CN110322206A - A kind of reagent information input method and device based on OCR identification - Google Patents
A kind of reagent information input method and device based on OCR identification Download PDFInfo
- Publication number
- CN110322206A CN110322206A CN201910680984.5A CN201910680984A CN110322206A CN 110322206 A CN110322206 A CN 110322206A CN 201910680984 A CN201910680984 A CN 201910680984A CN 110322206 A CN110322206 A CN 110322206A
- Authority
- CN
- China
- Prior art keywords
- reagent
- image
- label
- text
- storage
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/23—Updating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
- G06Q10/087—Inventory or stock management, e.g. order filling, procurement or balancing against orders
- G06Q10/0875—Itemisation or classification of parts, supplies or services, e.g. bill of materials
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Economics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Human Resources & Organizations (AREA)
- Finance (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Accounting & Taxation (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of reagent information input method and device based on OCR identification, whether the reagent label for detecting reagent to be put in storage first is directed at image acquisition units, obtains testing result；If then the testing result is that reagent label has been directed at image acquisition units, the image of corresponding reagent label is acquired by described image acquisition unit；OCR identification is carried out to the image of corresponding reagent label collected, obtains the content of text of corresponding reagent label；Classification prediction further is carried out to the content of text of obtained corresponding reagent label, obtains classification prediction result；The finally reagent information of the reagent to be put in storage according to the classification prediction result typing.In this way, the present invention is by way of OCR identification plus artificial intelligence, the manual operations before eliminating during reagent storage, whole process is all automatically performed by system, the operating time is greatly shortened, to largely improve warehouse-in efficiency.
Description
Technical field
The present invention relates to artificial neural network field more particularly to a kind of reagent information input methods based on OCR identification
And device.
Background technique
Stock control is the work that Most current unit all suffers from, and arrives all kinds of manufacturing enterprises greatly, sells manufacturer, small
To great quantity of small foundation unit, can all there be the demand of stock control, such as the raw materials for production of various specifications type, various office consumptions
The stock control etc. of material.However, the management of chemical reagent used in Scientific Research in University Laboratory or various research units, then be inventory's pipe
A more special type in reason.To find out its cause, essentially consisting in the generally existing certain risk of chemical reagent, or even have
It is much easy system poison, severe toxicity or easily makes quick-fried etc., therefore has higher requirement to the fineness of Reagent management.
In actual Reagent management, in order to realize the fine-grained management of reagent, can such as track reagent from buying to
Using to the whole life cycle informations scrapped, such as whose buying, why project purchasing, what who was used, the amount used,
The whereabouts etc. of remaining reagent, it is necessary to which stock control can accurately record the details of reagent.Therefore, reagent is entering library
When deposit system (subsequent to be known as in-stockroom operation), it is necessary to the complete information of detailed typing reagent, such as reagent name, purity,
Current amount, supplier etc..
In the related technology, reagent information storage mainly includes following several implementations: 1) being manually entered, i.e., according to arrival
Reagent information, manually check with by way of typing, by reagent information input system；2) procurement information is switched into inventory's letter
Breath, i.e., user is when purchasing reagent, it is necessary to provide the information of a part of reagent, such as title, required specification, purity
Deng can be using this partial information in purchasing system as storage information；3) supplier is required to provide goods information, i.e. user exists
When purchasing reagent, supplier can be required to provide corresponding merchandise news, storage when can import these information
System.
However, above-mentioned several reagent information storage implementations have the following deficiencies: 1) mode being manually entered is imitated
A possibility that rate is relatively low, and by hand in Input Process, error is bigger；2) there is the reagent of buying arrival and submit and purchase
The reagent information of typing is not consistent when request, such as packing specification, such as supplier etc., results in the need for after remodifying information
The problem of storage；3) supplier can may not necessarily provide the file of electronic edition, and the information provided generally can not be also introduced directly into
Inventory system could import after needing to handle data.
Summary of the invention
The embodiment of the present invention is put in storage various present in implementation ask to solve the above-mentioned available reagent information referred to
Topic, creative provides a kind of reagent information input method and device based on OCR identification.
According to embodiments of the present invention in a first aspect, providing a kind of reagent information input method based on OCR identification, this method
Include: to detect the reagent label of reagent to be put in storage whether to be directed at image acquisition units, obtains testing result；If the testing result
Image acquisition units have been directed at for reagent label, then have acquired the image of corresponding reagent label by described image acquisition unit；It is right
The image of corresponding reagent label collected carries out OCR identification, obtains the content of text of corresponding reagent label；To obtained right
It answers the content of text of reagent label to carry out classification prediction, obtains classification prediction result；According to classification prediction result typing institute
State the reagent information of reagent to be put in storage.
