CN106096659A - Image matching method and device - Google Patents
Image matching method and device Download PDFInfo
- Publication number
- CN106096659A CN106096659A CN201610436897.1A CN201610436897A CN106096659A CN 106096659 A CN106096659 A CN 106096659A CN 201610436897 A CN201610436897 A CN 201610436897A CN 106096659 A CN106096659 A CN 106096659A
- Authority
- CN
- China
- Prior art keywords
- resolution
- image
- template image
- identified
- template
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/24—Character recognition characterised by the processing or recognition method
- G06V30/248—Character recognition characterised by the processing or recognition method involving plural approaches, e.g. verification by template match; Resolving confusion among similar patterns, e.g. "O" versus "Q"
- G06V30/2504—Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
Abstract
The present invention provides a kind of image matching method and device, determined by the second resolution of the first resolution according to image to be identified and the first template image and the first template image is zoomed in and out the scale coefficient that process is used, only the first template image need to be carried out a scaling, i.e. can get the second template image that resolution is identical with the first resolution of image to be identified, avoid using the multiple scale coefficients in yardstick list that the first template image carries out repeatedly scaling to process, thus, improve the efficiency of images match.Further, determine that according to the first resolution of image to be identified and the second resolution of the first template image that the first template image zooms in and out the scale coefficient that process used is more accurate, it is thus possible to improve the accuracy of images match.
Description
Technical field
The present invention relates to computer technology, particularly relate to a kind of image matching method and device.
Background technology
Can identify, from an image to be identified, the target figure matched with template image by image matching technology
Picture, it is thus possible to process accordingly target image.
Image matching technology can be applicable to several scenes, as a example by mobile phone test scene based on images match, at this
Jing Zhong, sets up the corresponding relation of template image and operation, passes through from image to be identified (such as truncated picture mobile phone games)
Template matching identifies the target image matched with template image, and target image is carried out the operation that template image is corresponding, from
And, it is achieved automatization's mobile phone test.
In prior art, usually there will be the situation that the resolution of template image is different from the resolution of image to be identified,
Then according to multiple scale coefficients fixing in yardstick list, each scale coefficient is used template image to be zoomed in and out, according to chi
Template image after degree scaling carries out images match, obtains the matching result that each scale coefficient is corresponding, from multiple scale coefficients
Corresponding matching result obtains the matching result of optimum as target image.But, use prior art method, such as when
When there is multiple target area, it is impossible to determine that providing matching result is correct that, causes the inefficient of images match.
Summary of the invention
The present invention provides a kind of image matching method and device, to improve the efficiency of images match.
First aspect, the present invention provides a kind of image matching method, including:
First resolution according to image to be identified and the second resolution of the first template image, determine scale coefficient;
Use scale coefficient that the first template image is zoomed in and out process, obtain the second template image, the second template image
Resolution identical with first resolution;
Use the second template image to mate with image to be identified, obtain and the second template image from image to be identified
The target image matched.
Alternatively, according to first resolution and the second resolution of the first template image of image to be identified, yardstick is determined
Coefficient, including:
Obtain scaling pattern;
In conjunction with scaling pattern, according to first resolution and second resolution of the first template image of image to be identified
Rate, determines scale coefficient.
Alternatively, described scaling pattern is first mode;
Scaling pattern described in described combination, according to the first resolution and the of the first template image of image to be identified
Two resolution, determine scale coefficient, including:
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution
Rate, [RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described
The width of two resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first point
The height of resolution, DW is the width of described design resolution, and DH is the height of described design resolution;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image,
Including:
According to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] described first template image is zoomed in and out process,
[RWP2',RHP2'] it is the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'
Height for the resolution of described second template image.
Alternatively, described scaling pattern is the second pattern;
Scaling pattern described in described combination, according to the first resolution and the of the first template image of image to be identified
Two resolution, determine scale coefficient, including:
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution
Rate, [RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described
The width of two resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first point
The height of resolution, DW is the width of described design resolution, and DH is the height of described design resolution;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image,
Including:
According to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] described first template image is zoomed in and out process,
[RWP2',RHP2'] it is the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'
Height for the resolution of described second template image.
