CN107517272A - A kind of device, the system and method for automatic data collection set form data - Google Patents
A kind of device, the system and method for automatic data collection set form data Download PDFInfo
- Publication number
- CN107517272A CN107517272A CN201710828713.0A CN201710828713A CN107517272A CN 107517272 A CN107517272 A CN 107517272A CN 201710828713 A CN201710828713 A CN 201710828713A CN 107517272 A CN107517272 A CN 107517272A
- Authority
- CN
- China
- Prior art keywords
- data
- enterprise
- terminal
- module
- scan
- 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
- 238000013480 data collection Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000013527 convolutional neural network Methods 0.000 claims description 24
- 238000012549 training Methods 0.000 claims description 7
- 238000013461 design Methods 0.000 claims description 6
- 238000006073 displacement reaction Methods 0.000 claims description 6
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 238000010408 sweeping Methods 0.000 claims 1
- 238000012552 review Methods 0.000 abstract description 2
- 238000013507 mapping Methods 0.000 description 14
- 238000005516 engineering process Methods 0.000 description 11
- 210000002569 neuron Anatomy 0.000 description 11
- 230000006870 function Effects 0.000 description 10
- 238000000605 extraction Methods 0.000 description 8
- 239000000284 extract Substances 0.000 description 6
- 238000011160 research Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000007935 neutral effect Effects 0.000 description 5
- 230000006872 improvement Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 241001269238 Data Species 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000006855 networking Effects 0.000 description 3
- PCTMTFRHKVHKIS-BMFZQQSSSA-N (1s,3r,4e,6e,8e,10e,12e,14e,16e,18s,19r,20r,21s,25r,27r,30r,31r,33s,35r,37s,38r)-3-[(2r,3s,4s,5s,6r)-4-amino-3,5-dihydroxy-6-methyloxan-2-yl]oxy-19,25,27,30,31,33,35,37-octahydroxy-18,20,21-trimethyl-23-oxo-22,39-dioxabicyclo[33.3.1]nonatriaconta-4,6,8,10 Chemical group C1C=C2C[C@@H](OS(O)(=O)=O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2.O[C@H]1[C@@H](N)[C@H](O)[C@@H](C)O[C@H]1O[C@H]1/C=C/C=C/C=C/C=C/C=C/C=C/C=C/[C@H](C)[C@@H](O)[C@@H](C)[C@H](C)OC(=O)C[C@H](O)C[C@H](O)CC[C@@H](O)[C@H](O)C[C@H](O)C[C@](O)(C[C@H](O)[C@H]2C(O)=O)O[C@H]2C1 PCTMTFRHKVHKIS-BMFZQQSSSA-N 0.000 description 2
- 241000282326 Felis catus Species 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000013529 biological neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- WPWLFFMSSOAORQ-UHFFFAOYSA-N 5-bromo-4-chloro-3-indolyl acetate Chemical compound C1=C(Br)C(Cl)=C2C(OC(=O)C)=CNC2=C1 WPWLFFMSSOAORQ-UHFFFAOYSA-N 0.000 description 1
- 210000004556 brain Anatomy 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 150000001875 compounds Chemical class 0.000 description 1
- 238000004925 denaturation Methods 0.000 description 1
- 230000036425 denaturation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000004321 preservation Methods 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/06—Protocols specially adapted for file transfer, e.g. file transfer protocol [FTP]
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Character Discrimination (AREA)
Abstract
The invention discloses a kind of device, the system and method for automatic data collection set form data, wherein system divides terminal including end of scan, master server, enterprise's master terminal, enterprise;End of scan is connected with device by data interface module;Master server is attached with device by network, and master server is at least connected with a device, and master server is used to provide network support for device, receives data scanning result, merges scanning result and is supplied to enterprise point terminal;Enterprise's master terminal is connected with master server, and enterprise's master terminal is scanned Data Styles configuration for operating personnel;Enterprise point terminal is connected with master server by the device, and enterprise divides terminal quantity to be at least 1, and enterprise point terminal is used for the scan data pattern that operating personnel configure according to enterprise's master terminal and carries out data acquisition, edits and upload to master server.The present invention requires low, use cost is cheap, and picking rate is fast, and manual review's speed is fast using simply to user.
Description
Technical field
The present invention relates to data acquisition technology field, and in particular to a kind of device of automatic data collection set form data, is
System and method.
