CN116630982A - Scanning area positioning method based on AI dictionary pen - Google Patents
Scanning area positioning method based on AI dictionary pen Download PDFInfo
- Publication number
- CN116630982A CN116630982A CN202310547120.2A CN202310547120A CN116630982A CN 116630982 A CN116630982 A CN 116630982A CN 202310547120 A CN202310547120 A CN 202310547120A CN 116630982 A CN116630982 A CN 116630982A
- Authority
- CN
- China
- Prior art keywords
- character
- outline
- image
- distribution
- text
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 26
- 230000008859 change Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 5
- 230000009467 reduction Effects 0.000 claims description 4
- 230000001174 ascending effect Effects 0.000 claims description 3
- 230000000630 rising effect Effects 0.000 claims description 3
- 230000001419 dependent effect Effects 0.000 claims description 2
- 238000000605 extraction Methods 0.000 description 13
- 230000000694 effects Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 238000013519 translation Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/1444—Selective acquisition, locating or processing of specific regions, e.g. highlighted text, fiducial marks or predetermined fields
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/142—Image acquisition using hand-held instruments; Constructional details of the instruments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/18—Extraction of features or characteristics of the image
- G06V30/1801—Detecting partial patterns, e.g. edges or contours, or configurations, e.g. loops, corners, strokes or intersections
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Character Input (AREA)
Abstract
The invention relates to the technical field of character recognition, in particular to a scanning area positioning method based on an AI dictionary pen.
Description
Technical Field
The invention relates to the technical field of character recognition, in particular to a scanning area positioning method based on an AI dictionary pen.
Background
The dictionary pen is a scanning electronic dictionary, adopts a technology and an embedded translation system, replaces key input of the traditional electronic dictionary in a mode of scanning word searching at a high speed, and plays an important role in learning language culture.
For example, chinese patent publication No.: the invention discloses a camera calibration device and a camera calibration method applied to a scanning translation pen and a dictionary pen, and discloses a CN111783769A, which relate to the technical field of the scanning translation pen. According to the technical scheme, the effective window boundary line of the calibration camera can be automatically and accurately calibrated, the operation is quick, and the efficiency and the product quality are improved.
However, the prior art has the following problems:
in the prior art, under the condition that incomplete character outlines appear in a scanning image generated by a dictionary pen, incomplete character outlines after character outlines are extracted are also taken into character outline content recognition judgment, a judgment result is affected, character outlines in the image are extracted without consideration of selectivity, extraction of the incomplete character outlines is reduced, and the efficiency and effect of character recognition by the dictionary pen are improved.
Disclosure of Invention
In order to solve the above problems, the present invention provides a scanning area positioning method based on an AI dictionary pen, which includes,
step S1, acquiring a scanning image of a dictionary pen in a single scanning process, and dividing the scanning image into an initial image section and a plurality of subdivision image sections;
step S2, judging the distribution state of the character outlines in the scanned image based on the average height of the character outlines of each row in the initial image section, wherein the distribution state comprises a complete distribution state and a incomplete distribution state;
step S3, when the distribution state of the character outlines in the scanned image is a complete distribution state, extracting all character outlines in an initial image section and each subdivision image section, and carrying out character outline identification on each character outline;
setting a scanning area for extracting character outline when the distribution state of the character outline in the scanned image is incomplete, carrying out character outline recognition on the character outline extracted in the scanning area, determining the width of the scanning area based on the average height of the character outline of the sample line in the initial image section, the character outline line with the maximum average height of the sample line, and adjusting the distribution height of the scanning area in each sub-section based on the trend state of the character outline of the sample line in each sub-section,
the first adjustment mode is to increase the distribution height of the scanning area in the subdivided image segment;
the second adjustment mode is to reduce the distribution height of the scanning area in the subdivided image segment;
the distribution height is the height of the central line of the scanning area and the edge of the scanning image.
