CN108846339A - Character recognition method and device, electronic equipment and storage medium - Google Patents
Character recognition method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN108846339A CN108846339A CN201810563940.XA CN201810563940A CN108846339A CN 108846339 A CN108846339 A CN 108846339A CN 201810563940 A CN201810563940 A CN 201810563940A CN 108846339 A CN108846339 A CN 108846339A
- Authority
- CN
- China
- Prior art keywords
- text
- dimensional coordinate
- output
- determining
- stroke
- 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 93
- 238000003860 storage Methods 0.000 title claims abstract description 16
- 238000004321 preservation Methods 0.000 claims description 57
- 238000001514 detection method Methods 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 23
- 238000004891 communication Methods 0.000 claims description 19
- 238000013507 mapping Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 9
- 235000013399 edible fruits Nutrition 0.000 claims 3
- 238000005266 casting Methods 0.000 claims 1
- 238000012545 processing Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 230000004048 modification Effects 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010422 painting Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The invention discloses a character recognition method, a character recognition device, electronic equipment and a storage medium, which are used for solving the problems of inaccuracy and inconvenience in character input in the prior art. The method comprises the following steps: detecting whether a preset target gesture for writing exists in a current frame image in a video stream acquired by a camera; if so, determining and storing the coordinate information of the target gesture; and determining and outputting the target characters corresponding to the characters to be output according to each coordinate information corresponding to the stored characters to be output. When the user writes in the near-empty space, the electronic equipment determines the characters written by the user according to each three-dimensional coordinate of the target gesture when the user writes, so that the characters can be quickly and accurately input.
Description
Technical field
The present invention relates to technical field of character recognition, in particular to a kind of character recognition method, device, electronic equipment and deposit
Storage media.
Background technique
With science and technology and social progress, people's living standard is continuously improved, and various electronic equipments progress into people's
Life, provides various conveniences for user.
Existing electronic equipment is when typing text is used to input information, typically by way of keyboard or touch screen
Carry out typing.But in life have many people for the lack of skill of keyboard, the typing text speed by way of keyboard
It is very slow.The touch screen of electronic equipment is easily damaged or insensitive, is caused user when passing through touch screen input character, also can
It is many inconvenient to occur.
User generally for the writing of text be it is very familiar, how to utilize this advantage, rapidly and accurately typing text
Word is to need the problem of studying.
Summary of the invention
The embodiment of the invention discloses a kind of character recognition method, device, electronic equipment and storage mediums, existing to solve
Have inaccurate in typing text in technology, it has not been convenient to the problem of.
In order to achieve the above objectives, the embodiment of the invention discloses a kind of character recognition methods, are applied to electronic equipment, described
Method includes:
It detects in the current frame image in the video flowing of camera acquisition with the presence or absence of the preset target hand for writing
Gesture;If it is, determining the coordinate information of the target gesture, and save;
According to the corresponding each coordinate information of the text to be output of preservation, the corresponding target text of the text to be output is determined
Word, and export.
Further, it is used in the current frame image in the video flowing of the detection camera acquisition with the presence or absence of preset
The target gesture of writing includes:
The current frame image in video flowing acquired respectively according to two cameras, determines the overlapping in two current frame images
Region;It detects in overlapping region with the presence or absence of the preset target gesture for writing;
The coordinate information of the determination target gesture includes:
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
Further, after the first three-dimensional coordinate for determining the target gesture, the method also includes:
It is three-dimensional by described first according to the mapping relations of the writing display area of the overlapping region and display screen pre-saved
Coordinate is mapped as first two-dimensional coordinate for writing display area, according to first two-dimensional coordinate in the writing viewing area
Domain is shown.
Further, after the first three-dimensional coordinate for determining the target gesture, in the text to be output according to preservation
The corresponding each coordinate information of word, before determining the corresponding target text of the text to be output, the method also includes:
According to first three-dimensional coordinate, judges whether current stroke is write and finish;
If it is, carrying out subsequent step.
Further, if there is no the preset target gesture for writing in overlapping region, according to preservation to
The corresponding each coordinate information of output character, before determining the corresponding target text of the text to be output, the method is also wrapped
It includes:
Judgement is directed to the previous frame image of current frame image, if preserves the second three-dimensional coordinate of the target gesture;
If it is, carrying out subsequent step.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
The corresponding target text of word includes:
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, corresponding first text of the text to be output is determined
Word;
First text is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
Further, described according to matching result, determine that the corresponding target text of the text to be output includes:
Highest second text of matching degree is determined as the corresponding target text of the text to be output;Or,
Show the second text of preset quantity from high to low by matching degree;The second text that user selects is determined as described
The corresponding target text of text to be output.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
Corresponding first text of word includes:
According to the determining corresponding every unicursal of text to be output, corresponding first text of text to be output is determined, wherein
The corresponding every unicursal of text to be output is to determine the stroke after finishing writing, according to the corresponding each three-dimensional coordinate of the stroke
Determining.
Further, according to the first three-dimensional coordinate, determine whether current stroke writes the process finished and include:
According to first three-dimensional coordinate and the second three-dimensional coordinate of the previous frame image preservation for current frame image
First distance determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge that described second is three-dimensional
Coordinate whether be determining stroke effective three-dimensional coordinate, if so, using second three-dimensional coordinate as the end of current stroke
Three-dimensional coordinate determines that current stroke writing finishes.
Further, using first three-dimensional coordinate as determine stroke effective three-dimensional coordinate after, the method
Further include:
Judge second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
Further, after determining the corresponding every unicursal of text to be output, according to determining text to be output
Corresponding every unicursal, before determining corresponding first text of text to be output, the method also includes:
For every unicursal, determine whether the used time of the stroke is less than preset first duration threshold value;
If so, the stroke is filtered out.
Further, in the corresponding each coordinate information of text to be output according to preservation, the text to be output is determined
After corresponding target text, the method also includes:
Delete the corresponding each coordinate information of text to be output.
Further, if being directed to the previous frame image of current frame image, the two or three of the target gesture is not preserved
Coordinate is tieed up, the method also includes:
To the overlapping region pre-saved, there is no the frame numbers of the image of the target gesture to be updated.
Further, the method also includes:
When identification, which meets text, writes termination condition, at least one text of writing is subjected to voice broadcast.
Further, identify that meeting text writing termination condition includes:
Identification receives text and writes END instruction;Or
The frame number for recognizing the image that the target gesture is not present in the overlapping region pre-saved is greater than preset quantity.
The embodiment of the invention discloses a kind of character recognition device, described device includes:
Target gesture detection module, for detecting in the current frame image in the video flowing that camera acquires with the presence or absence of pre-
If for writing target gesture;
Coordinate information determining module is when being for the testing result in the target gesture detection module, it is determined that institute
The coordinate information of target gesture is stated, and is saved;
Text determining module determines described to defeated for the corresponding each coordinate information of text to be output according to preservation
The corresponding target text of text out;
Display module, for showing determining target text.
Further, the target gesture detection module, specifically for the video flowing acquired respectively according to two cameras
In current frame image, determine the overlapping region in two current frame images;Detection is used in overlapping region with the presence or absence of preset
The target gesture of writing;
The coordinate information determining module, when being specifically for the testing result in the target gesture detection module,
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
Further, the display module is also used to after the first three-dimensional coordinate for determining the target gesture, according to
The mapping relations of the writing display area of the overlapping region and display screen that pre-save, are mapped as institute for first three-dimensional coordinate
The first two-dimensional coordinate for writing display area is stated, is shown according to first two-dimensional coordinate in the writing display area.
Further, described device further includes:
First judgment module, for determining the first three-dimensional seat of the target gesture in the coordinate information determining module
After mark, according to first three-dimensional coordinate, judges whether current stroke is write and finish;
If the judging result of first judgment module be it is yes, trigger the text determining module.
Further, described device further includes:
When second judgment module for the testing result in the target gesture detection module is no, judgement is for current
The previous frame image of frame image, if preserve the second three-dimensional coordinate of the target gesture;If the second judgment module is sentenced
Disconnected result be it is yes, then trigger the text determining module.
Further, the text determining module, specifically for the corresponding each three-dimensional of text to be output according to preservation
Coordinate determines corresponding first text of the text to be output;
First text is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
Further, the text determining module, specifically for by highest second text of matching degree be determined as it is described to
The corresponding target text of output character;Or, showing the second text of preset quantity from high to low by matching degree;By user's selection
Second text is determined as the corresponding target text of the text to be output.