Alternatively, described scaling pattern is the 3rd pattern;
Scaling pattern described in described combination, according to the first resolution and the of the first template image of image to be identified
Two resolution, determine scale coefficient, including:
According toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, RWP1For described
The width of second resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first
The height of resolution;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image,
Including:
According to [RWP2',RHP2']=[αWRWP2,αHRHP2] described first template image is zoomed in and out process, [RWP2',
RHP2'] it is the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'For described
The height of the resolution of the second template image.
Alternatively, use scale coefficient that the first template image is zoomed in and out process, obtain the second template image, including:
The scale coefficient using at least two scaling pattern corresponding zooms in and out process to the first template image respectively,
Obtain at least two the second template image;
Use the second template image to mate with image to be identified, obtain and the second template image from image to be identified
The target image matched, including:
Use at least two the second template image mate with image to be identified respectively, obtain from image to be identified to
The first candidate image that few two the second template images match;
Candidate's figure that matching result is the highest is determined from the first candidate image that at least two the second template image matches
Picture is target image.
Alternatively, use the second template image to mate with image to be identified, obtain and second from image to be identified
The target image that template image matches, including:
Described second template image is used to mate in described image to be identified, according to best matching result with default
Error threshold, obtains N number of second candidate image, and described N is the integer more than or equal to 2;
In conjunction with coordinate and the coordinate of the second template image of N number of second candidate image, determine and the second template image phase
The target image joined.
Second aspect, the present invention provides a kind of image matching apparatus, including:
Processing module, for the first resolution according to image to be identified and the second resolution of the first template image, really
Dimensioning coefficient;
Scaling module, is used for using scale coefficient that the first template image is zoomed in and out process, obtains the second template
Image, the resolution of the second template image is identical with first resolution;
Matching module, for using the second template image to mate with image to be identified, obtains from image to be identified
The target image matched with the second template image.
Alternatively, processing module is specifically for obtaining scaling pattern;In conjunction with scaling pattern, according to figure to be identified
The first resolution of picture and the second resolution of the first template image, determine scale coefficient.
Alternatively, described scaling pattern is first mode;
Described processing module is specifically for basisDetermine scale coefficient, wherein, [RWP1,
RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, [DW, DH] is the design of described image to be identified
Resolution, RWP1For the width of described second resolution, RHP1For the height of described second resolution, RWP2For described first resolution
Width, RHP2For the height of described first resolution, DW is the width of described design resolution, and DH is the height of described design resolution;
Described scaling module is specifically for according to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] to described first mould
Plate image zooms in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'For described second Prototype drawing
The width of the resolution of picture, RHP2'Height for the resolution of described second template image.
Alternatively, described scaling pattern is the second pattern;
Described processing module specifically for
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution
Rate, [RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described
The width of two resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first point
The height of resolution, DW is the width of described design resolution, and DH is the height of described design resolution;
Described scaling module is specifically for according to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] to described first mould
Plate image zooms in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'For described second Prototype drawing
The width of the resolution of picture, RHP2'Height for the resolution of described second template image.
Alternatively, described scaling pattern is the 3rd pattern;
Described processing module specifically for
According toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, RWP1For described
The width of second resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first
The height of resolution;
Described scaling module is specifically for according to [RWP2',RHP2']=[αWRWP2,αHRHP2] to described first template
Image zooms in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'For described second template image
The width of resolution, RHP2'Height for the resolution of described second template image.
Alternatively, scaling module is distinguished specifically for the scale coefficient using at least two scaling pattern corresponding
First template image is zoomed in and out process, obtains at least two the second template image;
Matching module mates with image to be identified, from treating respectively specifically for using at least two the second template image
Identify and image obtains the first candidate image that at least two the second template image matches;From at least two the second template image
The candidate image matched determining, the first candidate image that matching result is the highest is target image.
Alternatively, matching module is carried out in described second template image and described image to be identified specifically for using
Joining, according to best matching result and default error threshold, obtain N number of second candidate image, described N is the integer more than or equal to 2;
In conjunction with coordinate and the coordinate of the second template image of N number of second candidate image, determine the target matched with the second template image
Image.
The image matching method of present invention offer and device, by the first resolution according to image to be identified and the first mould
The second resolution of plate image determines the first template image is zoomed in and out the scale coefficient that process is used, only need to be to the first mould
Plate image carries out a scaling, i.e. can get the second Prototype drawing that resolution is identical with the first resolution of image to be identified
Picture, it is to avoid use the multiple scale coefficients in yardstick list that the first template image carries out repeatedly scaling and process, thus, improve figure
Efficiency as coupling.Further, determine according to the first resolution of image to be identified and the second resolution of the first template image right
It is more accurate that first template image zooms in and out the scale coefficient that process used, it is thus possible to improve the accurate of images match
Degree.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, also may be used
To obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the schematic flow sheet of image matching method embodiment one of the present invention;
Fig. 2 is the schematic flow sheet of image matching method embodiment three of the present invention;
Fig. 3 is the schematic flow sheet of image matching method embodiment four of the present invention;
Fig. 4 is the schematic flow sheet of image matching method embodiment five of the present invention;
Fig. 5 is the schematic flow sheet of image matching method embodiment six of the present invention;
Fig. 6 is the schematic flow sheet of image matching method embodiment seven of the present invention;
Fig. 7 a is the second template image one schematic diagram;
Fig. 7 b is image one schematic diagram to be identified;
Fig. 8 a is the another kind of schematic diagram of the second template image;
Fig. 8 b is the another kind of schematic diagram of image to be identified;
Fig. 9 is the structural representation of image matching apparatus embodiment of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is carried out clear, complete
Describe, it is clear that described embodiment is only some embodiments of the present application rather than whole embodiments wholely.Based on
Embodiment in the application, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of the application protection.
Here will illustrate exemplary embodiment in detail, its example represents in the accompanying drawings.Explained below relates to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represents same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the present invention.On the contrary, they are only with the most appended
The example of the apparatus and method that some aspects that described in detail in claims, the present invention are consistent.
It will be appreciated that described embodiment be only the present invention a part of embodiment rather than whole embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art obtained under not making creative work premise all its
Its embodiment, broadly falls into the scope of protection of the invention.
The term used in embodiments of the present invention is only merely for describing the purpose of specific embodiment, and is not intended to be limiting
The present invention." a kind of ", " described " and " being somebody's turn to do " of singulative used in the embodiment of the present invention and appended claims
It is also intended to include most form, unless context clearly shows that other implications.
Should be appreciated that term "and/or" used herein is only a kind of incidence relation describing affiliated partner, represent
Three kinds of relations, such as, A and/or B can be there are, can represent: individualism A, there is A and B, individualism B these three simultaneously
Situation.It addition, character "/" herein, typically represent forward-backward correlation to as if the relation of a kind of "or".
Although should be appreciated that possible employing term first, second, third, etc. are to describe XXX in embodiments of the present invention, but
These XXX should not necessarily be limited by these terms.These terms are only used for XXX is distinguished from each other out.Such as, implementing without departing from the present invention
In the case of example scope, an XXX can also be referred to as the 2nd XXX, and similarly, the 2nd XXX can also be referred to as an XXX.
Depend on linguistic context, word as used in this " if ", " if " can be construed to " ... time " or
" when ... " or " in response to determining " or " in response to detection ".Similarly, depend on linguistic context, phrase " if it is determined that " or " as
Fruit detection (condition of statement or event) " " when determining " or " in response to determining " can be construed to or " when detection (statement
Condition or event) time " or " in response to detection (condition of statement or event) ".
Also, it should be noted term " includes ", " comprising " or its any other variant are intended to nonexcludability
Comprise, so that include that the commodity of a series of key element or system not only include those key elements, but also include the most clearly
Other key elements listed, or also include the key element intrinsic for this commodity or system.In the feelings not having more restriction
Under condition, statement " including ... " key element limited, it is not excluded that in the commodity including described key element or system also
There is other identical element.
The present invention is mainly by true according to the first resolution of image to be identified and the second resolution of the first template image
Determine the first template image is zoomed in and out the scale coefficient that process is used, only the first template image need to be carried out a subdimension contracting
Put, i.e. can get the second template image that resolution is identical with the second resolution of image to be identified, it is to avoid use yardstick list
In multiple scale coefficients the first template image is carried out repeatedly scaling process, thus, improve images match efficiency.Further,
First resolution according to image to be identified and the second resolution of the first template image determine and contract the first template image
Put the scale coefficient that process used more accurate, it is thus possible to improve the accuracy of images match.
The first resolution of image to be identified described in the invention refers to show the display device of described image to be identified
Screen resolution.
When the second resolution of the first template image described in the invention refers to obtain the first template image, show first
The screen resolution of the display device of template image.
Wherein, image to be identified and the first template image are typically derived from same application (such as: certain money is played)
Display image.
The design resolution of image to be identified described in the invention, refers to that art designing is in design a set of UI picture resource time institute
The resolution determined, generally, the design resolution of a application program is constant.
With specifically embodiment, technical scheme is described in detail below.These concrete enforcements below
Example can be combined with each other, and may repeat no more in some embodiment for same or analogous concept or process.
Embodiment one
Fig. 1 is the schematic flow sheet of image matching method embodiment one of the present invention, and the method for the present embodiment is as follows:
S101: according to first resolution and the second resolution of the first template image of image to be identified, determine yardstick system
Number.
Such as: image to be identified shows on electronic equipment 1, then first resolution is the resolution of electronic equipment 1;
First template image is to intercept from the image of electronic equipment 2 display, then second resolution is electronic equipment 2
Resolution;
When carrying out images match, the resolution of two images carrying out mating is identical, just can be mated knot accurately
Really, therefore, when first resolution is different with second resolution, need the first template image is zoomed in and out process, the present invention
First resolution according to image to be identified and the second resolution of the first template image, determine and contract the first template image
Put and process scale coefficient to be taken.
S102: use scale coefficient that the first template image is zoomed in and out process, obtain the second template image.
Wherein, the resolution of the second template image is identical with first resolution.
S103: use the second template image to mate with image to be identified, obtains and the second mould from image to be identified
The target image that plate image matches.
Specifically, it is by second template image that slides on image to be identified, to the second template image and figure to be identified
As mating.
Such as: the size of image to be identified is 100*100, the size of the second template image is 10*10, and matching process is:
(1) start from the upper left corner (0,0) of image to be identified, cut the intermediate images of a piece (0,0) to (10,10);
(2) mating with intermediate images and the second template image, matching result is designated as c;
(3) matching result c, is stored as the pixel value at result images (0,0) place;
(4) cut image to be identified from (0,1) to the intermediate images of (10,11), contrast, and be recorded as result images (0,
1) pixel value at place;
(5) (1)~(4) step is repeated until the intermediate images of cutting has been located in the lower right corner of image to be identified.
(6) according to the matching result of each intermediate images, obtain from image to be identified and match with the second template image
Target image.Wherein, target image is typically the region that the best intermediate images of matching result is corresponding.
The present embodiment, is determined by the second resolution of the first resolution according to image to be identified and the first template image
First template image is zoomed in and out the scale coefficient that process is used, only the first template image need to be carried out a subdimension contracting
Put, i.e. can get the second template image that resolution is identical with the first resolution of image to be identified, it is to avoid use yardstick list
In multiple scale coefficients the first template image is carried out repeatedly scaling process, thus, improve images match efficiency.Further,
First resolution according to image to be identified and the second resolution of the first template image determine and contract the first template image
Put the scale coefficient that process used more accurate, it is thus possible to improve the accuracy of images match.
Embodiment two
In the S101 of embodiment one, a kind of possible implementation is: obtain scaling pattern, in conjunction with described yardstick
Zoom mode, according to first resolution and the second resolution of the first template image of image to be identified, determines scale coefficient.When
During scaling pattern different, determining that the mode of scale coefficient is the most different, wherein, scaling pattern includes that equal proportion scales
Scaling with non-equal proportion, equal proportion scaling is that finger beam is identical with the scale coefficient that height zooms in and out, and non-equal proportion scaling is finger beam
The scale coefficient zoomed in and out with height is different, and equal proportion scaling is divided into again maximum scaling and minima to scale, and the present invention describes
First mode correspondence maximum zoom mode, the second pattern to minima zoom mode, the corresponding non-equal proportion contracting of the 3rd pattern
Mode playback.
Embodiment three
Fig. 2 is the schematic flow sheet of image matching method embodiment three of the present invention, and the present embodiment describes scaling pattern
For the technical scheme that minima zoom mode is corresponding, as shown in Figure 2:
S201: combine minima zoom mode, according to the first resolution and the of the first template image of image to be identified
Two resolution, determine scale coefficient.
Specifically, according toDetermine scale coefficient, wherein, [RWP1,RHP1] it is first point
Resolution, [RWP2,RHP2] it is second resolution, [DW, DH] is the design resolution of image to be identified, RWP1For second resolution
Width, RHP1For the height of second resolution, RWP2For the width of first resolution, RHP2For the height of first resolution, DW differentiates for design
The width of rate, DH is the height of design resolution;
S202: the scale coefficient using S201 to determine zooms in and out process to the first template image, obtains the second Prototype drawing
Picture.
Specifically, according to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] the first template image is zoomed in and out process,
[RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the width of the resolution of the second template image, RHP2'It it is the second mould
The height of the resolution of plate image.
S203: use the second template image to mate with image to be identified, obtains and the second mould from image to be identified
The target image that plate image matches.
This step sees the detailed description in S103, and here is omitted.
The present embodiment, by combining minima zoom mode, according to first resolution and first template of image to be identified
The second resolution of image, determines scale coefficient, uses the scale coefficient determined that the first template image is zoomed in and out process,
To the second template image, use the second template image to mate with image to be identified, obtain and second from image to be identified
The target image that template image matches, owing to the first template image is zoomed in and out the scale coefficient that process is used, is knot
Close minima zoom mode, directly determine according to the first resolution of image to be identified and the second resolution of the first template image
, only the first template image need to be carried out a scaling, i.e. can get the second resolution of resolution and image to be identified
The second identical template image, therefore, improves the efficiency of images match.Further, according to the first resolution of image to be identified and
The second resolution of the first template image determines that the first template image zooms in and out the scale coefficient that process used is more accurate
Really, it is thus possible to improve the accuracy of images match.
Embodiment four
Fig. 3 is the schematic flow sheet of image matching method embodiment four of the present invention, and the present embodiment describes scaling pattern
For the technical scheme that maximum zoom mode is corresponding, as shown in Figure 3:
S301: combine maximum zoom mode, according to the first resolution and the of the first template image of image to be identified
Two resolution, determine scale coefficient.
Specifically, according toDetermine scale coefficient, wherein, [RWP1,RHP1] it is first
Resolution, [RWP2,RHP2] it is second resolution, [DW, DH] is the design resolution of image to be identified, RWP1For second resolution
Width, RHP1For the height of second resolution, RWP2For the width of first resolution, RHP2For the height of first resolution, DW is design point
The width of resolution, DH is the height of design resolution.
S302: the scale coefficient using S301 to determine zooms in and out process to the first template image, obtains the second Prototype drawing
Picture.