Further, in the step S2, an average height of each line of text outline in the initial image segment is determined, the average heights of each line of text outline are compared, and a distribution state of text outlines in the scanned image is determined according to a comparison result, wherein,
if the comparison result meets the first difference condition, judging that the distribution state of the character outlines in the scanned image is a complete distribution state;
if the comparison result meets a second difference condition, judging that the distribution state of the character outlines in the scanned image is a incomplete distribution state;
the first difference condition is that the difference value of the average heights of any two lines of character outlines in the initial image section is smaller than a preset difference threshold value, and the second difference condition is that the difference value of the average heights of any two lines of character outlines is larger than or equal to the preset difference threshold value.
Further, in the step S3, a width of the scanning area is determined based on an average height of text contours in the sample line in the initial image segment, wherein,
the width of the scan area is determined to be proportional to the average height of the text outline of the sample line in the initial image segment.
Further, in the step S3, the first adjustment mode needs to satisfy the trend state of the text outline as an ascending trend state, and the second adjustment mode needs to satisfy the trend state of the text outline as a descending trend state.
Further, in the step S3, a text outline change curve of the sample line is constructed, wherein,
and constructing a character outline change curve of the sample line by taking the sequence of each character outline in the sample line as an independent variable and the distance from the midpoint of each character outline to the edge of the scanned image as a dependent variable.
Further, in the step S3, the method further includes determining a trend state of the text outline in the subdivided image segment based on the slope parameter of the text outline change curve of the sample line, wherein,
calculating a slope parameter delta K of the character profile variation curve of the sample line according to a formula (1),
in the formula (1), K (i) represents the distance between the midpoint of the ith text outline and the edge of the scanned image, K (i+1) represents the distance between the midpoint of the ith text outline and the edge of the scanned image, z represents the number of text outlines in a sample row, and i represents an integer greater than 0;
and determining the positive and negative conditions of the slope parameter,
if the slope parameter is greater than 0, judging that the trend state of the character outline in the scanned image is a rising trend state;
if the slope parameter is smaller than 0, judging that the trend state of the character outline in the scanned image is a descending trend state.
Further, in the step S3, the distribution height of the scanning area in the subdivided image segment is increased based on the slope parameter of the character profile change curve of the sample line in the first adjustment mode, wherein,
determining that the increase in the distribution height of the scan region is proportional to the absolute value of the slope parameter.
Further, in the step S3, the distribution height of the scanning area in the subdivided image segment is reduced based on the slope parameter of the character profile variation curve of the sample line in the second adjustment mode, wherein,
determining the reduction of the distribution height of the scanning area in direct proportion to the absolute value of the slope parameter.
Further, in the step S3, an initial distribution height of the scanning area is determined based on a width of the scanned image, wherein the distribution height of the scanning area is half of the width of the scanning area.
Further, in the step S3, text outline recognition is performed on the text outline, wherein,
and matching the text outline with the sample text outline in the text outline sample database, and identifying text contents represented by the text outline according to a matching result.
Compared with the prior art, the method has the advantages that the initial image segment is determined by taking the starting end of the scanned image as the starting point, the distribution state of the character contours in the scanned image is judged based on the average height of the character contours of each row in the initial image segment, the scanning area is set for extracting the character contours when the character contours are in incomplete distribution state, character contour recognition is carried out on the character contours extracted in the scanning area, the width of the scanning area is determined based on the average height of the character contours of the sample rows in the initial image segment, and the distribution height of the scanning area in each fine segment is adjusted based on the trend state of the character contours of the sample rows in each fine image segment.
Particularly, in the invention, the distribution state of the character outlines in the scanned image is judged based on the average height of each row of character outlines in the initial image section, in the practical situation, the average height of the incomplete character outlines is greatly different from the average height of the complete character outlines, the incomplete character outlines can be quickly identified on the premise of occupying less calculation force through the average height of each row of character outlines, the corresponding processing of different distribution states is convenient to follow-up, the extraction of the incomplete character outlines is reduced, the incomplete character outlines are prevented from interfering the normal character outline identification, and the identification effect of dictionary pens is improved.
