts the drug area in which the drug appears from the target image, and the first drug from the drug database.
  • First recognition that performs similarity recognition between the registered image acquisition unit that acquires the first registered image and the second registered image and the first registered image that corresponds to the drug region extracted by the drug region extraction unit.
  • a second recognition unit, a first recognition unit, and a first recognition unit that perform similar recognition with the second registered image for a drug region recognized to be similar to the first registered image in the unit and the first recognition unit.
  • a drug identification unit that identifies whether or not the drug is the first drug based on the recognition result of the recognition unit 2 is provided.
  • identification is performed based on the result of similarity recognition between the first registered image and the second registered image, and the similarity recognition with the second registered image is the same as that of the second registered image. Since it is performed on drug areas that are considered to be similar, accurate and efficient identification can be performed.
  • the first recognition unit calculates the first similarity between the drug region and the first registered image, and performs similarity recognition by comparing the first similarity with the threshold value.
  • the second recognition unit calculates the second similarity between the drug region and the second registered image.
  • the drug identification unit identifies the drug in the drug region as the first drug when the first similarity is greater than the second similarity, and the first similarity is the first.
  • the degree of similarity is 2 or less, the drug shown in the drug area is identified as not the first drug.
  • the first agent has a plurality of similar second agents.
  • the second recognition unit calculates the second similarity between the drug region and each of the plurality of second registered images.
  • the drug identification unit identifies the drug in the drug region as the first drug when the first similarity is greater than each of the plurality of second similarities, and the first When the similarity is at least one or less of the plurality of second similarity, the drug shown in the drug region is identified as not the first drug.
  • the similarity recognition of the first recognition unit and the second recognition unit is performed by template matching.
  • the drug identification method is a drug identification method for identifying a first drug when the first drug has a similar second drug, and is an image of the first drug.
  • the first A registered image acquisition step for acquiring a first registered image and a second registered image corresponding to a drug
  • an image input step for inputting a target image of the drug to be identified, and a drug being captured from the target image.
  • the first recognition step for performing similarity recognition between the drug region extracted in the drug region extraction step and the corresponding first registered image, and the first recognition step.
  • a drug identification step for identifying whether or not the drug is a first drug.
  • a program according to another aspect of the present invention is a program that causes a computer to perform a drug identification step of identifying a first drug when the first drug has a similar second drug, and is a first drug. From the drug database in which the first registered image, which is an image of, and the second registered image, which is an image of the second drug when the second drug is present, are stored in relation to the first registered image. From the registered image acquisition step of acquiring the first registered image and the second registered image corresponding to the first drug, the image input step of inputting the target image of the drug to be identified, and the target image.
  • identification is performed based on the result of similarity recognition between the first registered image and the second registered image, and the similarity recognition with the second registered image is the same as that of the second registered image. Since it is performed on drug regions that are considered to be similar, accurate and efficient identification can be performed.
  • FIG. 1 is a diagram showing a similarity distribution.
  • FIG. 2 is a diagram showing a similarity distribution.
  • FIG. 3 is a diagram showing the configuration of a packaged audit support system.
  • FIG. 4 is a diagram showing a state of acquiring an image of a sachet under illumination.
  • FIG. 5 is another diagram showing how an image of the packaging bag is acquired under illumination.
  • FIG. 6 is a block diagram showing a configuration example of the drug identification device.
  • FIG. 7 is a flowchart illustrating a drug identification method.
  • FIG. 8 is a diagram showing an example of a memory configuration of a drug database.
  • FIG. 9 is a diagram showing a target image.
  • FIG. 10 is a flowchart illustrating the calculation of the degree of similarity.
  • FIG. 10 is a flowchart illustrating the calculation of the degree of similarity.
  • FIG. 11 is a flowchart showing the similarity recognition of the first recognition unit, the similarity recognition of the second recognition unit, and the drug identification.
  • FIG. 12 is a diagram showing a target image.
  • FIG. 13 is a flowchart showing the similarity recognition of the first recognition unit, the similarity recognition of the second recognition unit, and the drug identification.
  • Drug ⁇ has drug ⁇ , which is a similar drug with similar size, color, and marking (printing). At this time, the drug ⁇ should be prescribed, but even if the drug ⁇ is mistakenly contained in the sachet, the drug ⁇ may be mistakenly identified as the drug ⁇ .
  • FIGS. 1 and 2 are diagrams showing the similarity distribution between the image of the drug to be identified and the registered image (master image).
  • the vertical axis of FIGS. 1 and 2 indicates the application frequency, and the horizontal axis indicates the degree of similarity with the registered image.
  • the registered image is an image of a drug prepared in advance for collating the drug.
  • the registered image is stored in the drug database, and the drug can be identified by collating the registered image with the captured image of the drug or the image of the enhanced drug.
  • FIG. 1 is a diagram showing the similarity distribution between the drug ⁇ and the drug ⁇ that is not similar to the drug ⁇ .
  • the similarity distribution 53 between the image of the drug ⁇ (drug region) and the registered image ⁇ of the drug ⁇ is shown, and the similarity between the image of the drug ⁇ (drug region) and the registered image ⁇ of the drug ⁇ is shown.
  • Distribution 51 is shown.
  • the registered image ⁇ and the image obtained by taking the drug ⁇ have a certain range of calculated similarity values, but the similarity is calculated to be relatively high (see similarity distribution 53).
  • the registered image ⁇ and the image obtained by photographing the drug ⁇ have a certain range of calculated similarity values, but the similarity is calculated to be relatively low (see the similarity distribution 51).
  • the similarity distribution 51 and the similarity distribution 53 do not overlap.
  • FIG. 2 is a diagram showing the similarity distribution between the registered image ⁇ and the drug ⁇ , and the registered image ⁇ and the drug ⁇ .
  • the similarity distribution 57 shows the similarity distribution between the image (drug region) in which the drug ⁇ is photographed and the registered image ⁇
  • the similarity distribution 55 is the image (drug region) in which the drug ⁇ is photographed and the registered image ⁇ . Shows the similarity distribution.
  • the registered image ⁇ and the image obtained by taking the drug ⁇ have a certain range in the calculated similarity values, but the similarity is calculated to be relatively high (see the similarity distribution 57).
  • the registered image ⁇ and the image obtained by photographing the drug ⁇ have a certain range of calculated similarity values, but the similarity is calculated to be relatively low (see the similarity distribution 55). Further, since the drug ⁇ and the drug ⁇ are similar in size, color, and marking (printing), the similarity distribution 55 and the similarity distribution 57 partially overlap.
  • the threshold Th1 cannot be set between the two similarity distributions as shown in FIG.
  • the threshold value Th2 for identifying the drug ⁇
  • the drug ⁇ can be identified as the drug ⁇ , but the drug ⁇ may also be erroneously identified as the drug ⁇ . It ends up.
  • the similarity threshold Th3 see FIG. 2 is set, there may be a case where the drug A is not identified as the drug A even though it is the drug A.
  • the present invention even when identifying a drug having a similar drug, the identification can be performed accurately and efficiently.
  • FIG. 3 is a diagram showing a configuration of a packaged audit support system 1 equipped with the drug identification device 2 (FIG. 6) of the present invention.
  • the one-pack audit support system 1 includes a processing unit 10, a storage unit (drug database 20A (see FIG. 6)) 20, a display unit 30, an operation unit 40, a transport mechanism 50, an illumination unit 12, and a camera 15A.
  • a camera 15B and a prescription reader 16 are provided.
  • the one-pack audit support system 1 provides support for auditing whether or not the drug is properly packaged according to the prescription.
  • the drug identification device 2 identifies the packaged drug, and the identification result is displayed on the display unit 30 (monitor 32) to support the user's audit.
  • the example in which the drug identification device 2 is used is not limited to the one-pack audit.
  • the drug identification device 2 may be used to discriminate the drug brought.
  • the drug identified by the drug identification device 2 is not limited to the drug prescribed by the prescription.
  • the drug identification device 2 can use the drug. Can be identified.
  • the camera 15A and the camera 15B are composed of a digital camera. As shown in FIG. 4, the camera 15A is arranged on the upper side (+ Z side of FIG. 4) of the medicine bandage PB in which the sachet TP (medicine package) is continuously formed, and the camera 15A is arranged on the lower side of the medicine bandage PB (FIG. 4).
  • the camera 15B is arranged on the -Z side of the bag, and the medicines packaged in one packaging bag TP are photographed from above and below (plurality in different directions).
  • the sachet TP (medicine bandage PB) is transported by the transport mechanism 50 in the + X direction of FIG. 4 (the axis along the longitudinal direction of the drug bandage PB; the arrow direction of FIG.
  • a plurality of light sources 13 having the light sources illuminate the packaging bag TP from four directions.
  • the prescription reader 16 reads the prescription information.
  • OCR Optical Character Recognition
  • information such as the prescribed drug and its quantity may be read from the bar code.
  • the user may read the prescription and input the prescription information by an input device such as a keyboard included in the operation unit 40.
  • FIG. 6 is a block diagram showing a configuration example of the drug identification device 2.
  • the drug identification device 2 is mainly composed of a processing unit 10 and a storage unit 20.
  • the processing unit 10 includes an image input unit 10A, a drug region extraction unit 10B, a registered image acquisition unit 10C, a first recognition unit 10D, a second recognition unit 10E, and a drug identification unit 10F.
  • the CPU Central Processing Unit
  • devices such as various electronic circuits refer to the data stored in EEPROM (Electronically Erasable and Programmable Read Only Memory: non-temporary recording medium), etc. Will be realized.
  • EEPROM Electrically Erasable and Programmable Read Only Memory: non-temporary recording medium
  • these functions are performed by executing a drug identification program stored in a ROM (Read Only Memory: non-temporary recording medium) or the like of the storage unit 20.
  • RAM Random Access Memory
  • processing unit 10 is connected to the storage unit 20, and can freely retrieve and rewrite the data of the program and drug database 20A stored in the storage unit 20.
  • FIG. 7 is a flowchart illustrating a drug identification method (drug identification step) using the drug identification device 2. A detailed explanation of each step will be given later.
  • the prescription is read by the prescription reader 16 and the prescription information is acquired. Then, based on the prescription information, the registered image acquisition unit 10C acquires the first registered image and the second registered image of the drug that can be dispensed by the prescription (registered image acquisition step: step S10).
  • the first registered image is a registered image of the first drug dispensed based on the prescription information.
  • the second registered image is a registered image of the second drug similar to the first drug.
  • the target image PI (see FIG. 9) obtained by photographing one packaging bag TP is input to the image input unit 10A (image input step: step S11).
  • the drug region extraction unit 10B extracts the drug region R1 and the drug region R2 in the target image PI (see FIG. 9) (drug region extraction step: step S12).
  • the first recognition unit 10D performs similar recognition with the first registered image with respect to the drug region R1 and the drug region R2 (first recognition step: step S13).
  • the first recognition unit 10D performs similar recognition between the drug region R1 and the drug region R2 and the first registered image (step S14).
  • the drug identification unit 10F identifies the drug based on the recognition result of the first recognition unit 10D.
  • the second recognition unit 10E recognizes that the drug region is similar to the first registered image. Similar recognition with the registered image of (second recognition step: step S15). After that, the drug identification unit 10F identifies the drug based on the recognition results of the first recognition unit 10D and the second recognition unit 10E (drug identification step: step S16).
  • the registered image acquisition step (step S10) is performed by the registered image acquisition unit 10C.
  • the registered image acquisition unit 10C acquires the first registered image and the second registered image corresponding to the first drug from the drug database 20A based on the prescription information.
  • the registered image acquisition unit 10C acquires the prescription information acquired by the prescription reader 16, and is similar to the first registered image of the first drug dispensed by the prescription information and the first drug.
  • a second registered image of the second drug is obtained from the drug database 20A.
  • the registered image acquisition unit 10C has described an example in which the prescription reader 16 acquires the first registered image and the second registered image corresponding to the first drug based on the prescription information.
  • the mode in which the registered image acquisition unit 10C acquires the first registered image and the second registered image from the drug database 20A is not limited to the above example. For example, even if the user inputs the prescription information in the operation unit 40 and the registered image acquisition unit 10C acquires the first registered image and the second registered image from the drug database 20A based on the input prescription information. Good.
  • FIG. 8 is a diagram showing an example of a memory configuration of the drug database 20A.
  • drug name "drug name”, "registered image” and “similar drug” are stored in association with each other.
  • the drug name drug A is remembered in relation to the surface image of drug A and the similar drug being drug X.
  • the surface image of the drug A is marked with "A" on the surface.
  • the similar drug of the drug A is the drug X.
  • the drug name drug B is remembered in relation to the surface image of drug B and the absence of similar drugs.
  • the surface image of the drug B is marked with "B" on the surface.
  • the drug name drug C is remembered in relation to the surface image of drug C and the similar drugs being drug Y and drug Z.
  • the surface image of the drug C is marked with "C” on the surface.
  • the similar drugs of the drug C are the drug Y and the drug Z.
  • the drug name, the registered image, and the similar drug are stored in association with each other.
  • the memory structure of the drug database 20A shown in FIG. 8 is an example and is not limited thereto.
  • the registered image may have a front surface image of the drug and a back surface image of the drug. By having the front side image and the back side image in the registered image in this way, it is possible to perform similar recognition in more detail.
  • the registered image acquisition unit 10C may use the registered image A (first registered image) of the drug A. And the registered image B (first registered image) of the drug B is acquired. Further, since the drug X is similar to the drug A, the registered image acquisition unit 10C acquires the registered image X (second registered image) of the drug X.
  • the image input step (step S11) is performed by the image input unit 10A.
  • the target image of the drug to be identified is input to the image input unit 10A.
  • the target image is an image of one sachet TP containing the drug in order to identify that the drug is packaged according to the prescription.
  • the target image may be composed of one photographed image obtained by photographing the sachet TP, or an image in which the marking or printing of the drug A and the drug B is emphasized by image processing in one or more photographed images. There may be.
  • the drug region extraction step (step S12) is performed by the drug region extraction unit 10B.
  • the drug region extraction unit 10B extracts (cuts out) the drug region in which the drug to be identified is shown from the target image for each drug.
  • the drug region extraction unit 10B extracts the drug region, for example, based on the difference in color between the drug and the background.
  • FIG. 9 is a diagram showing a target image PI input to the image input unit 10A.
  • the target image PI is photographed by the camera 15A, and enhancement processing is performed by the marking (printing) enhancement processing unit (not shown) of the processing unit 10.
  • the drug region extraction unit 10B extracts the drug region R1 of the drug A and the drug region R2 of the drug B, respectively.
  • the camera 15A and the camera 15B can photograph the packaging bag TP from both sides, and images of the front and back surfaces of the drug can be acquired without omission.
  • the example shown in FIG. 9 was taken with the camera 15A, and is an example in which the images of the surfaces of the drug A and the drug B could be acquired.
  • the first recognition step (step S13) is performed by the first recognition unit 10D.
  • the first recognition unit 10D performs similar recognition of the drug region extracted by the drug region extraction unit 10B and the corresponding first registered image. Specifically, the first recognition unit 10D calculates the first similarity between the drug region and the first registered image, and compares the first similarity with the threshold value to perform similarity recognition.
  • the second recognition step (step S15) is performed by the second recognition unit 10E.
  • the second recognition unit 10E performs similar recognition with the second registered image for the drug region recognized by the first recognition unit 10D to be similar to the first registered image. Specifically, the second recognition unit 10E calculates the second similarity between the drug region and the second registered image.
  • the drug identification step (step S16) is performed by the drug identification unit 10F.
  • the drug identification unit 10F discriminates whether or not the drug is the first drug based on the recognition results of the first recognition unit 10D and the second recognition unit 10E. Specifically, the drug identification unit 10F identifies the drug shown in the drug region as the first drug when the first similarity is larger than the second similarity, and the first similarity. When is less than or equal to the second similarity, the drug shown in the drug area is identified as not the first drug.
  • FIG. 10 is a flowchart illustrating the calculation of the degree of similarity performed by the first recognition unit 10D.
  • the second recognition unit 10E also calculates the similarity, the description will be omitted because the similarity is calculated in the same manner as the first recognition unit 10D.
  • the first recognition unit 10D recognizes whether or not the size or shape (circumscribed rectangle) of the drug region and the first registered image is different (step S20). Here, if either the size or shape (circumscribed rectangle) of the drug region and the first registered image is different, the similarity is calculated as "0" zero (step S24).
  • the first recognition unit 10D recognizes whether or not the size or shape (circumscribed rectangle) is different based on the difference in the feature amount between the drug region and the first registered image. For example, the first recognition unit 10D recognizes that the difference between the drug region and the first registered image is different when the difference between the feature amounts is equal to or larger than the threshold value.
  • the first recognition unit 10D recognizes whether or not the brightness (brightness) of the drug region and the first registered image is different (step S21).
  • the first recognition unit 10D calculates that the similarity is “0” zero if the brightness (brightness) of the drug region and the first registered image are different (step S24).
  • the first recognition unit 10D recognizes whether or not the brightness is different based on the difference in brightness between the drug region and the first registered image. For example, the first recognition unit 10D recognizes that when the difference in brightness between the drug region and the first registered image is equal to or greater than the threshold value, the two are different.
  • the first recognition unit 10D recognizes whether or not the color of the drug region and the first registered image are different (step S22).
  • the first recognition unit 10D recognizes that the similarity is “0” zero (step S24).
  • the first recognition unit 10D recognizes whether or not the colors are different based on the difference in the R (red), G (green), and B (blue) values between the drug region and the first registered image. To do. For example, the first recognition unit 10D recognizes that the two are different when the color difference between the drug region and the first registered image is equal to or greater than the threshold value.
  • the first recognition unit 10D calculates the degree of similarity between the drug region and the first registered image (step S22).
  • the first recognition unit 10D calculates the similarity by template matching. For example, the similarity is when 1.0 is the most similar (same case), and is calculated lower than 1.0 depending on the degree of different points.
  • the first recognition unit 10D calculates the first similarity between the drug region and the first registered image.
  • the second recognition unit 10E may calculate the similarity in the same manner as the first recognition unit 10D, but the second recognition unit 10E is a drug region recognized to be similar by the first recognition unit 10D. Since the similarity is calculated for the above, the similarity may be directly calculated by template matching from the beginning (step S23 in FIG. 10).
  • FIG. 11 is a flowchart showing the similarity recognition of the first recognition unit 10D, the similarity recognition of the second recognition unit 10E, and the drug identification.
  • the first recognition unit 10D performs similar recognition between the drug region R1 and the registered image A and the registered image B (step S30). Specifically, since the drug region R1 and the registered image A have the same size (shape), brightness (brightness), and color, the first recognition unit 10D calculates the similarity by template matching ( See step S23 in FIG. 10). On the other hand, the first recognition unit 10D calculates the similarity as zero because the size of the drug region R1 and the registered image B are different.
  • the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R1 and the registered image A is equal to or higher than the threshold value ThA, it is recognized that the drug region R1 is similar to the registered image A.
  • the threshold ThA is preferably set as low as possible from the viewpoint of extracting the drug A without omission. For example, the threshold ThA is set as the threshold Th2 described with reference to FIG.
  • the first recognition unit 10D recognizes that the drug region R1 and the registered image B are not similar to each other because the similarity is zero and is smaller than the threshold value ThA, and the drug identification unit 10F recognizes that the drug region R1 is not similar. It is identified as not drug B (step S35).
  • the second recognition unit 10E determines the presence or absence of a similar drug of the drug A (step S31), and if there is a similar drug, calculates the second degree of similarity between the drug region R1 and the registered image X. (Step S32). Then, the drug identification unit 10F discriminates whether or not the drug region R1 is the drug A based on the first similarity and the second similarity (step S33). Specifically, the drug identification identifies the drug region R1 from the drug A when the first similarity is greater than the second similarity (step S34). On the other hand, when the first similarity is equal to or less than the second similarity, the drug region R1 is identified as not the drug A (step S34).
  • the first recognition unit 10D performs similar recognition between the drug region R2 and the registered image A and the registered image B (step S30). Specifically, since the drug region R2 and the registered image B have the same size (shape), color, and marking (printing), the first recognition unit 10D calculates the similarity by template matching (FIG. 10). (See step S23). On the other hand, since the size of the drug region R2 and the registered image A are different in the first recognition unit 10D, the similarity is calculated as zero.
  • the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R2 and the registered image B is equal to or higher than the threshold value ThA, it is recognized that the drug region R2 is similar to the registered image B. Then, the second recognition unit 10E determines the presence or absence of a similar drug of the drug B (step S31), and since there is no similar drug in the drug B, the second recognition unit 10E calculates the second similarity degree. do not do. After that, the drug identification unit 10F identifies the drug shown in the drug region R2 from the drug B based on the similarity recognition of the first recognition unit 10D (step S34).
  • the first recognition unit 10D recognizes that the drug region R2 and the registered image A are not similar because the similarity is zero and is smaller than the threshold value ThA
  • the drug identification unit 10F recognizes that the drug region R2 is not similar. It is identified as not drug A (step S35).
  • the first recognition unit 10D performs similar recognition of the drug region and the corresponding first registered image. Then, in this aspect, the first recognition unit 10D performs similar recognition with the second registered image for the drug region recognized to be similar to the first registered image. Thereby, in this aspect, since the similarity recognition with the second registered image is performed for the drug region considered to be similar to the second registered image, accurate and efficient identification can be performed.
  • FIG. 12 is a diagram showing a target image PJ input to the image input unit 10A.
  • the target image PJ is acquired in the same manner as the target image PI described with reference to FIG. Then, the drug region extraction unit 10B extracts the drug region R3 of the drug C.
  • FIG. 13 is a flowchart showing the similarity recognition of the first recognition unit 10D, the similarity recognition of the second recognition unit 10E, and the drug identification of the drug identification unit 10F.
  • the identification of the drug C will be described with reference to the flowchart of FIG.
  • the drug C which is the first drug
  • the registration image acquisition unit 10C is registered to correspond to the first drug based on the prescription information.
  • Image C, registered image Y, and registered image Z are acquired from the drug database 20A.
  • the first recognition unit 10D performs similar recognition between the drug region R3 and the registered image C (step S40). Specifically, since the drug region R3 and the registered image C have the same size (shape), brightness (brightness), and color, the first recognition unit 10D calculates the similarity by template matching ( See step S23 in FIG. 10).
  • the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R3 and the registered image C is equal to or higher than the threshold value ThA, it is recognized that the drug region R3 is similar to the registered image C.
  • the second recognition unit 10E calculates the second similarity (1) between the drug region R3 and the registered image Y (step S41). In addition, the second recognition unit 10E calculates the second similarity (2) between the drug region R3 and the registered image Z (step S42). In this way, the second recognition unit 10E calculates the degree of similarity for each of the registered images of similar drugs. Then, the drug identification unit 10F performs identification based on each of the first similarity and the second similarity. Specifically, the drug identification unit 10F determines whether or not the first similarity is greater than the second similarity (1) (step S43). The drug identification unit 10F identifies that the drug region R3 is not drug C when the first similarity is not greater than the second similarity (1).
  • the drug identification unit 10F determines whether the first similarity is larger than the second similarity (2). Is determined (step S44). The drug identification unit 10F identifies that the drug region R3 is not drug C when the first similarity is not greater than the second similarity (2) (step S46). On the other hand, when the first similarity is larger than the second similarity (2), the drug region R3 is distinguished from the drug C (step S45).
  • the drug is accurately identified based on the recognition results of the first recognition unit 10D and the second recognition unit 10E. Efficient drug identification can be performed.
  • Drug identification device 10 Processing unit 10A: Image input unit 10B: Drug area extraction unit 10C: Registered image acquisition unit 10D: First recognition unit 10E: Second recognition unit 10F: Drug identification unit 12: Lighting unit 13: Light source 15A: Camera 15B: Camera 16: Prescription reader 20: Storage unit 20A: Drug database 30: Display unit 40: Operation unit 50: Conveyance mechanism

