CN109522932A - Medicine-chest drug intelligent positioning management method, system, storage medium and equipment - Google Patents
Medicine-chest drug intelligent positioning management method, system, storage medium and equipment Download PDFInfo
- Publication number
- CN109522932A CN109522932A CN201811213445.2A CN201811213445A CN109522932A CN 109522932 A CN109522932 A CN 109522932A CN 201811213445 A CN201811213445 A CN 201811213445A CN 109522932 A CN109522932 A CN 109522932A
- Authority
- CN
- China
- Prior art keywords
- drug
- medicine
- cargo path
- shipment
- coordinate
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
- 239000003814 drug Substances 0.000 title claims abstract description 277
- 229940079593 drug Drugs 0.000 title claims abstract description 200
- 238000007726 management method Methods 0.000 title claims abstract description 25
- 238000000605 extraction Methods 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 9
- 238000012937 correction Methods 0.000 claims abstract description 5
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 21
- 238000010191 image analysis Methods 0.000 claims description 13
- 238000012986 modification Methods 0.000 claims description 13
- 230000004048 modification Effects 0.000 claims description 13
- 230000004913 activation Effects 0.000 claims description 7
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000002790 cross-validation Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000000644 propagated effect Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000004880 explosion Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000009885 systemic effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 229940046236 drug tradename Drugs 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 238000012856 packing Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 239000011800 void material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
- G06K17/0022—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device
- G06K17/0025—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations arrangements or provisious for transferring data to distant stations, e.g. from a sensing device the arrangement consisting of a wireless interrogation device in combination with a device for optically marking the record carrier
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07F—COIN-FREED OR LIKE APPARATUS
- G07F17/00—Coin-freed apparatus for hiring articles; Coin-freed facilities or services
- G07F17/0092—Coin-freed apparatus for hiring articles; Coin-freed facilities or services for assembling and dispensing of pharmaceutical articles
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Networks & Wireless Communication (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Medical Preparation Storing Or Oral Administration Devices (AREA)
- Warehouses Or Storage Devices (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
A kind of medicine-chest drug intelligent positioning management method, system, storage medium and the equipment in medicine retail technology field, comprising the following steps: S1, establish drug identification model；S2, obtain the images to be recognized of demand drug；S3, pre- shipment medicine information extraction；S4, position correction barcode scanning drug.The present invention is matched using expection drug of the image detecting technique to client with to shipment drug, prevents mistake from selling drug, and medicine safety is sold in raising；Simultaneously can medicine information in real-time update medicine-chest, improve drug control efficiency in medicine-chest, reduce management cost.
Description
Technical field
The present invention relates to a kind of technology of medical retail domain, specifically a kind of medicine-chest drug intelligent positioning managers
Method, system, storage medium and equipment.
Background technique
With the continuous iteration of self-service machine technology and development, Pharmaceutical retail can be applied to now.In existing market
Automatic medicine selling machine is manually got in stocks mode using traditional self-service machine, that is, the personnel that get in stocks are according to host computer presupposed information by similar medicine
Product are manually placed into specified cargo path (host computer presupposed information can carry out artificial regeneration), and the drug that client buys as needed selects goods
Road shipment, equipment then pass through laser barcode scanning technology scanning drug trace to the source code confirmation shipment medicine information it is errorless after shipment.
The problem of in view of selling medicine and drug safety, it is necessary to it traces to the source barcode scanning to the drug sold, and drug is traced to the source code scanning
Identifying has strict requirements for the trace to the source position of code of barcode scanning equipment and drug, otherwise will be unable to complete the movement and smoothly go out
Medicine.Therefore above-mentioned drug process faces two tests: one is the practical placement position of drug answered with systemic presupposition position it is identical, cannot
There is deviation；The second is barcode position must be with systemic presupposition position consistency when putting drug by the personnel that replenish, there can be no deviations.So
And medicine-chest equipment sells powder to open environment, drug variety can not remain constant in same cargo path, and the people that replenishes
Member will load thousands of box drugs daily to make a fault.Thus how to be accurately positioned drug trace to the source code position so as into
The identification of row barcode scanning is a problem to be solved.