Background technology
Set form data, generally referring to data has identical fixed type template, exists between different sample datas
Certain repeated content, such as physical examination table, certificate.Mainly there is in the following manner currently for the data acquisition of set form:
First, by manually inputting original document into information system.Shortcoming:Speed is slow, and accuracy is difficult to ensure that;It is original
Papery document can not be input in system, and electronic data is easily tampered;Data can only collect afterwards, poor in timeliness.
Second, by the device scan original document with document identification function, then upload scanning result and original paper.Lack
Point:Equipment price is expensive;Scanning accuracy rate is not high, and processing speed is slow;General scanning device takes the mode scanned in full, no
The effective information that can be directed in result is retrieved;Scanning result needs artificial check and correction accuracy, and workload is big, and easily goes out
It is wrong;No data report interface to scanning device in itself, it is necessary to which collocation personnel and computer could realize data acquisition.
3rd, by the way that manually raw information is transcribed on special input medium, such as answering card or radio-frequency card, then
By the electronization that information is realized after proprietary recording device scanning input medium.Shortcoming:Equipment price is expensive, input medium
(answering card/radio-frequency card etc.) is expensive with respect to paper delivery medium, and needs to prepare in advance;High to the competency profiling of personnel, system must
Must be by the human users of familiar with computers system;It is cumbersome using preceding preparation, using it is preceding must be by the knot of tissue of acquisition target
Structure is input to system;Equipment fault-tolerant ability is poor, once occur unregistered personnel in system or input numerical fault system
It can be stopped;It is cumbersome, to medium on the speed of input data be less than handwriting input speed;System onboard data form and
Cuit renewal is complicated, it is necessary to which special technical staff updates manually, and maintainability is poor.
At present, there is problems with collection set form data:
Hardware contained by existing scheme can be only done single data acquisition function, need manually to go up count off after data acquisition
According to only reporting scanning result, can not form corresponding relation between original same scanning result can only abandon;
After existing scheme user's data acquisition, user needs to enter data by computer or third party's tool software
Row collection processing, it is cumbersome;
Prior art is all using a single point as service object, is supported without enterprise's networking data collection.
The content of the invention
It is an object of the invention to provide a kind of device, the system and method for automatic data collection set form data, gathering
After the completion of can externally provide can be with complete paired data by computer or mobile phone based on browser data accession page, user
Operations such as modification, error correction, electrical form export, while can carry out store function to data, user, can be with without another deposit data
Data acquisition renewal is carried out as needed, is easy to data to report and safeguard.
To achieve the above object, the technical scheme is that:A kind of device of automatic data collection set form data, it is described
Device includes microprocessor, and described device also includes data interface module, hardware control module, intelligent identification module, data and deposited
Store up module, template configuration module;The data interface module is used for described device connection end of scan;The hardware control module
The data interface module is connected with, hardware control module is used to be controlled end of scan, and hardware control module passes through number
The scan data of end of scan is obtained according to interface module;The intelligent identification module is connected with the hardware control module, intelligence
Identification module is used for the scan data for identifying end of scan;The data memory module is connected with the hardware control module, number
It is connected according to memory module with the intelligent identification module, data memory module is used for the recognition result for storing intelligent identification module;
The template configuration module is connected with the intelligent identification module, and template configuration module is used to configure intelligent identification module to scanning
The recognition template of data.
A kind of device of automatic data collection set form data as described above, described device operation have (SuSE) Linux OS,
Operation has primary control program in device, and the primary control program is used for the hardware control module, intelligent identification module, template configuration
Module is controlled.Hardware control module uses hardware control, and hardware control module belongs to prior art.Intelligent identification module
Using the handwriting recognition technology based on convolutional neural networks, relative to the prior-generation handwriting recognition technology based on image recognition technology
It is more accurate quick.
A kind of device of automatic data collection set form data as described above, the data interface module use usb data
Interface.The device can connect end of scan, and then the data letter between realization device and end of scan by USB data interface
Breath interaction.
The present invention provides a kind of system of automatic data collection set form data, and the system uses said apparatus, the system
System also divides terminal including end of scan, master server, enterprise's master terminal and enterprise;The end of scan passes through number with described device
Connected according to interface module;The master server is attached with described device by network, and master server is at least connected with an institute
Device is stated, master server is used to provide network support for described device, receives data scanning result, merges scanning result and is supplied to
Enterprise divides terminal;Enterprise's master terminal is connected with the master server, and enterprise's master terminal is scanned number for operating personnel
According to style configuration;The enterprise point terminal is connected with the master server by described device, and enterprise divides terminal quantity to be at least 1
Individual, enterprise divides terminal to be used for the scan data pattern progress data acquisition, Bian Jiyu that operating personnel configure according to enterprise's master terminal
Upload to the master server.