In particular, in the invention, when the distribution state of the character outlines in the scanned image is the incomplete distribution state, the width of the scanning area is determined based on the average height of the character outlines of the sample lines in the initial image section, in the practical situation, the character outlines in the sample lines are usually complete character outlines, and the width of the extraction area is determined based on the average height of the character outlines, so that the incomplete character outlines distributed on the upper side and the lower side of the sample lines can be prevented from being recognized by the extraction area too wide on the premise of ensuring that the complete character outlines are recognized, the extraction of the incomplete character outlines is reduced, the interference of the incomplete character outlines on the normal character outline recognition is avoided, and the character recognition effect of the dictionary pen is improved.
Particularly, in the invention, the distribution height of the scanning area in each fine segment is adjusted according to the trend state of the character outline of the sample line in each fine segment, in the practical situation, the dictionary pen is used manually, the scanning image cannot be ensured to be an ideal straight line in the scanning process, and offset can possibly occur under the condition of handwriting of characters to be scanned, and incomplete images can further occur in the scanning image, so that the invention adjusts the distribution height of the extraction area adaptively according to the trend state of the character outline, so that the extraction area can adapt to the trend state of the character outline, the extraction of the incomplete character outline can be reduced, the character outline recognition effect is ensured, and the missing recognition condition is reduced.
Drawings
FIG. 1 is a schematic diagram of steps of a scanning area positioning method based on an AI dictionary pen in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a scanning area structure according to an embodiment of the invention;
FIG. 3 is a schematic view of a distribution height structure of a scanning area according to an embodiment of the invention;
in the figure, 1: scanning an image, 2: the area is scanned.
Detailed Description
In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
Referring to fig. 1, fig. 2, and fig. 3, which are schematic diagrams of steps, a schematic diagram of a scanning area structure, and a schematic diagram of a distribution height structure of a scanning area based on an AI dictionary pen according to an embodiment of the present invention, the scanning area positioning method based on the AI dictionary pen includes:
step S1, acquiring a scanning image 1 of a dictionary pen in a single scanning process, and dividing the scanning image 1 into an initial image section and a plurality of subdivision image sections;
step S2, judging the distribution state of the character outlines in the scanned image 1 based on the average height of each row of character outlines in the initial image segment, wherein the distribution state comprises a complete distribution state and a incomplete distribution state;
step S3, when the distribution state of the character outlines in the scanned image 1 is a complete distribution state, extracting all the character outlines in the initial image section and each subdivision image section, and carrying out character outline identification on each character outline;
when the distribution state of the character outlines in the scanned image 1 is the incomplete distribution state, setting a scanned area 2 to extract the character outlines, carrying out character outline recognition on the character outlines extracted in the scanned area 2, determining the width of the scanned area 2 based on the average height of the character outlines of sample lines in the initial image section, the character outline line with the maximum average height of the sample lines, and adjusting the distribution height of the scanned area 2 in each sub-section based on the trend state of the character outlines of the sample lines in each sub-section, wherein,
the first adjustment mode is to increase the distribution height of the scanning area 2 in the subdivided image segments;
the second adjustment mode is to reduce the distribution height of the scanning area 2 in the subdivided image segments;
the distribution height is the height of the center line of the scanning area 2 and the edge of the scanned image 1.
Specifically, the specific mode of recognizing the text outline in the image is not limited, and the text outline can be recognized by a pre-trained model, and the related recognition algorithm and model are of the prior art and are not repeated here.
Specifically, the specific structure of the dictionary pen is not limited, the dictionary pen can be a dictionary pen with a scanning head, the scanning head can be an image acquisition unit for acquiring images of a scanning area, and the related structure is mature and is not described herein.
Specifically, in the step S2, the average height of each line of text outline in the initial image segment is determined, the average heights of each line of text outline are compared, and the distribution state of the text outline in the scanned image 1 is determined according to the comparison result, wherein,
if the comparison result meets a first difference condition, judging that the distribution state of the character outlines in the scanned image 1 is a complete distribution state;
if the comparison result meets a second difference condition, judging that the distribution state of the character outlines in the scanned image 1 is a incomplete distribution state;
the first difference condition is that the difference value of the average heights of any two lines of character outlines in the initial image section is smaller than a preset difference threshold value, and the second difference condition is that the difference value of the average heights of any two lines of character outlines is larger than or equal to the preset difference threshold value.