Summary of the invention
The present invention In view of the above shortcomings of the prior art, propose a kind of medicine-chest drug intelligent positioning management method,
System, storage medium and equipment are matched with to shipment drug using expection drug of the vision detection technology to client, are prevented
Mistake sells drug, and medicine safety is sold in raising.
The present invention is achieved by the following technical solutions:
The present invention relates to a kind of medicine-chest drug intelligent positioning management methods, comprising the following steps:
S1, establish drug identification model；
Initial coordinate is set to each cargo path in medicine-chest when not placing drug, later by different types of medicine for sale
Product and the cargo path, which correspond, places, using cargo path front end the first box drug as default drug, with cargo path where default drug
As default cargo path；For each cargo path, according to the initial coordinate mobile image capture device of the cargo path to normal place,
To be obtained using image capture device include default cargo path, cargo path number, presets drug and initial pictures that drug is traced to the source including code,
Using the initial pictures as standard picture, then depth residual error network is used to carry out including default drug kind to the standard picture
Class, the identification of default cargo path number and default drug boundary coordinate, drug are traced to the source a yard boundary coordinate, default cargo path number boundary seat
It is labeled in interior training, obtains drug identification model and default medicine information；
S2, obtain the images to be recognized of demand drug；
The demand information for obtaining drug, based on default where default medicine information mobile image capture device to demand drug
Cargo path position, image capture device acquisition are traced to the source including pre- shipment drug, drug including code, pre- shipment cargo path and cargo path number
Images to be recognized, the coordinate (x before record image capture device is mobilel, zl) and it is mobile after coordinate (x, z)；
S3, pre- shipment medicine information extraction；
Feature extraction is carried out using images to be recognized of the drug identification model to acquisition, obtains processed image and packet
It includes pre- shipment drug variety and trace to the source a yard boundary coordinate, pre- shipment cargo path number of boundary coordinate, drug exists with number boundary coordinate
Interior pre- shipment medicine information；
S4, position correction barcode scanning drug；
If pre- shipment medicine information is numbered with cargo path in default medicine information and drug variety matches simultaneously, calculate wait know
Other image center coordinate (xp, yp) with processed image on coordinate (xc, yc) deviant (Δ x, Δ y), mobile image are caught
Device is caught to correcting coordinate (x+ Δ x, z- Δ z), wherein xcFor pre- shipment drug central point abscissa, ycFor pre- shipment cargo path
Central point ordinate, Δ x=xc-xp, Δ y=yp-yc, Δ z=Δ ysin α, α are plane where drug placement plate in cargo path
With the angle of horizontal plane (drug placement plate is inclined relative to horizontal setting)；Scanning later knows drug code of tracing to the source and obtains number of tracing to the source
According to if data of tracing to the source are matched with demand medicine information, to the progress shipment of pre- shipment drug；If data of tracing to the source and demand medicine information
Mismatch then recycles drug, then trace to the source of data and demand medicine information to next box drug barcode scanning in same cargo path
Match；
If cargo path number can not be matched with drug variety at least one, it is defeated as abnormal image to store images to be recognized
Enter drug identification model and carries out model enhancing improvement, and image capture device returns to a upper coordinate (xl, zl), skip changing coordinates
(x, z) successively obtains the i.e. linear spy of drug image corresponding to changing coordinates (x, z) within the scope of cargo path or so adjacent cargo path
Rope carries out medicine information matching, and drug boundary coordinate matrix and drug are updated if it can find demand drug；If can not find needs
Drug is sought, then carries out alert process, medicine is sold in termination.
The drug variety is classified according to drug brand, Drug generic name, the drug trade name on Key works Drug packing.