A kind of system of automatic data collection set form data as described above, the end of scan use scanner, scan
Terminal is connected with described device by USB data line.When setting up the system of automatic data collection set form data, by end of scan
It is connected with said apparatus by USB data line.
A kind of system of automatic data collection set form data as described above, the master server are located in data cloud, main
Server by cloud computing platform expand while provides service at least two said apparatus.Master server can pass through cloud
Calculating platform is expanded provides service for the collecting device of magnanimity simultaneously.Master server is to be previously mentioned automatic data collection in multiple present invention
The device of set form data provides network support, receives scanning result, merges scanning result and provides the support of enterprise customer's data,
Remote technology is also provided simultaneously and supports that company provides data access interface for data acquisition.The data of collection can real-time report, just
Safeguarded in remote upgrade.
A kind of system of automatic data collection set form data as described above, the master server connect provided with mobile data
Mouthful, master server is connected with mobile terminal, and mobile terminal uses mobile phone terminal or tablet personal computer.Enterprise operations personnel can pass through
Mobile phone terminal or tablet personal computer carry out data access.
The present invention also provides a kind of method of automatic data collection set form data, and methods described is using said apparatus and passes through
Said system, it the described method comprises the following steps:
Enterprise's master terminal data acquisition board according to corresponding to data acquisition needs design, on data acquisition module board interface
Alignment code is configured, and data acquisition board is issued to enterprise point terminal by server;
Corporate boss's terminal-pair designs corresponding data recognition template according to data collection form, and configuration intelligent identification module is pair
The data recognition template pattern answered;
The data gathered according to data acquisition board are scanned by enterprise point terminal, and end of scan passes through Intelligent Recognition mould
The data of scanning are identified block according to data recognition template;
The scanning result of identification is stored by data memory module, and by scanning result upload onto the server for
Use.
A kind of method of automatic data collection set form data as described above, the alignment code are arranged on data acquisition board
Interface upper end, the quantity of alignment code is at least 2, and the pattern of each alignment code is different, is identified by the alignment code
Displacement, rotation and the scalability information of paper.Alignment code strengthens data acquisition equipment pair by increasing special marking on scanned copy
The stationkeeping ability of original contents, improve accuracy, speed.Traditional mode sets anchor point, general type at headerfooter
It is easy to the figure of identification for straight line either round dot etc. of right-angled intersection, only has position different between anchor point, other outward appearances are complete
It is complete consistent, it is impossible to provide the information such as rotation, direction, deformation, each alignment code contains unique id in this method, so multiple fixed
Site (more than 2) can provide more auxiliary informations such as the direction of whole input data, rotation, irregular deformation together, right
There is more preferable error correcting capability in being stained etc. for input data.
Existing acquisition method can only be once acquired for a kind of data format, it is impossible to multiple format mixing collection, this
Unique id of the method based on alignment code, can identify the classification of raw information, even if the initial data of multiple format is entered together
Row data acquisition, equipment can also automatically classify, extract respectively.
Prior art does not support the raw information typing of multipage, and even if information, all collection also can not be being originally inputted
It is divided into multiple set according to user, unique alignment code technology that this method uses, user id can be included, so as to realize multi-user
The collection of more page datas.
