CN101770569A - Dish name recognition method based on OCR - Google Patents
Dish name recognition method based on OCR Download PDFInfo
- Publication number
- CN101770569A CN101770569A CN200810246630A CN200810246630A CN101770569A CN 101770569 A CN101770569 A CN 101770569A CN 200810246630 A CN200810246630 A CN 200810246630A CN 200810246630 A CN200810246630 A CN 200810246630A CN 101770569 A CN101770569 A CN 101770569A
- Authority
- CN
- China
- Prior art keywords
- dish
- name
- identified
- ocr
- recognition
- 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
Images
Landscapes
- Character Discrimination (AREA)
- Character Input (AREA)
Abstract
The invention provides a dish name recognition method based on OCR, which belongs to the technical field of OCR. The method comprises the following steps: taking a photo of a menu; lining a region to be recognized on the menu photo conforming to the definition requirement; carrying out intelligent regulation on the region to be recognized; and using an OCR engine for recognizing the regulated recognized region to obtain recognition texts. The method utilizes a dish name knowledge base for correcting the recognition texts, selects results with the maximum matching degree as dish name recognition results, and filters out meaningless recognition results. The rectangular region to be recognized is determined according to specified coordinate points, and the intelligent regulation is carried out on the basis, so the noise in the recognition process is filtered out to the maximum degree, and the recognition effect of the OCR engine is improved. Users can upload the recognition results to the web pages of catering industry through network for propaganda and reservation after the web pages issue more dish name information, and at the same time, the invention also brings convenience for lots of gastronomists to know and select the respective delicacies.
Description
Technical field
The invention belongs to the OCR technical field, relate to a kind of recognition methods, relate in particular to a kind of name of the dish recognition methods based on OCR based on OCR.
Background technology
The OCR technology is the abbreviation (Optical Character Recognition) of optical character identification, be the literal of various books, manuscript, newpapers and periodicals, bill and other printed matter to be converted into image information, and then utilize character recognition technology image information to be converted into the computer input technology of the literal code character stream that can edit by optics input modes such as scanning, shootings.Can be applicable to the typing and the bank money process field of a large amount of written historical materialss, archives folder, official documents and correspondence.
Shooting mobile phone has been popularized now, but it is still very poor to be based on the application of camera, has not given play to due effect.The OCR technology is very ripe at present, but it uses the identification that still is confined to the scanner image, though using to some extent on present high-end smartphones, such as Samsung SCH-i819 and LG G832 etc., though these mobile phones can be discerned at business card etc. by camera, and recognition result stored in the address list, but also only be simple text identification, also need proofread and correct recognition result by manual.
Along with the development of catering trade and the quickening of rhythm of life, issue the trend that catering information becomes a kind of catering trade on the net, but need be published to a large amount of dish information on the net, no small burden is provided for the website worker.In addition, client at mealtime, foreign friend particularly, often wish to note the relevant information of some dish interested, but because the singularity of relevant informations such as name of the dish, present smart mobile phone is not fine for the effect of name of the dish information Recognition, often exists than mistake.
Summary of the invention
The invention provides a kind of name of the dish recognition methods based on OCR, by the camera on the mobile phone name of the dish information on the menu is taken, and by user's appointment intelligence is carried out in the name of the dish zone and locate, then the OCR recognition engine is discerned this zone, call the name of the dish knowledge base then and carry out verification, obtain correct name of the dish information.Name of the dish zone accurate and effective by user's appointment, and adopt the name of the dish knowledge base that OCR identification text is carried out verification, improved the accuracy of name of the dish recognition result greatly, can discern name of the dish information effectively, made things convenient for the collection of user to name of the dish information, and do not increase any hardware device, and reduced cost, have excellent application value and marketable value.
Name of the dish recognition methods based on OCR comprises the steps:
Step 1: the camera by mobile phone is taken menu, obtains menu image;
Step 2: menu image is carried out rim detection,, then again menu is taken if image detection is a blurred picture; Otherwise execution in step 3;
Step 3: meeting delimitation zone to be identified on the menu image of sharpness, and treating identified region and carry out intelligence adjustment;
Step 4: call the OCR recognition engine adjusted zone to be identified is discerned, obtain discerning text;
Step 5: the identification text is proofreaied and correct, obtained the name of the dish recognition result.
In the rim detection of described step 2, menu image being asked edge gradient, and all edge gradient values are averaged, if this average gradient value is then thought picture rich in detail greater than setting threshold, otherwise is blurred picture.