The number boundary coordinate that cargo path is preset in the pre- shipment medicine information is expressed as (xmin1, xmax1, ymin1, ymax1),
The boundary coordinate of the pre- shipment drug is expressed as (xmin2, xmax2, ymin2, ymax2), the drug is traced to the source the boundary coordinate table of code
It is shown as (xmin3, xmax3, ymin3, ymax3)；The boundary coordinate is code void that default cargo path number, pre- shipment drug, drug are traced to the source
Intend the coordinate on four vertex of rectangular wire frame after rectangular wire frame, the rectangular wire frame be include default cargo path number, pre- shipment
Drug or drug are traced to the source the minimum rectangle of code.
The central point abscissa x of the pre- shipment drugc=xmin2+(xmax2-xmin2)/2, the default cargo path number
Central point ordinate yc=ymin1+(ymax1-ymin1)/2。
The depth residual error network includes several residual blocks, several maximum pond layers and a full articulamentum；
The residual block includes an identical mapping and a convolutional layer, and the identical mapping of any one residual block is by institute
The input terminal for stating any one residual block is directed toward the output end of any one residual block；The convolutional layer extracts figure to be processed
As characteristic value obtains node diagnostic image；
Any one maximum pond layer is connect with two adjacent residual blocks, any one described maximum pond layer is to node spy
Sign image carries out feature information extraction, and reduction image size makes drug identification model be easier to restrain simultaneously；
High-resolution node diagnostic image and low resolution of the full articulamentum to convolutional layer generation in different residual blocks
Characteristic image after node diagnostic image mosaic is handled, and extraction obtains the full connection layer parameter about medicine information；
The drug identification model carries out medicine information classification according to full connection layer parameter, obtains the classification knot of medicine information
Fruit.
The convolutional layer uses LeakyRelu activation primitive, and is joined before activation using batch standardization feature
Number, characteristic parameter is normalized, and gradient is avoided to disappear and gradient explosion issues.
Because the drug in medicine-chest is in enclosed environment, light is only derived from the headlamp of image capturing device, therefore is carrying out
It needs to carry out enhancing processing to image dead color system when described trained, it is right on the basis of setting loss function and evaluation index later
The different characteristic information of drug assigns different weights and carries out changing including propagated forward, backpropagation, cross validation again
Generation, which calculates, to be realized.
The parameter of the loss function is whether to have the combination of drug, three drug variety, drug boundary coordinate variables.
The evaluation index is the combination of accurate rate and recall rate.
The LeakyRelu activation primitive that the convolutional layer uses is avoided by setting lower weight to negative sample in model
Cause certain nodes permanently dead due to negative sample in training process.
The present invention relates to a kind of medicine-chest drug intelligent positioning management systems, comprising: database module, image capture device,
Image analysis module, coordinate modification module and control module, wherein database module is connected with coordinate modification module and exports pre-
If demand medicine information, image capture device is connected with image analysis module and exports including default cargo path, cargo path number, pre-
Shipment drug and drug are traced to the source the images to be recognized including code, and image analysis module is connected with database module and exports identified
The medicine information of drug, image analysis module are connected with coordinate modification module and export pre- shipment medicine information, coordinate modification mould
Block is connected with control module and locks demand drug position to control according to pre- shipment medicine information and preset demand medicine information
Molding block exports barcode scanning coordinate information, and control module is connected with image capture device and exports barcode scanning position control information.
A kind of storage medium is stored at least one instruction in the storage medium, and at least one instruction is by handling
Device is loaded and is executed to realize the medicine-chest drug intelligent positioning management method.
A kind of equipment for medicine-chest drug intelligent positioning management method, the equipment include processor and memory, institute
It states and is stored at least one instruction in memory, at least one instruction is loaded as the processor and executed described in realization
Medicine-chest drug intelligent positioning management method.
Technical effect
Compared with prior art, the present invention is carried out using expection drug of the image detecting technique to client with to shipment drug
Matching, prevents mistake from selling drug, and medicine safety is sold in raising；Simultaneously can medicine information in real-time update medicine-chest, improve medicine-chest Chinese medicine
The product efficiency of management reduces management cost.
