WO2015143471A1 - Verfahren zur optischen erkennung von zeichen - Google Patents
Verfahren zur optischen erkennung von zeichen Download PDFInfo
- Publication number
- WO2015143471A1 WO2015143471A1 PCT/AT2015/050080 AT2015050080W WO2015143471A1 WO 2015143471 A1 WO2015143471 A1 WO 2015143471A1 AT 2015050080 W AT2015050080 W AT 2015050080W WO 2015143471 A1 WO2015143471 A1 WO 2015143471A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- image data
- display
- recognized
- characters
- outline
- 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.)
- Ceased
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Definitions
- the invention relates to a method for the optical recognition of characters displayed on a display, wherein image data is obtained, which contain a representation of the display, as well as a mobile terminal with an image sensor for receiving the image data and with a processor and a
- Character recognition method a measured value in the image data taken so ⁇ identified, translated into machine-coded text and transmitted to a data acquisition device.
- a preferred application of the invention is the acquisition, detection and processing of blood glucose readings from a blood glucose meter.
- a blood glucose meter For this purpose, apart from a manual transmission of the displayed data, sometimes cable-based solutions are marketed and used.
- the QR code is significantly easier to recognize electronically compared to human-readable characters because it has special recognition features for this purpose, which also allow for unambiguous alignment and equalization of the code.
- a QR code usually contains redundant information, so that some bad or unreadable parts of the Readability of the code as a whole does not affect.
- a completely wireless data transmission is only possible with the most expensive measuring devices, which for example offer a Bluetooth interface for data transmission. The vast majority of blood glucose meters, however, has no such interface .
- the invention provides a mobile terminal with an image sensor for receiving the image data and a
- An advantageous method for detecting an outline represented in the image data is that at least one contiguous sequence of adjacent pixels with the same or similar color value is recognized as an outline.
- the image data is first searched on a pixel not yet assigned to an outline and then, starting from this pixel, the color values of the adjacent pixels are compared in sequence until a pixel with the same or a similar color value is found, and then becomes with this pixel continues until no adjacent pixel can be found that is not already part of the outline.
- Pixels of the number of corner points of the display corresponds to be recognized as an outline of the display area.
- other or additional features may also be used in the recognition, such as e.g. an aspect ratio of the display.
- the number of vertices i. usually four vertices, allows a sufficiently good
- Outlines with a larger enclosed area which could be confused with the display may e.g. be counteracted by optical focusing on objects at a distance of a maximum of 1 m.
- a histogram of a horizontal projection of the color values of the processed image data can advantageously be generated for optical character recognition and the histogram can be used to generate the histogram
- the Recognition of the signs is based on the recognition that the vertical segments - in analogy to the above explanation for the horizontal segments - cause local maxima in the vertical projection that are detected and used to determine the
- the respectively displayed character can be determined from the color values of a portion of the processed image data corresponding to the character, preferably determining the states of a plurality of segments representing a character and from these states a displayed character.
- This method provides an efficient alternative to comparing the image data with all the possible states of the segment display corresponding comparison images. Instead, the knowledge about the positions of the individual segments within a character is used to separately represent, for each segment, the one represented in the image data
- Display area is not recognized, for example, because no outline is detected or none of the detected outlines the properties of the display, can be discarded before the character recognition and even before the preparation, so that the available resources not with a superfluous preparation or a probably futile - and so that particularly complicated - character recognition are unnecessarily burdened.
- the user does not incur any additional operating effort due to this over ⁇ jumping of individual images. It only slightly increases the duration of the recording, which is not or barely perceptible in a few discarded single images.
- FIG. 2 is a schematic block diagram of a mobile terminal and a measuring device according to FIG. 1;
- FIG. 2 is a schematic block diagram of a mobile terminal and a measuring device according to FIG. 1;
- FIG. 4 shows in more detail a method section for loading and converting acquired image data according to item IV in FIG. 3;
- Fig. 8 is a flowchart of a subroutine for simplifying an outline of Fig. 7;
- FIG. 11 shows a method section for identifying an identified character according to the number XI in FIG. 3.
- Fig. 1 illustrates an application situation in which a user uses a mobile terminal 1, to detect a on a display 2 of a measuring device 3, such as a blood glucose measuring device, readout and wear on the mobile terminal 1 to about ⁇ .