When described step 3 delimited zone to be identified, specify one or two coordinate points, or draw straight line, determine a rectangular area roughly according to the position of name of the dish to be identified at the middle part of name of the dish to be identified.
When described step 3 is determined the rectangular area,, to expansion all around, obtain comprising the circumscribed rectangular region of name of the dish to be identified with this coordinate points position initial point if specify a coordinate points; If when specifying two coordinate points, be angle point to be determined to comprise the circumscribed rectangular region of name of the dish to be identified with coordinate points; If draw straight line at the middle part of name of the dish to be identified, this straight line is expanded to both sides, what obtain comprising name of the dish to be identified is the circumscribed rectangular region of the length of side with straight length and extended amplitude sum.
When described step 3 intelligence is adjusted, treat the border of identified region and adjust, reduce name of the dish to be identified white space on every side, the zone to be identified that closely comprises name of the dish that obtains.
Described step 5 identification text carries out timing, to discern text and in the name of the dish knowledge base, carry out fuzzy matching, choosing the highest name of the dish of matching degree and confirm for the user, is final recognition result if the user confirms this matching result, then this recognition result is stored or Web publishing; Otherwise recognition result is stored and Web publishing after by the manual editing, is added to simultaneously in the name of the dish knowledge base.
Described step 5 identification text carries out timing, and the name of the dish knowledge base is supported multilingual.
The present invention is based on the name of the dish recognition methods of OCR, advantage compared with prior art is:
1, this method is when treating identified region and discern, at first formulated coordinate points, determine the zone to be identified of rectangle according to specified coordinate points, and carry out intelligence on this basis and adjust, effectively like this determined zone to be identified, filtering to the full extent the noise in the identifying, improved the discrimination of OCR engine.
2, this method is proofreaied and correct for the identification text, and the result who chooses the matching degree maximum in the name of the dish knowledge base has improved the accuracy of identification greatly as the result of name of the dish identification, filtering insignificant recognition result, convenient for users to use.
3, the new application direction of a kind of OCR has been proposed, earlier carry out the menu image collection with shooting mobile phone, then the OCR recognition engine is known name of the dish, utilize knowledge base that the identification text is proofreaied and correct then, the user can upload on the website of catering trade to recognition result by network, to propagate after the more dish name information issue or make a reservation for the website, also convenient simultaneously vast cuisines fan understands and selects their delicacies.
Description of drawings
Fig. 1 is the process flow diagram that the present invention is based on the name of the dish recognition methods of OCR;
Fig. 2 the present invention is based on the synoptic diagram that coordinate points of formulation of the name of the dish recognition methods of OCR delimited zone to be identified;
Fig. 3 the present invention is based on the synoptic diagram that two coordinate points of formulation of the name of the dish recognition methods of OCR delimited zone to be identified;
Fig. 4 the present invention is based on the synoptic diagram that the formulation straight line of the name of the dish recognition methods of OCR delimited zone to be identified;
Fig. 5 is the synoptic diagram through the adjusted zone to be identified of intelligence that the present invention is based on the name of the dish recognition methods of OCR;
Fig. 6 is the identification text of OCR engine output that the present invention is based on the name of the dish recognition methods of OCR;
Fig. 7 the present invention is based on the recognition result that obtains after the correction of name of the dish recognition methods of OCR.
Embodiment
Below in conjunction with accompanying drawing method of the present invention is elaborated.
Mobile phone in the present embodiment is equipped with the camera head and the OCR recognition engine of 5,000,000 pixels, and the SDK of employing is Chinese king OCR6.0 SDK, and can surf the Net by the WAP mode.
This method as shown in Figure 1, comprises the steps: based on the name of the dish recognition methods of OCR
Step 1: the camera by mobile phone is taken menu, obtains the menu photo, as pending menu image;
Step 2: by the photo that mobile phone cam is taken, sometimes because shake will cause image very empty, character is fuzzy, does not reach the requirement of OCR identification.So the photo of taking is carried out the judgement of sharpness, is for further processing determining whether.The average edge strength of menu image as detected value, is carried out rim detection, when average edge strength during less than pre-set threshold T, image blurring, need take again, greater than pre-set threshold T, judge till the clear picture until the average edge strength of image.Average edge strength described here be meant by edge algorithms such as Sobel, Roberts etc. calculate to Grad, employing Sobel algorithm in the present embodiment.As shown in Figure 2, preset threshold is T=60, and the average edge strength of this image is 100, and greater than certain threshold level, this image is a picture rich in detail.
Step 3: on the menu image that meets the sharpness requirement, delimit the zone to be identified of a rectangle, specify the rough position of name of the dish to be identified.In the present embodiment, specify a coordinate points, as shown in Figure 2, this coordinate points is a cross-shaped cursor, is initial point left and right sides four direction expansion up and down respectively with this coordinate points, obtains comprising the circumscribed rectangular region of name of the dish to be identified.The tracking cross that is positioned at " meat sauce face " printed words place among Fig. 2 is exactly user's specified coordinate point.
If when specifying two coordinate points, as shown in Figure 3, be angle point to be determined to comprise the circumscribed rectangular region of name of the dish to be identified with two coordinate points.The border of same this rectangular area is outside name of the dish to be identified, and promptly this rectangular area must comprise name of the dish fully, obtains comprising the circumscribed rectangular region of name of the dish to be identified.The rectangular area that is positioned at user's appointment that " meat sauce face " locate among Fig. 3 is exactly the rectangular areas of angle point being determined by two, and this name of the dish " meat sauce face " is exactly a name of the dish to be identified.
If straight line is drawn at the middle part at name of the dish image to be identified, this rectilinear direction is parallel with the character arranging direction.The rectilinear direction that in the present embodiment, as shown in Figure 4, runs through " meat sauce face " is parallel with the character arranging direction, and to its both sides, i.e. above-below direction expansion is till white space with straight line.Finally obtain comprising the circumscribed rectangular region of name of the dish to be identified, this circumscribed rectangular region is long with straight length, is wide with extended amplitude sum up and down.
In the time of user interactions, can not navigate to the exact position that needs to handle name of the dish, just provide one or two coordinate points.By these coordinate points, navigate to the name of the dish character zone that needs processing intelligently.Behind the selected zone to be identified, treat identified region and carry out intelligence adjustment, adjust the border of this rectangular area, reduce name of the dish to be identified white space on every side, obtain a zone to be identified that closely comprises name of the dish to be identified.
For the situation of the given point coordinate of user in the present embodiment, obtain rectangular area to be identified as shown in Figure 5 through intelligence adjustment.
Step 4: call optical character identification (OCR) engine modules,, send into the OCR recognition engine and handle, treat identified region and carry out character segmentation, carry out the coupling of monocase feature extraction and template then with the name of the dish character image zone that navigates to; The final identification text (but editing character stream) that forms, the identification text that obtains as shown in Figure 6, the recognition result that obtains is " a meat stir-fry department face ".
Step 5: identification text correction.Carry out fuzzy matching in the name of the dish knowledge base with identification text and the storage of this machine, find the name of the dish information relevant with discerning text.Choosing the name of the dish of matching degree the highest (if a plurality of identical matching degrees are arranged, then all listing) and confirm for the user, is final recognition result if the user confirms this matching result, then this recognition result is stored or Web publishing.Otherwise the user finds that the result of mating is incorrect, and the identification text is wrong, and then the user need carry out the manual editing to the OCR recognition result, stores then or Web publishing, and Automatic Program is increased to this name of the dish in the name of the dish knowledge base simultaneously.In the present embodiment, as shown in Figure 7, the name of the dish that matching degree is the highest is " a meat sauce face ", and matching degree is 0.98.The user thinks that recognition result is correct, and this result is uploaded in the specific food and drink website by the WAP module on the mobile phone, and as the issue of the information of website, the business of grade is used for propagating or makes a reservation.
Described knowledge Kuku is also supported multilingual, and the user can select recognition result is converted to the linguistic form output that mobile phone is provided with.If the language in the user mobile phone is English, then will discerns text " meat sauce face " and will be converted into English " Meat-sauce Spaghetti " and show.
Though part preferred embodiment of the present invention only has been described here, its meaning be not limit the scope of the invention, applicability and configuration.On the contrary, those skilled in the art are implemented, and will be understood that, can make suitable change and modification some details not departing under the spirit and scope of the invention situation that appended claims determines to the detailed description of embodiment.
Claims (7)
1. the name of the dish recognition methods based on OCR is characterized in that, comprises the steps:
Step 1: the camera by mobile phone is taken menu, obtains menu image;
Step 2: menu image is carried out rim detection,, then again menu is taken if image detection is a blurred picture; Otherwise execution in step three;
Step 3: meeting delimitation zone to be identified on the menu image of sharpness, and treating identified region and carry out intelligence adjustment;
Step 4: call the OCR recognition engine adjusted zone to be identified is discerned, obtain discerning text;
Step 5: the identification text is proofreaied and correct, obtained the name of the dish recognition result.
2. according to the described a kind of name of the dish recognition methods of claim 1 based on OCR, it is characterized in that: in the rim detection of described step 2, menu image is asked edge gradient, and all edge gradient values are averaged, if this average gradient value is greater than setting threshold, then think picture rich in detail, otherwise be blurred picture.
3. according to the described a kind of name of the dish recognition methods of claim 1 based on OCR, it is characterized in that: when described step 3 delimited zone to be identified, specify one or two coordinate points, or draw a straight line, determine a rectangular area roughly according to the position of name of the dish to be identified at the middle part of name of the dish to be identified.
4. according to the described a kind of name of the dish recognition methods of claim 3 based on OCR, it is characterized in that: when described step 3 is determined the rectangular area, if specify a coordinate points, to expansion all around, obtain comprising the circumscribed rectangular region of name of the dish to be identified with this coordinate points position initial point; If when specifying two coordinate points, be angle point to be determined to comprise the circumscribed rectangular region of name of the dish to be identified with coordinate points; If draw a straight line at the middle part of name of the dish to be identified, this straight line is expanded to both sides, what obtain comprising name of the dish to be identified is the circumscribed rectangular region of the length of side with straight length and extended amplitude sum.
5. according to the described a kind of name of the dish recognition methods of claim 1 based on OCR, it is characterized in that: when described step 3 intelligence is adjusted, treat the border of identified region and adjust, reduce name of the dish to be identified white space on every side, the zone to be identified that closely comprises name of the dish that obtains.
6. according to the described a kind of name of the dish recognition methods of claim 1 based on OCR, it is characterized in that: described step 5 utilizes the name of the dish knowledge base that the identification text is carried out timing, to discern text and in the name of the dish knowledge base, carry out fuzzy matching, choosing the highest name of the dish of matching degree confirms for the user, if it is final recognition result that the user confirms this matching result, then this recognition result is stored or Web publishing; Otherwise recognition result is stored and Web publishing after by the manual editing, is added to simultaneously in the name of the dish knowledge base.
7. according to the described a kind of name of the dish recognition methods based on OCR of claim 1, it is characterized in that: described step 5 identification text carries out timing, and the name of the dish knowledge base is supported multilingual.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810246630A CN101770569A (en) | 2008-12-31 | 2008-12-31 | Dish name recognition method based on OCR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200810246630A CN101770569A (en) | 2008-12-31 | 2008-12-31 | Dish name recognition method based on OCR |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101770569A true CN101770569A (en) | 2010-07-07 |
Family
ID=42503421
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200810246630A Pending CN101770569A (en) | 2008-12-31 | 2008-12-31 | Dish name recognition method based on OCR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101770569A (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102354365A (en) * | 2011-09-23 | 2012-02-15 | 上海合合信息科技发展有限公司 | Recognition method and system for accurately recognizing images |
| CN102968638A (en) * | 2011-08-31 | 2013-03-13 | 上海夏尔软件有限公司 | Image sharpness judgment method based on keyword optical character recognition |
| CN103488983A (en) * | 2013-09-13 | 2014-01-01 | 复旦大学 | Business card OCR data correction method and system based on knowledge base |
| CN103854024A (en) * | 2012-12-04 | 2014-06-11 | 百度国际科技（深圳）有限公司 | Method and device for extracting characters in image |
| CN103927552A (en) * | 2014-04-23 | 2014-07-16 | 北京奇虎科技有限公司 | Method and device for matching answers of target test questions |