Detailed description of the invention
Fig. 1 is Chinese medicine cabinet drug intelligent positioning management method flow chart of the present invention；
Fig. 2 is images to be recognized schematic diagram in the present invention；
Fig. 3 be preset cargo path number boundary coordinate in the present invention, pre- shipment drug boundary coordinate and drug are traced to the source code seat table
Schematic diagram；
Fig. 4 is the schematic diagram that barcode scanning center point coordinate is corrected in the present invention；
Fig. 5 is system topology schematic diagram in the present invention；
In figure: database module 1, image capture device 2, image analysis module 3, coordinate modification module 4, control module 5.
Specific embodiment
With reference to the accompanying drawing and specific embodiment the present invention will be described in detail.
Embodiment 1
As shown in Figure 1, the present embodiment is related to a kind of medicine-chest drug intelligent positioning management method, comprising the following steps:
S1, establish drug identification model；
Initial coordinate is set to each cargo path in medicine-chest when not placing drug, later by different types of medicine for sale
Product and the cargo path, which correspond, places, using cargo path front end the first box drug as default drug, with cargo path where default drug
As default cargo path；For each cargo path, according to the initial coordinate mobile image capture device of the cargo path to normal place,
To be obtained using image capture device include default cargo path, cargo path number, presets drug and initial pictures that drug is traced to the source including code,
Using the initial pictures as standard picture, then depth residual error network is used to carry out including default drug kind to the standard picture
Class, the identification of default cargo path number and default drug boundary coordinate, drug are traced to the source a yard boundary coordinate, default cargo path number boundary seat
It is labeled in interior training, obtains drug identification model and default medicine information；
S2, obtain the images to be recognized of demand drug；
The demand information for obtaining drug, based on default where default medicine information mobile image capture device to demand drug
Cargo path position, image capture device acquisition are traced to the source including pre- shipment drug, drug including code, pre- shipment cargo path and cargo path number
Images to be recognized, as shown in Figure 2；Coordinate (x before recording image capture device movementl, yl) and it is mobile after coordinate (x, y)；
S3, pre- shipment medicine information extraction；
Feature extraction is carried out using images to be recognized of the drug identification model to acquisition, obtains processed image and packet
It includes pre- shipment drug variety and trace to the source a yard boundary coordinate, pre- shipment cargo path number of boundary coordinate, drug exists with number boundary coordinate
Interior pre- shipment medicine information；
S4, position correction barcode scanning drug；
If pre- shipment medicine information is numbered with cargo path in default medicine information and drug variety matches simultaneously, calculate wait know
Other image center coordinate (xp, yp) with processed image on coordinate (xc, yc) deviant (Δ x, Δ y), mobile image are caught
Device is caught to correcting coordinate (x+ Δ x, z- Δ z), wherein Δ z=Δ ysin α, α are plane where drug placement plate in cargo path
With the angle of horizontal plane, xcFor pre- shipment drug central point abscissa, ycFor the central point ordinate of pre- shipment cargo path, Δ x=
xc-xp, Δ y=yp-yc, as shown in Figure 4；Scanning later knows drug code of tracing to the source and obtains data of tracing to the source, if tracing to the source data and demand medicine
Product information matches carry out shipment to pre- shipment drug；Drug is recycled if trace to the source data and demand medicine information mismatch, then right
Next box drug barcode scanning in same cargo path trace to the source the matchings of data and demand medicine information；
If cargo path number can not be matched with drug variety at least one, it is defeated as abnormal image to store images to be recognized
Enter drug identification model and carries out model enhancing improvement, and image capture device returns to a upper coordinate (xl, zl), skip changing coordinates
(x, z) successively obtains the i.e. linear spy of drug image corresponding to changing coordinates (x, z) within the scope of cargo path or so adjacent cargo path
Rope carries out medicine information matching, and drug boundary coordinate matrix and drug are updated if it can find demand drug；If can not find needs
Drug is sought, then carries out alert process, medicine is sold in termination.
