CN109726711A - A kind of text information identification extracting method based on machine learning - Google Patents
A kind of text information identification extracting method based on machine learning Download PDFInfo
- Publication number
- CN109726711A CN109726711A CN201811648483.0A CN201811648483A CN109726711A CN 109726711 A CN109726711 A CN 109726711A CN 201811648483 A CN201811648483 A CN 201811648483A CN 109726711 A CN109726711 A CN 109726711A
- Authority
- CN
- China
- Prior art keywords
- text
- user
- machine learning
- scan camera
- extracting method
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The text information that the invention discloses a kind of based on machine learning identifies extracting method, specifically includes connection database, text identification extracts, establishes model, the selection result.The invention has the advantages that effectively increasing the acquisition efficiency of text information, the fluency of user's operation is also improved.
Description
Technical field
The present invention relates to Text Feature Extraction field, especially a kind of text information based on machine learning identifies extracting method.
Background technique
There are some text informations sometimes on some pictures, these text informations include some spcial characters sometimes,
Be inconvenient to typewrite, automatic identification can be carried out by scanner, common text information identification extracting method is typically all
Scanner is attached on picture and is scanned, but text is not of uniform size, and the text information to be extracted is also not and connects very much
It passes through, extracts difficulty, need to retrieve it after extraction, but the result retrieved is many kinds of, is not easy to be searched.
Summary of the invention
The purpose of the present invention is to solve the above problem, devises a kind of text information identification based on machine learning and mention
Take method.
Realize above-mentioned purpose the technical scheme is that, a kind of text information based on machine learning identifies extraction side
Method specifically comprises the following steps:
Step 1: connection database: downloading corresponding port on mobile terminals, using port and pass through wireless network and service
Device terminal is attached, and corresponding database is established in server terminal；
Step 2: text identification is extracted: when the text on picture in photo, paper, mobile phone is carried out identification extraction, being placed
On the table, the part for removing text on text is blocked using blank sheet of paper, is fixed using two elastic webbings, by crane
It is placed on above text, textual scan camera lower end is stretched out by the square aperture on crane, by textual scan camera
Be directed at word segment, movetext scan camera shooting head and push switch are scanned, by Bluetooth signal by scan come text
Word is sent in mobile terminal, is identified；
Step 3: establishing model: different users' use is formed by historical record storage information and passed through wirelessly by mobile terminal
Network is sent on service terminal, carries out taxonomic revision to these information by the intelligent program on service terminal, and pass through machinery
Learning method establishes different models, so that the search habit of user is simulated, it, can when user searches again for certain information
Optimize it and search for target, improves the search efficiency of user；
Step 4: the selection result: by after Text region in step 2, intelligence system can be according to keyword message and simulation at
The use habit automatic screening of user, information required for user is shown, by user by required search
As a result it opens, and the usage history of user is recorded, and uploads again, constantly optimize.
When the text on photo or paper in the step 2 is excessive, picture can be clapped by mobile phone, be reduced
It is scanned again afterwards.
Crane upper surface two sides in the step 2 are equipped with slideway, and slideway is located at square aperture two sides, textual scan
Camera lower end two sides are equipped with connecting shaft, and connection shaft end is equipped with pulley, and pulley lower end is located in slideway.
Textual scan camera lower end in the step 2 is wide-mouth camera lens, is equipped on the downside of textual scan camera front end
Infrared laser lamp, infrared laser lamp Laser emission end are equipped with conical outlet, the inclination angle of tilt angle and wide-mouth camera lens
Degree is consistent.
Switch in the step 2 is located at textual scan camera upper end, is push switch, starts after pressing, pine
It is closed after opening.
Extracting method is identified using the text information based on machine learning of technical solution of the present invention production, can freely be adjusted
The height and angle of its whole scanner, and moved, it is convenient for scan text, sends retrieval machine for the text after scanning
After structure, the information of user's needs can be judged automatically out, convenient for searching according to the retrieval record of oneself and other people.
Detailed description of the invention
Fig. 1 is the flow diagram of the text information identification extracting method of the present invention based on machine learning；
Fig. 2 is the flow diagram that text identification of the present invention is extracted；
Fig. 3 is the flow diagram of the present invention for establishing model.
Specific embodiment
The present invention is specifically described with reference to the accompanying drawing, as shown in Figs. 1-3.
In the present embodiment, the first step, connection database: corresponding port is downloaded on mobile terminals, utilizes port
And be attached by wireless network and server terminal, corresponding database is established in server terminal；
Second step, text identification are extracted: will need the text extracted includes that photo, paper and other items are placed on the table, is used
Person, which picks up blank sheet of paper, blocks other useless information on text, is fixed using elastic webbing, crane is put
It sets above text, the adjustment of height is carried out according to the size of text, so that the text that square aperture alignment needs, text is swept
It retouches camera scanning end to be stretched out by square aperture, and pulley is fallen on slideway, after finding position, user presses lower switch,
Start textual scan camera and infrared laser lamp, laser lamp is radiated on text, and movetext scan camera shooting head, is led to
Crossing laser lamp can be shown that its scanning direction；
Third step, the selection result: intelligence system according to the usage record of user, can simulate the search habit of user, and
Automatic screening goes out specifying information required for user, and shows after these information are arranged.