Further, the text determining module, specifically for according to the determining corresponding every unicursal of text to be output,
Determine corresponding first text of text to be output, wherein the corresponding every unicursal of text to be output is to determine that the stroke writes
After, it is determined according to the corresponding each three-dimensional coordinate of the stroke.
Further, the text determining module is specifically used for according to first three-dimensional coordinate and is directed to present frame figure
The first distance for the second three-dimensional coordinate that the previous frame image of picture saves, determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge that described second is three-dimensional
Coordinate whether be determining stroke effective three-dimensional coordinate, if so, using second three-dimensional coordinate as the end of current stroke
Three-dimensional coordinate determines that current stroke writing finishes.
Further, the text determining module is also used in having using first three-dimensional coordinate as determining stroke
Imitate three-dimensional coordinate after, judge second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
Further, the text determining module is also used to after determining the corresponding every unicursal of text to be output,
According to the determining corresponding every unicursal of text to be output, before determining corresponding first text of text to be output, for every
Unicursal, determines whether the used time of the stroke is less than preset first duration threshold value;If so, the stroke is filtered out.
Further, described device further includes:
Removing module, for deleting text to be output corresponding each seat after the completion of the text determining module executes
Mark information.
Further, the text determining module does not save if being also used to the previous frame image for current frame image
There is the second three-dimensional coordinate of the target gesture, there is no the frames of the image of the target gesture to the overlapping region pre-saved
Number is updated.
Further, described device further includes:
Voice broadcast module, for when identification meet text write termination condition when, by least one text of writing into
Row voice broadcast.
Further, the voice broadcast module receives text specifically for identification and writes END instruction;Or it recognizes
There is no the frame numbers of the image of the target gesture to be greater than preset quantity for the overlapping region pre-saved.
The embodiment of the invention discloses a kind of electronic equipment, including:Processor, communication interface, memory and communication bus,
Wherein, processor, communication interface, memory complete mutual communication by communication bus;
It is stored with computer program in the memory, when described program is executed by the processor, so that the place
Manage the step of device executes any of the above-described the method.
The embodiment of the invention discloses a kind of computer readable storage medium, it is stored with the meter that can be executed by electronic equipment
Calculation machine program, when described program is run on the electronic equipment, so that the electronic equipment executes described in any of the above-described
The step of method.
The invention discloses a kind of character recognition method, device, electronic equipment and storage medium, method includes:Root detection
With the presence or absence of the preset target gesture for writing in current frame image in the video flowing of camera acquisition;If it is,
It determines the coordinate information of the target gesture, and saves;According to the corresponding each coordinate information of the text to be output of preservation, determine
The corresponding target text of the text to be output, and export.Since when user, which faces sky, to write, electronic equipment is according to determining user's book
Each three-dimensional coordinate of target gesture when writing determines the text that user writes, realizes rapidly and accurately typing text.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Figure 1A is a kind of Text region process schematic provided in an embodiment of the present invention;
Figure 1B is a kind of Text region process schematic provided in an embodiment of the present invention;
Fig. 2 is a kind of strokes of characters determination process schematic diagram provided in an embodiment of the present invention;
Fig. 3 is a kind of character recognition device structure chart provided in an embodiment of the present invention;
Fig. 4 is a kind of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of character recognition method, this method is applied to electronic equipment, this method include with
Lower step:
It detects in the current frame image in the video flowing of camera acquisition with the presence or absence of the preset target hand for writing
Gesture;If it is, determining the coordinate information of the target gesture, and save;
According to the corresponding each coordinate information of the text to be output of preservation, the corresponding target text of the text to be output is determined
Word, and export.
Character recognition method provided in an embodiment of the present invention is applied to electronic equipment, which can be computer, hand
Machine etc..Camera on electronic equipment can acquire image information in the form of video recording, and electronic equipment is being video according to video recording
When stream information carries out Text region, every frame image according to acquisition can be, real-time perfoming Text region is also possible to first adopt
The video flowing of collection saves, and carries out Text region to the video flowing of preservation.
In embodiments of the present invention, for user when facing empty write, first typing is used for the gesture write in the electronic device for meeting,
The gesture can be palm, can be fist, can be finger, and the gesture for being used to write is known as target gesture.
Electronic equipment analyzes every frame image in video flowing, determines in every frame image with the presence or absence of the target hand
Gesture, if there is the target gesture, it is determined that the corresponding coordinate information of target gesture, and corresponding seat is saved for each frame
Mark information.
Electronic equipment can determine the text to be output according to the corresponding each coordinate information of text to be output of preservation
Corresponding target text, and export.
Electronic equipment is after being aware of target gesture and what is, it is determined whether there are the process of target gesture belong to it is existing
Technology, in embodiments of the present invention without repeating.
Since in embodiments of the present invention, user faces sky and writes, electronic equipment, which passes through, determines target gesture when user writes
Coordinate information, determine that user faces the text of empty writing, realize and rapidly and accurately determine text.
Embodiment 1:
In order to make determining text more accurately, in the current frame image in the video flowing of the detection camera acquisition
Include with the presence or absence of the preset target gesture for writing:
The current frame image in video flowing acquired respectively according to two cameras, determines the overlapping in two current frame images
Region;It detects in overlapping region with the presence or absence of the preset target gesture for writing;
The coordinate information of the determination target gesture includes:
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
Character recognition method provided in an embodiment of the present invention is applied to electronic equipment, which takes the photograph including at least two
As head, and there are there are overlapping regions in the picture frame of at least two cameras acquisition.
The embodiment of the present invention is that there are the video flowings of two cameras of overlapping region acquisition in the picture frame for acquisition
Carry out Text region.
Two cameras for carrying out video flowing acquisition can be any position setting, as long as the acquisition of two cameras
There are overlapping regions in region, and preferably, which forms binocular camera.Two cameras are in entire text
Position in identification process will not be moved under normal circumstances, herein under the premise of, two cameras during Text region,
The overlapping region of the image of any moment acquisition is generally identical.
Each camera continuous acquisition image, acquiring in each camera of synchronization has image, then two cameras
Two field pictures are acquired, electronic equipment can analyze the two field pictures of synchronization, determine two frame figures of synchronization
Overlapping region as in.
It will currently analyze, determine that the image of overlapping region is known as current frame image.
Determine that the process of overlapping region belongs to the prior art in two field pictures, in embodiments of the present invention without superfluous
It states.
Electronic equipment can determine the weight of two current frame images when analyzing the current frame image that camera acquires
Target gesture in folded region with the presence or absence of this for writing, if it does, it can be said that bright user is currently facing empty writing,
If it is not, illustrating user currently without carrying out facing empty writing.
Electronic equipment detection overlapping region belongs to the prior art with the presence or absence of the process of target gesture, in the embodiment of the present invention
In without repeating.
The text of electronic equipment output is known as target text, which can be understood as the text finally identified
Word.
Under normal circumstances, when electronic equipment in detecting overlapping region there are when target gesture, then can be heavy according to this
Folded region determines the three-dimensional coordinate of target gesture, and is directed to every frame image, saves the three-dimensional coordinate of target gesture.It will be according to working as
The three-dimensional coordinate that the corresponding overlapping region of previous frame determines is known as the first three-dimensional coordinate.
A moment, the three-dimensional coordinate of target gesture is one, and target gesture may be palm, it may be possible to fist, it can
It can be finger.Palm, fist, finger can generally occupy multiple coordinate points, can choose one of those, such as central point conduct
First three-dimensional coordinate of the target gesture.
The process of the above-mentioned three-dimensional coordinate that target gesture is determined according to overlapping region belongs to the prior art, implements in the present invention
Without repeating in example.
In embodiments of the present invention, under what conditions according to the corresponding each coordinate information of the text to be output of preservation,
Determine that a kind of specific implementation method of the corresponding target text of the text to be output can be:
If there is no the preset target gesture for writing in overlapping region, judgement is upper for current frame image
One frame image, if preserve the second three-dimensional coordinate of the target gesture;If it is, determining that currently to meet target text true
Fixed condition determines the corresponding target of the text to be output that is, according to the corresponding each coordinate information of the text to be output of preservation
Text.