A kind of method of automatic data collection set form data as described above, intelligent identification module pass through convolutional neural networks
Recognizer carries out data training to identification object.Convolutional neural networks recognizer convolutional neural networks (CNN), it is to send out in recent years
Exhibition is got up, and causes a kind of efficient identification method paid attention to extensively.In the 1960s, Hubel and Wiesel is in research cat brain
For finding that its unique network structure can be effectively reduced feedback during the neuron of local sensitivity and set direction in cortex
The complexity of neutral net, then proposes convolutional neural networks.Now, CNN has become the research heat of numerous scientific domains
One of point,, can be directly defeated because the network avoids the pretreatment complicated early stage to image particularly in pattern classification field
Enter original image, thus obtained more being widely applied.K.Fukushima is convolution god in the new cognitron proposed in 1980
First through network is realized network.Then, more researchers are improved the network.Wherein, have and represent
Property achievement in research be that Alexander and Taylor propose " improvement cognitron ", this method combines various improved methods
Advantage simultaneously avoids time-consuming error back propagation.CNN basic structure includes two layers, and one is characterized extract layer, each god
Input through member is connected with the local acceptance region of preceding layer, and extracts the local feature.After the local feature is extracted,
Its position relationship between further feature is also decided therewith;The second is Feature Mapping layer, each computation layer of network is by more
Individual Feature Mapping composition, each Feature Mapping are a planes, and the weights of all neurons are equal in plane.Feature Mapping structure
Using activation primitive of the small sigmoid functions of influence function core as convolutional network so that Feature Mapping has shift invariant
Property.Further, since the neuron on a mapping face shares weights, thus reduce the number of network freedom parameter.Convolution god
It is this distinctive through followed by one computation layer for being used for asking local average and second extraction of each convolutional layer in network
Feature extraction structure reduces feature resolution twice.CNN is mainly used to identify displacement, scaling and other forms distortion consistency
X-Y scheme.Because CNN feature detection layer is learnt by training data, so when using CNN, display is avoided
Feature extraction, and implicitly learnt from training data;Furthermore due to the neuron weights on same Feature Mapping face
Identical, so network can be with collateral learning, this is also that convolutional network is connected with each other a big advantage of network relative to neuron.Volume
The special construction that product neutral net is shared with its local weight has the superiority of uniqueness in terms of speech recognition and image procossing,
It is laid out the biological neural network closer to reality, the shared complexity for reducing network of weights, particularly multidimensional input to
The image of amount can directly input the complexity that network this feature avoids data reconstruction in feature extraction and assorting process.
The invention has the advantages that:Using simple, can be automatically configured using preceding equipment, it is not necessary to manual intervention, also without
It need to prepare in advance;Low is required to user, without professional computer technical ability, as long as can correctly write with regard to data acquisition can be carried out;
Use cost is cheap, without being especially tailored collection form either equipment card, uses common print paper can normal work;Adopt
It is fast to collect speed, more than 1500 parts data can be gathered per hour per table apparatus;Hand-written data recognition accuracy is high, to common book
95% degree of accuracy can be reached by writing custom, good to the adaptability of different user hand-written data;Manual review's speed is fast, is running into
Accuracy rate is estimated that during unascertainable input data, manually emphasis checks these fuzzy results i.e. during compound result
Can, artificial recombination velocity is greatly speeded up, and other technical schemes are needed to manually checking recognition result in the whole text at present;Function is complete
Face, data acquisition, processing, preservation can be achieved, the multi-point data acquisition of enterprise customer can be supported after networking;Safeguard simply, only
Access technical problem more than technical staff's can processing most (95%) after network;System can be with online upgrading.
Brief description of the drawings
Fig. 1 is the schematic device of automatic data collection set form data;
Fig. 2 is the systematic schematic diagram of automatic data collection set form data;
Fig. 3 is the system buildup schematic diagram of automatic data collection set form data;
Fig. 4 is the method flow diagram of automatic data collection set form data.
Embodiment
Following examples are used to illustrate the present invention, but are not limited to the scope of the present invention.
As shown in Figure 1 and Figure 2, a kind of device of automatic data collection set form data, described device includes microprocessor 1, micro-
Processor 1 is used for device and completes instruction fetch, execute instruction, and the logic fortune between extraneous end of scan and itself each modular member
Operation is calculated, is the operation control part of device.Described device also includes data interface module 2, hardware control module 3, intelligence and known
Other module 4, data memory module 5, template configuration module 6;The data interface module 2 is used for described device connection end of scan
7;The hardware control module 3 is connected with the data interface module 2, and hardware control module 3 is used to control end of scan 7
System, hardware control module 3 obtain the scan data of end of scan 7 by data interface module 2;The intelligent identification module 4 with
The hardware control module 3 connects, and intelligent identification module 4 is used for the scan data for identifying end of scan 7;The data storage mould
Block 5 is connected with the hardware control module 3, and data memory module 5 is connected with the intelligent identification module 4, data memory module 5
For storing the recognition result of intelligent identification module 4;The template configuration module 6 is connected with the intelligent identification module 4, template
Configuration module 6 is used to configure recognition template of the intelligent identification module 4 to scan data.
In one embodiment of the device of automatic data collection set form data, described device operation has (SuSE) Linux OS,
Operation has primary control program in device, and the primary control program is used to match somebody with somebody the hardware control module 3, intelligent identification module 4, template
Module 6 is put to be controlled.Hardware control module 3 uses hardware control, and hardware control module 3 belongs to prior art.Intelligence is known
Other module 4 uses the handwriting recognition technology based on convolutional neural networks, hand-written relative to the prior-generation based on image recognition technology
Identification technology is more accurate quick.