Specifically, according to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] the first template image is zoomed in and out process,
[RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the width of the resolution of the second template image, RHP2'It it is the second mould
The height of the resolution of plate image.
S303: use the second template image to mate with image to be identified, obtains and the second mould from image to be identified
The target image that plate image matches.
This step sees the detailed description in S103, and here is omitted.
The present embodiment, by combining maximum zoom mode, according to first resolution and first template of image to be identified
The second resolution of image, determines scale coefficient, uses the scale coefficient determined that the first template image is zoomed in and out process,
To the second template image, use the second template image to mate with image to be identified, obtain and second from image to be identified
The target image that template image matches, owing to the first template image is zoomed in and out the scale coefficient that process is used, is knot
Close maximum zoom mode, directly determine according to the first resolution of image to be identified and the second resolution of the first template image
, only the first template image need to be carried out a scaling, i.e. can get the second resolution of resolution and image to be identified
The second identical template image, therefore, improves the efficiency of images match.Further, according to the first resolution of image to be identified and
The second resolution of the first template image determines that the first template image zooms in and out the scale coefficient that process used is more accurate
Really, it is thus possible to improve the accuracy of images match.
Embodiment five
Fig. 4 is the schematic flow sheet of image matching method embodiment five of the present invention, and the present embodiment describes scaling pattern
For the technical scheme that non-equal proportion zoom mode is corresponding, as shown in Figure 4:
S401: combine non-equal proportion zoom mode, according to the first resolution of image to be identified and the first template image
Second resolution, determines scale coefficient.
Specifically, according toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is first resolution, [RWP2,RHP2] it is second resolution, RWP1For second resolution
Width, RHP1For the height of second resolution, RWP2For the width of first resolution, RHP2Height for first resolution.
S402: the scale coefficient using S401 to determine zooms in and out process to the first template image, obtains the second Prototype drawing
Picture.
Specifically, according to [RWP2',RHP2']=[αWRWP2,αHRHP2] the first template image is zoomed in and out process,
[RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the width of the resolution of the second template image, RHP2'It it is the second mould
The height of the resolution of plate image.
S403: use the second template image to mate with image to be identified, obtains and the second mould from image to be identified
The target image that plate image matches.
This step sees the detailed description in S103, and here is omitted.
The present embodiment, by combining non-equal proportion zoom mode, according to first resolution and first mould of image to be identified
The second resolution of plate image, determines scale coefficient, uses the scale coefficient determined that the first template image is zoomed in and out process,
Obtain the second template image, use the second template image to mate with image to be identified, obtain and the from image to be identified
The target image that two template images match, owing to the first template image is zoomed in and out the scale coefficient that process is used, be
In conjunction with non-equal proportion zoom mode, directly according to first resolution and the second resolution of the first template image of image to be identified
Determine, only the first template image need to be carried out a scaling, i.e. can get second point of resolution and image to be identified
The second template image that resolution is identical, therefore, improves the efficiency of images match.Further, according to the first resolution of image to be identified
The second resolution of rate and the first template image determines and the first template image zooms in and out scale coefficient that process used more
Add accurately, it is thus possible to improve the accuracy of images match.
Embodiment six
When which kind of scaling pattern of employing cannot be determined, then use multiple scaling pattern, determine multiple yardstick
The scale coefficient that zoom mode is the most corresponding, uses the scale coefficient of multiple scaling pattern correspondence respectively to the first Prototype drawing
As zooming in and out process, obtain the second template image that multiple scaling pattern is the most corresponding, use multiple scaling mould
Corresponding respectively the second template image of formula mates with image to be identified, obtains the of multiple scaling pattern correspondence respectively
One candidate image, determines that from the first candidate image the first candidate image that matching result is the highest is target image.Such as Fig. 5 institute
Showing, Fig. 5 is the schematic flow sheet of image matching method embodiment six of the present invention, as follows:
S501: combine multiple scaling pattern, according to the first resolution of image to be identified and the first template image
Second resolution, determines the scale coefficient that every kind of scaling pattern is corresponding.