If do not saved, analysis identification can be carried out using next frame image as current frame image.
If user does not have started writing or user has write some text at the time of upper picture frame corresponds to
It is complete, and currently when not carrying out the writing of next text also, current frame image will not detect target hand in corresponding overlapping region
Gesture, and target gesture will not be detected in the corresponding overlapping region of previous frame image of current frame image.
Target gesture, i.e. user are not present in determining the corresponding overlapping region of present frame currently without carrying out facing sky book
After writing, user can be determined according to whether there is target gesture in the corresponding overlapping region of previous frame image of current frame image
Whether a word is just write at the time of previous frame image corresponds to, if it is, thinking that meeting target text determines condition, need
Determine target text.
When whether there is target gesture in the corresponding overlapping region of previous frame image for judging current frame image, it can be
Judgement is directed to the previous frame image of current frame image, if the second three-dimensional coordinate for preserving the target gesture, by present frame figure
The three-dimensional coordinate of the corresponding target gesture of the previous frame image of picture is known as the second three-dimensional coordinate.If preserving the second three-dimensional seat
Mark, then can determine at the time of previous frame corresponds to, user just writes a word, then can determine the text.
It is pre-saved in electronic equipment and needs the corresponding each coordinate information of output character i.e. three-dimensional coordinate, it is current determining
When meeting target text and determining condition, can according to the corresponding each three-dimensional coordinate of text to be output of preservation, determine it is described to
The corresponding target text of output character.After determining target text, it can also be shown on the screen of the electronic device.
When user writes a text, the target gesture for writing can work as overlapping region always in overlapping region
In be not present target gesture when, then it is assumed that user has write a text.Target gesture is not present in overlapping region, can be
Hand is had left overlapping region to user or user has left overlapping region or user has changed a hand different from target gesture
Gesture.The corresponding each three-dimensional coordinate of text to be output can be each three-dimensional coordinate determined in first time period, wherein the
There is target gesture in the corresponding overlapping region of each frame image in one period, and each frame figure in first time period
There is the picture frame adjacent with current frame image as in.
Assuming that being not present in the 3rd frame, the 5th frame-there are target gesture in the corresponding overlapping region of the 1st the-the 4 frame image of frame
In the corresponding overlapping region of 10th frame there are target gestures, it has been determined that go out target gesture in the corresponding overlay region of each frame image
Three-dimensional coordinate in domain.Present frame is the 11st frame, and target gesture is not present in the corresponding overlapping region of current frame image, then can be with
According to each of determination in the 5th the-the 10 frame of frame, other three-dimensional coordinates determine text, and first time period is the 5th the-the 10 frame structure of frame
At period.1st the-the 4 frame of frame is a word writing of user, because target gesture is not present in the 3rd frame, it has been determined that goes out the
Text in 1 the-the 4 frame of frame, so without determining again.
In embodiments of the present invention, there are overlapping region in the pickup area of two cameras, electronic equipment can be preparatory
Determine the corresponding space lattice in overlapping region.Electronic equipment can determine that the three-dimensional is sat after determining three-dimensional coordinate every time
It is marked on corresponding pixel in space lattice, according to the corresponding each three-dimensional coordinate of text to be output, is determined described to be output
When the corresponding target text of text, it can be the three-dimensional of each pixel in the dot matrix according to the overlapping region pre-saved and sit
Mark, determines the corresponding each three-dimensional coordinate of text to be output corresponding target pixel points in dot matrix, according to determining each mesh
Pixel is marked, determines the corresponding target text of the text to be output, that is, the text that each target pixel points are formed is true
It is set to target text.The three-dimensional coordinate they target pixel points the corresponding three-dimensional coordinate in dot matrix and determined according to overlapping region
It is identical.
Figure 1A is a kind of Text region process schematic provided in an embodiment of the present invention, which includes the following steps:
S101:According to the current frame image that two cameras acquire respectively, the overlapping region in two current frame images is determined.
S102:It detects with the presence or absence of the preset target gesture for writing in overlapping region, if it is, carrying out
S103, if it is not, then carrying out S104.
S103:According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined, and save.
S104:Judgement is directed to the previous frame image of current frame image, if preserve the target gesture second is three-dimensional
Coordinate;If it is, S105 is carried out, if it is not, then carrying out analysis identification using next frame image as current frame image.
S105:According to the corresponding each three-dimensional coordinate of the text to be output of preservation, determine that the text to be output is corresponding
Target text, and export.
Since in embodiments of the present invention, user faces sky and writes, electronic equipment, which passes through, determines target gesture when user writes
Three-dimensional coordinate, determine that user faces the text of empty writing, realize and rapidly and accurately determine text.
Embodiment 2:
On the basis of the above embodiments, in embodiments of the present invention, in the first three-dimensional for determining the target gesture
After coordinate, in the corresponding each coordinate information of text to be output according to preservation, the corresponding mesh of the text to be output is determined
Before marking text, the method also includes:
According to first three-dimensional coordinate, judges whether current stroke is write and finish;
If it is, carrying out subsequent step.
In embodiments of the present invention, when user writes a text, the target gesture for writing can be overlapped always
In region, user is typically write according to stroke in writing words, and user writes a stroke of some text
Afterwards, in the next stroke for writing the text, target gesture will not leave overlapping region, then there are target gestures in overlapping region.
Electronic equipment after a three-dimensional coordinate of every determining target gesture, can according to the determining three-dimensional coordinate,
It determines whether stroke writes completion, is completed if stroke is write, then it is assumed that meet target text and determines condition, it can be according to preservation
The corresponding each coordinate information of text to be output, determine the corresponding target text of the text to be output.
By taking the first three-dimensional coordinate as an example, if detecting that there are target gestures in overlapping region, according to the first three-dimensional seat
Mark, determines whether current stroke writes completion, if it is, determining that currently meeting target text determines condition;If it is not, then really
It is unsatisfactory for target text before settled and determines condition.
Figure 1B is a kind of Text region process schematic provided in an embodiment of the present invention, which includes the following steps:
S106:According to the current frame image that two cameras acquire respectively, the overlapping region in two current frame images is determined.
S107:It detects with the presence or absence of the preset target gesture for writing in overlapping region, if it is, carrying out
S108。
S108:According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined, and save.
S109:According to the corresponding each three-dimensional coordinate of the text to be output of preservation, determine that the text to be output is corresponding
Target text, and export.
Embodiment 3:
The writing situation that oneself is checked in order to facilitate user, on the basis of the various embodiments described above, in the embodiment of the present invention
In, after the first three-dimensional coordinate for determining the target gesture, the method also includes:
It is three-dimensional by described first according to the mapping relations of the writing display area of the overlapping region and display screen pre-saved
Coordinate is mapped as first two-dimensional coordinate for writing display area, according to first two-dimensional coordinate in the writing viewing area
Domain is shown.
In embodiments of the present invention, there is writing display area in the display screen of electronic equipment, which can be with
Show the writing situation of user, the mapping that the writing display area of overlapping region and display screen is pre-saved in electronic equipment is closed
Three-dimensional coordinate, can be mapped to two-dimensional coordinate according to the mapping relations, that is, after determining the first three-dimensional coordinate by system, can
First three-dimensional coordinate according to the mapping relations, to be mapped as to writing the first two-dimensional coordinate of display area, and then according to the
One two-dimensional coordinate is shown in writing display area, specifically be can be and is shown first two-dimensional coordinate pair in writing display area
The pixel answered.
It under normal circumstances, can be in two after writing display area and showing the corresponding each three-dimensional coordinate mapping of text to be output
Tie up the corresponding pixel of coordinate.The then pixel that user can show according to display area is write, checks the writing situation of oneself,
It corrects a mistake promptly, improves and write accuracy.
Embodiment 4:
In order to keep the target text of output more accurate, on the basis of the various embodiments described above, in embodiments of the present invention,
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, determine that the corresponding target text of the text to be output includes:
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, corresponding first text of the text to be output is determined
Word;
First text is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
In embodiments of the present invention, literal pool has been pre-saved in electronic equipment, has determined each of text to be output
Coordinate information that is, after three-dimensional coordinate, can determine corresponding first text of text to be output, specifically according to each three-dimensional coordinate
Process is similar with the process in above-described embodiment, is no longer repeated herein.