Abstract

Provided are a drug identification device, a drug identification method, and a program with which it is possible for efficient identification to be carried out accurately, even when identifying drugs including drugs that are similar to one another. A drug identification device (2) is provided with: a drug database (20A); an image input unit (10A) into which a target image is input; a drug region extracting unit (10B) which extracts from the target image a drug region in which a drug appears; a registered image acquiring unit (10C) which acquires a first registered image and a second registered image; a first recognizing unit (10D) which performs similarity recognition between the drug region and the corresponding first registered image; a second recognizing unit (10E) which performs similarity recognition with the second registered image, for a drug region that has been recognized by the first recognizing unit as being similar to the first registered image; and a drug identifying unit (10F) which identifies whether the drug is a first drug on the basis of the recognition results from the first recognizing unit and the second recognizing unit.

Description

薬剤識別装置、薬剤識別方法及びプログラムDrug identification device, drug identification method and program
　本発明は、薬剤識別装置、薬剤識別方法及びプログラムに関し、特に類似する薬剤を有する薬剤を識別する薬剤識別装置、薬剤識別方法及びプログラムに関する。 The present invention relates to a drug identification device, a drug identification method and a program, and particularly to a drug identification device, a drug identification method and a program for identifying a drug having a similar drug.
　従来より、処方箋通りに薬剤が調剤されているか否かの監査を行うために、調剤された薬剤を撮影し、その撮影画像に基づいて薬剤を識別する技術が用いられてきた。 Conventionally, in order to audit whether or not a drug is dispensed according to a prescription, a technique of photographing a dispensed drug and identifying the drug based on the photographed image has been used.
　例えば、特許文献１では、形状、色及び大きさが類似する薬剤の誤判定を防ぐことを目的とし、薬剤を撮影した撮影画像を用いて処方箋通りに調剤されているか否かの検査を行う技術が記載されている。具体的には、特許文献１に記載された技術では、処方箋に従って調剤されるべき薬剤及びその薬剤に類似する薬剤の薬剤画像（登録画像）を薬剤データベースから取得し、取得した薬剤画像と調剤された薬剤を撮影した撮影画像との照合を行っている。 For example, in Patent Document 1, for the purpose of preventing erroneous determination of drugs having similar shapes, colors, and sizes, a technique for inspecting whether or not a drug is dispensed according to a prescription using a photographed image of the drug. Is described. Specifically, in the technique described in Patent Document 1, a drug image (registered image) of a drug to be dispensed according to a prescription and a drug similar to the drug is acquired from a drug database and dispensed with the acquired drug image. It is collated with the photographed image of the drug.
特開２０１４－６４８３６号公報Japanese Unexamined Patent Publication No. 2014-64836
　ここで、特許文献１に記載された技術では、分包袋に分包された薬剤の各々とデータベースから取得した薬剤画像の各々の組み合わせを全て照合する必要がある。例えば特許文献１に記載の技術では、一つの分包袋に薬剤Ａ及び薬剤Ｂの２つの薬剤が調剤される場合であって、Ａに類似する薬剤αが存在する場合には、薬剤データベースから薬剤画像は３枚（薬剤画像Ａ、Ｂ及びα）取得され、薬剤Ａ及び薬剤Ｂと３枚の薬剤画像との計１２（２×３）回の照合が必要になり、効率的な照合が行われていない。 Here, in the technique described in Patent Document 1, it is necessary to collate all the combinations of each of the drugs packaged in the packaging bag and each of the drug images obtained from the database. For example, in the technique described in Patent Document 1, when two drugs, drug A and drug B, are dispensed in one packaging bag and a drug α similar to A is present, the drug database is used. Three drug images (drug images A, B and α) are acquired, and a total of 12 (2 × 3) collations between the drug A and drug B and the three drug images are required, which enables efficient collation. Not done.
　本発明はこのような事情に鑑みてなされたもので、その目的は、類似する薬剤を有する薬剤を識別する場合であっても、正確に効率的な識別を行うことができる薬剤識別装置、薬剤識別方法及びプログラムを提供することである。 The present invention has been made in view of such circumstances, and an object of the present invention is a drug identification device and a drug capable of accurately and efficiently identifying drugs having similar drugs. To provide identification methods and programs.
　上記目的を達成するための本発明の一の態様である薬剤識別装置は、第１の薬剤が類似する第２の薬剤を有する場合の第１の薬剤を識別する薬剤識別装置であって、第１の薬剤の画像である第１の登録画像、及び第２の薬剤が存在する場合に第２の薬剤の画像である第２の登録画像が第１の登録画像に関連して記憶されている薬剤データベースと、識別の対象である薬剤の対象画像が入力される画像入力部と、対象画像から薬剤が写っている薬剤領域を抽出する薬剤領域抽出部と、薬剤データベースから第１の薬剤に対応する、第１の登録画像及び第２の登録画像を取得する登録画像取得部と、薬剤領域抽出部で抽出された薬剤領域と対応する第１の登録画像との類似認識を行う第１の認識部と、第１の認識部において、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識を行う第２の認識部と、第１の認識部及び第２の認識部の認識結果に基づいて、薬剤が第１の薬剤であるか否かの識別を行う薬剤識別部と、を備える。 The drug identification device according to one aspect of the present invention for achieving the above object is a drug identification device that identifies a first drug when the first drug has a similar second drug. A first registered image, which is an image of the first drug, and a second registered image, which is an image of the second drug when the second drug is present, are stored in relation to the first registered image. Corresponds to the drug database, the image input unit in which the target image of the drug to be identified is input, the drug area extraction unit that extracts the drug area in which the drug appears from the target image, and the first drug from the drug database. First recognition that performs similarity recognition between the registered image acquisition unit that acquires the first registered image and the second registered image and the first registered image that corresponds to the drug region extracted by the drug region extraction unit. A second recognition unit, a first recognition unit, and a first recognition unit that perform similar recognition with the second registered image for a drug region recognized to be similar to the first registered image in the unit and the first recognition unit. A drug identification unit that identifies whether or not the drug is the first drug based on the recognition result of the recognition unit 2 is provided.
　本態様によれば、第１の登録画像と第２の登録画像との類似認識の結果に基づいて識別が行われ、また、第２の登録画像との類似認識は、第２の登録画像と類似すると考えられる薬剤領域に対して行われるので、正確で効率的な識別を行うことができる。 According to this aspect, identification is performed based on the result of similarity recognition between the first registered image and the second registered image, and the similarity recognition with the second registered image is the same as that of the second registered image. Since it is performed on drug areas that are considered to be similar, accurate and efficient identification can be performed.
　好ましくは、第１の認識部は、薬剤領域と第１の登録画像との第１の類似度を算出し、第１の類似度と閾値とを比較することにより、類似認識を行う。 Preferably, the first recognition unit calculates the first similarity between the drug region and the first registered image, and performs similarity recognition by comparing the first similarity with the threshold value.
　好ましくは、第２の認識部は、薬剤領域と第２の登録画像との第２の類似度を算出する。 Preferably, the second recognition unit calculates the second similarity between the drug region and the second registered image.
　好ましくは、薬剤識別部は、第１の類似度が第２の類似度よりも大きい場合に、薬剤領域に写っている薬剤を第１の薬剤であると識別し、第１の類似度が第２の類似度以下である場合に、薬剤領域に写っている薬剤を第１の薬剤でないと識別する。 Preferably, the drug identification unit identifies the drug in the drug region as the first drug when the first similarity is greater than the second similarity, and the first similarity is the first. When the degree of similarity is 2 or less, the drug shown in the drug area is identified as not the first drug.
　好ましくは、第１の薬剤は、類似する複数の第２の薬剤を有する。 Preferably, the first agent has a plurality of similar second agents.
　好ましくは、第２の認識部は、薬剤領域と複数の第２の登録画像の各々との第２の類似度を算出する。 Preferably, the second recognition unit calculates the second similarity between the drug region and each of the plurality of second registered images.
　好ましくは、薬剤識別部は、第１の類似度が複数の第２の類似度の各々よりも大きい場合に、薬剤領域に写っている薬剤を第１の薬剤であると識別し、第１の類似度が複数の第２の類似度の少なくとも一つ以下である場合に、薬剤領域に写っている薬剤を第１の薬剤でないと識別する。 Preferably, the drug identification unit identifies the drug in the drug region as the first drug when the first similarity is greater than each of the plurality of second similarities, and the first When the similarity is at least one or less of the plurality of second similarity, the drug shown in the drug region is identified as not the first drug.
　好ましくは、第１の認識部及び第２の認識部の類似認識は、テンプレートマッチングにより行われる。 Preferably, the similarity recognition of the first recognition unit and the second recognition unit is performed by template matching.
　本発明の他の態様である薬剤識別方法は、第１の薬剤が類似する第２の薬剤を有する場合の第１の薬剤を識別する薬剤識別方法であって、第１の薬剤の画像である第１の登録画像、及び第２の薬剤が存在する場合に第２の薬剤の画像である第２の登録画像が第１の登録画像に関連して記憶されている薬剤データベースから、第１の薬剤に対応する、第１の登録画像及び第２の登録画像を取得する登録画像取得ステップと、識別の対象である薬剤の対象画像が入力される画像入力ステップと、対象画像から薬剤が写っている薬剤領域を抽出する薬剤領域抽出ステップと、薬剤領域抽出ステップで抽出された薬剤領域と対応する第１の登録画像との類似認識を行う第１の認識ステップと、第１の認識ステップにおいて、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識を行う第２の認識ステップと、第１の認識ステップ及び第２の認識ステップの認識結果に基づいて、薬剤が第１の薬剤であるか否かの識別を行う薬剤識別ステップと、を含む。 The drug identification method according to another aspect of the present invention is a drug identification method for identifying a first drug when the first drug has a similar second drug, and is an image of the first drug. From the drug database in which the first registered image and the second registered image, which is the image of the second drug when the second drug is present, are stored in relation to the first registered image, the first A registered image acquisition step for acquiring a first registered image and a second registered image corresponding to a drug, an image input step for inputting a target image of the drug to be identified, and a drug being captured from the target image. In the drug region extraction step for extracting the existing drug region, the first recognition step for performing similarity recognition between the drug region extracted in the drug region extraction step and the corresponding first registered image, and the first recognition step. Based on the recognition results of the second recognition step, the first recognition step, and the second recognition step, in which the drug region recognized to be similar to the first registered image is recognized to be similar to the second registered image. , A drug identification step for identifying whether or not the drug is a first drug.
　本発明の他の態様であるプログラムは、第１の薬剤が類似する第２の薬剤を有する場合の第１の薬剤を識別する薬剤識別工程をコンピュータに実行させるプログラムであって、第１の薬剤の画像である第１の登録画像、及び第２の薬剤が存在する場合に第２の薬剤の画像である第２の登録画像が第１の登録画像に関連して記憶されている薬剤データベースから、第１の薬剤に対応する、第１の登録画像及び第２の登録画像を取得する登録画像取得ステップと、識別の対象である薬剤の対象画像が入力される画像入力ステップと、対象画像から薬剤が写っている薬剤領域を抽出する薬剤領域抽出ステップと、薬剤領域抽出ステップで抽出された薬剤領域と対応する第１の登録画像との類似認識を行う第１の認識ステップと、第１の認識ステップにおいて、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識を行う第２の認識ステップと、第１の認識ステップ及び第２の認識ステップの認識結果に基づいて、薬剤が第１の薬剤であるか否かの識別を行う薬剤識別ステップと、を含む薬剤識別工程をコンピュータに実行させる。 A program according to another aspect of the present invention is a program that causes a computer to perform a drug identification step of identifying a first drug when the first drug has a similar second drug, and is a first drug. From the drug database in which the first registered image, which is an image of, and the second registered image, which is an image of the second drug when the second drug is present, are stored in relation to the first registered image. From the registered image acquisition step of acquiring the first registered image and the second registered image corresponding to the first drug, the image input step of inputting the target image of the drug to be identified, and the target image. A drug region extraction step for extracting a drug region in which a drug is shown, a first recognition step for performing similarity recognition between the drug region extracted in the drug region extraction step and the corresponding first registered image, and a first In the recognition step, the drug region recognized to be similar to the first registered image is recognized in the second recognition step for performing similarity recognition with the second registered image, and the recognition of the first recognition step and the second recognition step. Based on the result, the computer is made to perform a drug identification step including a drug identification step of identifying whether or not the drug is the first drug.