According to an embodiment of the present invention, before the image to corresponding reagent label collected carries out OCR identification,
The method also includes: image preprocessing is carried out to the image of corresponding reagent label collected；Wherein, described image pre-processes
Including at least one following processing operation: adjustment image resolution ratio or the direction of rotation for adjusting image.
According to an embodiment of the present invention, whether the reagent label for detecting reagent to be put in storage is directed at Image Acquisition list
Member, comprising: whether the key information area for detecting the reagent label of reagent to be put in storage is directed at the central data of image acquisition units
Region.
According to an embodiment of the present invention, the image to corresponding reagent label collected carries out OCR identification, obtains
The content of text of corresponding reagent label, comprising: the image of corresponding reagent label collected is subjected to OCR knowledge as a whole
Not, the OCR recognition result including at least a text field is obtained；By the obtained OCR for including at least a text field
Recognition result is integrally as the content of text for corresponding to reagent label.
According to an embodiment of the present invention, classification prediction is carried out to the content of text of obtained corresponding reagent label, obtained
To classification prediction result, comprising: carry out word segmentation processing to the content of text of obtained corresponding reagent label, obtain word segmentation processing
As a result；Attribute class prediction is carried out to each participle in the word segmentation processing result using classification prediction model, to be divided
Class prediction result, the classification prediction result include that each segments corresponding attribute classification.
According to an embodiment of the present invention, the method also includes: each point is determined according to the classification prediction result
Word corresponds to the other probability value of Attribute class；Highest attribute classification in the corresponding other probability value of Attribute class of all participles is determined as most
Classification prediction result eventually.
According to an embodiment of the present invention, the reagent letter of the reagent to be put in storage according to the classification prediction result typing
Breath, comprising: replace corresponding participle using the highest other full name of Attribute class in the corresponding other probability value of Attribute class of all participles
Mode carry out the reagent information of reagent to be put in storage described in typing.
According to an embodiment of the present invention, the reagent information to be put in storage according to the classification prediction result typing it
Afterwards, the method also includes: the code shape label for identifying reagent to be put in storage is generated according to the reagent information of institute's typing.
Second aspect according to embodiments of the present invention also provides a kind of reagent information input device based on OCR identification, the dress
Setting includes: detection unit, and whether the reagent label for detecting reagent to be put in storage is directed at image acquisition units, obtains detection knot
Fruit；Described image acquisition unit acquires the figure of corresponding reagent label if being that reagent label has been aligned for the testing result
Picture；OCR recognition unit carries out OCR identification for the image to corresponding reagent label collected, obtains corresponding reagent label
Content of text；Classification predicting unit, carries out classification prediction for the content of text to obtained corresponding reagent label, is divided
Class prediction result；Data input unit, the reagent information for the reagent to be put in storage according to the classification prediction result typing.
According to an embodiment of the present invention, described device further includes image pre-processing unit, for identifying list by OCR
Before member carries out OCR identification to the image of corresponding reagent label collected, to the image of corresponding reagent label collected into
Row image preprocessing；Wherein, described image pretreatment includes at least one following processing operation: adjustment image resolution ratio or adjustment
The direction of rotation of image.
According to an embodiment of the present invention, the detection unit is specifically used for, and detects the reagent label of reagent to be put in storage
Whether key information area is directed at the central data region of image acquisition units.
According to an embodiment of the present invention, the OCR recognition unit is specifically used for, by corresponding reagent label collected
Image carries out OCR identification as a whole, obtains the OCR recognition result including at least a text field；It will be obtained
Including at least the whole content of text as corresponding reagent label of OCR recognition result an of the text field.
According to an embodiment of the present invention, the classification predicting unit is specifically used for, to obtained corresponding reagent label
Content of text carry out word segmentation processing, obtain word segmentation processing result；Using classification prediction model in the word segmentation processing result
Each participle carries out attribute class prediction, and to obtain classification prediction result, the classification prediction result includes each participle
Corresponding attribute classification.
According to an embodiment of the present invention, the classification predicting unit is also used to, and is determined according to the classification prediction result
Each segments the corresponding other probability value of Attribute class；By highest attribute classification in the corresponding other probability value of Attribute class of all participles
It is determined as final classification prediction result.
According to an embodiment of the present invention, the data input unit is specifically used for, and utilizes the corresponding Attribute class of all participles
The mode that the highest other full name of Attribute class replaces corresponding participle in other probability value carrys out the examination of reagent to be put in storage described in typing
Agent information.