After determining the first text, which can be matched with the text in literal pool, according to matching
As a result, determining the corresponding target text of text to be output, matching result be can be without matched text, then can not export target text
Word, it is also possible to which matched text has multiple.
It can be when according to matching result, determining the corresponding target text of the text to be output matching degree is highest
Second text is determined as the corresponding target text of the text to be output;It is also possible to show present count from high to low by matching degree
Second text of amount;The second text that user selects is determined as the corresponding target text of the text to be output.
First text is matched with literal pool, determines of the second text and the first text determined in literal pool
Process with degree height belongs to the prior art, in embodiments of the present invention without repeating.
A kind of specific embodiment:There is no the preset target gestures for writing in overlapping region, and are directed to and work as
The previous frame image of prior image frame preserves the second three-dimensional coordinate of the target gesture, it is determined that meets target text and determines
Condition, the corresponding each three-dimensional coordinate of the subsequent text to be output according to preservation, determines the text to be output corresponding first
Text;First text is matched with the literal pool pre-saved, highest second text of matching degree is determined as institute
The corresponding target text of text to be output is stated, or shows the second text of preset quantity from high to low by matching degree;By user
Second text of selection is determined as the corresponding target text of the text to be output.
Another specific embodiment:There are the preset target gestures for writing in overlapping region, and according to institute
The first three-dimensional coordinate is stated, determines that current stroke writing finishes, it is determined that meet target text and determines condition, it is subsequent according to preservation
The corresponding each three-dimensional coordinate of text to be output determines corresponding first text of the text to be output, that is, according to defeated
The corresponding every unicursal of text out determines corresponding first text of the text to be output;By first text and preparatory guarantor
The literal pool deposited is matched, and shows the second text of preset quantity from high to low by matching degree;By the second text of user's selection
Word is determined as the corresponding target text of the text to be output.
Due in embodiments of the present invention, the first text being matched with the second text in literal pool, determines mesh
Text is marked, realizes and more accurately determines text.
Embodiment 5:
User is typically write according to fixed stroke in writing words, in the base of the various embodiments described above
On plinth, in embodiments of the present invention, according to the corresponding each three-dimensional coordinate of the text to be output of preservation, the text to be output is determined
Corresponding first text of word includes:
According to the determining corresponding every unicursal of text to be output, corresponding first text of text to be output is determined, wherein
The corresponding every unicursal of text to be output is to determine the stroke after finishing writing, according to the corresponding each three-dimensional coordinate of the stroke
Determining.
In embodiments of the present invention, it when determining the first text, can be according to the corresponding each three-dimensional of text to be output
Coordinate determines the corresponding every unicursal of text to be output, subsequent to determine text pair to be output according to determining every unicursal
The first text answered.
Whether electronic equipment can determine stroke in the three-dimensional coordinate for determining the corresponding target gesture of every frame image every time
It writes and completes, if it is determined that after the completion of stroke is write, be determined the process of stroke, be also possible to determining that user write
After a complete text, i.e., the second three-dimensional seat of the target gesture is preserved in the determining previous frame image for current frame image
After mark, it is determined the process of stroke.It can also be that any moment before determining text determines stroke.
In embodiments of the present invention, after determining each three-dimensional coordinate, it can determine whether stroke writes completion, with
For determining the first three-dimensional coordinate, be illustrated, as shown in Fig. 2, according to the first three-dimensional coordinate, determine current stroke whether book
Complete process is write to include the following steps:
S201:According to the first distance of first three-dimensional coordinate and the second three-dimensional coordinate saved, current write is determined
Speed.
Wherein, the second three-dimensional coordinate is the three-dimensional coordinate saved for the previous frame image of current frame image.
S202:Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;If it is, into
Row S203, if it is not, then carrying out S207.
S203:Using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke.
S207:Using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;And judge that described second is three-dimensional
Whether coordinate is effective three-dimensional coordinate of determining stroke, if it is, carrying out S208.
S208:Using second three-dimensional coordinate as the end three-dimensional coordinate of current stroke, determine that current stroke has been write
Finish.
It is possible to further the corresponding each effective three-dimensional coordinate of the current stroke according to preservation, text to be output is determined
Corresponding current stroke.
In embodiments of the present invention, for general user in writing words, the writing speed for having oneself conventional is writing a certain pen
When picture, usually at the uniform velocity.
Electronic equipment after the three-dimensional coordinate for determining target gesture every time, can according to this time determine three-dimensional coordinate with
The three-dimensional coordinate that last time determines compares, it is determined whether has completed the writing of unicursal.
Firstly, by taking current frame image as an example, electronic equipment can be according to corresponding first three-dimensional coordinate of present frame and current
Corresponding second three-dimensional coordinate of the previous frame of frame, determines current writing speed, the specific can be that, determine the first three-dimensional coordinate with
The first distance of second three-dimensional coordinate is determined currently to write speed according to the time interval of first distance and adjacent two picture frame
Degree.
The velocity interval of stroke writing can be pre-saved in electronic equipment, which can be user in electronics
It is arranged in equipment, is also possible to what electronic equipment was determined according to the painting and calligraphy writing speed in preset time period.
After determining current writing speed, it can be determined that whether current writing speed, which is located at the stroke pre-saved, is write
In velocity interval, if it is, thinking that the writing of user is that normal stroke is write, then using the first three-dimensional coordinate as determining pen
The effective three-dimensional coordinate drawn is write in range if being not located at stroke, it is determined that the improper stroke of the writing of user is write, then
Using the first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke.
If the first three-dimensional coordinate be determine stroke effective three-dimensional coordinate, can also according to the second three-dimensional coordinate whether be
Effective three-dimensional coordinate determines effect of first three-dimensional coordinate in stroke, specifically, using first three-dimensional coordinate as really
After the effective three-dimensional coordinate for determining stroke, as shown in Fig. 2, the method also includes following steps:
S204:Judge second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;If it is, carrying out
S205, if it is not, then carrying out S206.
S206:Using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
S205:Using first three-dimensional coordinate as the intermediate three-dimensional coordinate of current stroke.
It whether can also be to have according to two-dimensional coordinate if the first three-dimensional coordinate is the invalid three-dimensional coordinate for determining stroke
Effect three-dimensional coordinate determines whether the stroke writes, specifically, judging whether second three-dimensional coordinate is the effective of determining stroke
Three-dimensional coordinate is sat if it is, determining that the stroke writes using second three-dimensional coordinate as the end three-dimensional of current stroke
Mark, if the second three-dimensional coordinate is the invalid three-dimensional coordinate for determining stroke, then it is assumed that current writing is not the normal pen of text
It draws.
Using the second three-dimensional coordinate as the end three-dimensional coordinate of current stroke after, so that it may according to the current pen of preservation
Corresponding each effective three-dimensional coordinate is drawn, determines the corresponding current stroke of text to be output.
The corresponding each effectively three-dimensional coordinate of current stroke includes the beginning three-dimensional coordinate of current stroke, terminates three-dimensional coordinate
And intermediate three-dimensional coordinate.Start the corresponding three-dimensional coordinate of starting point that three-dimensional coordinate is stroke, terminates the end that three-dimensional coordinate is stroke
The corresponding unit coordinate of point, start three-dimensional coordinate fix time and terminate really three-dimensional coordinate fix time really between period it is true
Fixed is the intermediate three-dimensional coordinate of stroke, either starting point, and terminal or the corresponding point of intermediate three-dimensional coordinate, are all in dot matrix
Pixel.
Above-mentioned every unicursal represents the motion profile of target gesture, and the starting point of stroke is the starting point of motion profile, pen
The end point of picture is the end point of motion profile.
Embodiment 6:
When facing sky and writing, each movement of target gesture can be recorded user, will lead to be not belonging in this way
The stroke of text can also can be regarded as a part of text, in order to improve the accuracy of determining text, in the base of the various embodiments described above
On plinth, in embodiments of the present invention, after determining the corresponding every unicursal of text to be output, according to determining to be output
The corresponding every unicursal of text, before determining corresponding first text of text to be output, the method also includes:
For every unicursal, determine whether the used time of the stroke is less than preset first duration;
If it is, the stroke is filtered out.