- a measuring device 3 such as a blood glucose measuring device
- an image sensor (not shown) of the mobile terminal 1 on the display 2 of the blood sugar measuring device 3 ge ⁇ directed.
- the mobile terminal 1 is set up, a method for optically recognizing displayed on the display 2
- the mobile terminal 1 in turn likewise has a display 8 on which the image data 9 obtained by means of the image sensor ⁇ which contain a representation of the display 2 are displayed. Specifically, a cutout 10 of the image data 9, which represents a detected ⁇ display area of the display 2, in color on the display 8 of the mobile terminal 1, for example, highlighted. In this way, the mobile terminal 1 signals the user to successfully recognize the display area of the display 2 of the measuring device 3.
- FIG. 2 shows a schematic block diagram of the mobile terminal 1 and of the measuring device 3 with the display 2.
- the mobile terminal 1 has an image sensor 11, a processor 12 and a memory 13.
- the processor 12 is configured to execute processing modules 14-19 and connected to both the image sensor 11 and the memory 13.
- the processing modules 14-19 comprise an edge detection module 14, a display extractor module 15, a line identification module 16, a character identification module 17, a character recognition module 18 and a comparison module 19.
- the edge detection module 14 is connected to the image sensor 11 and configured to receive the image sensor 11 for recording and To instruct transmission of image data and to recognize any edges represented in the thus obtained image data (see Fig. 6).
- the display extractor module 15 is set up to process the edges and outlines recognized by the edge recognition module 14 and to therefrom the edges or the outline of a display area represented in the image data and to extract the image data representing the recognized display area.
- the Displayextraktormodul 15 is adapted to process the extracted image data by correcting, for example, distortion and color values normali ⁇ Siert be.
- the 15 lines connected to the Displayextraktormodul identification module 16 is adapted for processing the sto ⁇ ready before image data, represented in the sto any ⁇ ready before image data lines of a text display are identified.
- semantic checks can also be carried out in the comparison processor 19, for example with regard to a possible range of values of measured values or with regard to a chronological order of recognized dates.
- the recognized and checked display text is then passed to one or more other application modules 20, which include, for example, a recording and management of the recognized data and a preparation for evaluation by the user.
- the processing modules 14-19 executed on the processor 12 access the memory 13 or databases 21-23 stored therein in order to perform their function.
- the processor 12 access the memory 13 or databases 21-23 stored therein in order to perform their function.
- Memory 13 general configuration parameters 21 for the processing modules 14-19 and a meter memory 22, which specific configuration parameters for one or more types of meters, such as an aspect ratio of the display 2 of jewei ⁇ term meter 3 or information on the semantic meaning one or more display lines.
- the already recognized display texts for checking or for further transmission in the value memory 23 can be temporarily stored.
- FIG. 3 The present method for optically recognizing a display content will now be described in overview with reference to the flow chart shown in FIG. 3, and in detail with reference to the more detailed flow charts shown in FIGS. 4-11, which illustrate the operations shown in FIG. A start (beginning 24) of the implementation of the present method may e.g. by the user through interaction with a mobile
- Method image data of a color image 26 are first obtained or loaded by an image sensor 11 and converted into a black and white image 27. On the basis of the black and white image 27 are then represented in the image data outlines 28 im
- one of the outlines 28 is selected as the display outline 31, and thus a display area of the display is recognized in the acquired image data.
- Display Outline 31 is known, corresponding to the display outline 31 image data of the color image 26, which thus represent the recognized display area in a
- a charging process 37 is executed at the beginning 24 of the present process initially, which activates, for example, an image sensor 11 and a measured by the image sensor 11 or captured Color 26 loads for further proces ⁇ processing.
- a black-and-white image 27 corresponding to the color image 26 is subsequently calculated in a conversion process 38 (see FIG.
- the Conversion ⁇ process 38 lies a pre-determined threshold 39 based on which, for example, from the configuration Para ⁇ meters 21 or - if the type of the depicted in the color image 26 meter 3 is known, for example, because it has been pre-entered manually by the user - from the device memory 22 is loaded.
- the process continues at the connection point A (see Fig. 6).
- a grayscale copy 41 is first of all calculated from the image data obtained, which correspond, for example, to the color image 26 or a section thereof in the context of a gray scale conversion 40.