In one embodiment of the device of automatic data collection set form data, the data interface module 2 uses usb data
Interface.The device can connect end of scan 7, and then the data between realization device and end of scan 7 by USB data interface
Information exchange.
As shown in Figure 1, Figure 2, Figure 3 shows, the present invention provides a kind of system of automatic data collection set form data, and the system is adopted
With said apparatus, the system also divides terminal 10 including end of scan 7, master server 8, enterprise's master terminal 9 and enterprise;It is described to sweep
Terminal 7 is retouched to be connected by data interface module 2 with described device;The master server 8 is connected with described device by network
Connect, master server 8 is at least connected with a described device, and master server 8 is used to provide network support for described device, receives data
Scanning result, merge scanning result and be supplied to enterprise point terminal 10;Enterprise's master terminal 9 is connected with the master server 8, enterprise
Owner's terminal 9 is scanned Data Styles configuration for operating personnel;The enterprise point terminal 10 passes through with the master server 8
Described device connects, and the quantity of enterprise point terminal 10 is at least 1, and enterprise divides terminal 10 to be used for operating personnel according to enterprise's master terminal
The scan data pattern of 9 configurations carries out data acquisition, edits and upload to the master server 8.
In one embodiment of the system of automatic data collection set form data, the end of scan 7 uses scanner, scanning
Terminal 7 is connected with described device by USB data line.When setting up the system of automatic data collection set form data, by scanning eventually
End 7 is connected with said apparatus by USB data line.
In one embodiment of the system of automatic data collection set form data, the master server 8 is located in data cloud 11,
Master server 8 by cloud computing platform expand while is the offer service of at least two end of scan 7.Master server 8 can lead to
Cloud computing platform is crossed to expand while provide service for the collecting device of magnanimity.Master server 8 is to be previously mentioned certainly in multiple present invention
The device of dynamic collection set form data provides network support, receives scanning result, merges scanning result and provides enterprise customer's number
According to support, while remote technology is also provided and supports that company provides data access interface for data acquisition.The data of collection can be real-time
Report, be easy to remote upgrade to safeguard.
In one embodiment of the system of automatic data collection set form data, the master server 8 connects provided with mobile data
Mouthful, master server 8 is connected with mobile terminal 12, and mobile terminal 12 uses mobile phone terminal or tablet personal computer.Enterprise operations personnel can
To carry out data access by mobile phone terminal or tablet personal computer.
As shown in figure 4, the present invention also provides a kind of method of automatic data collection set form data, methods described is using above-mentioned
Device and by said system, the described method comprises the following steps:
S1:The data acquisition board according to corresponding to data acquisition needs design of enterprise's master terminal 9, in data acquisition board circle
Alignment code is configured on face, and data acquisition board is issued to enterprise point terminal 10 by server;
S2:The contrasting data of enterprise's master terminal 9 gathers the corresponding data recognition template of Table Design, configures intelligent identification module
4 be corresponding data recognition template pattern;
S3:The data gathered according to data acquisition board are scanned by enterprise point terminal 10, and end of scan 7 passes through intelligence
The data of scanning are identified identification module 4 according to data recognition template;
S4:The scanning result of identification is stored by data memory module 5, and scanning result is uploaded onto the server
For using.
In one embodiment of the method for automatic data collection set form data, the alignment code is arranged on data acquisition board
Interface on, for example corner location can be arranged on, the quantity of alignment code is at least 2, the mutual not phase of the pattern of each alignment code
Together, displacement, rotation and the scalability information of paper are identified by the alignment code.Alignment code on scanned copy by increasing special mark
Note enhancing data acquisition equipment improves accuracy, speed to the stationkeeping ability of original contents.Traditional mode is at headerfooter
Anchor point is set, and general type is easy to the figure of identification for straight line either round dot etc. of right-angled intersection, only had between anchor point
Position is different, and other outward appearances are completely the same, it is impossible to provides the information such as rotation, direction, deformation, each alignment code contains in this method
There is unique id, so multiple anchor points (more than 2) can provide the direction of whole input data, irregularly rotation, change together
More auxiliary informations such as shape, there is more preferable error correcting capability for being stained etc. for input data.