According to an embodiment of the present invention, described device further includes generation unit, for the reagent information according to institute's typing
Generate the code shape label for identifying reagent to be put in storage.
The embodiment of the present invention detects reagent to be put in storage based on the OCR reagent information input method identified and device first
Whether reagent label is directed at image acquisition units, obtains testing result；If then the testing result is that reagent label has been aligned
Image acquisition units then acquire the image of corresponding reagent label by described image acquisition unit；To corresponding reagent collected
The image of label carries out OCR identification, obtains the content of text of corresponding reagent label；Further to obtained corresponding reagent label
Content of text carry out classification prediction, obtain classification prediction result；Finally wait enter according to the classification prediction result typing
The reagent information of library reagent.In this way, the present invention is by way of OCR identification plus artificial intelligence, reagent was put in storage before eliminating
Manual operations in journey, whole process are all automatically performed by system, greatly shorten the operating time, to largely mention
High warehouse-in efficiency.In this way, due to the raising of warehouse-in efficiency, personnel needed for Reagent management are reduced, and the examination automated
Agent is put in storage process, and by the application program Auto-writing of mobile terminal, operating process is very simple, and the chemical knowledge without profession is
It is achievable, the requirement of depositary management personnel is reduced, therefore the cost of chemical reagent management can be reduced.
It is to be appreciated that the teachings of the present invention does not need to realize whole beneficial effects recited above, but it is specific
Technical solution may be implemented specific technical effect, and other embodiments of the invention can also be realized and not mentioned above
Beneficial effect.
Detailed description of the invention
The following detailed description is read with reference to the accompanying drawings, above-mentioned and other mesh of exemplary embodiment of the invention
, feature and advantage will become prone to understand.In the accompanying drawings, if showing by way of example rather than limitation of the invention
Dry embodiment, in which:
In the accompanying drawings, identical or corresponding label indicates identical or corresponding part.
Fig. 1 shows implementation process schematic diagram of the embodiment of the present invention based on the OCR reagent information input method identified；
Concrete operations process signal Fig. 2 shows an application example of the invention based on the OCR reagent information typing identified
Figure；
Fig. 3 shows composed structure schematic diagram of the embodiment of the present invention based on the OCR reagent information input device identified.
Specific embodiment
The principle and spirit of the invention are described below with reference to several illustrative embodiments.It should be appreciated that providing this
A little embodiments are only to be to make those skilled in the art can better understand that realize the present invention in turn, and not with any side
Formula limits the scope of the invention.On the contrary, it is of the invention more thorough and complete to make for providing these embodiments, and it can incite somebody to action this
The range of invention is completely communicated to those skilled in the art.
The technical solution of the present invention is further elaborated in the following with reference to the drawings and specific embodiments.
Fig. 1 shows implementation process schematic diagram of the embodiment of the present invention based on the OCR reagent information input method identified；Figure
2 show concrete operations flow diagram of the application example based on the OCR reagent information typing identified of the invention.
With reference to Fig. 1, the embodiment of the present invention includes: operation 101 based on the reagent information input method that OCR is identified, detection to
Whether the reagent label of storage reagent is directed at image acquisition units, obtains testing result；Operation 102, if the testing result is
Reagent label has been directed at image acquisition units, then the image of corresponding reagent label is acquired by described image acquisition unit；Operation
103, OCR identification is carried out to the image of corresponding reagent label collected, obtains the content of text of corresponding reagent label；Operation
104, classification prediction is carried out to the content of text of obtained corresponding reagent label, obtains classification prediction result；Operation 105, root
According to the reagent information of reagent to be put in storage described in the classification prediction result typing.
With reference to Fig. 2, in operation 101~102, in image acquisition units (such as camera mould that front-end interface passes through mobile device
Block) the reagent label identified bat will be needed to store at image, and in a device.Since most of reagent bottle is circular, reagent
Label, that is, label paper is attached on bottle, can be presented in a manner of arc, it is thus possible under whole reagent labels can not all being shot
Come, or even if being filmed, the text of two side portions also will appear more serious deformation, be unfavorable for Text region.Cause
This, specifically when Image Acquisition, first detects the reagent label alignment image acquisition units of reagent to be put in storage, passes through the figure later
As acquisition unit acquires the image of corresponding reagent label.
According to an embodiment of the present invention, in operation 101, it can detecte the crucial letter of the reagent label of reagent to be put in storage
Whether breath region is directed at the central data region of image acquisition units.I.e. when Image Acquisition, key information area is made
For the central data region of shooting, key message is clearly presented in photo without deformation.Here key message is main
For " title ", " No. CAS ", " specification ", the information such as " purity "；Correspondingly, key information area mainly includes Name area, CAS
Number region, regular domain, purity region etc..