As shown in figure 3, the number boundary coordinate of the default cargo path is expressed as (x11, x21, y11, y21), the pre- shipment medicine
Product boundary coordinate is expressed as (x12, x22, y12, y22), a drug yard boundary coordinate of tracing to the source is expressed as (x13, x23, y13, y23), side
Boundary's coordinate is respective image；
The central point abscissa x of the pre- shipment drugc=x12+(x22-x12)/2, the center of the default cargo path number
Point ordinate yc=y11+(y21-y11)/2。
The depth residual error network includes several residual blocks, several maximum pond layers and a full articulamentum；
The residual block includes an identical mapping and a convolutional layer, and the identical mapping of any one residual block is by institute
The input terminal for stating any one residual block is directed toward the output end of any one residual block；The convolutional layer extracts figure to be processed
As characteristic value obtains node diagnostic image；
Any one maximum pond layer is connect with two adjacent residual blocks, any one described maximum pond layer is to node spy
Sign image carries out feature information extraction, and reduction image size makes drug identification model be easier to restrain simultaneously；
High-resolution node diagnostic image and low resolution of the full articulamentum to convolutional layer generation in different residual blocks
Characteristic image after node diagnostic image mosaic is handled, and extraction obtains the full connection layer parameter about medicine information；
The drug identification model carries out medicine information classification according to full connection layer parameter, obtains the classification knot of medicine information
Fruit.
The convolutional layer uses LeakyRelu activation primitive, by setting lower weight to negative sample, avoids instructing in model
Cause certain nodes permanently dead due to negative sample during white silk；And it is special using batch standardization before activation
Parameter is levied, characteristic parameter is normalized, gradient is avoided to disappear and gradient explosion issues.
The training is by carrying out enhancing processing to image dead color system, later in the base of setting loss function and evaluation index
Assigning different weights to the different characteristic information of drug on plinth and carrying out again includes that propagated forward, backpropagation, cross validation exist
Interior iterative calculation is realized.
The parameter of the loss function is whether to have the combination of drug, three drug variety, drug boundary coordinate variables.
The evaluation index is the combination of accurate rate and recall rate.
As shown in figure 5, the present embodiment is related to a kind of medicine-chest drug intelligent positioning management system, comprising: database module 1,
Image capture device 2, image analysis module 3, coordinate modification module 4 and control module 5, wherein database module 1 is repaired with coordinate
Positive module 4 is connected and exports preset demand medicine information, and image capture device 2 is connected with image analysis module 3 and exports packet
It includes default cargo path, the images to be recognized that cargo path number, pre- shipment drug and drug are traced to the source including code, image analysis module 3 and sits
Mark correction module 4 is connected and exports pre- shipment medicine information, and image analysis module 3 is connected with database module 1 and exports through knowing
The medicine information of other drug, coordinate modification module 4 are connected with control module 5 and according to pre- shipment medicine informations and preset demand
Medicine information locks demand drug position and exports barcode scanning coordinate information, control module 5 and image capture device 2 to control module 5
It is connected and output position controls information.
It is emphasized that: the above is only presently preferred embodiments of the present invention, not make in any form to the present invention
Limitation, any simple modification, equivalent change and modification to the above embodiments according to the technical essence of the invention,
All of which are still within the scope of the technical scheme of the invention.