Even if existing table style has alignment code, and the positioning of the identical alignment code this programme of multiple profiles
Each of code is different (profile either interior details), based on different alignment codes, scaling, rotation with regard to paper can be deduced
Turn and breakage recovers, existing positioning pattern is AAAA, if paper have rotated, then the location information extracted is not
This change can be shown.The positioning pattern of this programme is ABCD (being not necessarily clock-wise order), if can after paper rotation
To find that the position of alignment code becomes for BCDA, if paper sheet break, then such remaining information based on AB, ABC, ACD ...
It can also infer that direction, scaling and the data of original list relative to the position of alignment code.
It should be further stated that using thinking of the present invention, any distinguishing figure of outward appearance can serve as positioning
Code, the ABCD being not limited only to used in the present embodiment, the quantity of alignment code not-be 4 surely, as long as can be realized more than 2
Positioning function.It is pointed out that positioning number of codes is more, it is better for the big sheet positioning effect of Area comparison.
Meanwhile the arrangement of alignment code is not necessarily rectangle, straight line (2), triangle (3), polygon (multiple), arbitrarily
The shape appointed in advance can.It is to be noted that be previously mentioned in this implementation it is " more " not hand over unclear, but be
Preferably illustrate the thinking and implementation of the present invention.
Existing acquisition method can only be once acquired for a kind of data format, it is impossible to multiple format mixing collection, this
Unique id of the method based on alignment code, can identify the classification of raw information, even if the initial data of multiple format is entered together
Row data acquisition, equipment can also automatically classify, extract respectively.Prior art does not support the raw information typing of multipage,
Even if information, all collection also can not be divided into multiple set, unique alignment code skill that this method uses being originally inputted according to user
Art, user id can be included, so as to realize the collection of the more page datas of multi-user.
In one embodiment of the method for automatic data collection set form data, intelligent identification module 4 passes through convolutional Neural net
Network recognizer carries out data training to identification object.Convolutional neural networks recognizer convolutional neural networks (CNN), are in recent years
Grow up, and cause a kind of efficient identification method paid attention to extensively.In the 1960s, Hubel and Wiesel is in research cat
For finding that its unique network structure can be effectively reduced anti-during the neuron of local sensitivity and set direction in cortex
The complexity of neutral net is presented, then proposes convolutional neural networks.Now, CNN has become the research of numerous scientific domains
One of focus,, can be direct because the network avoids the pretreatment complicated early stage to image particularly in pattern classification field
Original image is inputted, thus has obtained more being widely applied.K.Fukushima is convolution in the new cognitron proposed in 1980
First of neutral net realizes network.Then, more researchers are improved the network.Wherein, there is generation
The achievement in research of table is that Alexander and Taylor proposes " improvement cognitron ", and this method combines various improved methods
The advantages of and avoid time-consuming error back propagation.CNN basic structure includes two layers, and one is characterized extract layer, each
The input of neuron is connected with the local acceptance region of preceding layer, and extracts the local feature.Once the local feature is extracted
Afterwards, its position relationship between further feature is also decided therewith;The second is Feature Mapping layer, each computation layer of network by
Multiple Feature Mappings composition, each Feature Mapping are a planes, and the weights of all neurons are equal in plane.Feature Mapping knot
Structure is using activation primitive of the small sigmoid functions of influence function core as convolutional network so that Feature Mapping has displacement not
Denaturation.Further, since the neuron on a mapping face shares weights, thus reduce the number of network freedom parameter.Convolution
Each convolutional layer followed by one in neutral net is used for asking the computation layer of local average and second extraction, this peculiar
The structure of feature extraction twice reduce feature resolution.CNN is mainly used to identify that displacement, scaling and other forms distortion are constant
The X-Y scheme of property.Because CNN feature detection layer is learnt by training data, so when using CNN, avoid aobvious
The feature extraction shown, and implicitly learnt from training data;Furthermore due to the neuron power on same Feature Mapping face
It is worth identical, so network can be with collateral learning, this is also that convolutional network is connected with each other a big advantage of network relative to neuron.
The special construction that convolutional neural networks are shared with its local weight has the superior of uniqueness in terms of speech recognition and image procossing
Property, it is laid out the biological neural network closer to reality, and the shared complexity for reducing network of weights, particularly multidimensional input
The image of vector can directly input the complexity that network this feature avoids data reconstruction in feature extraction and assorting process.