The preset difference threshold is determined based on the average height of the character outlines of each row, and is aimed at characterizing the difference, so that the person skilled in the art can set the difference in the interval [0.3d,0.5d ] in order to avoid the excessive difference on the premise of guaranteeing the distinction, and d represents the average height of the character outlines of each row.
Specifically, in the invention, the distribution state of the character outlines in the scanned image 1 is judged based on the average height of each row of character outlines in the initial image section, in the practical situation, the average height of the incomplete character outlines is greatly different from the average height of the complete character outlines, the incomplete character outlines can be quickly identified on the premise of occupying less calculation force through the average height of each row of character outlines, the corresponding processing of different distribution states is convenient to follow, the extraction of the incomplete character outlines is reduced, the incomplete character outlines are prevented from interfering the normal character outline identification, and the identification effect of dictionary pens is improved.
Specifically, in step S3, the width of the scanning area 2 is determined based on the average height of the text outline in the sample line in the initial image segment, wherein,
the width of the scan area 2 is determined in direct proportion to the average height of the text outline of the sample line in the initial image segment.
In order to ensure that the scanning area 2 can scan the complete sample line and avoid the area from scanning too wide to the peripheral residual text outline, in this embodiment, the width dh=1.2h of the scanning area 2 is set, where h represents the average height of the text outline in the sample line.
Specifically, in the invention, when the distribution state of the character outlines in the scanned image 1 is the incomplete distribution state, the width of the scanned area 2 is determined based on the average height of the character outlines of the sample lines in the initial image segment, in the practical situation, the character outlines in the sample lines are usually complete character outlines, and the width of the extracted area is determined according to the average height, so that the incomplete character outlines distributed on the upper side and the lower side of the sample lines can be prevented from being recognized by the excessive width of the extracted area on the premise of ensuring that the complete character outlines are recognized, the extraction of the incomplete character outlines is reduced, the interference of the incomplete character outlines on the normal character outline recognition is avoided, and the character recognition effect of a dictionary pen is improved.
Specifically, in the step S3, the first adjustment mode needs to satisfy the trend state of the text outline as an ascending trend state, and the second adjustment mode needs to satisfy the trend state of the text outline as a descending trend state.
Specifically, in the step S3, a text outline change curve of the sample line is constructed, wherein,
and constructing a character outline change curve of the sample line by taking the sequence of each character outline in the sample line as an independent variable and taking the distance from the midpoint of each character outline to the edge of the scanning image 1 as an independent variable.
Specifically, in the step S3, the method further includes determining a trend state of the text outline in the subdivided image segment based on the slope parameter of the text outline change curve of the sample line, wherein,
calculating a slope parameter delta K of the character profile variation curve of the sample line according to a formula (1),
in the formula (1), K (i) represents the distance between the midpoint of the ith text outline and the edge of the scanned image 1, K (i+1) represents the distance between the midpoint of the ith text outline and the edge of the scanned image 1, z represents the number of text outlines in a sample line, and i represents an integer greater than 0;
and determining the positive and negative conditions of the slope parameter,
if the slope parameter is greater than 0, judging that the trend state of the character outline in the scanned image 1 is a rising trend state;
if the slope parameter is smaller than 0, judging that the trend state of the character outline in the scanned image 1 is a descending trend state.
Specifically, as shown in fig. 3, in the step S3, the distribution height of the scanning area 2 in the subdivided image segment is increased based on the slope parameter of the character profile curve of the sample line in the first adjustment mode, wherein,
the increase in the height of the distribution of the scanning area 2 is determined in direct proportion to the absolute value of the slope parameter.
In the present embodiment, an increase hd=hf×|Δk|/H is set, where Hf represents the initial distribution height of the scanning area 2 and H represents the average width of the text outline of the sample line.