According to an embodiment of the present invention, before operation 103, image that can first to corresponding reagent label collected
Carry out image preprocessing；Wherein, described image pretreatment includes at least one following processing operation: adjustment image resolution ratio or tune
The direction of rotation of whole image.Specifically, subsequent to image to meet after the image storage in a mobile device of corresponding reagent label
The requirement transmitted, stored and identified, needs to carry out image certain processing, including adjustment image resolution ratio is to suitable big
Small or adjustment image direction of rotation.
In operation 103, with reference to Fig. 2, the image that previous step is handled well is committed to the function that OCR identification is carried out to image
Interface, by the algorithm of text identification, by the content recognitions that can be identified all in image at the text field, and will corresponding reagent mark
The content of text of label returns to mobile terminal.
In actual operation, operation 103 is without by region division and doing different disposal to each region for image, and only need by
Image is transmitted to text identification interface as a whole, and the full text of acquisition is passed back.Therefore, according to the present invention
The image of corresponding reagent label collected specifically can be carried out OCR knowledge by one embodiment, operation 103 as a whole
Not, the OCR recognition result including at least a text field is obtained；By the obtained OCR for including at least a text field
Recognition result is integrally as the content of text for corresponding to reagent label.
Here, the content of text identified from image by operation 103, may include Chinese, English, number,
Corresponding meaning is also different, if any the Chinese and English title of reagent, brand, the packing specification of reagent, there is the title of supplier,
There is the information etc. of production firm.
Operation of the present invention 104 does the purpose of classification prediction, the text exactly identified, by its stroke to content of text
It is divided into word, then speculates that word may corresponding real meaning.Therefore pre- firstly the need of the classification for establishing word category analysis
Model is surveyed, the mode of convolutional neural networks (CNN) is used herein as, it is pre- to establish the classification that category analysis is carried out to the word of input
Model is surveyed, so which attribute that this word belongs in reagent information can be judged after one word of input.
According to an embodiment of the present invention, in operation 104, can content of text to obtained corresponding reagent label into
Row word segmentation processing obtains word segmentation processing result；Using classification prediction model to each in the word segmentation processing result segment into
Row attribute class prediction, to obtain classification prediction result, the classification prediction result includes that each segments corresponding attribute classification.
With reference to Fig. 2, application example of the present invention passes through the technology of artificial intelligence neural networks (CNN), establishes text information point
The model of class, after the adjustment of model and training, the accuracy rate of identification reaches 90% or more, therefore will identify in picture
Text disposably judges the field of multiple attributes, to be accurately filled up in the field of respective attributes, completes storage letter
The automation of breath is filled in, and directly can connect printer, print label by mobile phone.
According to an embodiment of the present invention, operation 104 further includes determining each participle according to the classification prediction result
The corresponding other probability value of Attribute class；Highest attribute classification in the corresponding other probability value of Attribute class of all participles is determined as finally
Classification prediction result.In the process, due to needing multiple attributes of identification agent, the title including reagent, the CAS of compound
Number, specification, purity, supplier etc., it is therefore desirable to judge a possibility that each text belongs to some attribute, take wherein that possibility is most
High one is as final prediction result.
Further, the reagent information of the reagent to be put in storage according to the classification prediction result typing, comprising: utilize institute
There is the mode that the highest other full name of Attribute class replaces corresponding participle in the corresponding other probability value of Attribute class of participle to come typing institute
State the reagent information of reagent to be put in storage.I.e. equipment can also do the adjustment of primary intelligence to prediction result by some judgment rules,
Such as vendor name is supplemented, after recognizing the keyword in vendor name, to being adjusted to the complete of supplier
Claim, to obtain more accurate result.
In operation 105, after the text identified in picture is carried out attributive classification by above-mentioned classification prediction model, i.e.,
Word can be automatically filled in storage information in corresponding input frame.Certainly in practical applications, user can be according to the true of reagent
Real information makes a decision and modifies to the data that system is automatically filled in, information is then done in-stockroom operation.Believe in acquired image
For breath than more visible, in the case that recognition accuracy is relatively high, it is not necessary to modify information by user, can directly be put in storage.
According to an embodiment of the present invention, after operation 105, the method also includes: believed according to the reagent of institute's typing
Breath generates the code shape label for identifying reagent to be put in storage.Specifically, after storage information is filled in completely, system can be automatically generated often
The corresponding uniqueness bar code of a reagent, i.e. code shape label, and printer can be directly connected to by mobile device, by its label
It prints, is attached on reagent bottle.