Case study on implementation one,
In the text information on extracting wall or on big poster, taken pictures using smart phone to it, then in smart phone
On picture is zoomed in or out, make its text information be unlikely to be distorted, by smart phone place on the table, use blank sheet of paper
Position in addition to text information is subjected to folding gear, and is fixed, crane is placed on above smart phone, text is swept
It retouches camera to be placed on crane, opens infrared laser lamp, adjust the height of crane, determine scanning range, pin
Switch, movetext scan camera shooting head is scanned, and after scanned, trip switch is sent out text information by Bluetooth signal
It is sent in mobile terminal, after the intelligent search of mobile terminal, show that its is desired as a result, and according to pair of user's selection
Information is answered, usage record is subjected to upload preservation again.
Case study on implementation two,
When extracting the text information on books, since its text is smaller, directly textual scan camera can be attached on books,
Switch is pinned to be scanned, after scanned, trip switch is sent text information in mobile terminal by Bluetooth signal,
After the intelligent search of mobile terminal, obtain its it is desired as a result, and according to user choose corresponding informance, will make again
Upload preservation is carried out with record.
Case study on implementation three,
When extracting the text information on cylindrical object, first crane is set, height is adjusted according to the diameter of object, it will be literary
Word scan camera shooting head is placed on crane and fixes, and opens infrared laser lamp, makes on the outside of text information on cylindrical object
It is wrapped up with blank sheet of paper, text information is placed below textual scan camera, carried out according to the light source of infrared laser lamp transmitting
Positioning rotates cylindrical object, and pins switch, is scanned to it, after scanned, trip switch passes through Bluetooth signal
It sends text information in mobile terminal, after the intelligent search of mobile terminal, show that its is desired as a result, and according to making
The corresponding informance that user chooses, carries out upload preservation for usage record again.
Above-mentioned technical proposal only embodies the optimal technical scheme of technical solution of the present invention, those skilled in the art
The principle of the present invention is embodied to some variations that some of them part may be made, belongs to the scope of protection of the present invention it
It is interior.
Claims (5)
1. a kind of text information based on machine learning identifies extracting method, which is characterized in that specifically comprise the following steps:
Step 1: connection database: downloading corresponding port on mobile terminals, using port and pass through wireless network and service
Device terminal is attached, and corresponding database is established in server terminal；
Step 2: text identification is extracted: when the text on picture in photo, paper, mobile phone is carried out identification extraction, being placed
On the table, the part for removing text on text is blocked using blank sheet of paper, is fixed using two elastic webbings, by crane
It is placed on above text, textual scan camera lower end is stretched out by the square aperture on crane, by textual scan camera
Be directed at word segment, movetext scan camera shooting head and push switch are scanned, by Bluetooth signal by scan come text
Word is sent in mobile terminal, is identified；
Step 3: establishing model: different users' use is formed by historical record storage information and passed through wirelessly by mobile terminal
Network is sent on service terminal, carries out taxonomic revision to these information by the intelligent program on service terminal, and pass through machinery
Learning method establishes different models, so that the search habit of user is simulated, it, can when user searches again for certain information
Optimize it and search for target, improves the search efficiency of user；
Step 4: the selection result: by after Text region in step 2, intelligence system can be according to keyword message and simulation at
The use habit automatic screening of user, information required for user is shown, by user by required search
As a result it opens, and the usage history of user is recorded, and uploads again, constantly optimize.
2. a kind of text information based on machine learning according to claim 1 identifies extracting method, which is characterized in that institute
State text on the photo or paper in step 2 it is excessive when, picture can be clapped by mobile phone, be carried out again after being reduced
Scanning.
3. a kind of text information based on machine learning according to claim 1 identifies extracting method, which is characterized in that institute
The crane upper surface two sides stated in step 2 are equipped with slideway, and slideway is located at square aperture two sides, textual scan camera lower end
Two sides are equipped with connecting shaft, and connection shaft end is equipped with pulley, and pulley lower end is located in slideway.
4. a kind of text information based on machine learning according to claim 1 identifies extracting method, which is characterized in that institute
Stating the textual scan camera lower end in step 2 is wide-mouth camera lens, is equipped with infrared laser on the downside of textual scan camera front end
Lamp, infrared laser lamp Laser emission end are equipped with conical outlet, and tilt angle is consistent with the tilt angle of wide-mouth camera lens.
5. a kind of text information based on machine learning according to claim 1 identifies extracting method, which is characterized in that institute
The switch stated in step 2 is located at textual scan camera upper end, is push switch, starts after pressing, close after release.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811648483.0A CN109726711A (en) | 2018-12-30 | 2018-12-30 | A kind of text information identification extracting method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811648483.0A CN109726711A (en) | 2018-12-30 | 2018-12-30 | A kind of text information identification extracting method based on machine learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109726711A true CN109726711A (en) | 2019-05-07 |
Family
ID=66298107
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811648483.0A Pending CN109726711A (en) | 2018-12-30 | 2018-12-30 | A kind of text information identification extracting method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109726711A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110673778A (en) * | 2019-09-23 | 2020-01-10 | 联想(北京)有限公司 | Output control method and device, electronic equipment, terminal and server |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101582083A (en) * | 2008-05-15 | 2009-11-18 | 株式会社理光 | Web-based detection in image, extraction and recognition |
| CN105323488A (en) * | 2015-11-24 | 2016-02-10 | 成都九十度工业产品设计有限公司 | An intelligent camera device |
| CN108418993A (en) * | 2018-04-16 | 2018-08-17 | 安徽宝葫芦信息科技集团股份有限公司 | A kind of portable apparatus for scanning archives |
| CN208058336U (en) * | 2017-12-28 | 2018-11-06 | 智能投影谷（威海）科技有限公司 | A kind of video camera with rotary lifting frame |
| CN208273074U (en) * | 2018-06-06 | 2018-12-21 | 广州磐信计算机科技有限公司 | A kind of completely new image modalities |
-
2018
- 2018-12-30 CN CN201811648483.0A patent/CN109726711A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101582083A (en) * | 2008-05-15 | 2009-11-18 | 株式会社理光 | Web-based detection in image, extraction and recognition |
| CN105323488A (en) * | 2015-11-24 | 2016-02-10 | 成都九十度工业产品设计有限公司 | An intelligent camera device |
| CN208058336U (en) * | 2017-12-28 | 2018-11-06 | 智能投影谷（威海）科技有限公司 | A kind of video camera with rotary lifting frame |
| CN108418993A (en) * | 2018-04-16 | 2018-08-17 | 安徽宝葫芦信息科技集团股份有限公司 | A kind of portable apparatus for scanning archives |
| CN208273074U (en) * | 2018-06-06 | 2018-12-21 | 广州磐信计算机科技有限公司 | A kind of completely new image modalities |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110673778A (en) * | 2019-09-23 | 2020-01-10 | 联想(北京)有限公司 | Output control method and device, electronic equipment, terminal and server |
| CN110673778B (en) * | 2019-09-23 | 2021-11-16 | 联想(北京)有限公司 | Output control method and device, electronic equipment, terminal and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US8873853B2 (en) | Methods and systems for content processing | |
| JP6229656B2 (en) | Control device and storage medium | |
| US20050001024A1 (en) | Electronic apparatus, electronic camera, electronic device, image display apparatus, and image transmission system | |
| CN107358135A (en) | A kind of Quick Response Code barcode scanning method and device | |
| CN112702521B (en) | Image shooting method and device, electronic equipment and computer readable storage medium | |
| CN101609505B (en) | Method and apparatus for recognizing characters | |
| CN101057491A (en) | Wireless image capture device with biometric readers | |
| CN109035908B (en) | Interactive reading method | |
| CN108563702B (en) | Voice explanation data processing method and device based on exhibit image recognition | |
| JP4949924B2 (en) | Imaging system and imaging apparatus | |
| US10133932B2 (en) | Image processing apparatus, communication system, communication method and imaging device | |
| CN103795934A (en) | Image processing method and electronic device | |
| CN102201051A (en) | Text excerpting device, method and system | |
| WO2017166802A1 (en) | Method and device for classifying and storing photos, and mobile terminal | |
| CN109726711A (en) | A kind of text information identification extracting method based on machine learning | |
| CN106534608A (en) | Foldable authentication high-speed image shooting instrument | |
| CN110378289B (en) | Reading and identifying system and method for vehicle identification code | |
| CN108171231A (en) | A kind of communication means and device based on image identification | |
| CN109377795A (en) | Learning interaction method of intelligent equipment and intelligent equipment | |
| CN109377834A (en) | A kind of text conversion method and system of helping blind people read | |
| CN107438156A (en) | A kind of image pickup method and mobile terminal | |
| CN105898137A (en) | Image collection and information push methods, image collection and information push devices and mobile phone | |
| CN105320242A (en) | Photographing method and photographing terminal | |
| CN206684750U (en) | A kind of image recognition and information query system based on exhibition activity | |
| CN103605687A (en) | Photographing and image recognizing system and method of mobile terminal |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190507 |