In embodiments of the present invention, the first duration has been pre-saved in electronic equipment, every unicursal in normal words exists
When writing, the used time be it is longer, when there is the stroke for being not belonging to text, it is believed that the writing duration is shorter, in order to more accurate
Ground determines text, useless stroke can be filtered out, and specifically can be for every unicursal, when determine the stroke used time whether
When less than preset first duration, which is filtered out, the first text is determined according to remaining stroke.
The process for determining the used time of stroke includes:The frame number for determining the corresponding picture frame of the stroke, according to the frame number and phase
The time interval of adjacent two field pictures determines the used time of the stroke.
Embodiment 7:
In order to keep determining text more accurate, and for electronic equipment memory space is saved, in the various embodiments described above
On the basis of, in embodiments of the present invention, in the corresponding each coordinate information of text to be output according to preservation, determine described to defeated
Out after the corresponding target text of text, the method also includes:
Delete the corresponding each coordinate information of text to be output.
Electronic equipment, can be by the corresponding coordinate information of the text, such as three-dimensional coordinate after determining a text every time
Delete, in this way when determining next text, can understand accurately know which the corresponding three-dimensional coordinate of next text has, i.e., to
The corresponding each three-dimensional coordinate of output character can be the whole three-dimensional coordinates currently saved in electronic equipment, improves and determines text
The accuracy of word.On the other hand, memory space also is saved for electronic equipment.
Embodiment 8:
On the basis of the various embodiments described above, in embodiments of the present invention, if being directed to the previous frame figure of current frame image
Picture does not preserve the second three-dimensional coordinate of the target gesture, the method also includes:
To the overlapping region pre-saved, there is no the frame numbers of the image of the target gesture to be updated.
When target gesture is not present in electronic equipment in determining the corresponding overlapping region of a certain frame, record overlapping can be
There is no the frame numbers of the image of target gesture in region, if target hand is not present in the corresponding overlapping region of the previous frame of present frame
Gesture can then be updated the frame number of preservation, under normal circumstances to add 1.
Embodiment 9:
User is generally not only to write a text when face empty write, but multiple words, and electronic equipment can be with
Whether identification user writes completion, on the basis of the various embodiments described above, in embodiments of the present invention, the method also includes:
When identification, which meets text, writes termination condition, at least one text of writing is subjected to voice broadcast.
In embodiments of the present invention, text is pre-saved in electronic equipment and writes the condition terminated, when identification meets text
When word writes termination condition, the text write can be shown for user, in order to further increase user experience, electronic equipment may be used also
It is broadcasted with the text for writing user.
Electronic equipment can be when identifying whether that meeting text writes termination condition and identify whether to receive text writing
END instruction;Termination condition is write if it is, determining and meeting text.
Text, which writes END instruction, can be end operation or electronic equipment reception of the user on electronic equipment screen
The END instruction sent to other equipment.
Electronic equipment can also be when identifying whether to meet text and writing termination condition and detect in overlapping region and not deposit
Whether it is more than preset second duration in the preset target gesture for writing, writes knot if it is, determining and meeting text
Beam condition.It specifically can be, the frame number for recognizing the image that the target gesture is not present in the overlapping region pre-saved is greater than
Preset quantity, then it is assumed that meet text and write termination condition.The quantity is generally higher than 2, can be 5, can be 10.
Embodiment 10:
Fig. 3 is a kind of character recognition device structure chart provided in an embodiment of the present invention, is applied to include at least two camera shootings
The character recognition device of head, described device include:
Target gesture detection module 31 whether there is for detecting in the current frame image in the video flowing that camera acquires
The preset target gesture for writing;
Coordinate information determining module 32 is when being for the testing result in the target gesture detection module, it is determined that
The coordinate information of the target gesture, and save;
Text determining module 33, for the corresponding each coordinate information of text to be output according to preservation, determine it is described to
The corresponding target text of output character;
Display module 34, for showing determining target text.
Further, the target gesture detection module 31, specifically for the video acquired respectively according to two cameras
Current frame image in stream determines the overlapping region in two current frame images;It detects and whether there is preset use in overlapping region
In the target gesture of writing;
The coordinate information determining module 32 is yes specifically for the testing result in the target gesture detection module
When, according to the overlapping region, determine the first three-dimensional coordinate of the target gesture.
Further, the display module 34 is also used to after the first three-dimensional coordinate for determining the target gesture, root
According to the mapping relations of the writing display area of the overlapping region and display screen pre-saved, first three-dimensional coordinate is mapped as
First two-dimensional coordinate for writing display area, is shown according to first two-dimensional coordinate in the writing display area
Show.
Further, described device further includes:
First judgment module 35, for determining the 1st of the target gesture the in the coordinate information determining module 32
After tieing up coordinate, according to first three-dimensional coordinate, judges whether current stroke is write and finish;
If the judging result of first judgment module 35 be it is yes, trigger the text determining module 33.
Further, described device further includes:
Second judgment module 36, for the testing result in the target gesture detection module 31 be it is no when, judgement is directed to
The previous frame image of current frame image, if preserve the second three-dimensional coordinate of the target gesture;If the second judgment module
36 judging result be it is yes, then trigger the text determining module 33.
Further, the text determining module 33, specifically for the text to be output corresponding each three according to preservation
Coordinate is tieed up, determines corresponding first text of the text to be output;
The candidate character is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
Further, the text determining module 33, it is described specifically for highest second text of matching degree to be determined as
The corresponding target text of text to be output;Or, showing the second text of preset quantity from high to low by matching degree;User is selected
The second text be determined as the corresponding target text of the text to be output.
Further, the text determining module 33, specifically for according to the determining corresponding each pen of text to be output
It draws, determines corresponding first text of text to be output, wherein the corresponding every unicursal of text to be output is to determine the stroke book
After writing, determined according to the corresponding each three-dimensional coordinate of the stroke.
Further, the text determining module 33 is specifically used for according to first three-dimensional coordinate and is directed to present frame
The first distance for the second three-dimensional coordinate that the previous frame image of image saves, determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge that described second is three-dimensional
Coordinate whether be determining stroke effective three-dimensional coordinate, if so, using second three-dimensional coordinate as the end of current stroke
Three-dimensional coordinate determines that current stroke writing finishes.
Further, the text determining module 33 is also used to using first three-dimensional coordinate as determining stroke
After effective three-dimensional coordinate, judge second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
Further, the text determining module 33, be also used to determine the corresponding every unicursal of text to be output it
Afterwards, according to the determining corresponding every unicursal of text to be output, before determining corresponding first text of text to be output, for
Every unicursal, determines whether the used time of the stroke is less than preset first duration threshold value;If so, the stroke is filtered out.
Further, described device further includes:
Removing module 37, it is corresponding every for after the completion of the text determines 33 pieces of execution of mould, deleting text to be output
A coordinate information.
Further, the text determining module 33 is not protected if being also used to the previous frame image for current frame image
There is the second three-dimensional coordinate of the target gesture, there is no the images of the target gesture to the overlapping region pre-saved
Frame number is updated.
Further, described device further includes:
Voice broadcast module 38, for when identification meet text write termination condition when, by least one text of writing
Carry out voice broadcast.
Further, the voice broadcast module 38 receives text specifically for identification and writes END instruction;Or identification
To the overlapping region pre-saved, there is no the frame numbers of the image of the target gesture to be greater than preset quantity.
Embodiment 11:
Fig. 4 is a kind of electronic equipment provided in an embodiment of the present invention, including:Processor 41, communication interface 42, memory 43
With communication bus 44, wherein processor 41, communication interface 42, memory 43 complete mutual communication by communication bus 44;
It is stored with computer program in the memory, when described program is executed by the processor, so that the place
It manages device and executes following steps:
It detects in the current frame image in the video flowing of camera acquisition with the presence or absence of the preset target hand for writing
Gesture;If it is, determining the coordinate information of the target gesture, and save;
According to the corresponding each coordinate information of the text to be output of preservation, the corresponding target text of the text to be output is determined
Word, and export.
Further, it is used in the current frame image in the video flowing of the detection camera acquisition with the presence or absence of preset
The target gesture of writing includes:
The current frame image in video flowing acquired respectively according to two cameras, determines the overlapping in two current frame images
Region;It detects in overlapping region with the presence or absence of the preset target gesture for writing;
The coordinate information of the determination target gesture includes:
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
Further, after the first three-dimensional coordinate for determining the target gesture, according to the overlapping region pre-saved
With the mapping relations of the writing display area of display screen, first three-dimensional coordinate is mapped as described to write the of display area
One two-dimensional coordinate is shown according to first two-dimensional coordinate in the writing display area.
Further, after the first three-dimensional coordinate for determining the target gesture, in the text to be output according to preservation
The corresponding each coordinate information of word, before determining the corresponding target text of the text to be output, the method also includes:
According to first three-dimensional coordinate, judges whether current stroke is write and finish;
If it is, carrying out subsequent step.
Further, if there is no the preset target gesture for writing in overlapping region, according to preservation to
The corresponding each coordinate information of output character, before determining the corresponding target text of the text to be output, the method is also wrapped
It includes:
Judgement is directed to the previous frame image of current frame image, if preserves the second three-dimensional coordinate of the target gesture;
If it is, carrying out subsequent step.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
The corresponding target text of word includes:
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, corresponding first text of the text to be output is determined
Word;
First text is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
Further, described according to matching result, determine that the corresponding target text of the text to be output includes:
Highest second text of matching degree is determined as the corresponding target text of the text to be output;Or,
Show the second text of preset quantity from high to low by matching degree;The second text that user selects is determined as described
The corresponding target text of text to be output.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
Corresponding first text of word includes:
According to the determining corresponding every unicursal of text to be output, corresponding first text of text to be output is determined, wherein
The corresponding every unicursal of text to be output is to determine the stroke after finishing writing, according to the corresponding each three-dimensional coordinate of the stroke
Determining.
Further, according to the first three-dimensional coordinate, determine whether current stroke writes the process finished and include:
According to first three-dimensional coordinate and the second three-dimensional coordinate of the previous frame image preservation for current frame image
First distance determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge that described second is three-dimensional
Coordinate whether be determining stroke effective three-dimensional coordinate, if so, using second three-dimensional coordinate as the end of current stroke
Three-dimensional coordinate determines that current stroke writing finishes.
Further, using first three-dimensional coordinate as determine stroke effective three-dimensional coordinate after, described in judgement
Second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
Further, after determining the corresponding every unicursal of text to be output, according to determining text to be output
Corresponding every unicursal, for every unicursal, determines the used time of the stroke before determining corresponding first text of text to be output
Whether preset first duration threshold value is less than;
If so, the stroke is filtered out.
Further, in the corresponding each coordinate information of text to be output according to preservation, the text to be output is determined
After corresponding target text, the corresponding each coordinate information of text to be output is deleted.
Further, if being directed to the previous frame image of current frame image, the two or three of the target gesture is not preserved
Coordinate is tieed up, there is no the frame numbers of the image of the target gesture to be updated to the overlapping region pre-saved.
Further, when identification, which meets text, writes termination condition, at least one text of writing is subjected to voice and is broadcast
Report.
Further, identify that meeting text writing termination condition includes:
Identification receives text and writes END instruction;Or
The frame number for recognizing the image that the target gesture is not present in the overlapping region pre-saved is greater than preset quantity.
The communication bus that electronic equipment in the various embodiments described above is mentioned can be Peripheral Component Interconnect standard
(Peripheral Component Interconnect, PCI) bus or expanding the industrial standard structure (Extended
Industry Standard Architecture, EISA) bus etc..It is total that the communication bus can be divided into address bus, data
Line, control bus etc..Only to be indicated with a thick line in figure, it is not intended that an only bus or a type convenient for indicating
The bus of type.
Communication interface, for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit, network processing unit (Network
Processor, NP) etc.;It can also be digital signal processor (Digital Signal Processing, DSP), dedicated collection
At circuit, field programmable gate array or other programmable logic device, discrete gate or transistor logic, discrete hard
Part component etc..
Embodiment 10:
The embodiment of the invention provides a kind of computer readable storage medium, it is stored with the meter that can be executed by electronic equipment
Calculation machine program, when described program is run on the electronic equipment, so that the electronic equipment executes following steps:
It detects in the current frame image in the video flowing of camera acquisition with the presence or absence of the preset target hand for writing
Gesture;If it is, determining the coordinate information of the target gesture, and save;
According to the corresponding each coordinate information of the text to be output of preservation, the corresponding target text of the text to be output is determined
Word, and export.
Further, it is used in the current frame image in the video flowing of the detection camera acquisition with the presence or absence of preset
The target gesture of writing includes:
The current frame image in video flowing acquired respectively according to two cameras, determines the overlapping in two current frame images
Region;It detects in overlapping region with the presence or absence of the preset target gesture for writing;
The coordinate information of the determination target gesture includes:
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
Further, after the first three-dimensional coordinate for determining the target gesture, according to the overlapping region pre-saved
With the mapping relations of the writing display area of display screen, first three-dimensional coordinate is mapped as described to write the of display area
One two-dimensional coordinate is shown according to first two-dimensional coordinate in the writing display area.
Further, after the first three-dimensional coordinate for determining the target gesture, in the text to be output according to preservation
The corresponding each coordinate information of word, before determining the corresponding target text of the text to be output, the method also includes:
According to first three-dimensional coordinate, judges whether current stroke is write and finish;
If it is, carrying out subsequent step.
Further, if there is no the preset target gesture for writing in overlapping region, according to preservation to
The corresponding each coordinate information of output character, before determining the corresponding target text of the text to be output, the method is also wrapped
It includes:
Judgement is directed to the previous frame image of current frame image, if preserves the second three-dimensional coordinate of the target gesture;
If it is, carrying out subsequent step.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
The corresponding target text of word includes:
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, corresponding first text of the text to be output is determined
Word;
First text is matched with the literal pool pre-saved, according to matching result, is determined described to be output
The corresponding target text of text.
Further, described according to matching result, determine that the corresponding target text of the text to be output includes:
Highest second text of matching degree is determined as the corresponding target text of the text to be output;Or,
Show the second text of preset quantity from high to low by matching degree;The second text that user selects is determined as described
The corresponding target text of text to be output.
Further, the corresponding each three-dimensional coordinate of the text to be output according to preservation, determines the text to be output
Corresponding first text of word includes:
According to the determining corresponding every unicursal of text to be output, corresponding first text of text to be output is determined, wherein
The corresponding every unicursal of text to be output is to determine the stroke after finishing writing, according to the corresponding each three-dimensional coordinate of the stroke
Determining.
Further, according to the first three-dimensional coordinate, determine whether current stroke writes the process finished and include:
According to first three-dimensional coordinate and the second three-dimensional coordinate of the previous frame image preservation for current frame image
First distance determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge that described second is three-dimensional
Coordinate whether be determining stroke effective three-dimensional coordinate, if so, using second three-dimensional coordinate as the end of current stroke
Three-dimensional coordinate determines that current stroke writing finishes.
Further, using first three-dimensional coordinate as determine stroke effective three-dimensional coordinate after, described in judgement
Second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
Further, after determining the corresponding every unicursal of text to be output, according to determining text to be output
Corresponding every unicursal, for every unicursal, determines the used time of the stroke before determining corresponding first text of text to be output
Whether preset first duration threshold value is less than;
If so, the stroke is filtered out.
Further, in the corresponding each coordinate information of text to be output according to preservation, the text to be output is determined
After corresponding target text, the corresponding each coordinate information of text to be output is deleted.
Further, if being directed to the previous frame image of current frame image, the two or three of the target gesture is not preserved
Coordinate is tieed up, there is no the frame numbers of the image of the target gesture to be updated to the overlapping region pre-saved.
Further, when identification, which meets text, writes termination condition, at least one text of writing is subjected to voice and is broadcast
Report.
Further, identify that meeting text writing termination condition includes:
Identification receives text and writes END instruction;Or
The frame number for recognizing the image that the target gesture is not present in the overlapping region pre-saved is greater than preset quantity.
Computer readable storage medium in above-described embodiment can be times that the processor in electronic equipment can access
What usable medium or data storage device, including but not limited to magnetic storage such as floppy disk, hard disk, tape, magneto-optic disk (MO) etc.,
Optical memory such as CD, DVD, BD, HVD etc. and semiconductor memory such as ROM, EPROM, EEPROM, nonvolatile memory
(NAND FLASH), solid state hard disk (SSD) etc..