Specifically, as shown in fig. 3, in the step S3, the distribution height of the scanning area 2 in the subdivided image segment is reduced based on the slope parameter of the character profile curve of the sample line in the second adjustment mode, wherein,
the reduction in the height of the distribution of the scanning area 2 is determined in direct proportion to the absolute value of the slope parameter.
The reduction amount hc=hf×|Δk|/H is set in the present embodiment, where Hf represents the initial distribution height of the scanning area 2, and H represents the average width of the text outline of the sample line.
Specifically, in the invention, the distribution height of the scanning area 2 in each fine segment is adjusted according to the trend state of the character outline of the sample line in each fine segment, in the practical situation, the dictionary pen is used manually, the scanning image 1 cannot be ensured to be an ideal straight line in the scanning process, and offset can possibly occur under the condition of handwriting of characters to be scanned, and further, incomplete images appear in the scanning image 1, so that the invention adaptively adjusts the distribution height of the extraction area based on the trend state of the character outline, so that the extraction area can adapt to the trend state of the character outline, the extraction of the incomplete character outline can be reduced, the character outline recognition effect is ensured, and the missing recognition condition is reduced.
Specifically, in the step S3, the initial distribution height of the scanning area 2 is determined based on the width of the scanned image 1, wherein the initial distribution height of the scanning area 2 is half the width of the scanning area 2.
Specifically, in the step S3, the text outline is identified, wherein the text outline is matched with the sample text outline in the text outline sample database, and text content represented by the text outline is identified according to the matching result, wherein each sample text outline is associated with the represented text content, and if the similarity of the text outline and the similarity of the sample text outline exceeds the similarity threshold, the text outline is determined to be matched with the sample text outline, and further, the content represented by the text outline is determined to be the content associated with the sample text outline.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
Claims (10)
1. The scanning area positioning method based on the AI dictionary pen is characterized by comprising the following steps:
step S1, acquiring a scanning image of a dictionary pen in a single scanning process, and dividing the scanning image into an initial image section and a plurality of subdivision image sections;
step S2, judging the distribution state of the character outlines in the scanned image based on the average height of the character outlines of each row in the initial image section, wherein the distribution state comprises a complete distribution state and a incomplete distribution state;
step S3, when the distribution state of the character outlines in the scanned image is a complete distribution state, extracting all character outlines in an initial image section and each subdivision image section, and carrying out character outline identification on each character outline;
setting a scanning area for extracting character outline when the distribution state of the character outline in the scanned image is incomplete, carrying out character outline recognition on the character outline extracted in the scanning area, determining the width of the scanning area based on the average height of the character outline of the sample line in the initial image section, the character outline line with the maximum average height of the sample line, and adjusting the distribution height of the scanning area in each sub-section based on the trend state of the character outline of the sample line in each sub-section,
the first adjustment mode is to increase the distribution height of the scanning area in the subdivided image segment;
the second adjustment mode is to reduce the distribution height of the scanning area in the subdivided image segment;
the distribution height is the height of the central line of the scanning area and the edge of the scanning image.
2. The AI-dictionary-pen-based scanning area positioning method of claim 1, wherein in the step S2, an average height of each line of character outlines in the initial image segment is determined, the average heights of each line of character outlines are compared, and a distribution state of the character outlines in the scanned image is determined according to a comparison result, wherein,
if the comparison result meets the first difference condition, judging that the distribution state of the character outlines in the scanned image is a complete distribution state;
if the comparison result meets a second difference condition, judging that the distribution state of the character outlines in the scanned image is a incomplete distribution state;
the first difference condition is that the difference value of the average heights of any two lines of character outlines in the initial image section is smaller than a preset difference threshold value, and the second difference condition is that the difference value of the average heights of any two lines of character outlines is larger than or equal to the preset difference threshold value.
3. The AI dictionary pen-based scan area positioning method according to claim 2, wherein in the step S3, the width of the scan area is determined based on the average height of the character outline in the sample line in the initial image segment, wherein,
the width of the scan area is determined to be proportional to the average height of the text outline of the sample line in the initial image segment.