Claims (10)
1. a kind of medicine-chest drug intelligent positioning management method, which comprises the following steps:
S1, establish drug identification model；
When not placing drug in medicine-chest each cargo path set initial coordinate, later will different types of drug for sale and
The cargo path, which corresponds, places, using cargo path front end the first box drug as default drug, using preset drug place cargo path as
Default cargo path；It is used for each cargo path according to the initial coordinate mobile image capture device of the cargo path to normal place
It includes default cargo path, cargo path number that image capture device, which obtains, presets drug and initial pictures that drug is traced to the source including code, by institute
Initial pictures are stated as standard picture, then using depth residual error network to the standard picture carry out include preset drug variety,
Default cargo path number identification and default drug boundary coordinate, drug are traced to the source a yard boundary coordinate, default cargo path number boundary coordinate
It is labeled in interior training, obtains drug identification model and default medicine information；
S2, obtain the images to be recognized of demand drug；
The demand information for obtaining drug, based on cargo path default where default medicine information mobile image capture device to demand drug
Position, image capture device obtain include pre- shipment drug, drug trace to the source code, pre- shipment cargo path and cargo path number including wait know
Other image, the coordinate (x before record image capture device is mobilel, zl) and it is mobile after coordinate (x, z)；
S3, pre- shipment medicine information extraction；
Feature extraction is carried out using images to be recognized of the drug identification model to acquisition, obtains processed image and including pre-
Shipment drug variety and boundary coordinate, drug are traced to the source including a yard boundary coordinate, pre- shipment cargo path number and number boundary coordinate
Pre- shipment medicine information；
S4, position correction barcode scanning drug；
If pre- shipment medicine information is numbered with cargo path in default medicine information and drug variety matches simultaneously, figure to be identified is calculated
Inconocenter point coordinate (xp, yp) with processed image on coordinate (xc, yc) deviant (Δ x, Δ y), mobile image capture dress
It sets to amendment coordinate (x+ Δ x, z- Δ z), wherein xcFor pre- shipment drug central point abscissa, ycFor the center of pre- shipment cargo path
Point ordinate, Δ x=xc-xp, Δ y=yp-yc, Δ z=Δ ysin α, α are plane and water where drug placement plate in cargo path
The angle of plane；Scanning later knows drug code of tracing to the source and obtains data of tracing to the source, if data of tracing to the source are matched with demand medicine information, to pre-
Shipment drug carries out shipment；Recycle drug if trace to the source data and demand medicine information mismatch, then under in same cargo path
One box drug barcode scanning trace to the source the matchings of data and demand medicine information；
If cargo path number can not be matched with drug variety at least one, images to be recognized is stored as abnormal image and inputs medicine
Product identification model carries out model enhancing and improves, and image capture device returns to a upper coordinate (xl, zl), skip changing coordinates (x,
Z), acquisition drug image progress medicine information matching within the scope of cargo path or so adjacent cargo path is being corresponded to changing coordinates (x, z),
Drug boundary coordinate matrix and drug are updated if it can find demand drug；If can not find demand drug, carry out at alarm
Medicine is sold in reason, termination.
2. medicine-chest drug intelligent positioning management method according to claim 1, characterized in that in the pre- shipment medicine information
The number boundary coordinate of default cargo path is expressed as (xmin1, xmax1, ymin1, ymax1), the boundary coordinate of the pre- shipment drug indicates
For (xmin2, xmax2, ymin2, ymax2), the trace to the source boundary coordinate of code of the drug is expressed as (xmin3, xmax3, ymin3, ymax3)；
The central point abscissa x of the pre- shipment drugc=xmin2+(xmax2-xmin2)/2, the central point of the default cargo path number
Ordinate yc=ymin1+(ymax1-ymin1)/2。
3. medicine-chest drug intelligent positioning management method according to claim 1, characterized in that the depth residual error network includes
Several residual blocks, several maximum pond layers and a full articulamentum；
The residual block includes an identical mapping and a convolutional layer, and the identical mapping of any one residual block is by described
The input terminal an of residual block of anticipating is directed toward the output end of any one residual block；It is special that the convolutional layer extracts image to be processed
Value indicative obtains node diagnostic image；
Any one maximum pond layer is connect with two adjacent residual blocks, any one described maximum pond layer is to node diagnostic figure
As carrying out feature information extraction reduction image size simultaneously；
The high-resolution node diagnostic image and low resolution node that the full articulamentum generates convolutional layer in different residual blocks
The spliced characteristic image of characteristic image is handled, and extraction obtains the full connection layer parameter about medicine information；
The drug identification model carries out medicine information classification according to full connection layer parameter, obtains the classification results of medicine information.
4. medicine-chest drug intelligent positioning management method according to claim 3, characterized in that the convolutional layer uses
LeakyRelu activation primitive, and batch standardization characteristic parameter is used before activation, characteristic parameter is normalized.
5. medicine-chest drug intelligent positioning management method according to claim 1, characterized in that the training passes through dark to image
Colour system carries out enhancing processing, assigns on the basis of setting loss function and evaluation index to the different characteristic information of drug later
Different weights carries out the realization of the iterative calculation including propagated forward, backpropagation, cross validation again.