The invention discloses a kind of character recognition method, device, electronic equipment and storage medium, method includes:Root detection
With the presence or absence of the preset target gesture for writing in current frame image in the video flowing of camera acquisition;If it is,
It determines the coordinate information of the target gesture, and saves;According to the corresponding each coordinate information of the text to be output of preservation, determine
The corresponding target text of the text to be output, and export.Since when user, which faces sky, to write, electronic equipment is according to determining user's book
Each three-dimensional coordinate of target gesture when writing determines the text that user writes, realizes rapidly and accurately typing text.
For systems/devices embodiment, since it is substantially similar to the method embodiment, so the comparison of description is simple
Single, the relevent part can refer to the partial explaination of embodiments of method.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or an operation are distinguished with another entity or another operation, without necessarily requiring or implying these entities
Or there are any actual relationship or orders between operation.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, the reality of complete hardware embodiment, complete Application Example or connected applications and hardware aspect can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although the preferred embodiment of the application has been described, it is created once a person skilled in the art knows basic
Property concept, then additional changes and modifications can be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as
It selects embodiment and falls into all change and modification of the application range.
Obviously, those skilled in the art can carry out various modification and variations without departing from the essence of the application to the application
Mind and range.In this way, if these modifications and variations of the application belong to the range of the claim of this application and its equivalent technologies
Within, then the application is also intended to include these modifications and variations.
Claims (32)
1. a kind of character recognition method, which is characterized in that it is applied to electronic equipment, the method includes:
It detects in the current frame image in the video flowing of camera acquisition with the presence or absence of the preset target gesture for writing;Such as
Fruit is, it is determined that the coordinate information of the target gesture, and save;
According to the corresponding each coordinate information of the text to be output of preservation, the corresponding target text of the text to be output is determined,
And it exports.
2. the method as described in claim 1, which is characterized in that the present frame figure in the video flowing of the detection camera acquisition
Include with the presence or absence of the preset target gesture for writing as in:
The current frame image in video flowing acquired respectively according to two cameras, determines the overlay region in two current frame images
Domain;It detects in overlapping region with the presence or absence of the preset target gesture for writing;
The coordinate information of the determination target gesture includes:
According to the overlapping region, the first three-dimensional coordinate of the target gesture is determined.
3. method according to claim 2, which is characterized in that after the first three-dimensional coordinate for determining the target gesture,
The method also includes:
According to the mapping relations of the writing display area of the overlapping region and display screen pre-saved, by first three-dimensional coordinate
Be mapped as it is described write display area the first two-dimensional coordinate, according to first two-dimensional coordinate the writing display area into
Row display.
4. method according to claim 2, which is characterized in that the first three-dimensional coordinate for determining the target gesture it
Afterwards, in the corresponding each coordinate information of text to be output according to preservation, the corresponding target text of the text to be output is determined
Before, the method also includes:
According to first three-dimensional coordinate, judges whether current stroke is write and finish;
If it is, carrying out subsequent step.
5. method according to claim 2, which is characterized in that if there is no the preset mesh for writing in overlapping region
It marks gesture and determines the corresponding target of the text to be output in the corresponding each coordinate information of text to be output according to preservation
Before text, the method also includes:
Judgement is directed to the previous frame image of current frame image, if preserves the second three-dimensional coordinate of the target gesture;If
It is then to carry out subsequent step.
6. method as described in claim 4 or 5, which is characterized in that according to the corresponding each three-dimensional of the text to be output of preservation
Coordinate determines that the corresponding target text of the text to be output includes:
According to the corresponding each three-dimensional coordinate of the text to be output of preservation, corresponding first text of the text to be output is determined;
First text is matched with the literal pool pre-saved, according to matching result, determines the text to be output
Corresponding target text.
7. method as claimed in claim 6, which is characterized in that it is described according to matching result, determine the text pair to be output
The target text answered includes:
Highest second text of matching degree is determined as the corresponding target text of the text to be output;Or,
Show the second text of preset quantity from high to low by matching degree;The second text that user selects is determined as described to defeated
The corresponding target text of text out.
8. method as claimed in claim 6, which is characterized in that the corresponding each three-dimensional of the text to be output according to preservation
Coordinate determines that corresponding first text of the text to be output includes:
According to the determining corresponding every unicursal of text to be output, corresponding first text of text to be output is determined, wherein to defeated
The corresponding every unicursal of text is to determine the stroke after finishing writing out, is determined according to the corresponding each three-dimensional coordinate of the stroke
's.
9. the method as described in claim 4 or 8, which is characterized in that according to the first three-dimensional coordinate, determine current stroke whether book
Writing complete process includes:
According to the first of the second three-dimensional coordinate that first three-dimensional coordinate is saved with the previous frame image for current frame image
Distance determines current writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge second three-dimensional coordinate
Whether be determining stroke effective three-dimensional coordinate, if so, second three-dimensional coordinate is three-dimensional as the end of current stroke
Coordinate determines that current stroke writing finishes.
10. method as claimed in claim 9, which is characterized in that in having using first three-dimensional coordinate as determining stroke
After imitating three-dimensional coordinate, the method also includes:
Judge second three-dimensional coordinate whether be determining stroke effective three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
11. method according to claim 8, which is characterized in that after determining the corresponding every unicursal of text to be output,
According to the determining corresponding every unicursal of text to be output, before determining corresponding first text of text to be output, the side
Method further includes:
For every unicursal, determine whether the used time of the stroke is less than preset first duration threshold value;
If so, the stroke is filtered out.
12. the method as described in claim 1, which is characterized in that in the corresponding each coordinate of text to be output according to preservation
Information, after determining the corresponding target text of the text to be output, the method also includes:
Delete the corresponding each coordinate information of text to be output.
13. method as claimed in claim 5, which is characterized in that if being directed to the previous frame image of current frame image, do not save
There is the second three-dimensional coordinate of the target gesture, the method also includes:
To the overlapping region pre-saved, there is no the frame numbers of the image of the target gesture to be updated.
14. method as claimed in claim 13, which is characterized in that the method also includes:
When identification, which meets text, writes termination condition, at least one text of writing is subjected to voice broadcast.
15. method as claimed in claim 14, which is characterized in that identification meets text writing termination condition and includes:
Identification receives text and writes END instruction;Or
The frame number for recognizing the image that the target gesture is not present in the overlapping region pre-saved is greater than preset quantity.
16. a kind of character recognition device, which is characterized in that described device includes:
Target gesture detection module, for detecting in the current frame image in the video flowing that camera acquires with the presence or absence of preset
Target gesture for writing;
Coordinate information determining module is when being for the testing result in the target gesture detection module, it is determined that the mesh
The coordinate information of gesture is marked, and is saved;
Text determining module determines the text to be output for the corresponding each coordinate information of text to be output according to preservation
The corresponding target text of word;
Display module, for showing determining target text.
17. device as claimed in claim 16, which is characterized in that the target gesture detection module is specifically used for according to two
The current frame image in video flowing that a camera acquires respectively, determines the overlapping region in two current frame images;Detection overlapping
With the presence or absence of the preset target gesture for writing in region;
The coordinate information determining module, when being specifically for the testing result in the target gesture detection module, according to
The overlapping region determines the first three-dimensional coordinate of the target gesture.
18. device as claimed in claim 16, which is characterized in that the display module is also used to determining the target hand
After first three-dimensional coordinate of gesture, according to the overlapping region and display screen pre-saved writing display area mapping relations,
First three-dimensional coordinate is mapped as first two-dimensional coordinate for writing display area, is existed according to first two-dimensional coordinate
The writing display area is shown.
19. device as claimed in claim 17, which is characterized in that described device further includes:
First judgment module, for determined in the coordinate information determining module target gesture the first three-dimensional coordinate it
Afterwards, according to first three-dimensional coordinate, judge whether current stroke is write and finish;
If the judging result of first judgment module be it is yes, trigger the text determining module.
20. device as claimed in claim 17, which is characterized in that described device further includes:
Second judgment module, for the testing result in the target gesture detection module be it is no when, judgement be directed to present frame figure
The previous frame image of picture, if preserve the second three-dimensional coordinate of the target gesture;If the judgement knot of the second judgment module
Fruit be it is yes, then trigger the text determining module.
21. the device as described in claim 19 or 20, which is characterized in that the text determining module is specifically used for according to guarantor
The corresponding each three-dimensional coordinate of the text to be output deposited determines corresponding first text of the text to be output;
First text is matched with the literal pool pre-saved, according to matching result, determines the text to be output
Corresponding target text.