- the gray level conversion 40 takes place in a manner known per se, for example by complete desaturation of the color data or by conversion of the individual color pixels of the image data to a corresponding brightness value (also called darkness level).
- an average pixel value of the pixel of the gray-scale copy 41 surrounding the current pixel is first calculated in a calculation step 46 in accordance with a block size 47 loaded from the configuration parameters 21.
- the average pixel value thus obtained is with a threshold 39 (see Fig. 4) or 48 (see Fig. 9) passed to the conversion.
- a threshold 39 see Fig. 4
- 48 see Fig. 9
- an assignment is made as black pixel 49 or white pixel 50 in the black and white copy 42.
- FIG. 6 shows the sequence of the contour recognition 29.
- a new outline is initialized at the location of the current pixel, the initialization 54 caches the Benach ⁇ showed white pixels as a starting point for the search 55 for a neighboring black pixels. If the above conditions are not satisfied, it is determined in a further checking step 56 whether a new outline starts at the pixel adjacent to the left. This is the case if the current pixel is assigned the color value white and the pixel adjacent to the left is assigned the value black, and if the pixel adjacent to the left is not part of an already recognized outline. If these three conditions are met, an initialization 57 of a new outline starts from the left pixel adjacent to the current pixel, the current pixel being used as the starting point for the subsequent search 55
- Area of all squares 62 is calculated and that square with the largest surface area is recognized as the display outline 31.
- the contour is divided at this pixel and the method recursively for each of the two parts, that is, for a portion of the outline from the first pixel to the image ⁇ point with the greatest distance, and for a second portion of this pixel to the last pixel, recursively repeated.
- the division 70 has led to a number of sections in which each pixel within the individual sections lies within the limit value ⁇ , all remaining pixels of the outline to be simplified, ie all those pixels which do not delimit at least one section, are deleted, s , Block 71, with the remaining pixels forming the simplified outline.
- the conversion 38 takes place analogously to FIG. 4 in accordance with the method illustrated in FIG. 5, wherein instead of the fixed threshold value 39, the calculated threshold value 48 is used.
- the black and white cutout 75 thus corresponds to one
- FIG. 10 shows the identification 34 (see FIG. 3) of the lines and of the characters forming the lines of the display 2.
- a calculation of a horizontalproduction profile of the black screen is made.
- White-section 75 completed.
- the horizontal projection profile essentially corresponds to a histogram of the proportion of black pixels in each pixel row of the black-and-white detail 75.
- the first derivative of the histogram is calculated, and the upper edges and lower edges respectively are calculated each line based on the local maxima or minima in the derived horizontal projection profile.
- an identification of the characters contained therein follows a similar procedure, wherein first a vertical
- Projection profile is calculated, block 78, and from the first derivative of the vertical projection profile or their local maxima and minima the left and right edges of each character is identified, block 79.
- corresponding display text 80 e.g. In the value memory 23 (see Fig. 2), stored for use in comparison module 19, block 81.
- field 82 the last stored display texts 80 are compared and for at least three matching display texts 80 this display text is e.g. to other application modules 20 (see Fig. 2) passed.
- Meter memory 22 (see Fig. 2) are loaded. It contains, for example, information about the position and extent of the individual segments within the black-and-white character 83. On the basis of this information, first a calculation, block 85, of a proportion of black pixels assigned to each segment is calculated. The black pixel portions per segment of the black-and-white character 83 thus obtained are sorted, block 86, and a position of the largest gap between two adjacent portions is sought, block 87. If the size of the gap, i. the maximum distance between two adjacent shares exceeds a fixed minimum, box 88, the segments assigned to the shares above the gap become active and the remaining ones
- Segments marked as inactive Otherwise, i. if the gap does not exceed the minimum, all segments are assigned the same status, i. either all are recognized as active (corresponding to the number eight shown) or all as inactive (which is practically considered to be zero).
- a predetermined amount of black pixels which is fixed or configurable, e.g. is loaded from the meter memory 22, used as a reference value to decide whether all segments are classified as active or inactive (depending on whether the fractions obtained are greater or less than the reference value).