4. The AI-dictionary-pen-based scanning area positioning method of claim 1, wherein in the step S3, the first adjustment mode is required to satisfy the trend state of the text outline as an ascending trend state, and the second adjustment mode is required to satisfy the trend state of the text outline as a descending trend state.
5. The AI-dictionary-pen-based scanning area positioning method of claim 4, further comprising constructing a sample line text contour change curve in step S3, wherein,
and constructing a character outline change curve of the sample line by taking the sequence of each character outline in the sample line as an independent variable and the distance from the midpoint of each character outline to the edge of the scanned image as a dependent variable.
6. The AI-dictionary pen-based scan area positioning method of claim 5, further comprising determining a trend state of the text outline in the subdivided image segment based on a slope parameter of the sample line text outline change curve in step S3, wherein,
calculating a slope parameter delta K of the character profile variation curve of the sample line according to a formula (1),
in the formula (1), K (i) represents the distance between the midpoint of the ith text outline and the edge of the scanned image, K (i+1) represents the distance between the midpoint of the ith text outline and the edge of the scanned image, z represents the number of text outlines in a sample row, and i represents an integer greater than 0;
and determining the positive and negative conditions of the slope parameter,
if the slope parameter is greater than 0, judging that the trend state of the character outline in the scanned image is a rising trend state;
if the slope parameter is smaller than 0, judging that the trend state of the character outline in the scanned image is a descending trend state.
7. The AI-dictionary pen-based scan area positioning method according to claim 6, wherein in the step S3, the distribution height of the scan area in the subdivided image segment is increased based on the slope parameter of the sample line text profile change curve in the first adjustment mode,
determining that the increase in the distribution height of the scan region is proportional to the absolute value of the slope parameter.
8. The AI-dictionary pen-based scan area positioning method according to claim 6, wherein in the step S3, the distribution height of the scan area in the subdivided image segment is reduced based on the slope parameter of the sample line text profile change curve in the second adjustment mode,
determining the reduction of the distribution height of the scanning area in direct proportion to the absolute value of the slope parameter.
9. The AI dictionary pen-based scan area positioning method according to claim 1, wherein in the step S3, an initial distribution height of the scan area is determined based on a width of the scan image, wherein the distribution height of the scan area is half of the width of the scan area.
10. The AI-dictionary-pen-based scanning area positioning method according to claim 1, wherein in the step S3, character contour recognition is performed on character contours, wherein,
and matching the text outline with the sample text outline in the text outline sample database, and identifying text contents represented by the text outline according to a matching result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310547120.2A CN116630982B (en) | 2023-05-16 | 2023-05-16 | Scanning area positioning method based on AI dictionary pen |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310547120.2A CN116630982B (en) | 2023-05-16 | 2023-05-16 | Scanning area positioning method based on AI dictionary pen |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116630982A true CN116630982A (en) | 2023-08-22 |
| CN116630982B CN116630982B (en) | 2024-09-03 |
Family
ID=87620576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310547120.2A Active CN116630982B (en) | 2023-05-16 | 2023-05-16 | Scanning area positioning method based on AI dictionary pen |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116630982B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120005054U (en) * | 2010-12-31 | 2012-07-10 | 정서연 | Scan reader |
| CN105512659A (en) * | 2015-12-10 | 2016-04-20 | 广东小天才科技有限公司 | Correction method and system for character line bending |