22. device as claimed in claim 21, which is characterized in that the text determining module is specifically used for matching degree most
The second high text is determined as the corresponding target text of the text to be output;Or, showing present count from high to low by matching degree
Second text of amount;The second text that user selects is determined as the corresponding target text of the text to be output.
23. device as claimed in claim 21, which is characterized in that the text determining module, specifically for according to determining
The corresponding every unicursal of text to be output, determines corresponding first text of text to be output, wherein text to be output is corresponding every
Unicursal is to determine the stroke after finishing writing, is determined according to the corresponding each three-dimensional coordinate of the stroke.
24. the device as described in claim 19 or 23, which is characterized in that the text determining module is specifically used for according to institute
The first distance for the second three-dimensional coordinate that the first three-dimensional coordinate is saved with the previous frame image for current frame image is stated, determination is worked as
Preceding writing speed;
Judge whether current writing speed is located within the scope of the stroke writing speed pre-saved;
If so, using first three-dimensional coordinate as the effective three-dimensional coordinate for determining stroke;
If not, using first three-dimensional coordinate as the invalid three-dimensional coordinate for determining stroke;Judge second three-dimensional coordinate
Whether be determining stroke effective three-dimensional coordinate, if so, second three-dimensional coordinate is three-dimensional as the end of current stroke
Coordinate determines that current stroke writing finishes.
25. device as claimed in claim 24, which is characterized in that the text determining module is also used to by described first
After three-dimensional coordinate is as the effective three-dimensional coordinate for determining stroke, judge whether second three-dimensional coordinate is having for determining stroke
Imitate three-dimensional coordinate;
If it is not, then using first three-dimensional coordinate as the beginning three-dimensional coordinate of current stroke.
26. device as claimed in claim 23, which is characterized in that the text determining module is also used to determining to defeated
Out after the corresponding every unicursal of text, according to the determining corresponding every unicursal of text to be output, text to be output is determined
Before corresponding first text, for every unicursal, determine whether the used time of the stroke is less than preset first duration threshold value;Such as
Fruit is to filter out the stroke.
27. device as claimed in claim 16, which is characterized in that described device further includes:
Removing module, for after the completion of the text determining module executes, deleting the corresponding each coordinate letter of text to be output
Breath.
28. device as claimed in claim 20, which is characterized in that the text determining module, if be also used to for current
The previous frame image of frame image does not preserve the second three-dimensional coordinate of the target gesture, not to the overlapping region pre-saved
There are the frame numbers of the image of the target gesture to be updated.
29. device as claimed in claim 28, which is characterized in that described device further includes:
Voice broadcast module, for when identification meets text and writes termination condition, at least one text of writing to be carried out language
Sound casting.
30. device as claimed in claim 29, which is characterized in that the voice broadcast module is received specifically for identification
Text writes END instruction;Or it recognizes the overlapping region that pre-saves there is no the frame numbers of the image of the target gesture to be greater than
Preset quantity.
31. a kind of electronic equipment, which is characterized in that including:Processor, communication interface, memory and communication bus, wherein place
Device, communication interface are managed, memory completes mutual communication by communication bus;
It is stored with computer program in the memory, when described program is executed by the processor, so that the processor
Perform claim requires the step of any one of 1-15 the method.
32. a kind of computer readable storage medium, which is characterized in that it is stored with the computer journey that can be executed by electronic equipment
Sequence, when described program is run on the electronic equipment, so that the electronic equipment perform claim requires any one of 1-15 institute
The step of stating method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810563940.XA CN108846339B (en) | 2018-06-04 | 2018-06-04 | Character recognition method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810563940.XA CN108846339B (en) | 2018-06-04 | 2018-06-04 | Character recognition method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108846339A true CN108846339A (en) | 2018-11-20 |
CN108846339B CN108846339B (en) | 2020-11-27 |
Family
ID=64210683
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810563940.XA Active CN108846339B (en) | 2018-06-04 | 2018-06-04 | Character recognition method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108846339B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110059678A (en) * | 2019-04-17 | 2019-07-26 | 上海肇观电子科技有限公司 | A kind of detection method, device and computer readable storage medium |
CN111031232A (en) * | 2019-04-24 | 2020-04-17 | 广东小天才科技有限公司 | Dictation real-time detection method and electronic equipment |
CN111081103A (en) * | 2019-05-17 | 2020-04-28 | 广东小天才科技有限公司 | Dictation answer obtaining method, family education equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060018546A1 (en) * | 2004-07-21 | 2006-01-26 | Hewlett-Packard Development Company, L.P. | Gesture recognition |
CN104793724A (en) * | 2014-01-16 | 2015-07-22 | 北京三星通信技术研究有限公司 | Sky-writing processing method and device |
CN106484108A (en) * | 2016-09-30 | 2017-03-08 | 天津大学 | Chinese characters recognition method based on double vision point gesture identification |
CN107728916A (en) * | 2017-09-20 | 2018-02-23 | 科大讯飞股份有限公司 | Every the display methods and device of empty handwriting tracks |
-
2018
- 2018-06-04 CN CN201810563940.XA patent/CN108846339B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20060018546A1 (en) * | 2004-07-21 | 2006-01-26 | Hewlett-Packard Development Company, L.P. | Gesture recognition |
CN104793724A (en) * | 2014-01-16 | 2015-07-22 | 北京三星通信技术研究有限公司 | Sky-writing processing method and device |
CN106484108A (en) * | 2016-09-30 | 2017-03-08 | 天津大学 | Chinese characters recognition method based on double vision point gesture identification |
CN107728916A (en) * | 2017-09-20 | 2018-02-23 | 科大讯飞股份有限公司 | Every the display methods and device of empty handwriting tracks |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110059678A (en) * | 2019-04-17 | 2019-07-26 | 上海肇观电子科技有限公司 | A kind of detection method, device and computer readable storage medium |
CN111031232A (en) * | 2019-04-24 | 2020-04-17 | 广东小天才科技有限公司 | Dictation real-time detection method and electronic equipment |
CN111081103A (en) * | 2019-05-17 | 2020-04-28 | 广东小天才科技有限公司 | Dictation answer obtaining method, family education equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108846339B (en) | 2020-11-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110738101B (en) | Behavior recognition method, behavior recognition device and computer-readable storage medium | |
CN110532984B (en) | Key point detection method, gesture recognition method, device and system | |
CN106934376B (en) | A kind of image-recognizing method, device and mobile terminal | |
CN107609489B (en) | Handwriting writing path evaluation device and method and electronic equipment | |
CN109635772A (en) | Dictation content correcting method and electronic equipment | |
CN108280843A (en) | A kind of video object detecting and tracking method and apparatus | |
CN103065134A (en) | Fingerprint identification device and method with prompt information | |
CN108846339A (en) | Character recognition method and device, electronic equipment and storage medium | |
CN106326853A (en) | Human face tracking method and device | |
CN111160288A (en) | Gesture key point detection method and device, computer equipment and storage medium | |
CN110287862B (en) | Anti-candid detection method based on deep learning | |
CN110969045B (en) | Behavior detection method and device, electronic equipment and storage medium | |
CN104268864B (en) | Card edge extracting method and device | |
CN113743177A (en) | Key point detection method, system, intelligent terminal and storage medium | |
CN110363790A (en) | Target tracking method, device and computer readable storage medium | |
CN107622497A (en) | Image cropping method, apparatus, computer-readable recording medium and computer equipment | |
CN104933145A (en) | Photograph processing method and device and mobile terminal | |
CN109284700A (en) | The method of multiple Face datections, storage medium, equipment and system in image | |
KR102440198B1 (en) | VIDEO SEARCH METHOD AND APPARATUS, COMPUTER DEVICE, AND STORAGE MEDIUM | |
CN109961103B (en) | Training method of feature extraction model, and image feature extraction method and device | |
CN113256683B (en) | Target tracking method and related equipment | |
CN106682669A (en) | Image processing method and mobile terminal | |
CN110222704A (en) | A kind of Weakly supervised object detection method and device | |
CN110765926B (en) | Picture book identification method, device, electronic equipment and storage medium | |
CN105551027B (en) | A kind of frontier tracing method and 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 |