　本発明によれば、第１の登録画像と第２の登録画像との類似認識の結果に基づいて識別が行われ、また、第２の登録画像との類似認識は、第２の登録画像と類似すると考えられる薬剤領域に対して行われるので、正確で効率的な識別を行うことができる。 According to the present invention, identification is performed based on the result of similarity recognition between the first registered image and the second registered image, and the similarity recognition with the second registered image is the same as that of the second registered image. Since it is performed on drug regions that are considered to be similar, accurate and efficient identification can be performed.
図１は、類似度分布を示す図である。FIG. 1 is a diagram showing a similarity distribution. 図２は、類似度分布を示す図である。FIG. 2 is a diagram showing a similarity distribution. 図３は、一包化監査支援システムの構成を示す図である。FIG. 3 is a diagram showing the configuration of a packaged audit support system. 図４は、照明下で分包袋の画像を取得する様子を示す図である。FIG. 4 is a diagram showing a state of acquiring an image of a sachet under illumination. 図５は、照明下で分包袋の画像を取得する様子を示す他の図である。FIG. 5 is another diagram showing how an image of the packaging bag is acquired under illumination. 図６は、薬剤識別装置の構成例を示すブロック図である。FIG. 6 is a block diagram showing a configuration example of the drug identification device. 図７は、薬剤識別方法に関して説明するフローチャートである。FIG. 7 is a flowchart illustrating a drug identification method. 図８は、薬剤データベースの記憶構成例を示す図である。FIG. 8 is a diagram showing an example of a memory configuration of a drug database. 図９は、対象画像を示す図である。FIG. 9 is a diagram showing a target image. 図１０は、類似度の算出に関して説明するフローチャートである。FIG. 10 is a flowchart illustrating the calculation of the degree of similarity. 図１１は、第１の認識部の類似認識、第２の認識部の類似認識及び薬剤識別を示すフローチャートである。FIG. 11 is a flowchart showing the similarity recognition of the first recognition unit, the similarity recognition of the second recognition unit, and the drug identification. 図１２は、対象画像を示す図である。FIG. 12 is a diagram showing a target image. 図１３は、第１の認識部の類似認識、第２の認識部の類似認識及び薬剤識別を示すフローチャートである。FIG. 13 is a flowchart showing the similarity recognition of the first recognition unit, the similarity recognition of the second recognition unit, and the drug identification.
　以下、添付図面に従って本発明にかかる薬剤識別装置、薬剤識別方法及びプログラムの好ましい実施の形態について説明する。 Hereinafter, preferred embodiments of the drug identification device, drug identification method, and program according to the present invention will be described with reference to the accompanying drawings.
　＜類似薬剤＞
　先ず、薬剤αが類似薬剤βを有する場合に、薬剤αの正確な識別が困難となることについて説明する。以下の例では、薬剤αを処方箋に従って分包袋に入れる場合について説明する。 <Similar drug>
First, it will be described that when the drug α has a similar drug β, it becomes difficult to accurately identify the drug α. In the following example, a case where the drug α is placed in a sachet according to a prescription will be described.
　薬剤αは、サイズ、色、刻印（印字）が似ている類似薬剤である薬剤βを有する。このとき、処方すべきは薬剤αであるが、分包袋内に誤って薬剤βが入っていても、薬剤βを薬剤αと誤って識別してしまう場合がある。 Drug α has drug β, which is a similar drug with similar size, color, and marking (printing). At this time, the drug α should be prescribed, but even if the drug β is mistakenly contained in the sachet, the drug β may be mistakenly identified as the drug α.
　図１及び図２は、識別対象の薬剤の画像と登録画像（マスタ画像）との類似度分布を示す図である。図１及び図２の縦軸は出願頻度を示し、横軸は登録画像との類似度を示す。なお、登録画像とはその薬剤を照合するために予め用意された薬剤の画像である。登録画像は薬剤データベースに記憶され、登録画像と撮影された薬剤の画像又は強調処理された薬剤の画像と照合することにより、薬剤を特定することができる。 1 and 2 are diagrams showing the similarity distribution between the image of the drug to be identified and the registered image (master image). The vertical axis of FIGS. 1 and 2 indicates the application frequency, and the horizontal axis indicates the degree of similarity with the registered image. The registered image is an image of a drug prepared in advance for collating the drug. The registered image is stored in the drug database, and the drug can be identified by collating the registered image with the captured image of the drug or the image of the enhanced drug.
　図１は、薬剤αと薬剤αと類似していない薬剤γの類似度分布を示す図である。薬剤αを撮影した画像（薬剤領域）と薬剤αの登録画像αとの類似度分布５３が示されており、薬剤γを撮影した画像（薬剤領域）と薬剤αの登録画像αとの類似度分布５１が示されている。 FIG. 1 is a diagram showing the similarity distribution between the drug α and the drug γ that is not similar to the drug α. The similarity distribution 53 between the image of the drug α (drug region) and the registered image α of the drug α is shown, and the similarity between the image of the drug γ (drug region) and the registered image α of the drug α is shown. Distribution 51 is shown.
　登録画像αと薬剤αを撮影した画像とは、算出される類似度の値に一定の幅があるが、比較的高く類似度が算出される（類似度分布５３を参照）。一方で、登録画像αと薬剤γを撮影した画像とは、算出される類似度の値に一定の幅があるが、比較的低く類似度が算出される（類似度分布５１を参照）。しかしながら、薬剤αと薬剤γとは、類似していないので、類似度分布５１と類似度分布５３とが重なることはない。 The registered image α and the image obtained by taking the drug α have a certain range of calculated similarity values, but the similarity is calculated to be relatively high (see similarity distribution 53). On the other hand, the registered image α and the image obtained by photographing the drug γ have a certain range of calculated similarity values, but the similarity is calculated to be relatively low (see the similarity distribution 51). However, since the drug α and the drug γ are not similar, the similarity distribution 51 and the similarity distribution 53 do not overlap.
　従って、薬剤αを類似度によって識別する場合の閾値Ｔｈ１を図１に示す様に類似度分布５１及び類似度分布５３との間に設定することにより、登録画像αにより薬剤αを正確に識別することができる。 Therefore, by setting the threshold Th1 when the drug α is identified by the similarity between the similarity distribution 51 and the similarity distribution 53 as shown in FIG. 1, the drug α is accurately identified by the registered image α. be able to.
　一方で、薬剤αと類似薬剤の薬剤βとの識別では、上述したように閾値を設定することは、以下に説明するように困難となる。 On the other hand, in distinguishing the drug α from the drug β of a similar drug, it is difficult to set the threshold value as described above as described below.
　図２は、登録画像αと薬剤α、及び登録画像αと薬剤βとの類似度分布を示す図である。類似度分布５７は、薬剤αを撮影した画像（薬剤領域）と登録画像αとの類似度分布を示し、類似度分布５５は、薬剤βを撮影した画像（薬剤領域）と登録画像αとの類似度分布を示す。登録画像αと薬剤αを撮影した画像とは、算出される類似度の値に一定の幅があるが、比較的高く類似度が算出される（類似度分布５７を参照）。一方で、登録画像αと薬剤βを撮影した画像とは、算出される類似度の値に一定の幅があるが、比較的低く類似度が算出される（類似度分布５５を参照）。また、薬剤αと薬剤βとは、サイズ、色、及び刻印（印字）が類似しているので、類似度分布５５と類似度分布５７とは一部で重なる。 FIG. 2 is a diagram showing the similarity distribution between the registered image α and the drug α, and the registered image α and the drug β. The similarity distribution 57 shows the similarity distribution between the image (drug region) in which the drug α is photographed and the registered image α, and the similarity distribution 55 is the image (drug region) in which the drug β is photographed and the registered image α. Shows the similarity distribution. The registered image α and the image obtained by taking the drug α have a certain range in the calculated similarity values, but the similarity is calculated to be relatively high (see the similarity distribution 57). On the other hand, the registered image α and the image obtained by photographing the drug β have a certain range of calculated similarity values, but the similarity is calculated to be relatively low (see the similarity distribution 55). Further, since the drug α and the drug β are similar in size, color, and marking (printing), the similarity distribution 55 and the similarity distribution 57 partially overlap.
　このように、類似度分布５５と類似度分布５７とが一部で重なっているので、図１に示したように、二つの類似度分布の間に閾値Ｔｈ１を設定することができない。一方で、薬剤αを識別する類似度の閾値Ｔｈ２（図２参照）を設定すると、薬剤αは薬剤αとして識別することができるが、薬剤βも薬剤αとして誤って識別される場合が発生してしまう。また、類似度の閾値Ｔｈ３（図２参照）を設定すると、薬剤Ａであるのにもかかわらず薬剤Ａとして識別されない場合が発生してしまう。 As described above, since the similarity distribution 55 and the similarity distribution 57 partially overlap each other, the threshold Th1 cannot be set between the two similarity distributions as shown in FIG. On the other hand, if the threshold value Th2 (see FIG. 2) for identifying the drug α is set, the drug α can be identified as the drug α, but the drug β may also be erroneously identified as the drug α. It ends up. Further, when the similarity threshold Th3 (see FIG. 2) is set, there may be a case where the drug A is not identified as the drug A even though it is the drug A.
　このように、類似度分布が一部で重なるような、類似薬剤βを有する薬剤αを正確に識別するための類似度の閾値を決定することは困難である。 As described above, it is difficult to determine the threshold value of similarity for accurately identifying the drug α having the similar drug β such that the similarity distributions partially overlap.
　そこで本発明を利用することにより、類似薬剤を有する薬剤を識別する場合であっても、正確に且つ効率的に識別を行うことができる。 Therefore, by using the present invention, even when identifying a drug having a similar drug, the identification can be performed accurately and efficiently.
　＜薬剤識別装置＞
　図３は、本発明の薬剤識別装置２（図６）を搭載する一包化監査支援システム１の構成を示す図である。一包化監査支援システム１は処理部１０と、記憶部（薬剤データベース２０Ａ（図６参照））２０と、表示部３０と、操作部４０と、搬送機構５０と、照明部１２と、カメラ１５Ａと、カメラ１５Ｂと、処方箋リーダ１６とを備える。なお、一包化監査支援システム１では、処方箋に従って適切に薬剤が一包化されているかの監査の支援が行われる。具体的には、薬剤識別装置２により一包化された薬剤を識別し、その識別結果を表示部３０（モニタ３２）に表示することにより、ユーザの監査の支援を行う。なお、薬剤識別装置２が利用される例は、一包化監査に限られるものではない。例えば患者が別の病院で処方されて所持していた薬を、入院時に持参した場合に、薬剤識別装置２を利用して、その持参した薬剤の鑑別を行ってもよい。また、薬剤識別装置２が識別する薬剤は、処方箋によって処方される薬剤に限られない。例えば、薬剤識別装置２は処方箋が無い場合であっても、識別対象（第１の薬剤）を特定することできる情報（第１の薬剤を特定する情報）があれば、薬剤識別装置２は薬剤を識別することができる。 <Drug identification device>
FIG. 3 is a diagram showing a configuration of a packaged audit support system 1 equipped with the drug identification device 2 (FIG. 6) of the present invention. The one-pack audit support system 1 includes a processing unit 10, a storage unit (drug database 20A (see FIG. 6)) 20, a display unit 30, an operation unit 40, a transport mechanism 50, an illumination unit 12, and a camera 15A. A camera 15B and a prescription reader 16 are provided. The one-pack audit support system 1 provides support for auditing whether or not the drug is properly packaged according to the prescription. Specifically, the drug identification device 2 identifies the packaged drug, and the identification result is displayed on the display unit 30 (monitor 32) to support the user's audit. The example in which the drug identification device 2 is used is not limited to the one-pack audit. For example, when a patient brings a drug prescribed and possessed by another hospital at the time of admission, the drug identification device 2 may be used to discriminate the drug brought. Further, the drug identified by the drug identification device 2 is not limited to the drug prescribed by the prescription. For example, even if the drug identification device 2 does not have a prescription, if there is information that can identify the identification target (first drug) (information that identifies the first drug), the drug identification device 2 can use the drug. Can be identified.