The reagent information input method that the embodiment of the present invention is identified based on OCR detects the reagent mark of reagent to be put in storage first
Whether label are directed at image acquisition units, obtain testing result；If then the testing result has been directed at image for reagent label and has adopted
Collect unit, then acquires the image of corresponding reagent label by described image acquisition unit；To corresponding reagent label collected
Image carries out OCR identification, obtains the content of text of corresponding reagent label；Further to the text of obtained corresponding reagent label
Content carries out classification prediction, obtains classification prediction result；The finally reagent to be put in storage according to the classification prediction result typing
Reagent information.In this way, the present invention OCR identification plus artificial intelligence by way of, before eliminating reagent storage during
Manual operations, whole process are all automatically performed by system, greatly shorten the operating time, to largely improve storage
Efficiency.In this way, due to the raising of warehouse-in efficiency, personnel needed for Reagent management are reduced, and the reagent storage automated
Process, by the application program Auto-writing of mobile terminal, operating process is very simple, and the chemical knowledge without profession can be complete
At reducing the requirement of depositary management personnel, therefore the cost of chemical reagent management can be reduced.
Based on the reagent information input method referred to above based on OCR identification, the embodiment of the present invention provides one kind again
Based on the reagent information input device of OCR identification, as shown in figure 3, the device 30 includes: detection unit 301, for detecting wait enter
Whether the reagent label of library reagent is directed at image acquisition units 302, obtains testing result；Described image acquisition unit 302, is used for
If the testing result is that reagent label has been aligned, the image of corresponding reagent label is acquired；OCR recognition unit 303, for pair
The image of corresponding reagent label collected carries out OCR identification, obtains the content of text of corresponding reagent label；Classification predicting unit
304, classification prediction is carried out for the content of text to obtained corresponding reagent label, obtains classification prediction result；Information record
Enter unit 305, the reagent information for the reagent to be put in storage according to the classification prediction result typing.
According to an embodiment of the present invention, described device 30 further includes image pre-processing unit, for identifying by OCR
Before unit 303 carries out OCR identification to the image of corresponding reagent label collected, to the figure of corresponding reagent label collected
As carrying out image preprocessing；Wherein, described image pretreatment includes at least one following processing operation: adjustment image resolution ratio or
Adjust the direction of rotation of image.
According to an embodiment of the present invention, the detection unit 301 is specifically used for, and detects the reagent label of reagent to be put in storage
Key information area whether be directed at the central data regions of image acquisition units.
According to an embodiment of the present invention, the OCR recognition unit 303 is specifically used for, by corresponding reagent mark collected
The image of label carries out OCR identification as a whole, obtains the OCR recognition result including at least a text field；By gained
The OCR recognition result including at least a text field arrived is integrally as the content of text for corresponding to reagent label.
According to an embodiment of the present invention, the classification predicting unit 304 is specifically used for, to obtained corresponding reagent mark
The content of text of label carries out word segmentation processing, obtains word segmentation processing result；Using classification prediction model to the word segmentation processing result
In each participle carry out attribute class prediction, with obtain classification prediction result, the classification prediction result includes each point
Word corresponds to attribute classification.
According to an embodiment of the present invention, the classification predicting unit 304 is also used to, true according to the classification prediction result
The fixed corresponding other probability value of Attribute class of each participle；By highest Attribute class in the corresponding other probability value of Attribute class of all participles
It is not determined as final classification prediction result.
According to an embodiment of the present invention, the data input unit 305 is specifically used for, and utilizes the corresponding attribute of all participles
The mode that the highest other full name of Attribute class replaces corresponding participle in the probability value of classification carrys out reagent to be put in storage described in typing
Reagent information.
According to an embodiment of the present invention, described device 30 further includes generation unit, for being believed according to the reagent of institute's typing
Breath generates the code shape label for identifying reagent to be put in storage.
It need to be noted that: it is and preceding above to the description of the reagent information input device embodiment identified based on OCR
State embodiment of the method shown in FIG. 1 description be it is similar, have with aforementioned embodiment of the method shown in FIG. 1 it is similar beneficial to effect
Fruit, therefore do not repeat them here.For the present invention to undisclosed technical detail in the reagent information input device identified based on OCR,
It please refers to the description of the aforementioned embodiment of the method shown in FIG. 1 of the present invention and understands, to save length, therefore repeat no more.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the device that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or device institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or device.