6. medicine-chest drug intelligent positioning management method according to claim 5, characterized in that the parameter of the loss function is
Whether the combination of drug, drug variety, drug boundary coordinate three variable is had.
7. medicine-chest drug intelligent positioning management method according to claim 5, characterized in that the evaluation index is accurate rate
With the combination of recall rate.
8. the system based on any one of the claim 1~7 medicine-chest drug intelligent positioning management method, which is characterized in that packet
It includes: database module, image capture device, image analysis module, coordinate modification module and control module, wherein database mould
Block is connected with coordinate modification module and exports preset demand medicine information, and image capture device is connected simultaneously with image analysis module
Output includes the images to be recognized that default cargo path, cargo path are numbered, pre- shipment drug and drug are traced to the source including code, image analysis module
It is connected and exports the medicine information of identified drug with database module, image analysis module is connected and defeated with coordinate modification module
Pre- shipment medicine information out, coordinate modification module are connected with control module and according to pre- shipment medicine informations and preset demand medicine
Product information locking demand drug position exports barcode scanning coordinate information to control module, and control module is connected simultaneously with image capture device
Export barcode scanning position control information.
9. a kind of storage medium, which is characterized in that be stored at least one instruction in the storage medium, described at least one refers to
Order is loaded by processor and is executed to realize such as medicine-chest drug intelligent positioning manager according to any one of claims 1 to 7
Method.
10. a kind of equipment for medicine-chest drug intelligent positioning management method, which is characterized in that the equipment include processor and
Memory, at least one instruction is stored in the memory, and at least one instruction is loaded and executed by the processor
To realize such as medicine-chest drug intelligent positioning management method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811213445.2A CN109522932B (en) | 2018-10-18 | 2018-10-18 | Intelligent positioning management method, system, storage medium and equipment for medicines in medicine cabinet |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811213445.2A CN109522932B (en) | 2018-10-18 | 2018-10-18 | Intelligent positioning management method, system, storage medium and equipment for medicines in medicine cabinet |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109522932A true CN109522932A (en) | 2019-03-26 |
| CN109522932B CN109522932B (en) | 2021-06-11 |
Family
ID=65771803
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811213445.2A Expired - Fee Related CN109522932B (en) | 2018-10-18 | 2018-10-18 | Intelligent positioning management method, system, storage medium and equipment for medicines in medicine cabinet |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109522932B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111192411A (en) * | 2019-12-30 | 2020-05-22 | 广州享药户联优选科技有限公司 | Method and device for realizing medicine sharing based on intelligent medicine chest |
| CN112542001A (en) * | 2019-09-20 | 2021-03-23 | 威海新北洋数码科技有限公司 | Medicine dispensing machine |
| CN113593128A (en) * | 2021-07-09 | 2021-11-02 | 广州云印信息科技有限公司 | Medicine selling management method, equipment and storage medium of automatic medicine selling machine |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5221024A (en) * | 1992-12-18 | 1993-06-22 | Campbell Gordon M | Programmable medicine dispenser with manual override and color coded medicine canisters |
| CN2585332Y (en) * | 2002-10-25 | 2003-11-05 | 成都三九数码医药设备有限公司 | Machine for automatically selling medicine packed in box |
| US7751932B1 (en) * | 1993-01-25 | 2010-07-06 | Automed Technologies, Inc. | Method for tracking and dispensing medical items |
| CN105094133A (en) * | 2015-08-24 | 2015-11-25 | 华南理工大学 | Health service robot based on multifunctional intelligent electronic medicine kit |
| CN107085722A (en) * | 2015-09-25 | 2017-08-22 | 阿里巴巴集团控股有限公司 | The method and device of specify information is positioned in the picture |