Data acquisition equipment can be provided externally after the completion of collection based on browser data accession page in the present invention, used
Family can be with operations such as modification, error correction, the electrical form export of complete paired data by computer or mobile phone, while data acquisition is set
Standby also to include data storage function, user is without another deposit data.With, using a single point as service object, not having in prior art
The mode for having the collection of enterprise networking data to support is compared, and all collecting devices of the present invention are all supported to network online reported data and more
New data collection form, for enterprise customer, it is only necessary to issue new data acquisition module on Centroid (such as parent company)
Plate, information can automatically update all child nodes (subsidiary, office etc.), improve enterprise customer's internal communication efficiency, keep away
Exempt from operational error.All nodes of the present invention all support real-time interconnection, and Centroid (such as parent company) can own with real-time collecting
The data of access device, for the user of this programme, complete Data Collection and also just while complete data and report.Adopt
Collection equipment uses networked devices, and the data collected and original input picture information can be uploaded directly into network, collecting device
Also there is local data disposal ability simultaneously, user completes most of data mart modeling operation without additional means can.
Although above with general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention, belong to the scope of protection of present invention.
Claims (10)
1. a kind of device of automatic data collection set form data, described device include microprocessor, it is characterised in that:Described device
Also include data interface module, hardware control module, intelligent identification module, data memory module, template configuration module;The number
It is used for described device connection end of scan according to interface module;The hardware control module is connected with the data interface module, firmly
Part control module is used to be controlled end of scan, and hardware control module obtains sweeping for end of scan by data interface module
Retouch data;The intelligent identification module is connected with the hardware control module, and intelligent identification module is used to identify end of scan
Scan data;The data memory module is connected with the hardware control module, data memory module and the Intelligent Recognition mould
Block connects, and data memory module is used for the recognition result for storing intelligent identification module;The template configuration module and the intelligence
Identification module connects, and template configuration module is used to configure recognition template of the intelligent identification module to scan data.
A kind of 2. device of automatic data collection set form data according to claim 1, it is characterised in that:Described device is transported
Row has (SuSE) Linux OS, and being run in device has primary control program, and the primary control program is used for the hardware control module, intelligence
Energy identification module, template configuration module are controlled.
A kind of 3. device of automatic data collection set form data according to claim 1, it is characterised in that:The data connect
Mouth mold block uses USB data interface.
4. a kind of system of automatic data collection set form data, the system includes the dress as described in any one of claims 1 to 3
Put, it is characterised in that:The system also divides terminal including end of scan, master server, enterprise's master terminal and enterprise;The scanning
Terminal is connected with described device by data interface module;The master server is attached with described device by network, main
Server is at least connected with a described device, and master server is used to provide network support for described device, receives data scanning knot
Fruit, merge scanning result and be supplied to enterprise point terminal;Enterprise's master terminal is connected with the master server, and enterprise's master terminal is used
Data Styles configuration is scanned in operating personnel;The enterprise point terminal is connected with the master server by described device,
Enterprise divides terminal quantity to be at least 1, and enterprise divides terminal to be used for the scan data sample that operating personnel configure according to enterprise's master terminal
Formula carries out data acquisition, edits and upload to the master server.
A kind of 5. system of automatic data collection set form data according to claim 4, it is characterised in that:The scanning is eventually
End uses scanner, and end of scan is connected with described device by USB data line.
A kind of 6. system of automatic data collection set form data according to claim 4, it is characterised in that:The main service
Device is located in data cloud, and master server by cloud computing platform expand while is at least two automatic data collection set form number
According to device provide service.
A kind of 7. system of automatic data collection set form data according to claim 4, it is characterised in that:The main service
Device is provided with mobile data interface, and master server is connected with mobile terminal, and mobile terminal uses mobile phone terminal or tablet personal computer.
8. a kind of method of automatic data collection set form data, methods described passes through the system, its feature using described device
It is:It the described method comprises the following steps:
Enterprise's master terminal data acquisition board according to corresponding to data acquisition needs design, is configured on data acquisition module board interface
Alignment code, and data acquisition board is issued to enterprise point terminal by server;
Corporate boss's terminal-pair designs corresponding data recognition template according to data collection form, and configuration intelligent identification module is corresponding
Data recognition template pattern;
The data gathered according to data acquisition board are scanned by enterprise point terminal, and end of scan passes through intelligent identification module pair
The data of scanning are identified according to data recognition template;
The scanning result of identification is stored by data memory module, and scanning result is uploaded onto the server for making
With.
A kind of 9. method of automatic data collection set form data according to claim 8, it is characterised in that:The alignment code
The interface upper end of data acquisition board is arranged on, the quantity of alignment code is at least 2, and the pattern of each alignment code is different,
Displacement, rotation and the scalability information of paper are identified by the alignment code.