　カメラ１５Ａ及びカメラ１５Ｂはデジタルカメラにより構成される。図４に示すように、分包袋ＴＰ（薬包）が連続して構成される薬包帯ＰＢの上側（図４の＋Ｚ側）にカメラ１５Ａが配置され、薬包帯ＰＢの下側（図４の－Ｚ側）にカメラ１５Ｂが配置されて、一つの分包袋ＴＰに分包された薬剤を上下（複数の異なる方向）から撮影する。分包袋ＴＰ（薬包帯ＰＢ）は搬送機構５０により図４の＋Ｘ方向（薬包帯ＰＢの長手方向に沿った軸；図４の矢印方向）に搬送され、撮影の際には照明部１２が有する複数の光源１３が４方向から分包袋ＴＰを照明する。図５において、複数の光源１３のそれぞれとカメラ１５Ａ、１５Ｂの撮影光軸ＰＡとの間隔（ｄ１、ｄ２、ｄ３、ｄ４）は同じである。つまり、複数の光源１３と撮影光軸ＰＡとが等間隔（ｄ１＝ｄ２＝ｄ３＝ｄ４）である。 The camera 15A and the camera 15B are composed of a digital camera. As shown in FIG. 4, the camera 15A is arranged on the upper side (+ Z side of FIG. 4) of the medicine bandage PB in which the sachet TP (medicine package) is continuously formed, and the camera 15A is arranged on the lower side of the medicine bandage PB (FIG. 4). The camera 15B is arranged on the -Z side of the bag, and the medicines packaged in one packaging bag TP are photographed from above and below (plurality in different directions). The sachet TP (medicine bandage PB) is transported by the transport mechanism 50 in the + X direction of FIG. 4 (the axis along the longitudinal direction of the drug bandage PB; the arrow direction of FIG. 4), and the illumination unit 12 is used for shooting. A plurality of light sources 13 having the light sources illuminate the packaging bag TP from four directions. In FIG. 5, the intervals (d1, d2, d3, d4) between the plurality of light sources 13 and the photographing optical axes PA of the cameras 15A and 15B are the same. That is, the plurality of light sources 13 and the photographing optical axis PA are at equal intervals (d1 = d2 = d3 = d4).
　処方箋リーダ１６は、処方箋情報を読み取る。例えばＯＣＲ（Optical Character Recognition）により、紙に記載された処方箋から患者氏名、処方された薬剤及びその数量等の情報を読み取る。処方された薬剤に関する情報を示すバーコード等が処方箋に記録されている場合は、処方された薬剤及びその数量等の情報をバーコードから読み取ってもよい。また、ユーザが処方箋を読み取り、操作部４０が有するキーボード等の入力デバイスにより処方箋情報を入力してもよい。 The prescription reader 16 reads the prescription information. For example, OCR (Optical Character Recognition) reads information such as the patient's name, the prescribed drug and its quantity from the prescription written on the paper. When a bar code or the like indicating information on the prescribed drug is recorded on the prescription, information such as the prescribed drug and its quantity may be read from the bar code. Further, the user may read the prescription and input the prescription information by an input device such as a keyboard included in the operation unit 40.
　図６は、薬剤識別装置２の構成例を示すブロック図である。薬剤識別装置２は、主に処理部１０及び記憶部２０で構成される。 FIG. 6 is a block diagram showing a configuration example of the drug identification device 2. The drug identification device 2 is mainly composed of a processing unit 10 and a storage unit 20.
　処理部１０は、画像入力部１０Ａ、薬剤領域抽出部１０Ｂ、登録画像取得部１０Ｃ、第１の認識部１０Ｄ、第２の認識部１０Ｅ及び薬剤識別部１０Ｆを備える。これらの機能は、コンピュータに搭載されるＣＰＵ（Central Processing Unit）、各種電子回路等のデバイスが、ＥＥＰＲＯＭ（Electronically Erasable and Programmable Read Only Memory：非一時的記録媒体）等に記憶されたデータを参照して実現される。また、これらの機能は記憶部２０のＲＯＭ（Read Only Memory：非一時的記録媒体）等に記憶された薬剤識別プログラムを実行することにより行われる。処理の際には、ＲＡＭ（Random Access Memory）等が一時記憶領域、作業領域として用いられる。なお図６ではこれらデバイスの図示は省略する。 The processing unit 10 includes an image input unit 10A, a drug region extraction unit 10B, a registered image acquisition unit 10C, a first recognition unit 10D, a second recognition unit 10E, and a drug identification unit 10F. For these functions, the CPU (Central Processing Unit) mounted on the computer, devices such as various electronic circuits refer to the data stored in EEPROM (Electronically Erasable and Programmable Read Only Memory: non-temporary recording medium), etc. Will be realized. Further, these functions are performed by executing a drug identification program stored in a ROM (Read Only Memory: non-temporary recording medium) or the like of the storage unit 20. At the time of processing, RAM (Random Access Memory) or the like is used as a temporary storage area or a work area. Note that the illustration of these devices is omitted in FIG.
　また、処理部１０は、記憶部２０に接続されており、記憶部２０に記憶されているプログラム及び薬剤データベース２０Ａのデータを自由に取り出し、及び書き換えを行うことができる。 Further, the processing unit 10 is connected to the storage unit 20, and can freely retrieve and rewrite the data of the program and drug database 20A stored in the storage unit 20.
　＜薬剤識別方法＞
　図７は、薬剤識別装置２を使用した薬剤識別方法（薬剤識別工程）に関して説明するフローチャートである。なお、各ステップの詳しい説明は後で行う。 <Drug identification method>
FIG. 7 is a flowchart illustrating a drug identification method (drug identification step) using the drug identification device 2. A detailed explanation of each step will be given later.
　先ず、処方箋リーダ１６により処方箋が読まれ処方箋情報が取得される。そして処方箋情報に基づいて登録画像取得部１０Ｃにより、処方箋により調剤され得る薬剤の第１の登録画像及び第２の登録画像が取得される（登録画像取得ステップ：ステップＳ１０）。ここで第１の登録画像とは、処方箋情報により調剤される第１の薬剤の登録画像である。また、第２の登録画像とは、第１の薬剤に類似する第２の薬剤の登録画像である。 First, the prescription is read by the prescription reader 16 and the prescription information is acquired. Then, based on the prescription information, the registered image acquisition unit 10C acquires the first registered image and the second registered image of the drug that can be dispensed by the prescription (registered image acquisition step: step S10). Here, the first registered image is a registered image of the first drug dispensed based on the prescription information. The second registered image is a registered image of the second drug similar to the first drug.
　次に、画像入力部１０Ａに、一つの分包袋ＴＰを撮影した対象画像ＰＩ（図９参照）が入力される（画像入力ステップ：ステップＳ１１）。その後、薬剤領域抽出部１０Ｂにより、対象画像ＰＩにおいて薬剤領域Ｒ１及び薬剤領域Ｒ２が抽出される（図９参照）（薬剤領域抽出ステップ：ステップＳ１２）。そして、第１の認識部１０Ｄにより、薬剤領域Ｒ１及び薬剤領域Ｒ２に関して、第１の登録画像との類似認識が行われる（第１の認識ステップ：ステップＳ１３）。そして、第１の認識部１０Ｄは、薬剤領域Ｒ１及び薬剤領域Ｒ２と第１の登録画像との類似認識を行う（ステップＳ１４）。第１の認識部１０Ｄにより、第１の登録画像と類似しないと認識された場合には、第１の認識部１０Ｄの認識結果に基づいて、薬剤識別部１０Ｆは薬剤の識別を行う。一方、第１の認識部１０Ｄにより、第１の登録画像と類似すると認識された場合には、第２の認識部１０Ｅにより、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識を行う（第２の認識ステップ：ステップＳ１５）。その後、薬剤識別部１０Ｆは、第１の認識部１０Ｄ及び第２の認識部１０Ｅの認識結果に基づいて、薬剤の識別を行う（薬剤識別ステップ：ステップＳ１６）。 Next, the target image PI (see FIG. 9) obtained by photographing one packaging bag TP is input to the image input unit 10A (image input step: step S11). Then, the drug region extraction unit 10B extracts the drug region R1 and the drug region R2 in the target image PI (see FIG. 9) (drug region extraction step: step S12). Then, the first recognition unit 10D performs similar recognition with the first registered image with respect to the drug region R1 and the drug region R2 (first recognition step: step S13). Then, the first recognition unit 10D performs similar recognition between the drug region R1 and the drug region R2 and the first registered image (step S14). When the first recognition unit 10D recognizes that the image is not similar to the first registered image, the drug identification unit 10F identifies the drug based on the recognition result of the first recognition unit 10D. On the other hand, when the first recognition unit 10D recognizes that the image is similar to the first registered image, the second recognition unit 10E recognizes that the drug region is similar to the first registered image. Similar recognition with the registered image of (second recognition step: step S15). After that, the drug identification unit 10F identifies the drug based on the recognition results of the first recognition unit 10D and the second recognition unit 10E (drug identification step: step S16).
　次に、上述した薬剤識別装置２を搭載した一包化監査支援システム１を使用した薬剤識別方法の各ステップに関して詳細に説明する。 Next, each step of the drug identification method using the packaged audit support system 1 equipped with the drug identification device 2 described above will be described in detail.
　＜＜登録画像取得ステップ＞＞
　登録画像取得ステップ（ステップＳ１０）は登録画像取得部１０Ｃにより行われる。登録画像取得部１０Ｃは、薬剤データベース２０Ａから処方箋情報に基づいて第１の薬剤に対応する、第１の登録画像及び第２の登録画像を取得する。具体的には、登録画像取得部１０Ｃは、処方箋リーダ１６で取得された処方箋情報を取得し、処方箋情報により調剤される第１の薬剤の第１の登録画像と、第１の薬剤に類似する第２の薬剤の第２の登録画像を薬剤データベース２０Ａから取得する。なお、上述の説明では、登録画像取得部１０Ｃは、処方箋リーダ１６で処方箋情報に基づいて第１の薬剤に対応する、第１の登録画像及び第２の登録画像を取得した例を説明した。しかし、登録画像取得部１０Ｃが薬剤データベース２０Ａから第１の登録画像及び第２の登録画像を取得する態様は上述の例に限定されるものではない。例えば、ユーザが処方箋情報を操作部４０で入力し、その入力された処方箋情報に基づいて、登録画像取得部１０Ｃは薬剤データベース２０Ａから第１の登録画像及び第２の登録画像を取得してもよい。 << Registration image acquisition step >>
The registered image acquisition step (step S10) is performed by the registered image acquisition unit 10C. The registered image acquisition unit 10C acquires the first registered image and the second registered image corresponding to the first drug from the drug database 20A based on the prescription information. Specifically, the registered image acquisition unit 10C acquires the prescription information acquired by the prescription reader 16, and is similar to the first registered image of the first drug dispensed by the prescription information and the first drug. A second registered image of the second drug is obtained from the drug database 20A. In the above description, the registered image acquisition unit 10C has described an example in which the prescription reader 16 acquires the first registered image and the second registered image corresponding to the first drug based on the prescription information. However, the mode in which the registered image acquisition unit 10C acquires the first registered image and the second registered image from the drug database 20A is not limited to the above example. For example, even if the user inputs the prescription information in the operation unit 40 and the registered image acquisition unit 10C acquires the first registered image and the second registered image from the drug database 20A based on the input prescription information. Good.
　図８は、薬剤データベース２０Ａの記憶構成例を示す図である。 FIG. 8 is a diagram showing an example of a memory configuration of the drug database 20A.
　薬剤データベース２０Ａでは、「薬剤名称」「登録画像」及び「類似薬剤」が関連して記憶されている。薬剤Ａという薬剤名称は、薬剤Ａの表面画像及び類似薬剤が薬剤Ｘであることが関連して記憶されている。薬剤Ａの表面画像は「Ａ」の刻印が表面に付されている。また、薬剤Ａの類似薬剤は薬剤Ｘである。薬剤Ｂという薬剤名称は、薬剤Ｂの表面画像及び類似薬剤が無いことが関連して記憶されている。薬剤Ｂの表面画像は「Ｂ」の刻印が表面に付されている。薬剤Ｃという薬剤名称は、薬剤Ｃの表面画像及び類似薬剤が薬剤Ｙ及び薬剤Ｚであることが関連して記憶されている。薬剤Ｃの表面画像は「Ｃ」の刻印が表面に付されている。また、薬剤Ｃの類似薬剤は薬剤Ｙ及び薬剤Ｚである。このように、薬剤データベース２０Ａでは、薬剤名称、登録画像及び類似薬剤が関連して記憶されている。なお、図８に示した薬剤データベース２０Ａの記憶構成は、一例であってこれに限定されるものではない。例えば登録画像は、薬剤の表面画像と薬剤の裏面画像とを有していてもよい。このように、登録画像に表面画像と裏面画像とを有することにより、より詳細に類似認識を行うことができる。 In the drug database 20A, "drug name", "registered image" and "similar drug" are stored in association with each other. The drug name drug A is remembered in relation to the surface image of drug A and the similar drug being drug X. The surface image of the drug A is marked with "A" on the surface. Further, the similar drug of the drug A is the drug X. The drug name drug B is remembered in relation to the surface image of drug B and the absence of similar drugs. The surface image of the drug B is marked with "B" on the surface. The drug name drug C is remembered in relation to the surface image of drug C and the similar drugs being drug Y and drug Z. The surface image of the drug C is marked with "C" on the surface. Further, the similar drugs of the drug C are the drug Y and the drug Z. As described above, in the drug database 20A, the drug name, the registered image, and the similar drug are stored in association with each other. The memory structure of the drug database 20A shown in FIG. 8 is an example and is not limited thereto. For example, the registered image may have a front surface image of the drug and a back surface image of the drug. By having the front side image and the back side image in the registered image in this way, it is possible to perform similar recognition in more detail.