In several embodiments provided herein, it should be understood that disclosed device and method can pass through it
Its mode is realized.Apparatus embodiments described above are merely indicative, for example, the division of the unit, only
A kind of logical function partition, there may be another division manner in actual implementation, such as: multiple units or components can combine, or
It is desirably integrated into another device, or some features can be ignored or not executed.In addition, shown or discussed each composition portion
Mutual coupling or direct-coupling or communication connection is divided to can be through some interfaces, the INDIRECT COUPLING of equipment or unit
Or communication connection, it can be electrical, mechanical or other forms.
Above-mentioned unit as illustrated by the separation member, which can be or may not be, to be physically separated, aobvious as unit
The component shown can be or may not be physical unit；Both it can be located in one place, and may be distributed over multiple network lists
In member；Some or all of units can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
In addition, each functional unit in various embodiments of the present invention can be fully integrated in one processing unit, it can also
To be each unit individually as a unit, can also be integrated in one unit with two or more units；It is above-mentioned
Integrated unit both can use formal implementation of hardware, also can use hardware and the form of SFU software functional unit is added to realize.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above method embodiment can pass through
The relevant hardware of program instruction is completed, and program above-mentioned can store in calculating machine read/write memory medium, which exists
When execution, step including the steps of the foregoing method embodiments is executed；And storage medium above-mentioned includes: movable storage device, read-only deposits
The various media that can store program code such as reservoir (Read Only Memory, ROM), magnetic or disk.
If alternatively, the above-mentioned integrated unit of the present invention is realized in the form of software function module and as independent product
When selling or using, it also can store in a calculating machine read/write memory medium.Based on this understanding, the present invention is implemented
Substantially the part that contributes to existing technology can be embodied in the form of software products the technical solution of example in other words,
The calculating machine software product is stored in a storage medium, including some instructions are used so that operation machine equipment (can be with
It is personal calculating machine, server or network equipment etc.) execute all or part of each embodiment the method for the present invention.
And storage medium above-mentioned includes: various Jie that can store program code such as movable storage device, ROM, magnetic or disk
Matter.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be based on the protection scope of the described claims.
Claims (10)
1. a kind of reagent information input method based on OCR identification, which is characterized in that the described method includes:
Whether the reagent label for detecting reagent to be put in storage is directed at image acquisition units, obtains testing result；
If the testing result is that reagent label has been directed at image acquisition units, corresponded to by the acquisition of described image acquisition unit
The image of reagent label；
OCR identification is carried out to the image of corresponding reagent label collected, obtains the content of text of corresponding reagent label；
Classification prediction is carried out to the content of text of obtained corresponding reagent label, obtains classification prediction result；
According to the reagent information of reagent to be put in storage described in the classification prediction result typing.
2. the method according to claim 1, wherein the image to corresponding reagent label collected carries out
Before OCR identification, the method also includes:
Image preprocessing is carried out to the image of corresponding reagent label collected；Wherein, described image pretreatment includes following place
At least one reason operation: adjustment image resolution ratio or the direction of rotation for adjusting image.
3. method according to claim 1 or 2, which is characterized in that whether the reagent label for detecting reagent to be put in storage
It is directed at image acquisition units, comprising:
Whether the key information area for detecting the reagent label of reagent to be put in storage is directed at the central data region of image acquisition units.
4. method according to claim 1 or 2, which is characterized in that the image to corresponding reagent label collected
OCR identification is carried out, the content of text of corresponding reagent label is obtained, comprising:
The image of corresponding reagent label collected is subjected to OCR identification as a whole, obtains including at least a text
The OCR recognition result of field；
It will be in the obtained OCR recognition result for including at least a text field integrally text as corresponding reagent label
Hold.
5. method according to claim 1 or 2, which is characterized in that the content of text of obtained corresponding reagent label
Classification prediction is carried out, classification prediction result is obtained, comprising:
Word segmentation processing is carried out to the content of text of obtained corresponding reagent label, obtains word segmentation processing result；
Attribute class prediction is carried out to each participle in the word segmentation processing result using classification prediction model, to be classified
Prediction result, the classification prediction result include that each segments corresponding attribute classification.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
Determine that each segments the corresponding other probability value of Attribute class according to the classification prediction result；
Highest attribute classification in the corresponding other probability value of Attribute class of all participles is determined as final classification prediction result.
7. according to the method described in claim 6, it is characterized in that, wait be put in storage examination according to the classification prediction result typing
The reagent information of agent, comprising:
Utilize all sides for segmenting the highest other full name of Attribute class in the corresponding other probability value of Attribute class and replacing corresponding participle
Formula carrys out the reagent information of reagent to be put in storage described in typing.