| CN207068142U (en) * | 2017-04-28 | 2018-03-02 | 中南大学湘雅三医院 | A kind of electronic medicine frame and its system |
-
2018
- 2018-10-18 CN CN201811213445.2A patent/CN109522932B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5221024A (en) * | 1992-12-18 | 1993-06-22 | Campbell Gordon M | Programmable medicine dispenser with manual override and color coded medicine canisters |
| US7751932B1 (en) * | 1993-01-25 | 2010-07-06 | Automed Technologies, Inc. | Method for tracking and dispensing medical items |
| CN2585332Y (en) * | 2002-10-25 | 2003-11-05 | 成都三九数码医药设备有限公司 | Machine for automatically selling medicine packed in box |
| CN105094133A (en) * | 2015-08-24 | 2015-11-25 | 华南理工大学 | Health service robot based on multifunctional intelligent electronic medicine kit |
| CN107085722A (en) * | 2015-09-25 | 2017-08-22 | 阿里巴巴集团控股有限公司 | The method and device of specify information is positioned in the picture |
| CN207068142U (en) * | 2017-04-28 | 2018-03-02 | 中南大学湘雅三医院 | A kind of electronic medicine frame and its system |
Non-Patent Citations (1)
| Title |
|---|
| 刘书恋: "医院自动化药房控制系统的设计", 《产业与科技论坛》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112542001A (en) * | 2019-09-20 | 2021-03-23 | 威海新北洋数码科技有限公司 | Medicine dispensing machine |
| CN112542001B (en) * | 2019-09-20 | 2023-04-07 | 威海新北洋数码科技有限公司 | Medicine dispensing machine |
| CN111192411A (en) * | 2019-12-30 | 2020-05-22 | 广州享药户联优选科技有限公司 | Method and device for realizing medicine sharing based on intelligent medicine chest |
| CN113593128A (en) * | 2021-07-09 | 2021-11-02 | 广州云印信息科技有限公司 | Medicine selling management method, equipment and storage medium of automatic medicine selling machine |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109522932B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109472297A (en) | Medicine-chest drug intelligent positioning management method and system | |
| CN109522932A (en) | Medicine-chest drug intelligent positioning management method, system, storage medium and equipment | |
| US11908125B2 (en) | Surgical kit inspection systems and methods for inspecting surgical kits having parts of different types | |
| US20110192893A1 (en) | Method and Apparatus for Handling Packages in an Automated Dispensary | |
| CN106573381A (en) | Truck unloader visualization | |
| CN113658325B (en) | Intelligent identification and early warning method for uncertain objects of production line in digital twin environment | |
| CN110245663A (en) | One kind knowing method for distinguishing for coil of strip information | |
| CN111753712A (en) | Method, system and equipment for monitoring safety of power production personnel | |
| CN106341682A (en) | Aging function detection method, system and device | |
| CN109086694A (en) | A kind of face identification system and method | |
| US7046377B2 (en) | Method for determining corresponding points in three-dimensional measurement | |
| CN107333123A (en) | Detecting system of focusing and focusing detection method | |
| CN109693140A (en) | A kind of intelligent flexible production line and its working method | |
| CN115115768A (en) | Object coordinate recognition system, method, device and medium based on stereoscopic vision | |
| CN110440764A (en) | A kind of secondary alignment methods of meter detection holder, device and equipment | |
| CN112214513A (en) | Intelligent manufacturing material management system | |
| CN117132653A (en) | Target-based machine vision displacement measurement method, system and equipment | |
| CN107462301A (en) | Liquid level monitoring method | |
| KR102520802B1 (en) | 3D bio-printer monitoring system and repairing system and method for correction of shape defects | |
| CN116256366A (en) | Chip defect detection method, detection system and storage medium | |
| CN114097003A (en) | Method and computer program product for measuring and verifying pharmaceutical portions | |
| CN112184673A (en) | Tablet target detection method for medication compliance management | |
| CN109975299A (en) | Light emitting source detection system and method | |
| CN111160330A (en) | Training method for improving image recognition accuracy with assistance of electronic tag recognition | |
| EP4019894B1 (en) | Method of measuring a workpiece by using a coordinate measuring machine, and arrangement comprising a coordinate measuring machine |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210611 Termination date: 20211018 |