　例えば、処方箋に基づいて、一つの分包袋ＴＰに薬剤Ａ及び薬剤Ｂを入れることとなっている場合には、登録画像取得部１０Ｃは、薬剤Ａの登録画像Ａ（第１の登録画像）及び薬剤Ｂの登録画像Ｂ（第１の登録画像）を取得する。また、登録画像取得部１０Ｃは、薬剤Ａには薬剤Ｘが類似するので、薬剤Ｘの登録画像Ｘ（第２の登録画像）を取得する。 For example, when the drug A and the drug B are to be put in one packaging bag TP based on the prescription, the registered image acquisition unit 10C may use the registered image A (first registered image) of the drug A. And the registered image B (first registered image) of the drug B is acquired. Further, since the drug X is similar to the drug A, the registered image acquisition unit 10C acquires the registered image X (second registered image) of the drug X.
　＜＜画像入力ステップ及び薬剤領域抽出ステップ＞＞
　画像入力ステップ（ステップＳ１１）は画像入力部１０Ａにより行われる。画像入力部１０Ａには、識別対象である薬剤の対象画像が入力される。例えば対象画像は、薬剤が処方箋通りに一包化されていることを識別する場合には、薬剤が入れられた一つの分包袋ＴＰを撮影した画像である。ここで対象画像は、分包袋ＴＰを撮影した１枚の撮影画像で構成されても良いし、単数又は複数の撮影画像において薬剤Ａ及び薬剤Ｂの刻印又は印字を画像処理で強調した画像であってもよい。 << Image input step and drug region extraction step >>
The image input step (step S11) is performed by the image input unit 10A. The target image of the drug to be identified is input to the image input unit 10A. For example, the target image is an image of one sachet TP containing the drug in order to identify that the drug is packaged according to the prescription. Here, the target image may be composed of one photographed image obtained by photographing the sachet TP, or an image in which the marking or printing of the drug A and the drug B is emphasized by image processing in one or more photographed images. There may be.
　薬剤領域抽出ステップ（ステップＳ１２）は薬剤領域抽出部１０Ｂにより行われる。薬剤領域抽出部１０Ｂは、対象画像から、識別の対象の薬剤が写っている薬剤領域を薬剤毎に抽出（切り出す）する。薬剤領域抽出部１０Ｂは、例えば薬剤と背景との色の違いに基づいて、薬剤領域を抽出する。 The drug region extraction step (step S12) is performed by the drug region extraction unit 10B. The drug region extraction unit 10B extracts (cuts out) the drug region in which the drug to be identified is shown from the target image for each drug. The drug region extraction unit 10B extracts the drug region, for example, based on the difference in color between the drug and the background.
　図９は、画像入力部１０Ａに入力された対象画像ＰＩを示す図である。対象画像ＰＩは、カメラ１５Ａで撮影され、処理部１０の刻印（印字）強調処理部（不図示）で強調処理が行われている。薬剤領域抽出部１０Ｂは、薬剤Ａの薬剤領域Ｒ１と薬剤Ｂの薬剤領域Ｒ２とをそれぞれ抽出する。なお、一包化監査支援システム１では、カメラ１５Ａ及びカメラ１５Ｂで分包袋ＴＰを両面から撮影をして、薬剤の表面及び裏面の画像を漏れなく取得することができる。しかし、図９に示した例では、カメラ１５Ａで撮影したものであり、薬剤Ａ及び薬剤Ｂの表面の画像を取得できた場合の例である。 FIG. 9 is a diagram showing a target image PI input to the image input unit 10A. The target image PI is photographed by the camera 15A, and enhancement processing is performed by the marking (printing) enhancement processing unit (not shown) of the processing unit 10. The drug region extraction unit 10B extracts the drug region R1 of the drug A and the drug region R2 of the drug B, respectively. In the one-pack audit support system 1, the camera 15A and the camera 15B can photograph the packaging bag TP from both sides, and images of the front and back surfaces of the drug can be acquired without omission. However, the example shown in FIG. 9 was taken with the camera 15A, and is an example in which the images of the surfaces of the drug A and the drug B could be acquired.
　＜＜第１の認識ステップ、第２の認識ステップ及び薬剤識別ステップ＞＞
　第１の認識ステップ（ステップＳ１３）は第１の認識部１０Ｄにより行われる。第１の認識部１０Ｄは、薬剤領域抽出部１０Ｂで抽出された薬剤領域と対応する第１の登録画像との類似認識を行う。具体的に、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との第１の類似度を算出し、第１の類似度と閾値とを比較することにより、類似認識を行う。 << 1st recognition step, 2nd recognition step and drug identification step >>
The first recognition step (step S13) is performed by the first recognition unit 10D. The first recognition unit 10D performs similar recognition of the drug region extracted by the drug region extraction unit 10B and the corresponding first registered image. Specifically, the first recognition unit 10D calculates the first similarity between the drug region and the first registered image, and compares the first similarity with the threshold value to perform similarity recognition.
　第２の認識ステップ（ステップＳ１５）は第２の認識部１０Ｅにより行われる。第２の認識部１０Ｅは、第１の認識部１０Ｄにおいて、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識を行う。具体的に、第２の認識部１０Ｅは、薬剤領域と第２の登録画像の第２の類似度を算出する。 The second recognition step (step S15) is performed by the second recognition unit 10E. The second recognition unit 10E performs similar recognition with the second registered image for the drug region recognized by the first recognition unit 10D to be similar to the first registered image. Specifically, the second recognition unit 10E calculates the second similarity between the drug region and the second registered image.
　薬剤識別ステップ（ステップＳ１６）は、薬剤識別部１０Ｆにより行われる。薬剤識別部１０Ｆは、第１の認識部１０Ｄ及び第２の認識部１０Ｅの認識結果に基づいて、薬剤が第１の薬剤であるか否かの識別を行う。具体的に、薬剤識別部１０Ｆは、第１の類似度が第２の類似度よりも大きい場合に、薬剤領域に写っている薬剤を第１の薬剤であると識別し、第１の類似度が第２の類似度以下である場合に、薬剤領域に写っている薬剤を第１の薬剤でないと識別する。 The drug identification step (step S16) is performed by the drug identification unit 10F. The drug identification unit 10F discriminates whether or not the drug is the first drug based on the recognition results of the first recognition unit 10D and the second recognition unit 10E. Specifically, the drug identification unit 10F identifies the drug shown in the drug region as the first drug when the first similarity is larger than the second similarity, and the first similarity. When is less than or equal to the second similarity, the drug shown in the drug area is identified as not the first drug.
　図１０は、第１の認識部１０Ｄで行われる類似度の算出に関して説明するフローチャートである。なお、第２の認識部１０Ｅでも類似度が算出されるが、第１の認識部１０Ｄと同様にして類似度の算出が行われるので説明は省略する。 FIG. 10 is a flowchart illustrating the calculation of the degree of similarity performed by the first recognition unit 10D. Although the second recognition unit 10E also calculates the similarity, the description will be omitted because the similarity is calculated in the same manner as the first recognition unit 10D.
　先ず、第１の認識部１０Ｄは、薬剤領域と第１の登録画像とのサイズ又は形（外接矩形）が異なるか否かを認識する（ステップＳ２０）。ここで、薬剤領域と第１の登録画像とのサイズ又は形（外接矩形）のいずれかが異なれば、類似度は「０」ゼロと算出する（ステップＳ２４）。なお、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との特徴量の差に基づいて、サイズ又は形（外接矩形）が異なるか否かを認識する。例えば、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との特徴量の差が閾値以上である場合には、両者は異なると認識する。 First, the first recognition unit 10D recognizes whether or not the size or shape (circumscribed rectangle) of the drug region and the first registered image is different (step S20). Here, if either the size or shape (circumscribed rectangle) of the drug region and the first registered image is different, the similarity is calculated as "0" zero (step S24). The first recognition unit 10D recognizes whether or not the size or shape (circumscribed rectangle) is different based on the difference in the feature amount between the drug region and the first registered image. For example, the first recognition unit 10D recognizes that the difference between the drug region and the first registered image is different when the difference between the feature amounts is equal to or larger than the threshold value.
　次に、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との輝度（明るさ）が異なるか否かを認識する（ステップＳ２１）。ここで、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との輝度（明るさ）が異なれば、類似度は「０」ゼロと算出する（ステップＳ２４）。なお、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との輝度の差に基づいて、輝度が異なるか否かを認識する。例えば、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との輝度の差が閾値以上である場合には、両者は異なると認識する。 Next, the first recognition unit 10D recognizes whether or not the brightness (brightness) of the drug region and the first registered image is different (step S21). Here, the first recognition unit 10D calculates that the similarity is “0” zero if the brightness (brightness) of the drug region and the first registered image are different (step S24). The first recognition unit 10D recognizes whether or not the brightness is different based on the difference in brightness between the drug region and the first registered image. For example, the first recognition unit 10D recognizes that when the difference in brightness between the drug region and the first registered image is equal to or greater than the threshold value, the two are different.
　次に、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との色が異なるか否かを認識する（ステップＳ２２）。ここで、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との色が異なる場合には、類似度は「０」ゼロと認識する（ステップＳ２４）。なお、第１の認識部１０Ｄは、薬剤領域と第１の登録画像とのＲ（赤）、Ｇ（緑）、Ｂ（青）の値の差に基づいて、色が異なるか否かを認識する。例えば、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との色の差が閾値以上である場合には、両者は異なると認識する。 Next, the first recognition unit 10D recognizes whether or not the color of the drug region and the first registered image are different (step S22). Here, when the color of the drug region and the first registered image are different, the first recognition unit 10D recognizes that the similarity is “0” zero (step S24). The first recognition unit 10D recognizes whether or not the colors are different based on the difference in the R (red), G (green), and B (blue) values between the drug region and the first registered image. To do. For example, the first recognition unit 10D recognizes that the two are different when the color difference between the drug region and the first registered image is equal to or greater than the threshold value.
　次に、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との類似度を算出する（ステップＳ２２）。第１の認識部１０Ｄは、テンプレートマッチングにより類似度を算出する。例えば類似度は、１．０が最も類似する場合（同一の場合）であり、異なる箇所の程度により、１．０から低く算出される。 Next, the first recognition unit 10D calculates the degree of similarity between the drug region and the first registered image (step S22). The first recognition unit 10D calculates the similarity by template matching. For example, the similarity is when 1.0 is the most similar (same case), and is calculated lower than 1.0 depending on the degree of different points.
　このようにして、第１の認識部１０Ｄは、薬剤領域と第１の登録画像との第１の類似度を算出する。なお、第２の認識部１０Ｅは第１の認識部１０Ｄと同様に類似度を算出してもよいが、第２の認識部１０Ｅは、第１の認識部１０Ｄで類似すると認識された薬剤領域について類似度を算出するので、初めからテンプレートマッチングにより類似度を直接算出（図１０のステップＳ２３）してもよい。 In this way, the first recognition unit 10D calculates the first similarity between the drug region and the first registered image. The second recognition unit 10E may calculate the similarity in the same manner as the first recognition unit 10D, but the second recognition unit 10E is a drug region recognized to be similar by the first recognition unit 10D. Since the similarity is calculated for the above, the similarity may be directly calculated by template matching from the beginning (step S23 in FIG. 10).
　次に、第１の認識部１０Ｄの類似認識、第２の認識部１０Ｅの類似認識及び薬剤識別部１０Ｆの薬剤識別に関して、薬剤Ａ及び薬剤Ｂのそれぞれの識別について具体的に説明する。 Next, regarding the similarity recognition of the first recognition unit 10D, the similarity recognition of the second recognition unit 10E, and the drug identification of the drug identification unit 10F, the identification of each of the drug A and the drug B will be specifically described.
　図１１は、第１の認識部１０Ｄの類似認識、第２の認識部１０Ｅの類似認識及び薬剤識別を示すフローチャートである。 FIG. 11 is a flowchart showing the similarity recognition of the first recognition unit 10D, the similarity recognition of the second recognition unit 10E, and the drug identification.
　先ずは、対象画像ＰＩにおける薬剤領域Ｒ１の識別に関して説明する。 First, the identification of the drug region R1 in the target image PI will be described.
　第１の認識部１０Ｄは、薬剤領域Ｒ１と登録画像Ａ及び登録画像Ｂとの類似認識を行う（ステップＳ３０）。具体的には、薬剤領域Ｒ１と登録画像Ａとは、サイズ（形）、輝度（明るさ）、及び色が同じであるので第１の認識部１０Ｄは、テンプレートマッチングにより類似度を算出する（図１０のステップＳ２３を参照）。一方、第１の認識部１０Ｄは、薬剤領域Ｒ１と登録画像Ｂとはサイズが異なるので類似度をゼロと算出する。 The first recognition unit 10D performs similar recognition between the drug region R1 and the registered image A and the registered image B (step S30). Specifically, since the drug region R1 and the registered image A have the same size (shape), brightness (brightness), and color, the first recognition unit 10D calculates the similarity by template matching ( See step S23 in FIG. 10). On the other hand, the first recognition unit 10D calculates the similarity as zero because the size of the drug region R1 and the registered image B are different.