8. the method according to claim 1, wherein wait be put in storage examination according to the classification prediction result typing
After the reagent information of agent, the method also includes:
The code shape label for identifying reagent to be put in storage is generated according to the reagent information of institute's typing.
9. a kind of reagent information input device based on OCR identification, which is characterized in that described device includes:
Whether detection unit, the reagent label for detecting reagent to be put in storage are directed at image acquisition units, obtain testing result；
Described image acquisition unit acquires corresponding reagent label if being that reagent label has been aligned for the testing result
Image；
OCR recognition unit carries out OCR identification for the image to corresponding reagent label collected, obtains corresponding reagent label
Content of text；
Classification predicting unit, carries out classification prediction for the content of text to obtained corresponding reagent label, and it is pre- to obtain classification
Survey result；
Data input unit, the reagent information for the reagent to be put in storage according to the classification prediction result typing.
10. device according to claim 9, which is characterized in that described device further include:
Image pre-processing unit, for carrying out OCR knowledge by image of the OCR recognition unit to corresponding reagent label collected
Before not, image preprocessing is carried out to the image of corresponding reagent label collected；Wherein, described image pretreatment includes as follows
At least one processing operation: adjustment image resolution ratio or the direction of rotation for adjusting image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910680984.5A CN110322206A (en) | 2019-07-26 | 2019-07-26 | A kind of reagent information input method and device based on OCR identification |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910680984.5A CN110322206A (en) | 2019-07-26 | 2019-07-26 | A kind of reagent information input method and device based on OCR identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110322206A true CN110322206A (en) | 2019-10-11 |
Family
ID=68124747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910680984.5A Pending CN110322206A (en) | 2019-07-26 | 2019-07-26 | A kind of reagent information input method and device based on OCR identification |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110322206A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652229A (en) * | 2020-05-25 | 2020-09-11 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111784115A (en) * | 2020-06-09 | 2020-10-16 | 岭东核电有限公司 | Nuclear power station chemical information management method, system, equipment and storage medium |
| CN111860263A (en) * | 2020-07-10 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Information input method and device and computer readable storage medium |
| CN113159559A (en) * | 2021-04-15 | 2021-07-23 | 杭州电子科技大学 | Automatic identification, classification, investigation and early warning method for chemical reagent storage |
| WO2022165692A1 (en) * | 2021-02-04 | 2022-08-11 | 深圳迈瑞生物医疗电子股份有限公司 | Reagent management method and related device |
| CN114911987A (en) * | 2022-05-24 | 2022-08-16 | 无锡市第五人民医院 | Autonomous analysis method and system for detection reagent strip |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105005793A (en) * | 2015-07-15 | 2015-10-28 | 广州敦和信息技术有限公司 | Method and device for automatically identifying and recording invoice character strip |
| CN108399516A (en) * | 2017-10-10 | 2018-08-14 | 居瑞雪 | A kind of monitoring system of reagent dosage of hospital clinical laboratories and application method |
| CN108470069A (en) * | 2018-03-29 | 2018-08-31 | 天津城建大学 | Scientific Research in University Laboratory reagent information querying method based on augmented reality and system |
| CN108876195A (en) * | 2018-07-17 | 2018-11-23 | 李小玲 | A kind of intelligentized teachers ' teaching quality evaluating system |
| CN109255290A (en) * | 2018-07-27 | 2019-01-22 | 北京三快在线科技有限公司 | Menu recognition methods, device, electronic equipment and storage medium |
| CN109784339A (en) * | 2018-12-13 | 2019-05-21 | 平安普惠企业管理有限公司 | Picture recognition test method, device, computer equipment and storage medium |
| CN109919147A (en) * | 2019-03-04 | 2019-06-21 | 上海宝尊电子商务有限公司 | The method of text identification in drop for clothing image |
| CN109993160A (en) * | 2019-02-18 | 2019-07-09 | 北京联合大学 | A kind of image flame detection and text and location recognition method and system |
-
2019
- 2019-07-26 CN CN201910680984.5A patent/CN110322206A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105005793A (en) * | 2015-07-15 | 2015-10-28 | 广州敦和信息技术有限公司 | Method and device for automatically identifying and recording invoice character strip |
| CN108399516A (en) * | 2017-10-10 | 2018-08-14 | 居瑞雪 | A kind of monitoring system of reagent dosage of hospital clinical laboratories and application method |