- a character recognition, block 89 for example by looking up the
- Activity in a segment table that associates each combination of activities with a meaning in the form of a letter or number that recognizes characters. This character recognition is repeated for each character of a line, as indicated by the connection points F and G in FIG. 11 and FIG. 10, until all identified characters have been recognized.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Character Discrimination (AREA)
- Image Analysis (AREA)
- Controls And Circuits For Display Device (AREA)
- Studio Devices (AREA)
- Geometry (AREA)
- Control Of Indicators Other Than Cathode Ray Tubes (AREA)
Abstract
Description
Claims
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020167030200A KR101842535B1 (ko) | 2014-03-27 | 2015-03-27 | 부호의 광학적 검출 방법 |
| US15/129,378 US10055668B2 (en) | 2014-03-27 | 2015-03-27 | Method for the optical detection of symbols |
| JP2017501433A JP6630341B2 (ja) | 2014-03-27 | 2015-03-27 | シンボルの光学的検出方法 |
| EP15715669.6A EP3123393B1 (de) | 2014-03-27 | 2015-03-27 | Verfahren zur optischen erkennung von zeichen |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| ATA50224/2014A AT515595A2 (de) | 2014-03-27 | 2014-03-27 | Verfahren zur optischen Erkennung von Zeichen |
| ATA50224/2014 | 2014-03-27 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2015143471A1 true WO2015143471A1 (de) | 2015-10-01 |
Family
ID=52823966
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/AT2015/050080 Ceased WO2015143471A1 (de) | 2014-03-27 | 2015-03-27 | Verfahren zur optischen erkennung von zeichen |
Country Status (6)
| Country | Link |
|---|---|
| US (1) | US10055668B2 (de) |
| EP (1) | EP3123393B1 (de) |
| JP (1) | JP6630341B2 (de) |
| KR (1) | KR101842535B1 (de) |
| AT (1) | AT515595A2 (de) |
| WO (1) | WO2015143471A1 (de) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6974128B2 (ja) * | 2017-11-14 | 2021-12-01 | アズビル株式会社 | セグメント表示読取装置および方法 |
| KR102045940B1 (ko) * | 2019-05-21 | 2019-11-18 | (주)케이테크놀로지 | 평판 디스플레이 액정 셀의 에지 검사방법 |
| CN110554991A (zh) * | 2019-09-03 | 2019-12-10 | 浙江传媒学院 | 一种文本图片的矫正与管理方法 |
| EP3798898A1 (de) * | 2019-09-30 | 2021-03-31 | Anyline GmbH | Computerimplementiertes verfahren zur optischen zeichenerkennung |
| JP7670321B2 (ja) * | 2021-05-21 | 2025-04-30 | Necプラットフォームズ株式会社 | 画像監視装置、方法及びプログラム |
| US20230186655A1 (en) * | 2021-12-14 | 2023-06-15 | Adalberto Maldonado Irizarry | Smart License Plate Detection System for Land and Sea Vehicles |
| US12406353B2 (en) * | 2022-12-22 | 2025-09-02 | Communications Test Design, Inc. | Systems and methods for defect detection on displays |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120092329A1 (en) * | 2010-10-13 | 2012-04-19 | Qualcomm Incorporated | Text-based 3d augmented reality |
| DE102011000516A1 (de) | 2011-02-04 | 2012-08-09 | Karlheinz Schmidt | Messdatenerfassungssystem |
| DE102012110273A1 (de) | 2011-10-28 | 2013-05-02 | General Electric Company | Prüfsystem und -verfahren zum Korrelieren von Daten von Sensoren und visuellen Anzeigeeinrichtungen |
| US20130265232A1 (en) * | 2012-04-08 | 2013-10-10 | Samsung Electronics Co., Ltd. | Transparent display apparatus and method thereof |
| US20140064559A1 (en) * | 2012-09-03 | 2014-03-06 | Toshiba Tec Kabushiki Kaisha | Commodity recognition apparatus and commodity recognition method |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS58213107A (ja) | 1982-06-02 | 1983-12-12 | Kawasaki Heavy Ind Ltd | 流動層自動灰排出装置 |
| JPS6459482A (en) | 1987-08-31 | 1989-03-07 | Toshiba Corp | Character recognizing device |