　そして、第１の認識部１０Ｄは、算出された類似度と閾値ＴｈＡとを比較して、類似認識を行う。薬剤領域Ｒ１と登録画像Ａとの類似度は、閾値ＴｈＡ以上であるので、薬剤領域Ｒ１と登録画像Ａと類似すると認識する。なお、閾値ＴｈＡは、できるだけ薬剤Ａを漏れなく抽出する観点より低めに設定されることが好ましい。例えば閾値ＴｈＡは、図２で説明した閾値Ｔｈ２のように設定される。一方、第１の認識部１０Ｄは、薬剤領域Ｒ１と登録画像Ｂとは、類似度がゼロであり閾値ＴｈＡより小さいので、類似していないと認識し、薬剤識別部１０Ｆは、薬剤領域Ｒ１は薬剤Ｂでないと識別する（ステップＳ３５）。 Then, the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R1 and the registered image A is equal to or higher than the threshold value ThA, it is recognized that the drug region R1 is similar to the registered image A. The threshold ThA is preferably set as low as possible from the viewpoint of extracting the drug A without omission. For example, the threshold ThA is set as the threshold Th2 described with reference to FIG. On the other hand, the first recognition unit 10D recognizes that the drug region R1 and the registered image B are not similar to each other because the similarity is zero and is smaller than the threshold value ThA, and the drug identification unit 10F recognizes that the drug region R1 is not similar. It is identified as not drug B (step S35).
　その後、第２の認識部１０Ｅは、薬剤Ａの類似薬剤の有無について判定し（ステップＳ３１）、類似薬剤が有る場合には、薬剤領域Ｒ１と登録画像Ｘとの第２の類似度を算出する（ステップＳ３２）。そして、薬剤識別部１０Ｆは、第１の類似度と第２の類似度に基づいて、薬剤領域Ｒ１が薬剤Ａであるか否かを識別する（ステップＳ３３）。具体的には、薬剤識別は、第１の類似度が第２の類似度よりも大きい場合には、薬剤領域Ｒ１を薬剤Ａと識別する（ステップＳ３４）。一方、第１の類似度が第２の類似度以下の場合には、薬剤領域Ｒ１は薬剤Ａではないと識別する（ステップＳ３４）。 After that, the second recognition unit 10E determines the presence or absence of a similar drug of the drug A (step S31), and if there is a similar drug, calculates the second degree of similarity between the drug region R1 and the registered image X. (Step S32). Then, the drug identification unit 10F discriminates whether or not the drug region R1 is the drug A based on the first similarity and the second similarity (step S33). Specifically, the drug identification identifies the drug region R1 from the drug A when the first similarity is greater than the second similarity (step S34). On the other hand, when the first similarity is equal to or less than the second similarity, the drug region R1 is identified as not the drug A (step S34).
　次に、対象画像ＰＩにおける薬剤領域Ｒ２の識別に関して、同様に図１１に沿って説明する。 Next, the identification of the drug region R2 in the target image PI will be similarly described with reference to FIG.
　第１の認識部１０Ｄは、薬剤領域Ｒ２と登録画像Ａ及び登録画像Ｂとの類似認識を行う（ステップＳ３０）。具体的には、薬剤領域Ｒ２と登録画像Ｂとは、サイズ（形）、色、刻印（印字）が同じであるので第１の認識部１０Ｄは、テンプレートマッチングにより類似度を算出する（図１０のステップＳ２３を参照）。一方、第１の認識部１０Ｄは、薬剤領域Ｒ２と登録画像Ａとはサイズが異なるので、類似度をゼロと算出する。 The first recognition unit 10D performs similar recognition between the drug region R2 and the registered image A and the registered image B (step S30). Specifically, since the drug region R2 and the registered image B have the same size (shape), color, and marking (printing), the first recognition unit 10D calculates the similarity by template matching (FIG. 10). (See step S23). On the other hand, since the size of the drug region R2 and the registered image A are different in the first recognition unit 10D, the similarity is calculated as zero.
　そして、第１の認識部１０Ｄは、算出された類似度と閾値ＴｈＡとを比較して、類似認識を行う。薬剤領域Ｒ２と登録画像Ｂとの類似度は、閾値ＴｈＡ以上であるので、薬剤領域Ｒ２と登録画像Ｂと類似すると認識する。そして、第２の認識部１０Ｅは、薬剤Ｂの類似薬剤の有無について判定し（ステップＳ３１）、薬剤Ｂには類似薬剤は存在しないので、第２の認識部１０Ｅは第２の類似度を算出しない。その後、薬剤識別部１０Ｆは、第１の認識部１０Ｄの類似認識に基づいて、薬剤領域Ｒ２に写っている薬剤は薬剤Ｂと識別する（ステップＳ３４）。一方、第１の認識部１０Ｄは、薬剤領域Ｒ２と登録画像Ａとは、類似度がゼロであり閾値ＴｈＡより小さいので、類似していないと認識し、薬剤識別部１０Ｆは、薬剤領域Ｒ２は薬剤Ａでないと識別する（ステップＳ３５）。 Then, the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R2 and the registered image B is equal to or higher than the threshold value ThA, it is recognized that the drug region R2 is similar to the registered image B. Then, the second recognition unit 10E determines the presence or absence of a similar drug of the drug B (step S31), and since there is no similar drug in the drug B, the second recognition unit 10E calculates the second similarity degree. do not do. After that, the drug identification unit 10F identifies the drug shown in the drug region R2 from the drug B based on the similarity recognition of the first recognition unit 10D (step S34). On the other hand, the first recognition unit 10D recognizes that the drug region R2 and the registered image A are not similar because the similarity is zero and is smaller than the threshold value ThA, and the drug identification unit 10F recognizes that the drug region R2 is not similar. It is identified as not drug A (step S35).
　以上で説明したように、本発明においては、先ず、第１の認識部１０Ｄにより、薬剤領域と対応する第１の登録画像との類似認識が行われる。そして、本態様では、第１の認識部１０Ｄにおいて、第１の登録画像と類似すると認識された薬剤領域について、第２の登録画像との類似認識が行われる。これにより、本態様は、第２の登録画像との類似認識は、第２の登録画像と類似すると考えられる薬剤領域に対して行われるので、正確で効率的な識別を行うことができる。 As described above, in the present invention, first, the first recognition unit 10D performs similar recognition of the drug region and the corresponding first registered image. Then, in this aspect, the first recognition unit 10D performs similar recognition with the second registered image for the drug region recognized to be similar to the first registered image. Thereby, in this aspect, since the similarity recognition with the second registered image is performed for the drug region considered to be similar to the second registered image, accurate and efficient identification can be performed.
　＜その他の例＞
　＜＜類似薬剤が複数ある場合＞＞
　上述した説明では、薬剤Ａ（第１の画像）が一つの類似薬剤（薬剤Ｘ）を有する場合について説明を行った。しかしながら、本発明はこの例に限定されず、類似薬剤が複数ある場合にも適用される。 <Other examples>
<< When there are multiple similar drugs >>
In the above description, the case where the drug A (first image) has one similar drug (drug X) has been described. However, the present invention is not limited to this example, and is also applied when there are a plurality of similar agents.
　図１２は、画像入力部１０Ａに入力された対象画像ＰＪを示す図である。対象画像ＰＪは図９で説明した対象画像ＰＩと同様にして取得されている。そして、薬剤領域抽出部１０Ｂは、薬剤Ｃの薬剤領域Ｒ３を抽出する。 FIG. 12 is a diagram showing a target image PJ input to the image input unit 10A. The target image PJ is acquired in the same manner as the target image PI described with reference to FIG. Then, the drug region extraction unit 10B extracts the drug region R3 of the drug C.
　図１３は、第１の認識部１０Ｄの類似認識、第２の認識部１０Ｅの類似認識及び薬剤識別部１０Ｆの薬剤識別を示すフローチャートである。薬剤Ｃの識別について、図１１のフローチャートに沿って説明をする。なお、処方箋情報では、一つの分包袋ＴＰに第１の薬剤である薬剤Ｃを入れることになっており、登録画像取得部１０Ｃは、処方箋情報に基づいて第１の薬剤に対応する、登録画像Ｃ、登録画像Ｙ及び登録画像Ｚを薬剤データベース２０Ａから取得する。 FIG. 13 is a flowchart showing the similarity recognition of the first recognition unit 10D, the similarity recognition of the second recognition unit 10E, and the drug identification of the drug identification unit 10F. The identification of the drug C will be described with reference to the flowchart of FIG. In the prescription information, the drug C, which is the first drug, is to be put in one packaging bag TP, and the registration image acquisition unit 10C is registered to correspond to the first drug based on the prescription information. Image C, registered image Y, and registered image Z are acquired from the drug database 20A.
　第１の認識部１０Ｄは、薬剤領域Ｒ３と登録画像Ｃとの類似認識を行う（ステップＳ４０）。具体的には、薬剤領域Ｒ３と登録画像Ｃとは、サイズ（形）、輝度（明るさ）、及び色が同じであるので第１の認識部１０Ｄは、テンプレートマッチングにより類似度を算出する（図１０のステップＳ２３を参照）。 The first recognition unit 10D performs similar recognition between the drug region R3 and the registered image C (step S40). Specifically, since the drug region R3 and the registered image C have the same size (shape), brightness (brightness), and color, the first recognition unit 10D calculates the similarity by template matching ( See step S23 in FIG. 10).
　その後、第１の認識部１０Ｄは、算出された類似度と閾値ＴｈＡとを比較して、類似認識を行う。薬剤領域Ｒ３と登録画像Ｃとの類似度は、閾値ＴｈＡ以上であるので、薬剤領域Ｒ３と登録画像Ｃと類似すると認識する。 After that, the first recognition unit 10D compares the calculated similarity with the threshold value ThA and performs similarity recognition. Since the degree of similarity between the drug region R3 and the registered image C is equal to or higher than the threshold value ThA, it is recognized that the drug region R3 is similar to the registered image C.
　その後、第２の認識部１０Ｅは、薬剤領域Ｒ３と登録画像Ｙとの第２の類似度（１）を算出する（ステップＳ４１）。また、第２の認識部１０Ｅは、薬剤領域Ｒ３と登録画像Ｚとの第２の類似度（２）を算出する（ステップＳ４２）。このように、第２の認識部１０Ｅは、類似薬剤の登録画像の各々について類似度を算出する。そして、薬剤識別部１０Ｆは、第１の類似度と第２の類似度の各々に基づいて識別を行う。具体的には、薬剤識別部１０Ｆは、第１の類似度が第２の類似度（１）よりも大きいか否かの判定を行う（ステップＳ４３）。薬剤識別部１０Ｆは、第１の類似度が第２の類似度（１）よりも大きくない場合には、薬剤領域Ｒ３は薬剤Ｃでないと識別する。一方、第１の類似度が第２の類似度（１）よりも大きい場合には、次に薬剤識別部１０Ｆは、第１の類似度が第２の類似度（２）よりも大きいか否かの判定を行う（ステップＳ４４）。薬剤識別部１０Ｆは、第１の類似度が第２の類似度（２）よりも大きくない場合には、薬剤領域Ｒ３は薬剤Ｃでないと識別する（ステップＳ４６）。一方、第１の類似度が第２の類似度（２）よりも大きい場合には、薬剤領域Ｒ３は薬剤Ｃと識別する（ステップＳ４５）。 After that, the second recognition unit 10E calculates the second similarity (1) between the drug region R3 and the registered image Y (step S41). In addition, the second recognition unit 10E calculates the second similarity (2) between the drug region R3 and the registered image Z (step S42). In this way, the second recognition unit 10E calculates the degree of similarity for each of the registered images of similar drugs. Then, the drug identification unit 10F performs identification based on each of the first similarity and the second similarity. Specifically, the drug identification unit 10F determines whether or not the first similarity is greater than the second similarity (1) (step S43). The drug identification unit 10F identifies that the drug region R3 is not drug C when the first similarity is not greater than the second similarity (1). On the other hand, when the first similarity is larger than the second similarity (1), the drug identification unit 10F then determines whether the first similarity is larger than the second similarity (2). Is determined (step S44). The drug identification unit 10F identifies that the drug region R3 is not drug C when the first similarity is not greater than the second similarity (2) (step S46). On the other hand, when the first similarity is larger than the second similarity (2), the drug region R3 is distinguished from the drug C (step S45).
　このように、２以上の類似薬剤を有する薬剤を識別する場合であっても、第１の認識部１０Ｄ及び第２の認識部１０Ｅの認識結果に基づいて、薬剤を識別することにより、正確で効率的な薬剤の識別を行うことができる。 In this way, even when identifying a drug having two or more similar drugs, the drug is accurately identified based on the recognition results of the first recognition unit 10D and the second recognition unit 10E. Efficient drug identification can be performed.
　以上で本発明の例に関して説明してきたが、本発明は上述した実施の形態に限定されず、本発明の精神を逸脱しない範囲で種々の変形が可能であることは言うまでもない。 Although the examples of the present invention have been described above, it goes without saying that the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.
１　　　：一包化監査支援システム
２　　　：薬剤識別装置
１０　　：処理部
１０Ａ　：画像入力部
１０Ｂ　：薬剤領域抽出部
１０Ｃ　：登録画像取得部
１０Ｄ　：第１の認識部
１０Ｅ　：第２の認識部
１０Ｆ　：薬剤識別部
１２　　：照明部
１３　　：光源
１５Ａ　：カメラ
１５Ｂ　：カメラ
１６　　：処方箋リーダ
２０　　：記憶部
２０Ａ　：薬剤データベース
３０　　：表示部
４０　　：操作部
５０　　：搬送機構 1: Packaged audit support system 2: Drug identification device 10: Processing unit 10A: Image input unit 10B: Drug area extraction unit 10C: Registered image acquisition unit 10D: First recognition unit 10E: Second recognition unit 10F: Drug identification unit 12: Lighting unit 13: Light source 15A: Camera 15B: Camera 16: Prescription reader 20: Storage unit 20A: Drug database 30: Display unit 40: Operation unit 50: Conveyance mechanism

Claims (11)

  1. 　第１の薬剤が類似する第２の薬剤を有する場合の前記第１の薬剤を識別する薬剤識別装置であって、
    　前記第１の薬剤の画像である第１の登録画像、及び前記第２の薬剤が存在する場合に前記第２の薬剤の画像である第２の登録画像が前記第１の登録画像に関連して記憶されている薬剤データベースと、
    　前記識別の対象である薬剤の対象画像が入力される画像入力部と、
    　前記対象画像から前記薬剤が写っている薬剤領域を抽出する薬剤領域抽出部と、
    　前記薬剤データベースから前記第１の薬剤に対応する、前記第１の登録画像及び前記第２の登録画像を取得する登録画像取得部と、
    　前記薬剤領域抽出部で抽出された前記薬剤領域と対応する前記第１の登録画像との類似認識を行う第１の認識部と、
    　前記第１の認識部において、前記第１の登録画像と類似すると認識された前記薬剤領域について、前記第２の登録画像との類似認識を行う第２の認識部と、
    　前記第１の認識部及び前記第２の認識部の認識結果に基づいて、前記薬剤が前記第１の薬剤であるか否かの識別を行う薬剤識別部と、
    　を備える薬剤識別装置。     A drug identification device that identifies the first drug when the first drug has a similar second drug.
    The first registered image, which is an image of the first drug, and the second registered image, which is an image of the second drug when the second drug is present, are related to the first registered image. The drug database that is stored in
    An image input unit in which a target image of the drug to be identified is input, and
    A drug region extraction unit that extracts a drug region in which the drug appears from the target image,
    A registered image acquisition unit that acquires the first registered image and the second registered image corresponding to the first drug from the drug database.
    A first recognition unit that performs similar recognition with the first registered image corresponding to the drug region extracted by the drug region extraction unit, and
    A second recognition unit that performs similar recognition with the second registered image for the drug region recognized to be similar to the first registered image in the first recognition unit.
    A drug identification unit that identifies whether or not the drug is the first drug based on the recognition results of the first recognition unit and the second recognition unit.
    A drug identification device comprising.
  2. 　前記第１の認識部は、前記薬剤領域と前記第１の登録画像との第１の類似度を算出し、前記第１の類似度と閾値とを比較することにより、前記類似認識を行う請求項１に記載の薬剤識別装置。 The first recognition unit calculates the first similarity between the drug region and the first registered image, and compares the first similarity with the threshold value to perform the similarity recognition. Item 2. The drug identification device according to item 1.
  3. 　前記第２の認識部は、前記薬剤領域と前記第２の登録画像との第２の類似度を算出する請求項２に記載の薬剤識別装置。 The drug identification device according to claim 2, wherein the second recognition unit calculates a second similarity between the drug region and the second registered image.
  4. 　前記薬剤識別部は、前記第１の類似度が前記第２の類似度よりも大きい場合に、前記薬剤領域に写っている前記薬剤を前記第１の薬剤であると識別し、前記第１の類似度が前記第２の類似度以下である場合に、前記薬剤領域に写っている前記薬剤を前記第１の薬剤でないと識別する請求項３に記載の薬剤識別装置。 When the first similarity is larger than the second similarity, the drug identification unit identifies the drug in the drug region as the first drug, and determines that the first drug is the first drug. The drug identification device according to claim 3, wherein when the degree of similarity is equal to or less than the second degree of similarity, the drug shown in the drug region is identified as not the first drug.
  5. 　前記第１の薬剤は、類似する複数の前記第２の薬剤を有する請求項２に記載の薬剤識別装置。 The drug identification device according to claim 2, wherein the first drug has a plurality of similar second drugs.
  6. 　前記第２の認識部は、前記薬剤領域と複数の前記第２の登録画像の各々との第２の類似度を算出する請求項５に記載の薬剤識別装置。 The drug identification device according to claim 5, wherein the second recognition unit calculates a second similarity between the drug region and each of the plurality of second registered images.
  7. 　前記薬剤識別部は、前記第１の類似度が複数の前記第２の類似度の各々よりも大きい場合に、前記薬剤領域に写っている前記薬剤を前記第１の薬剤であると識別し、前記第１の類似度が複数の前記第２の類似度の少なくとも一つ以下である場合に、前記薬剤領域に写っている前記薬剤を前記第１の薬剤でないと識別する請求項６に記載の薬剤識別装置。 When the first similarity is greater than each of the plurality of second similarity, the drug identification unit identifies the drug in the drug region as the first drug. The sixth aspect of claim 6, wherein when the first similarity is at least one or less than a plurality of the second similarity, the drug shown in the drug region is identified as not the first drug. Drug identification device.
  8. 　前記第１の認識部及び前記第２の認識部の前記類似認識は、テンプレートマッチングにより行われる請求項１から７のいずれか１項に記載の薬剤識別装置。 The drug identification device according to any one of claims 1 to 7, wherein the similarity recognition of the first recognition unit and the second recognition unit is performed by template matching.
  9. 　第１の薬剤が類似する第２の薬剤を有する場合の前記第１の薬剤を識別する薬剤識別方法であって、
    　前記第１の薬剤の画像である第１の登録画像、及び前記第２の薬剤が存在する場合に前記第２の薬剤の画像である第２の登録画像が前記第１の登録画像に関連して記憶されている薬剤データベースから、前記第１の薬剤に対応する、前記第１の登録画像及び前記第２の登録画像を取得する登録画像取得ステップと、
    　前記識別の対象である薬剤の対象画像が入力される画像入力ステップと、
    　前記対象画像から前記薬剤が写っている薬剤領域を抽出する薬剤領域抽出ステップと、
    　前記薬剤領域抽出ステップで抽出された前記薬剤領域と対応する前記第１の登録画像との類似認識を行う第１の認識ステップと、
    　前記第１の認識ステップにおいて、前記第１の登録画像と類似すると認識された前記薬剤領域について、前記第２の登録画像との類似認識を行う第２の認識ステップと、
    　前記第１の認識ステップ及び前記第２の認識ステップの認識結果に基づいて、前記薬剤が前記第１の薬剤であるか否かの識別を行う薬剤識別ステップと、
    　を含む薬剤識別方法。     A drug identification method for identifying the first drug when the first drug has a similar second drug.
    The first registered image, which is an image of the first drug, and the second registered image, which is an image of the second drug when the second drug is present, are related to the first registered image. A registered image acquisition step of acquiring the first registered image and the second registered image corresponding to the first drug from the drug database stored in the above.
    An image input step in which a target image of the drug to be identified is input, and
    A drug region extraction step for extracting a drug region in which the drug is shown from the target image,
    A first recognition step for performing similarity recognition between the drug region extracted in the drug region extraction step and the corresponding first registered image, and
    In the first recognition step, the second recognition step of performing the similarity recognition of the drug region recognized to be similar to the first registered image with the second registered image,
    A drug identification step for identifying whether or not the drug is the first drug based on the recognition results of the first recognition step and the second recognition step.
    Drug identification method including.
  10. 　第１の薬剤が類似する第２の薬剤を有する場合の前記第１の薬剤を識別する薬剤識別工程をコンピュータに実行させるプログラムであって、
    　前記第１の薬剤の画像である第１の登録画像、及び前記第２の薬剤が存在する場合に前記第２の薬剤の画像である第２の登録画像が前記第１の登録画像に関連して記憶されている薬剤データベースから、前記第１の薬剤に対応する、前記第１の登録画像及び前記第２の登録画像を取得する登録画像取得ステップと、
    　前記識別の対象である薬剤の対象画像が入力される画像入力ステップと、
    　前記対象画像から前記薬剤が写っている薬剤領域を抽出する薬剤領域抽出ステップと、
    　前記薬剤領域抽出ステップで抽出された前記薬剤領域と対応する前記第１の登録画像との類似認識を行う第１の認識ステップと、
    　前記第１の認識ステップにおいて、前記第１の登録画像と類似すると認識された前記薬剤領域について、前記第２の登録画像との類似認識を行う第２の認識ステップと、
    　前記第１の認識ステップ及び前記第２の認識ステップの認識結果に基づいて、前記薬剤が前記第１の薬剤であるか否かの識別を行う薬剤識別ステップと、
    　を含む薬剤識別工程をコンピュータに実行させるプログラム。     A program that causes a computer to perform a drug identification step for identifying the first drug when the first drug has a similar second drug.
    The first registered image, which is an image of the first drug, and the second registered image, which is an image of the second drug when the second drug is present, are related to the first registered image. A registered image acquisition step of acquiring the first registered image and the second registered image corresponding to the first drug from the drug database stored in the above.
    An image input step in which a target image of the drug to be identified is input, and
    A drug region extraction step for extracting a drug region in which the drug is shown from the target image,
    A first recognition step for performing similarity recognition between the drug region extracted in the drug region extraction step and the corresponding first registered image, and
    In the first recognition step, the second recognition step of performing the similarity recognition of the drug region recognized to be similar to the first registered image with the second registered image,
    A drug identification step for identifying whether or not the drug is the first drug based on the recognition results of the first recognition step and the second recognition step.
    A program that causes a computer to perform a drug identification process that includes.
  11. 　非一時的かつコンピュータ読取可能な記録媒体であって、前記記録媒体に格納された指令がコンピュータによって読み取られた場合に請求項１０に記載のプログラムをコンピュータに実行させる記録媒体。 A recording medium that is non-temporary and computer-readable, and causes the computer to execute the program according to claim 10 when a command stored in the recording medium is read by the computer.
PCT/JP2020/026850 2019-07-16 2020-07-09 Drug identification device, drug identification method, and program WO2021010277A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2021533015A JP7280361B2 (en) 2019-07-16 2020-07-09 Drug identification device, drug identification method and program

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2019131441 2019-07-16
JP2019-131441 2019-07-16

Publications (1)

Publication Number Publication Date
WO2021010277A1 true WO2021010277A1 (en) 2021-01-21

Family

ID=74210771

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/026850 WO2021010277A1 (en) 2019-07-16 2020-07-09 Drug identification device, drug identification method, and program

Country Status (2)

Country Link
JP (1) JP7280361B2 (en)
WO (1) WO2021010277A1 (en)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004167158A (en) * 2002-11-22 2004-06-17 Glory Ltd Medicine recognition device
JP2005056252A (en) * 2003-08-06 2005-03-03 Casio Comput Co Ltd Image collation device, method and program
WO2011145351A1 (en) * 2010-05-20 2011-11-24 パナソニック株式会社 Drug solution determination device and drug solution determination method
WO2012147907A1 (en) * 2011-04-28 2012-11-01 株式会社湯山製作所 Medicine checking device and apparatus for separately packaging medicines
JP2014064836A (en) * 2012-09-27 2014-04-17 Fujifilm Corp Medicine inspection device and method
JP2018128931A (en) * 2017-02-09 2018-08-16 キヤノン株式会社 Image processing device, image processing method, and program
WO2018173649A1 (en) * 2017-03-23 2018-09-27 富士フイルム株式会社 Drug recognizing device, drug recognizing method, and drug recognizing program
WO2018221065A1 (en) * 2017-05-30 2018-12-06 富士フイルム株式会社 Drug inspection assistance device and drug inspection assistance method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019039016A1 (en) 2017-08-25 2019-02-28 富士フイルム株式会社 Medicine inspection assistance device, image processing device, image processing method, and program
WO2019044439A1 (en) 2017-08-31 2019-03-07 富士フイルム株式会社 Medicine inspection device, image processing device, image processing method, and program

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004167158A (en) * 2002-11-22 2004-06-17 Glory Ltd Medicine recognition device
JP2005056252A (en) * 2003-08-06 2005-03-03 Casio Comput Co Ltd Image collation device, method and program
WO2011145351A1 (en) * 2010-05-20 2011-11-24 パナソニック株式会社 Drug solution determination device and drug solution determination method
WO2012147907A1 (en) * 2011-04-28 2012-11-01 株式会社湯山製作所 Medicine checking device and apparatus for separately packaging medicines
JP2014064836A (en) * 2012-09-27 2014-04-17 Fujifilm Corp Medicine inspection device and method
JP2018128931A (en) * 2017-02-09 2018-08-16 キヤノン株式会社 Image processing device, image processing method, and program
WO2018173649A1 (en) * 2017-03-23 2018-09-27 富士フイルム株式会社 Drug recognizing device, drug recognizing method, and drug recognizing program
WO2018221065A1 (en) * 2017-05-30 2018-12-06 富士フイルム株式会社 Drug inspection assistance device and drug inspection assistance method

Also Published As

Publication number Publication date
JPWO2021010277A1 (en) 2021-01-21
JP7280361B2 (en) 2023-05-23

Similar Documents

Publication Publication Date Title
WO2015165365A1 (en) Facial recognition method and system
US9836665B2 (en) Management system, list production device, method, computer readable recording medium, data structure, and printed label
WO2014050485A1 (en) Drug inspection device and method
US20160203291A1 (en) Drug verification device, drug verification system and drug verification method
JPS5959397A (en) Labeller for characteristic point
US10937152B2 (en) Inspection support method and inspection support device
JP4783331B2 (en) Face recognition device
JP2004318879A5 (en)
WO2015008732A1 (en) Optical character recognition device
US11423246B2 (en) Medicine inspection assistance device, image processing device, image processing method, and program
US11580762B2 (en) Training a card type classifier with simulated card images
JP6853891B2 (en) Drug audit equipment, image processing equipment, image processing methods and programs
WO2021065658A1 (en) Medicine collation device, medicine collation system, medicine management method, and program
WO2021010277A1 (en) Drug identification device, drug identification method, and program
JP6757851B2 (en) Dispensing audit support device and dispensing audit support method
US7346214B2 (en) Method for capturing a complete data set of forms provided with graphic characters
TWI684157B (en) A Smart Claims System Based on Action Vehicle
JP6975882B2 (en) Collation device and collation result display method
JP2017173925A (en) Optical character recognition device
US11817207B1 (en) Medication inventory system including image based boundary determination for generating a medication tray stocking list and related methods
US11721432B1 (en) Medication inventory system including boundary outline based medication tray stocking list and related methods
WO2021014527A1 (en) Comparison position output system
JP6941331B2 (en) Image recognition system
JP2019017913A (en) Information processing system, control method for information processing system, and program
US20170200383A1 (en) Automated review of forms through augmented reality

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20841267

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2021533015

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20841267

Country of ref document: EP

Kind code of ref document: A1