A kind of 10. method of automatic data collection set form data according to claim 8, it is characterised in that:Intelligent Recognition
Module carries out data training by convolutional neural networks recognizer to identification object.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710828713.0A CN107517272A (en) | 2017-09-14 | 2017-09-14 | A kind of device, the system and method for automatic data collection set form data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710828713.0A CN107517272A (en) | 2017-09-14 | 2017-09-14 | A kind of device, the system and method for automatic data collection set form data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107517272A true CN107517272A (en) | 2017-12-26 |
Family
ID=60724108
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710828713.0A Pending CN107517272A (en) | 2017-09-14 | 2017-09-14 | A kind of device, the system and method for automatic data collection set form data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107517272A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567711A (en) * | 2010-12-29 | 2012-07-11 | 方正国际软件(北京)有限公司 | Method and system for making and using scanning recognition template |
CN103617415A (en) * | 2013-11-19 | 2014-03-05 | 北京京东尚科信息技术有限公司 | Device and method for automatically identifying invoice |
CN204178382U (en) * | 2014-10-16 | 2015-02-25 | 上海京颐科技股份有限公司 | A kind of electronization of the paper-bill based on OCR device |
CN104915666A (en) * | 2015-06-02 | 2015-09-16 | 舟山奇普嘉教育科技有限公司 | Information card information positioning acquisition method based on paper-made image scanning |
CN106933761A (en) * | 2015-12-31 | 2017-07-07 | 航天信息股份有限公司 | Collecting method and the conversion equipment using the method gathered data |
-
2017
- 2017-09-14 CN CN201710828713.0A patent/CN107517272A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567711A (en) * | 2010-12-29 | 2012-07-11 | 方正国际软件(北京)有限公司 | Method and system for making and using scanning recognition template |
CN103617415A (en) * | 2013-11-19 | 2014-03-05 | 北京京东尚科信息技术有限公司 | Device and method for automatically identifying invoice |
CN204178382U (en) * | 2014-10-16 | 2015-02-25 | 上海京颐科技股份有限公司 | A kind of electronization of the paper-bill based on OCR device |
CN104915666A (en) * | 2015-06-02 | 2015-09-16 | 舟山奇普嘉教育科技有限公司 | Information card information positioning acquisition method based on paper-made image scanning |
CN106933761A (en) * | 2015-12-31 | 2017-07-07 | 航天信息股份有限公司 | Collecting method and the conversion equipment using the method gathered data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yang et al. | Deep plastic surgery: Robust and controllable image editing with human-drawn sketches | |
JP6980040B2 (en) | Medical report generation method and equipment | |
CN106202447B (en) | A kind of digital cartography system having production and management integration | |
CN101853299B (en) | Image searching result ordering method based on perceptual cognition | |
CN107368967A (en) | Engineering safety quality inspection intelligent management based on internet | |
CN103488711B (en) | A kind of method and system of quick Fabrication vector font library | |
CN107067060A (en) | The Quick Response Code RFID tag print system and book borrowing and reading method of self-service book cabinet | |
CN107169073A (en) | A kind of data managing method and management platform | |
CN108053530A (en) | A kind of intelligent access control system of the Yun Jiaduan based on face recognition | |
CN105373990A (en) | Drug intelligent management system | |
JPH09508719A (en) | Card making apparatus and method | |
CN105139149B (en) | A kind of method of quick counting server assets configuration information | |
CN107194835B (en) | Hotel reception robot system | |
CN112862024B (en) | Text recognition method and system | |
CN112232047A (en) | Method, system, equipment and medium for multi-dimensional data acquisition and automatic summarization | |
CN109086653A (en) | Handwriting model training method, hand-written character recognizing method, device, equipment and medium | |
CN113901933B (en) | Electronic invoice information extraction method, device and equipment based on artificial intelligence | |
CN206788898U (en) | The Quick Response Code RFID label tag print system of self-service book cabinet | |
CN109740579A (en) | A kind of fast food cash register system based on dish intelligent recognition | |
CN107248230A (en) | The online rent and sale method and system of intelligent graphic bookcase, books | |
CN107451942A (en) | A kind of bookshelf and corresponding bookshelf graphic software platform method | |
CN201590106U (en) | Integrated management information system for letters and visits | |
CN103576658A (en) | Production tracking control method for lead-acid cells | |
CN108038775A (en) | A kind of large enterprise's financial management method and system | |
CN110991279A (en) | Document image analysis and recognition method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20240621 |