| US5159667A (en) * | 1989-05-31 | 1992-10-27 | Borrey Roland G | Document identification by characteristics matching |
| JPH10320526A (ja) | 1997-05-14 | 1998-12-04 | Toppan Printing Co Ltd | 画像処理支援システムおよび画像処理支援方法並びに画像処理支援プログラムを記録した記録媒体 |
| US7221810B2 (en) * | 2000-11-13 | 2007-05-22 | Anoto Group Ab | Method and device for recording of information |
| WO2002063380A2 (en) * | 2001-02-06 | 2002-08-15 | Koninklijke Philips Electronics N.V. | Preventing green non-uniformity in image sensors |
| JP2008059081A (ja) * | 2006-08-29 | 2008-03-13 | Sony Corp | 画像処理装置及び画像処理方法、並びにコンピュータ・プログラム |
| JP5278093B2 (ja) * | 2009-03-26 | 2013-09-04 | 大日本印刷株式会社 | 記事関連情報提供方法、装置、プログラム、記録媒体 |
| JP4772894B2 (ja) * | 2009-08-03 | 2011-09-14 | シャープ株式会社 | 画像出力装置、携帯端末装置、撮像画像処理システム、画像出力方法、プログラムおよび記録媒体 |
-
2014
- 2014-03-27 AT ATA50224/2014A patent/AT515595A2/de not_active Application Discontinuation
-
2015
- 2015-03-27 JP JP2017501433A patent/JP6630341B2/ja active Active
- 2015-03-27 EP EP15715669.6A patent/EP3123393B1/de active Active
- 2015-03-27 KR KR1020167030200A patent/KR101842535B1/ko active Active
- 2015-03-27 WO PCT/AT2015/050080 patent/WO2015143471A1/de not_active Ceased
- 2015-03-27 US US15/129,378 patent/US10055668B2/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120092329A1 (en) * | 2010-10-13 | 2012-04-19 | Qualcomm Incorporated | Text-based 3d augmented reality |
| DE102011000516A1 (de) | 2011-02-04 | 2012-08-09 | Karlheinz Schmidt | Messdatenerfassungssystem |
| DE102012110273A1 (de) | 2011-10-28 | 2013-05-02 | General Electric Company | Prüfsystem und -verfahren zum Korrelieren von Daten von Sensoren und visuellen Anzeigeeinrichtungen |
| US20130265232A1 (en) * | 2012-04-08 | 2013-10-10 | Samsung Electronics Co., Ltd. | Transparent display apparatus and method thereof |
| US20140064559A1 (en) * | 2012-09-03 | 2014-03-06 | Toshiba Tec Kabushiki Kaisha | Commodity recognition apparatus and commodity recognition method |
Non-Patent Citations (5)
| Title |
|---|
| JAEYOUNG KIM ET AL: "Implementation of image processing and augmented reality programs for smart mobile device", PROCEEDINGS OF 2011 6TH INTERNATIONAL FORUM ON STRATEGIC TECHNOLOGY, 1 August 2011 (2011-08-01), pages 1070 - 1073, XP055173571, ISBN: 978-1-45-770398-0, DOI: 10.1109/IFOST.2011.6021205 * |
| JOAO PAOLO LIMA ET AL: "Real-Time Pattern Recognition using the OpenCV Library", SYMPOSIUM ON VIRTUAL AND AUGMENTED REALITY, 1 June 2007 (2007-06-01), pages 1 - 42, XP055199497, Retrieved from the Internet <URL:http://www.researchgate.net/profile/Thiago_Farias/publication/265311134_REAL-TIME_PATTERN_RECOGNITION_USING_THE_OPENCV_LIBRARY/links/54c8fbf70cf289f0ced13b28.pdf> [retrieved on 20150701] * |
| MARC PETTER ET AL: "Automatic text detection for mobile augmented reality translation", COMPUTER VISION WORKSHOPS (ICCV WORKSHOPS), 2011 IEEE INTERNATIONAL CONFERENCE ON, IEEE, 6 November 2011 (2011-11-06), pages 48 - 55, XP032095218, ISBN: 978-1-4673-0062-9, DOI: 10.1109/ICCVW.2011.6130221 * |
| XIAN LI ET AL: "LCD/LED DIGIT RECOGNITION BY IPHONE", M.SC. THESIS, 31 May 2011 (2011-05-31), pages 1 - 33, XP055199577, Retrieved from the Internet <URL:http://repositories.tdl.org/ttu-ir/bitstream/handle/2346/ETD-TTU-2011-05-1485/LI-THESIS.pdf?sequence=1&isAllowed=y> [retrieved on 20150701] * |
| YIBO LI ET AL: "Automatic Recognition System for Numeric Characters on Ammeter Dial Plate", YOUNG COMPUTER SCIENTISTS, 2008. ICYCS 2008. THE 9TH INTERNATIONAL CONFERENCE FOR, IEEE, PISCATAWAY, NJ, USA, 18 November 2008 (2008-11-18), pages 913 - 918, XP031373293, ISBN: 978-0-7695-3398-8 * |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2017521011A (ja) | 2017-07-27 |