| CN112153236A (en) * | 2020-11-05 | 2020-12-29 | 安徽淘云科技有限公司 | Control method of scanning pen, electronic device and readable storage medium |
| CN212659089U (en) * | 2020-07-14 | 2021-03-05 | 上海趣威文化发展有限公司 | Scanning translation point-reading pen |
| CN113723420A (en) * | 2021-09-03 | 2021-11-30 | 安徽淘云科技股份有限公司 | Scanning method and related equipment thereof |
| CN114550174A (en) * | 2022-02-07 | 2022-05-27 | 青岛海尔科技有限公司 | Reading assisting method and device |
| WO2023280173A1 (en) * | 2021-07-09 | 2023-01-12 | 华为技术有限公司 | Scanning method and electronic device |
| CN218587228U (en) * | 2022-11-17 | 2023-03-07 | 科大讯飞股份有限公司 | Scanning device |
| CN116110048A (en) * | 2021-11-08 | 2023-05-12 | 广东小天才科技有限公司 | Cursor generation method, cursor generation device and scanning pen |
-
2023
- 2023-05-16 CN CN202310547120.2A patent/CN116630982B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120005054U (en) * | 2010-12-31 | 2012-07-10 | 정서연 | Scan reader |
| CN105512659A (en) * | 2015-12-10 | 2016-04-20 | 广东小天才科技有限公司 | Correction method and system for character line bending |
| CN212659089U (en) * | 2020-07-14 | 2021-03-05 | 上海趣威文化发展有限公司 | Scanning translation point-reading pen |
| CN112153236A (en) * | 2020-11-05 | 2020-12-29 | 安徽淘云科技有限公司 | Control method of scanning pen, electronic device and readable storage medium |
| WO2023280173A1 (en) * | 2021-07-09 | 2023-01-12 | 华为技术有限公司 | Scanning method and electronic device |
| CN113723420A (en) * | 2021-09-03 | 2021-11-30 | 安徽淘云科技股份有限公司 | Scanning method and related equipment thereof |
| CN116110048A (en) * | 2021-11-08 | 2023-05-12 | 广东小天才科技有限公司 | Cursor generation method, cursor generation device and scanning pen |
| CN114550174A (en) * | 2022-02-07 | 2022-05-27 | 青岛海尔科技有限公司 | Reading assisting method and device |
| CN218587228U (en) * | 2022-11-17 | 2023-03-07 | 科大讯飞股份有限公司 | Scanning device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116630982B (en) | 2024-09-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7039235B1 (en) | Ruled line extracting apparatus for extracting ruled line from normal document image and method thereof | |
| El Abed et al. | Comparison of different preprocessing and feature extraction methods for offline recognition of handwritten arabicwords | |
| CN106650608B (en) | The recognition methods of rectangle posting in a kind of no anchor point paper | |
| Chan et al. | Searching off-line arabic documents | |
| CN106503711A (en) | A kind of character recognition method | |
| Tseng et al. | An efficient knowledge-based stroke extraction method for multi-font Chinese characters | |
| CN107122775A (en) | A kind of Android mobile phone identity card character identifying method of feature based matching | |
| CN104820827A (en) | Method for recognizing punctiform characters on surfaces of cables | |
| CN115497109B (en) | Character and image preprocessing method based on intelligent translation | |
| Bera et al. | Normalization of unconstrained handwritten words in terms of Slope and Slant Correction | |
| CN115713772A (en) | Transformer substation panel character recognition method, system, equipment and storage medium | |
| CN115880566A (en) | Intelligent marking system based on visual analysis | |
| CN107798355B (en) | Automatic analysis and judgment method based on document image format | |
| CN116630982A (en) | Scanning area positioning method based on AI dictionary pen | |
| JP2000331120A (en) | Device and method for recognizing character and recording medium stored with control program therefor | |
| CN113989485B (en) | Text character segmentation method and system based on OCR (optical character recognition) | |
| Kibria | Bengali optical character recognition using self organizing map | |
| CN115171133A (en) | Table structure detection method for leveling irregular table image | |
| US11210564B2 (en) | Information processing apparatus for character recognition and non-transitory computer readable medium | |
| CN112581487A (en) | Method for automatically extracting detection area and positioning kernel | |
| CN115995080B (en) | Archive intelligent management system based on OCR (optical character recognition) | |
| CN103679178A (en) | Identification method for manually filled information on colorful printed paper | |
| Ting et al. | A syntactic business form classifier | |
| CN116704507B (en) | Accurate recognition method for dictionary pen area content | |
| JPH02116987A (en) | Character recognizing device |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |