CN109409357A - A kind of method and device of numeric area in intelligent recognition video - Google Patents
A kind of method and device of numeric area in intelligent recognition video Download PDFInfo
- Publication number
- CN109409357A CN109409357A CN201810997149.XA CN201810997149A CN109409357A CN 109409357 A CN109409357 A CN 109409357A CN 201810997149 A CN201810997149 A CN 201810997149A CN 109409357 A CN109409357 A CN 109409357A
- Authority
- CN
- China
- Prior art keywords
- video
- rectangle
- horizontal line
- area
- intelligent recognition
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000004458 analytical method Methods 0.000 claims abstract description 16
- 238000001914 filtration Methods 0.000 claims description 10
- 239000000284 extract Substances 0.000 claims description 7
- 238000007781 pre-processing Methods 0.000 claims description 7
- 238000012790 confirmation Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 description 12
- 238000004590 computer program Methods 0.000 description 7
- 230000000875 corresponding effect Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 241001258866 Nardia Species 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/20—Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
- H04N21/23—Processing of content or additional data; Elementary server operations; Server middleware
- H04N21/234—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs
- H04N21/23418—Processing of video elementary streams, e.g. splicing of video streams or manipulating encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Character Input (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of method and devices of the numeric area in intelligent recognition video.The method of numeric area in the intelligent recognition video, comprising: video to be identified is pre-processed；It analyzes in the video frame of the video with the presence or absence of horizontal line；The horizontal line if it exists generates rectangle determinating area according to the horizontal line；The text information of the number of image in rectangle determinating area described in intelligent recognition；The text information of the number if it exists then confirms that the rectangle determinating area is numeric area.The present invention can identify the numeric area in video by the analysis mode of convenient and efficient in a relatively short period of time, provide reliable guarantee for subsequent intelligent information service.
Description
Technical field
The present invention relates to technical field of video processing, in particular to the method for the numeric area in a kind of intelligent recognition video
And device.
Background technique
Video, especially net cast are widely taken root in the daily life of cybertimes.Certain views
The numeric area of frequency live streaming, such as the score board of section of football match video live streaming, then the timing board of such as dash match net cast,
The secondary attack that kills of such as knife tower match net cast injures statistical regions again, these above-mentioned numeric areas are that have huge reality
Meaning, if it will be for subsequent extraction numeric area that computer can accurately get the numeric area in net cast in real time
Image in digital information provide reliable guarantee.But due to the characteristic of net cast, use existing characteristics of image
The technology of analysis identifies that operand required for numeric area is very huge in net cast, it is difficult to meet compared with
Reach the demand of higher recognition success rate in short recognition time.How the appropriate processing above problem, just become industry urgently
The project of solution.
Summary of the invention
The present invention provides a kind of method and device of the numeric area in intelligent recognition video, simultaneously to lower operand
The numeric area in video is identified in a relatively short period of time.
According to a first aspect of the embodiments of the present invention, a kind of method of the numeric area in intelligent recognition video, packet are provided
It includes:
Video to be identified is pre-processed；
It analyzes in the video frame of the video with the presence or absence of horizontal line；
The horizontal line if it exists generates rectangle determinating area according to the horizontal line；
The text information of the number of image in rectangle determinating area described in intelligent recognition；
The text information of the number if it exists then confirms that the rectangle determinating area is numeric area.
It is in one embodiment, described that video to be identified is pre-processed, comprising:
1 frame video frame of the video is obtained at interval of x frame, wherein the quantity of the video frame of acquisition is n；
Gray proces are carried out to the video frame；
High-pass filtering processing is carried out to the video after gray proces.
In one embodiment, it whether there is horizontal line in the video frame of the analysis video, comprising:
Analyze the gradient information of the y direction in the video frame of the video；
The gradient value for judging whether there is the corresponding y direction of continuous pixel mark of same ordinate of orthogonal axes is all larger than
Preset Grads threshold；
If above-mentioned be judged as true, it is determined that whether there is horizontal line in the video frame of the video.
In one embodiment, the horizontal line described if it exists generates rectangle determinating area, packet according to the horizontal line
It includes:
The horizontal line if it exists obtains the horizontal length and position；
It is that a long side generates rectangle with the horizontal line according to preset length-width ratio；
Confirm that the rectangle is rectangle determinating area.
In one embodiment, the text information of the number of the image in rectangle determinating area described in the intelligent recognition,
Include:
The rectangle determinating area is input in preset intelligent recognition model；
The preset intelligent recognition model extracts the text information of the image in the rectangle determinating area.
According to a second aspect of the embodiments of the present invention, a kind of device of the numeric area in intelligent recognition video, packet are provided
It includes:
Preprocessing module, for being pre-processed to video to be identified；
Analysis module whether there is horizontal line in the video frame for analyzing the video；
Generation module generates rectangle determinating area according to the horizontal line for the horizontal line if it exists；
Identification module, the text information of the number for the image in rectangle determinating area described in intelligent recognition
Confirmation module then confirms that the rectangle determinating area is digital block for the text information of the number if it exists
Domain.
In one embodiment, the preprocessing module, comprising:
First acquisition submodule, for obtaining 1 frame video frame of the video at interval of x frame, wherein the video frame of acquisition
Quantity be n；
Gray scale submodule, for carrying out gray proces to the video frame；
Submodule is filtered, for carrying out high-pass filtering processing to the video after gray proces.
In one embodiment, the analysis module, comprising:
Analyze submodule, the gradient information of the y direction in video frame for analyzing the video；
Judging submodule, for judging whether there is the corresponding y direction of continuous pixel mark of same ordinate of orthogonal axes
Gradient value be all larger than preset Grads threshold；
Submodule is determined, if being judged as true for above-mentioned, it is determined that whether there is horizontal line in the video frame of the video.
In one embodiment, the generation module, comprising:
Second acquisition submodule obtains the horizontal length and position for the horizontal line if it exists；
Submodule is generated, for being that a long side generates rectangle with the horizontal line according to preset length-width ratio；
Submodule is confirmed, for confirming that the rectangle is rectangle determinating area.
In one embodiment, the identification module, comprising:
Input submodule, for the rectangle determinating area to be input in preset intelligent recognition model；
Extracting sub-module extracts the image in the rectangle determinating area for the preset intelligent recognition model
Text information.
Other features and advantages of the present invention will be illustrated in the following description, also, partly becomes from specification
It obtains it is clear that understand through the implementation of the invention.The objectives and other advantages of the invention can be by written explanation
Specifically noted structure is achieved and obtained in book, claims and attached drawing.
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
Detailed description of the invention
Attached drawing is used to provide further understanding of the present invention, and constitutes part of specification, with reality of the invention
It applies example to be used to explain the present invention together, not be construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the stream of the method for the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
Cheng Tu；
Fig. 2 is the step of the method for the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The flow chart of rapid S11；
Fig. 3 is the step of the method for the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The flow chart of rapid S12；
Fig. 4 is the step of the method for the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The flow chart of rapid S13；
Fig. 5 is the step of the method for the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The flow chart of rapid S14；
Fig. 6 is the frame of the device of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
Figure；
Fig. 7 is the pre- of the device of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The block diagram of processing module 61；
Fig. 8 is point of the device of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
Analyse the block diagram of module 62；
Fig. 9 is the life of the device of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
At the block diagram of module 63；
Figure 10 is the device of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention
The block diagram of identification module 64；
Figure 11 is the schematic diagram of the numeric area in a kind of intelligent recognition video shown in an exemplary embodiment of the invention.
Specific embodiment
Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings, it should be understood that preferred reality described herein
Apply example only for the purpose of illustrating and explaining the present invention and is not intended to limit the present invention.
Fig. 1 is the method flow of the numeric area in a kind of intelligent recognition video shown according to an exemplary embodiment
Figure, as shown in Figure 1, the method for the numeric area in the intelligent recognition video, includes the following steps S11-S15:
In step s 11, video to be identified is pre-processed；
In step s 12, it analyzes in the video frame of the video with the presence or absence of horizontal line；
In step s 13, the horizontal line if it exists generates rectangle determinating area according to the horizontal line；
In step S14, the text information of the number of the image in rectangle determinating area described in intelligent recognition；
In step S15, the text information of the number, then confirm that the rectangle determinating area is digital block if it exists
Domain.
In one embodiment, video, especially net cast were widely taken root in the day of the people of cybertimes
Often in life.The numeric area of certain net casts, such as the score board of section of football match video live streaming, then such as dash match view
The timing board of frequency live streaming, then the secondary attack that kills of such as knife tower match net cast injure statistical regions, these above-mentioned numeric areas
There is huge practical significance, if computer can be accurately got in real time after the numeric area in net cast will be
Digital information in the image of continuous extraction numeric area provides reliable guarantee.But due to the characteristic of net cast, make
With the technology of existing image characteristic analysis, identify that operand required for numeric area is very huge in net cast
, it is difficult to meet the needs of reaching higher recognition success rate in shorter recognition time.Technical solution in the present embodiment can be appropriate
The kind processing above problem.
Video to be identified is pre-processed.Wherein, 1 frame video frame of the video is obtained at interval of x frame, wherein obtain
The quantity of the video frame taken is n.Gray proces are carried out to the video frame.Video after gray proces is carried out at high-pass filtering
Reason.
It analyzes in the video frame of the video with the presence or absence of horizontal line.Wherein, the longitudinal axis side in the video frame of the video is analyzed
To gradient information.The gradient value for judging whether there is the corresponding y direction of continuous pixel mark of same ordinate of orthogonal axes is equal
Greater than preset Grads threshold.If above-mentioned be judged as true, it is determined that whether there is horizontal line in the video frame of the video.
The horizontal line if it exists generates rectangle determinating area according to the horizontal line.Wherein, the horizontal line if it exists obtains
The horizontal length and position.It is that a long side generates rectangle with the horizontal line according to preset length-width ratio.Confirm that the rectangle is
Rectangle determinating area.
The text information of the number of image in the intelligent recognition rectangle determinating area.Wherein, by the rectangle determinating area
It is input in preset intelligent recognition model.The preset intelligent recognition model extracts the image in the rectangle determinating area
Text information.
The text information of the number if it exists then confirms that the rectangle determinating area is numeric area.
It can be identified in a relatively short period of time by the analysis mode of convenient and efficient using the technical solution in the present embodiment
Numeric area in video provides reliable guarantee for subsequent intelligent information service.
In one embodiment, as shown in Fig. 2, step S11 includes the following steps S21-S23:
In the step s 21,1 frame video frame of the video is obtained at interval of x frame, wherein the quantity of the video frame of acquisition
For n；
In step S22, gray proces are carried out to the video frame；
In step S23, high-pass filtering processing is carried out to the video after gray proces.
In one embodiment, when being sampled to video, without extracting whole video frames in some period,
Because the redundancy of consecutive frame is very high.It is to be processed to reduce by the way of extracting 1 frame video frame at interval of x frame in the present embodiment
Data volume.Wherein, the more excellent value interval of x is (1,11).
There are three the Color Channels of each of video frame pixel, is red, green, blue respectively.Each face
The value of chrominance channel is (0,255).Image is carried out gray proces by the scheme in the present embodiment, can be to avoid three colors of calculating
The operand in channel, it is only necessary to calculate single pass operand.The quantity for obtaining video frame is n, specific value and the interception of n
The duration of video is positively correlated.To which calculation amount can be effectively reduced using the technical solution in the present embodiment, thus significantly
Improve the arithmetic speed of the numeric area in system identification video.The video after gray proces is carried out at high-pass filtering later
Reason, to reach the technical purpose of the interference noise in removal image.
In one embodiment, as shown in figure 3, step S12 includes the following steps S31-S33:
In step S31, the gradient information of the y direction in the video frame of the video is analyzed；
In step s 32, the corresponding y direction of continuous pixel mark of same ordinate of orthogonal axes is judged whether there is
Gradient value is all larger than preset Grads threshold；
In step S33, if above-mentioned be judged as true, it is determined that whether there is horizontal line in the video frame of the video.
In one embodiment, the gradient information of the y direction in video frame is analyzed using Sobel edge detection algorithm,
When the gradient value that any pixel point corresponds to ordinate is greater than preset Grads threshold, confirm that the pixel is in horizontal edge
One point, constantly repeats above-mentioned process, to reach the technical purpose for detecting the horizontal edge of entire video frame.Because
Numeric area is covered in video frame on original picture, so the horizontal edge of the numeric area must be in same vertical
On axial coordinate, and the horizontal edge of the numeric area is at least one.It can be removed other horizontal sides in video frame in this way
The interference information of edge, to reduce the operand of the horizontal edge of detection numeric area.
In one embodiment, as shown in figure 4, step S13 includes the following steps S41-S43:
In step S41, the horizontal line, obtains the horizontal length and position if it exists；
It is that a long side generates rectangle with the horizontal line according to preset length-width ratio in step S42；
In step S43, confirm that the rectangle is rectangle determinating area.
In one embodiment, there are the horizontal lines in upper one embodiment out for system detection, by obtaining the horizontal line
Length and position, so that it may determine a horizontal sides in the numeric area, the horizontal sides, that is, long side.According to preset length
The length-width ratio on side and broadside, so that it may obtain the length of the broadside of the numeric area.In the coordinate information according to the both ends of the long side,
It is assured that out the location information at the both ends of broadside.Because the long side may be the long side of the top of the numeric area, can also
It can be the long side of the lower section of the numeric area, it is possible to construct two rectangles, which is zhou duicheng tuxing, symmetrically
Axis is the long side that system detection goes out.In order to express easily, it might as well assume the long side L detected1Two endpoints coordinate be (x1,
y1) and (x2, y1), the length-width ratio k of preset long side and broadside, the length of long side is (x1-x2) absolute value m.If long side L1 exists
The lower section of numeric area, the then coordinate on four vertex of the rectangular area A generated are (x1, y1)、(x2, y1)、(x2, y2) and (x1,
y2)；If long side L1In the top of numeric area, then the coordinate on four vertex of the rectangular area B generated is (x1, y3)、(x2,
y3)、(x2, y1) and (x1, y1).Wherein, the length of broadside is (y1-y2) the ratio of absolute value n, absolute value m and absolute value n be
Length-width ratio k.Rectangular area A and rectangular area B is zhou duicheng tuxing, and symmetry axis is long side L1.Rectangle determinating area D is rectangle region
The sum of the region of domain A and B.Specific figure is detailed in attached drawing 11.So as to determine rectangle determinating area in the video frame, the square
Shape determinating area includes two axisymmetric rectangles.The institute in the video frame other than the rectangle determinating area can be ignored in system
There is region, so that operand greatly reduce.
In one embodiment, as shown in figure 5, step S14 includes the following steps S51-S52:
In step s 51, the rectangle determinating area is input in preset intelligent recognition model；
In step S52, the preset intelligent recognition model extracts the text of the image in the rectangle determinating area
This information.
In one embodiment, in order to express easily, rectangle determinating area might as well be named as D, rectangle determinating area D
Include rectangular area A and rectangular area B.A square in rectangle determinating area D is analyzed in preset intelligent recognition model
It whether include digital information in the A of shape region.The preset intelligent recognition model is attempted to extract the text of the image in the A of rectangular area
Information, text information include digital information.If including digital information in the A of rectangular area, rectangular area A is exactly digital block
Domain；If not including digital information in the A of rectangular area, rectangular area B is exactly numeric area.
In one embodiment, Fig. 6 is the number in a kind of intelligent recognition video shown according to an exemplary embodiment
The device block diagram in region.If Fig. 6 shows, which includes preprocessing module 61, analysis module 62, generation module 63, identification module
64 and confirmation module 65.
The preprocessing module 61, for being pre-processed to video to be identified；
The analysis module 62 whether there is horizontal line in the video frame for analyzing the video；
The generation module 63 generates rectangle determinating area according to the horizontal line for the horizontal line if it exists；
The identification module 64, the text information of the number for the image in rectangle determinating area described in intelligent recognition；
The confirmation module 65 then confirms the rectangle determinating area for number for the text information of the number if it exists
Block domain.
As shown in fig. 7, the preprocessing module 61 includes the first acquisition submodule 71, gray scale submodule 72 and filtering submodule
73。
First acquisition submodule 71, for obtaining 1 frame video frame of the video at interval of x frame, wherein the view of acquisition
The quantity of frequency frame is n；
The gray scale submodule 72, for carrying out gray proces to the video frame；
The filtering submodule 73, for carrying out high-pass filtering processing to the video after gray proces.
As shown in figure 8, the analysis module 62 includes analysis submodule 81, judging submodule 82 and determines submodule 83.
The analysis submodule 81, the gradient information of the y direction in video frame for analyzing the video；
The judging submodule 82, for judging whether there is the corresponding longitudinal axis of continuous pixel mark of same ordinate of orthogonal axes
The gradient value in direction is all larger than preset Grads threshold；
Determination submodule 83, if being judged as true for above-mentioned, it is determined that whether there is water in the video frame of the video
Horizontal line.
As shown in figure 9, the generation module 63 includes the second acquisition submodule 91, generates submodule 92 and confirmation submodule
93。
Second acquisition submodule 91 obtains the horizontal length and position for the horizontal line if it exists；
The generation submodule 92, for being that a long side generates rectangle with the horizontal line according to preset length-width ratio；
The confirmation submodule 93, for confirming that the rectangle is rectangle determinating area.
As shown in Figure 10, which includes input submodule 101 and extracting sub-module 102.
The input submodule 101, for the rectangle determinating area to be input in preset intelligent recognition model；
The extracting sub-module 102 extracts in the rectangle determinating area for the preset intelligent recognition model
The text information of image.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The shape for the computer program product implemented in usable storage medium (including but not limited to magnetic disk storage and optical memory etc.)
Formula.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
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.
Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art
Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies
Within, then the present invention is also intended to include these modifications and variations.
Claims (10)
1. a kind of method of the numeric area in intelligent recognition video characterized by comprising
Video to be identified is pre-processed；
It analyzes in the video frame of the video with the presence or absence of horizontal line；
The horizontal line if it exists generates rectangle determinating area according to the horizontal line；
The text information of the number of image in rectangle determinating area described in intelligent recognition；
The text information of the number if it exists then confirms that the rectangle determinating area is numeric area.
2. the method as described in claim 1, which is characterized in that described to be pre-processed to video to be identified, comprising:
1 frame video frame of the video is obtained at interval of x frame, wherein the quantity of the video frame of acquisition is n；
Gray proces are carried out to the video frame；
High-pass filtering processing is carried out to the video after gray proces.
3. the method as described in claim 1, which is characterized in that with the presence or absence of level in the video frame of the analysis video
Line, comprising:
Analyze the gradient information of the y direction in the video frame of the video；
Judge whether there is the corresponding y direction of continuous pixel mark of same ordinate of orthogonal axes gradient value be all larger than it is default
Grads threshold；
If above-mentioned be judged as true, it is determined that whether there is horizontal line in the video frame of the video.
4. the method as described in claim 1, which is characterized in that the horizontal line described if it exists, it is raw according to the horizontal line
Rectangular determinating area, comprising:
The horizontal line if it exists obtains the horizontal length and position；
It is that a long side generates rectangle with the horizontal line according to preset length-width ratio；
Confirm that the rectangle is rectangle determinating area.
5. the method as described in claim 1, which is characterized in that image in rectangle determinating area described in the intelligent recognition
The text information of number, comprising:
The rectangle determinating area is input in preset intelligent recognition model；
The preset intelligent recognition model extracts the text information of the image in the rectangle determinating area.
6. a kind of device of the numeric area in intelligent recognition video characterized by comprising
Preprocessing module, for being pre-processed to video to be identified；
Analysis module whether there is horizontal line in the video frame for analyzing the video；
Generation module generates rectangle determinating area according to the horizontal line for the horizontal line if it exists；
Identification module, the text information of the number for the image in rectangle determinating area described in intelligent recognition；
Confirmation module then confirms that the rectangle determinating area is numeric area for the text information of the number if it exists.
7. device according to claim 6, which is characterized in that the preprocessing module, comprising:
First acquisition submodule, for obtaining 1 frame video frame of the video at interval of x frame, wherein the number of the video frame of acquisition
Amount is n；
Gray scale submodule, for carrying out gray proces to the video frame；
Submodule is filtered, for carrying out high-pass filtering processing to the video after gray proces.
8. device according to claim 6, which is characterized in that the analysis module, comprising:
Analyze submodule, the gradient information of the y direction in video frame for analyzing the video；
Judging submodule, the ladder of the corresponding y direction of continuous pixel mark for judging whether there is same ordinate of orthogonal axes
Angle value is all larger than preset Grads threshold；
Submodule is determined, if being judged as true for above-mentioned, it is determined that whether there is horizontal line in the video frame of the video.
9. device according to claim 6, which is characterized in that the generation module, comprising:
Second acquisition submodule obtains the horizontal length and position for the horizontal line if it exists；
Submodule is generated, for being that a long side generates rectangle with the horizontal line according to preset length-width ratio；
Submodule is confirmed, for confirming that the rectangle is rectangle determinating area.
10. device according to claim 6, which is characterized in that the identification module, comprising:
Input submodule, for the rectangle determinating area to be input in preset intelligent recognition model；
Extracting sub-module extracts the text of the image in the rectangle determinating area for the preset intelligent recognition model
Information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810997149.XA CN109409357A (en) | 2018-08-29 | 2018-08-29 | A kind of method and device of numeric area in intelligent recognition video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810997149.XA CN109409357A (en) | 2018-08-29 | 2018-08-29 | A kind of method and device of numeric area in intelligent recognition video |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109409357A true CN109409357A (en) | 2019-03-01 |
Family
ID=65463788
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810997149.XA Pending CN109409357A (en) | 2018-08-29 | 2018-08-29 | A kind of method and device of numeric area in intelligent recognition video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409357A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111405360A (en) * | 2020-03-25 | 2020-07-10 | 腾讯科技（深圳）有限公司 | Video processing method and device, electronic equipment and storage medium |
-
2018
- 2018-08-29 CN CN201810997149.XA patent/CN109409357A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111405360A (en) * | 2020-03-25 | 2020-07-10 | 腾讯科技（深圳）有限公司 | Video processing method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10896349B2 (en) | Text detection method and apparatus, and storage medium | |
| US20140253429A1 (en) | Visual language for human computer interfaces | |
| CN104182750A (en) | Extremum connected domain based Chinese character detection method in natural scene image | |
| JP7096904B2 (en) | Motion locus features Processing methods, devices, and computer storage media | |
| CN110348439A (en) | A kind of method, computer-readable medium and the system of automatic identification price tag | |
| CN110569774B (en) | Automatic line graph image digitalization method based on image processing and pattern recognition | |
| CN106023261B (en) | A kind of method and device of television video target following | |
| CN112001429A (en) | Depth forgery video detection method based on texture features | |
| CN103685866B (en) | video image stabilization method and device thereof | |
| CN109598271A (en) | A kind of character segmentation method and device | |
| CN111914818B (en) | Method for detecting forest fire smoke root nodes based on multi-frame discrete confidence | |
| JP2018018173A (en) | Image processing device, image processing method, and computer program | |
| CN103793923A (en) | Method and device for acquiring moving object in image | |
| Lin et al. | License plate recognition based on mathematical morphology and template matching | |
| JP2008210387A (en) | Noise elimination device and noise elimination program for improving binarization performance of document image | |
| CN104318240A (en) | Flower bud discriminating method based on computer vision | |
| CN103699876A (en) | Method and device for identifying vehicle number based on linear array CCD (Charge Coupled Device) images | |
| CN113486881B (en) | Text recognition method, device, equipment and medium | |
| CN109409357A (en) | A kind of method and device of numeric area in intelligent recognition video | |
| CN109241885A (en) | A kind of method and device of special area in intelligent recognition video | |
| EP3241153B1 (en) | Detection of solid color frames for determining transitions in video content | |
| CN104850819B (en) | Information processing method and electronic equipment | |
| CN103971376A (en) | Application program execution method and device | |
| JP7111599B2 (en) | Drawing recognition device, drawing recognition method and program | |
| US20170147895A1 (en) | Method and device for digit separation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190301 |