| CN108470069A (en) * | 2018-03-29 | 2018-08-31 | 天津城建大学 | Scientific Research in University Laboratory reagent information querying method based on augmented reality and system |
| CN108876195A (en) * | 2018-07-17 | 2018-11-23 | 李小玲 | A kind of intelligentized teachers ' teaching quality evaluating system |
| CN109255290A (en) * | 2018-07-27 | 2019-01-22 | 北京三快在线科技有限公司 | Menu recognition methods, device, electronic equipment and storage medium |
| CN109784339A (en) * | 2018-12-13 | 2019-05-21 | 平安普惠企业管理有限公司 | Picture recognition test method, device, computer equipment and storage medium |
| CN109993160A (en) * | 2019-02-18 | 2019-07-09 | 北京联合大学 | A kind of image flame detection and text and location recognition method and system |
| CN109919147A (en) * | 2019-03-04 | 2019-06-21 | 上海宝尊电子商务有限公司 | The method of text identification in drop for clothing image |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111652229A (en) * | 2020-05-25 | 2020-09-11 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111652229B (en) * | 2020-05-25 | 2023-09-12 | 泰康保险集团股份有限公司 | Information input method and device, electronic equipment and storage medium |
| CN111784115A (en) * | 2020-06-09 | 2020-10-16 | 岭东核电有限公司 | Nuclear power station chemical information management method, system, equipment and storage medium |
| CN111784115B (en) * | 2020-06-09 | 2024-06-11 | 岭东核电有限公司 | Nuclear power station chemical information management method, system, equipment and storage medium |
| CN111860263A (en) * | 2020-07-10 | 2020-10-30 | 海尔优家智能科技（北京）有限公司 | Information input method and device and computer readable storage medium |
| WO2022165692A1 (en) * | 2021-02-04 | 2022-08-11 | 深圳迈瑞生物医疗电子股份有限公司 | Reagent management method and related device |
| CN113159559A (en) * | 2021-04-15 | 2021-07-23 | 杭州电子科技大学 | Automatic identification, classification, investigation and early warning method for chemical reagent storage |
| CN113159559B (en) * | 2021-04-15 | 2024-02-09 | 杭州电子科技大学 | Automatic identification, classification, investigation and early warning method for chemical reagent storage |
| CN114911987A (en) * | 2022-05-24 | 2022-08-16 | 无锡市第五人民医院 | Autonomous analysis method and system for detection reagent strip |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110322206A (en) | A kind of reagent information input method and device based on OCR identification | |
| CN113822494B (en) | Risk prediction method, device, equipment and storage medium | |
| US20230206000A1 (en) | Data-driven structure extraction from text documents | |
| CN103699523B (en) | Product classification method and apparatus | |
| Vasudevan et al. | When does dough become a bagel? analyzing the remaining mistakes on imagenet | |
| US20230005286A1 (en) | Methods, systems, articles of manufacture, and apparatus for decoding purchase data using an image | |
| CN108830147A (en) | A kind of commodity on shelf price recognition methods based on image recognition, device and system | |
| CN116049397B (en) | Sensitive information discovery and automatic classification method based on multi-mode fusion | |
| CN111666766A (en) | Data processing method, device and equipment | |
| CN112115993A (en) | Zero sample and small sample evidence photo anomaly detection method based on meta-learning | |
| CN107169061A (en) | A kind of text multi-tag sorting technique for merging double information sources | |
| Ji et al. | Attention based meta path fusion for heterogeneous information network embedding | |
| CN112541077A (en) | Processing method and system for power grid user service evaluation | |
| CN115081025A (en) | Sensitive data management method and device based on digital middlebox and electronic equipment | |
| CN112183655A (en) | Document multi-label classification method and device | |
| Sun et al. | Indiscernible object counting in underwater scenes | |
| CN101213539A (en) | Cross descriptor learning system, method and program product therefor | |
| Belhadj et al. | Consideration of the word’s neighborhood in GATs for information extraction in semi-structured documents | |
| CN114708073B (en) | Intelligent detection method and device for surrounding mark and serial mark, electronic equipment and storage medium | |
| Mandal et al. | Improving it support by enhancing incident management process with multi-modal analysis | |
| Spichakova et al. | Using machine learning for automated assessment of misclassification of goods for fraud detection | |
| CN111737107B (en) | Repeated defect report detection method based on heterogeneous information network | |
| Gao et al. | Informative scene graph generation via debiasing | |
| CN117271713A (en) | Associated object recognition method, associated object recognition device, electronic equipment and storage medium | |
| CN115221323A (en) | Cold start processing method, device, equipment and medium based on intention recognition model |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191011 |