| JP6630341B2 (ja) | 2020-01-15 |
| KR20170010753A (ko) | 2017-02-01 |
| KR101842535B1 (ko) | 2018-03-28 |
| AT515595A2 (de) | 2015-10-15 |
| EP3123393B1 (de) | 2022-04-20 |
| US10055668B2 (en) | 2018-08-21 |
| EP3123393A1 (de) | 2017-02-01 |
| US20170177968A1 (en) | 2017-06-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3123393B1 (de) | Verfahren zur optischen erkennung von zeichen | |
| DE69600461T2 (de) | System und Verfahren zur Bewertung der Abbildung eines Formulars | |
| DE69226846T2 (de) | Verfahren zur Bestimmung von Wortgrenzen im Text | |
| EP1413972B1 (de) | Prüfung von Bildaufnahmen von Personen | |
| DE102009023756B4 (de) | Verfahren zum Herkunftsnachweis und zur Urheberschaft von Bildern | |
| EP0131676B1 (de) | Verfahren zum automatischen Digitalisieren des Umrisses von Strichgraphiken z.B. Buchstaben | |
| DE102011056660B4 (de) | Markierungsleser, der zum Priorisieren von Bildern eingerichtet ist und Verfahren zum Decodieren einer Markierung | |
| DE60307967T2 (de) | Bildverarbeitungsverfahren für die untersuchung des erscheinungsbildes | |
| DE102013206009A1 (de) | Robustes Zuschneiden von Nummernschildbildern | |
| DE10346690B4 (de) | Verfahren und Vorrichtung zum Detektieren eines Linearobjekts | |
| DE102018129863A1 (de) | Vorrichtung und Verfahren zur Datenverarbeitung | |
| DE102013210375A1 (de) | Auffinden von text in natürlichen szenen | |
| WO2010115464A1 (de) | Zweidimensionaler symbolcode und verfahren zum lesen des symbolcodes | |
| DE60020038T2 (de) | Verfahren zum Verarbeiten eines numerischen Bildes | |
| DE102009057884A1 (de) | Abbildungsvorrichtung | |
| DE69713977T2 (de) | Verfahren und Gerät zur Retuschierung eines digitalen Farbbildes | |
| DE60303138T2 (de) | Vergleichen von mustern | |
| EP2639738A1 (de) | Verfahren und Lesegerät zur Erfassung einer Mehrzahl nacheinander auf eine Anzeigevorrichtung dargestellter zweidimensionaler Codes | |
| BE1029597A1 (de) | Bildverarbeitungssysteme und -verfahren zum automatischen Erzeugen eines oder mehrerer Bildverarbeitungsaufträge auf Grundlage interessierender Regionen (ROIs) digitaler Bilder | |
| DE102015115394A1 (de) | Verfahren und Vorrichtung zum Überlagern eines Abbilds einer realen Szenerie mit einem virtuellen Bild und mobiles Gerät | |
| DE112022005421T5 (de) | Verfahren zur identifizierung einer zeichenorientierung und zur dekodierung von zeichen für maschinelle bildverarbeitungssysteme | |
| DE112016002252T5 (de) | Automatische erkennung von panoramagesten | |
| DE112024000706T5 (de) | Objektidentifikation basierend auf einer teilweisen dekodierung | |
| DE112019006414T5 (de) | Vorrichtung und Verfahren zur Detektion von Menschen | |
| EP3009984A1 (de) | Detektionssystem für optische Codes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 15715669 Country of ref document: EP Kind code of ref document: A1 |
|
| DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
| ENP | Entry into the national phase |
Ref document number: 2017501433 Country of ref document: JP Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 15129378 Country of ref document: US |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| ENP | Entry into the national phase |
Ref document number: 20167030200 Country of ref document: KR Kind code of ref document: A |
|
| REEP | Request for entry into the european phase |
Ref document number: 2015715669 Country of ref document: EP |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2015715669 Country of ref document: EP |