| CN103955525A (en) * | 2014-05-09 | 2014-07-30 | 北京奇虎科技有限公司 | Method and client for searching answer to test question |
| CN104346611A (en) * | 2013-08-06 | 2015-02-11 | 富士施乐株式会社 | Information processing apparatus and information processing method |
| CN105373790A (en) * | 2015-10-23 | 2016-03-02 | 北京汉王数字科技有限公司 | Layout analysis method and device |
| CN107609914A (en) * | 2017-09-22 | 2018-01-19 | 上海爱优威软件开发有限公司 | A kind of intelligent menu design method and system |
| CN107730402A (en) * | 2017-10-13 | 2018-02-23 | 李继泽 | Intelligent ordering system and method |
| CN108875697A (en) * | 2018-07-05 | 2018-11-23 | 南昌市微轲联信息技术有限公司 | Collecting vehicle information method for uploading, device, storage medium and computer equipment |
| 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 |
| CN109740597A (en) * | 2018-12-29 | 2019-05-10 | 口碑(上海)信息技术有限公司 | The recognition methods of menu information and device |
| CN109886263A (en) * | 2018-12-29 | 2019-06-14 | 口碑(上海)信息技术有限公司 | Menu configuration method, device, storage medium and computer equipment |
| CN109918416A (en) * | 2019-02-28 | 2019-06-21 | 生活空间(沈阳)数据技术服务有限公司 | A kind of method, device and equipment of document entry |
| CN110110577A (en) * | 2019-01-22 | 2019-08-09 | 口碑(上海)信息技术有限公司 | Identify method and device, the storage medium, electronic device of name of the dish |
| CN110942057A (en) * | 2018-09-25 | 2020-03-31 | 杭州海康威视数字技术股份有限公司 | Container number identification method and device and computer equipment |
| CN111192346A (en) * | 2019-12-26 | 2020-05-22 | 科大讯飞股份有限公司 | Electronic menu generation method, device and equipment |
| CN111275134A (en) * | 2020-02-26 | 2020-06-12 | 上海明略人工智能(集团)有限公司 | Image matching method and device, storage medium and electronic device |
| CN111385424A (en) * | 2018-12-25 | 2020-07-07 | 佳能株式会社 | Image processing system and image processing method |
| CN111898612A (en) * | 2020-06-30 | 2020-11-06 | 北京来也网络科技有限公司 | OCR recognition method and device combining RPA and AI, equipment and medium |
| CN112801232A (en) * | 2021-04-09 | 2021-05-14 | 苏州艾隆科技股份有限公司 | Scanning identification method and system applied to prescription entry |
| CN114022881A (en) * | 2021-10-24 | 2022-02-08 | 何雨萌 | AI + AR article counting system and article counting method |
| CN115439854A (en) * | 2022-09-05 | 2022-12-06 | 深圳市学之友科技有限公司 | Scanning display method based on interconnection of scanning pen and intelligent terminal |
-
2008
- 2008-12-31 CN CN200810246630A patent/CN101770569A/en active Pending
Cited By (29)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102968638A (en) * | 2011-08-31 | 2013-03-13 | 上海夏尔软件有限公司 | Image sharpness judgment method based on keyword optical character recognition |
| CN102354365A (en) * | 2011-09-23 | 2012-02-15 | 上海合合信息科技发展有限公司 | Recognition method and system for accurately recognizing images |
| CN103854024A (en) * | 2012-12-04 | 2014-06-11 | 百度国际科技（深圳）有限公司 | Method and device for extracting characters in image |
| CN104346611A (en) * | 2013-08-06 | 2015-02-11 | 富士施乐株式会社 | Information processing apparatus and information processing method |
| CN103488983B (en) * | 2013-09-13 | 2016-10-26 | 复旦大学 | The business card OCR data correcting method in a kind of knowledge based storehouse and system |
| CN103488983A (en) * | 2013-09-13 | 2014-01-01 | 复旦大学 | Business card OCR data correction method and system based on knowledge base |
| CN103927552A (en) * | 2014-04-23 | 2014-07-16 | 北京奇虎科技有限公司 | Method and device for matching answers of target test questions |
| CN103955525A (en) * | 2014-05-09 | 2014-07-30 | 北京奇虎科技有限公司 | Method and client for searching answer to test question |
| CN105373790A (en) * | 2015-10-23 | 2016-03-02 | 北京汉王数字科技有限公司 | Layout analysis method and device |
| CN105373790B (en) * | 2015-10-23 | 2019-02-05 | 北京汉王数字科技有限公司 | Printed page analysis method and apparatus |
| CN107609914A (en) * | 2017-09-22 | 2018-01-19 | 上海爱优威软件开发有限公司 | A kind of intelligent menu design method and system |
| CN107730402A (en) * | 2017-10-13 | 2018-02-23 | 李继泽 | Intelligent ordering system and method |
| CN108875697A (en) * | 2018-07-05 | 2018-11-23 | 南昌市微轲联信息技术有限公司 | Collecting vehicle information method for uploading, device, storage medium and computer equipment |
| 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 |
| CN110942057A (en) * | 2018-09-25 | 2020-03-31 | 杭州海康威视数字技术股份有限公司 | Container number identification method and device and computer equipment |
| CN111385424A (en) * | 2018-12-25 | 2020-07-07 | 佳能株式会社 | Image processing system and image processing method |
| US11216695B2 (en) | 2018-12-25 | 2022-01-04 | Canon Kabushiki Kaisha | Image processing system and image processing method |
| CN109886263B (en) * | 2018-12-29 | 2020-12-29 | 口碑(上海)信息技术有限公司 | Menu configuration method and device, storage medium and computer equipment |
| CN109886263A (en) * | 2018-12-29 | 2019-06-14 | 口碑(上海)信息技术有限公司 | Menu configuration method, device, storage medium and computer equipment |
| CN109740597A (en) * | 2018-12-29 | 2019-05-10 | 口碑(上海)信息技术有限公司 | The recognition methods of menu information and device |
| CN110110577A (en) * | 2019-01-22 | 2019-08-09 | 口碑(上海)信息技术有限公司 | Identify method and device, the storage medium, electronic device of name of the dish |
| CN109918416A (en) * | 2019-02-28 | 2019-06-21 | 生活空间(沈阳)数据技术服务有限公司 | A kind of method, device and equipment of document entry |
| CN111192346A (en) * | 2019-12-26 | 2020-05-22 | 科大讯飞股份有限公司 | Electronic menu generation method, device and equipment |
| CN111192346B (en) * | 2019-12-26 | 2024-02-13 | 科大讯飞股份有限公司 | Electronic menu generation method, device and equipment |
| CN111275134A (en) * | 2020-02-26 | 2020-06-12 | 上海明略人工智能(集团)有限公司 | Image matching method and device, storage medium and electronic device |
| CN111898612A (en) * | 2020-06-30 | 2020-11-06 | 北京来也网络科技有限公司 | OCR recognition method and device combining RPA and AI, equipment and medium |
| CN112801232A (en) * | 2021-04-09 | 2021-05-14 | 苏州艾隆科技股份有限公司 | Scanning identification method and system applied to prescription entry |
| CN114022881A (en) * | 2021-10-24 | 2022-02-08 | 何雨萌 | AI + AR article counting system and article counting method |
| CN115439854A (en) * | 2022-09-05 | 2022-12-06 | 深圳市学之友科技有限公司 | Scanning display method based on interconnection of scanning pen and intelligent terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101770569A (en) | Dish name recognition method based on OCR | |
| US10108860B2 (en) | Systems and methods for generating composite images of long documents using mobile video data | |
| US7729536B2 (en) | Boundary extracting method, program, and device using the same | |
| US8155442B2 (en) | Method and apparatus for modifying the histogram of an image | |
| KR101499379B1 (en) | System and method for recognizing text information in object | |
| Yang et al. | Binarization of low-quality barcode images captured by mobile phones using local window of adaptive location and size | |
| RU2631765C1 (en) | Method and system of correcting perspective distortions in images occupying double-page spread | |
| CN101751566B (en) | Method and device for identifying and annotating menu based on handheld device | |
| US20130251262A1 (en) | Text image trimming method | |
| CN107085699B (en) | Information processing apparatus, control method of information processing apparatus, and storage medium | |
| Wang et al. | An effective background estimation method for shadows removal of document images | |
| GB2517674A (en) | Image capture using client device | |
| CN103955499A (en) | Visual experience enhancing method based on instant calculation and dynamic tracing | |
| CN104765796A (en) | Image recognizing searching method and device | |
| WO2022111461A1 (en) | Recognition method and apparatus, and electronic device | |
| CN104408427A (en) | Method and device for recognizing quadrangle of image | |
| CN111669492A (en) | Method for processing shot digital image by terminal and terminal | |
| US10373329B2 (en) | Information processing apparatus, information processing method and storage medium for determining an image to be subjected to a character recognition processing | |
| CN109685079B (en) | Method and device for generating characteristic image category information | |
| Najman et al. | Indexing technical drawings using title block structure recognition | |
| CN112446297A (en) | Electronic typoscope and intelligent mobile phone text auxiliary reading method applicable to same | |
| Tam et al. | Quadrilateral Signboard Detection and Text Extraction. | |
| Chazalon et al. | Improving document matching performance by local descriptor filtering | |
| Bertucci et al. | Text selection by structured light marking for hand-held cameras | |
| Burian et al. | Document image binarization using the camera device in mobile phones |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20100707 |