Such as: first use maximum zoom mode, obtain scale coefficient corresponding to maximum zoom mode 1;See S201
Middle detailed description, here is omitted.
Use minima zoom mode again, obtain scale coefficient corresponding to minima zoom mode 2;See in S301 detailed
Describing, here is omitted.
Finally use non-equal proportion zoom mode, obtain the scale coefficient 3 that non-equal proportion zoom mode is corresponding;See S401
Middle detailed description, here is omitted.
It should be noted that for specifically first obtaining the scale coefficient which kind of scaling pattern is corresponding, the present invention does not does
Limit.
S502: the first template image is zoomed in and out by scale coefficient respectively that use at least two scaling pattern corresponding
Process, obtain at least two the second template image.
Such as: first use the scale coefficient 1 that maximum zoom mode is corresponding, the first template image is zoomed in and out process,
Obtaining the second template image A, see in S202 and describe in detail, here is omitted.
Use the scale coefficient 2 that minima zoom mode is corresponding again, the first template image is zoomed in and out process, obtain
Two template image B, see in S302 and describe in detail, and here is omitted.
Finally use the scale coefficient 3 that non-equal proportion zoom mode is corresponding, the first template image is zoomed in and out process,
To the second template image C;Seeing in S402 and describe in detail, here is omitted.
It should be noted that for specifically first using scale coefficient corresponding to which kind of scaling pattern to the first Prototype drawing
Shape zooms in and out, and the present invention does not limits.
S503: use at least two the second template image to mate with image to be identified respectively, from image to be identified
Obtain the first candidate image that at least two the second template image matches.
Such as: first use the second template image A to mate with image to be identified, obtain and second from image to be identified
The first candidate image that template image A matches, sees in S103 and describes in detail, and here is omitted.
Use the second template image B to mate with image to be identified again, obtain and the second template from image to be identified
The first candidate image that image B matches, sees in S103 and describes in detail, and here is omitted.
Finally use the second template image C to mate with image to be identified, obtain and the second mould from image to be identified
The first candidate image that plate image C matches, sees in S103 and describes in detail, and here is omitted.
It should be noted that for specifically first using the second template image corresponding to which kind of scaling pattern with to be identified
Image mates, and the present invention does not limits.
S504: determine from the first candidate image that at least two the second template image matches that matching result is the highest
One candidate image is target image.
Such as: the first time that the first candidate image of matching from the second template image A, the second template image B match
Select in the first candidate image that image and the second template image C match, determine that the first candidate image that matching result is the highest is
Target image.
The present embodiment, by combining multiple scaling pattern, according to first resolution and first mould of image to be identified
The second resolution of plate image, determines the scale coefficient that every kind of scaling pattern is corresponding, uses at least two scaling mould
Scale coefficient corresponding to formula zooms in and out process to the first template image respectively, obtains at least two the second template image, uses
At least two the second template image mates with image to be identified respectively, obtains at least two the second mould from image to be identified
The first candidate image that plate image matches, determines from the first candidate image that at least two the second template image matches
Joining the first the highest candidate image of result is target image, in the process using every kind of scaling pattern to carry out images match
In, only the first template image need to be carried out a scaling, i.e. can get the second resolution of resolution and image to be identified
The second identical template image, therefore, improves the efficiency of images match.Further, according to the first resolution of image to be identified and
The second resolution of the first template image determines that the first template image zooms in and out the scale coefficient that process used is more accurate
Really, it is thus possible to improve the accuracy of images match.
Embodiment seven
In the various embodiments described above, the second template image is used to mate with image to be identified, from image to be identified
Obtaining the target image one specifically implementation matched with the second template image as shown in Figure 6, Fig. 6 is image of the present invention
The schematic flow sheet of matching process embodiment seven, as follows:
S601: use the second template image to mate in image to be identified, misses with presetting according to best matching result
Difference limen value, obtains N number of second candidate image.
Wherein, N is the integer more than or equal to 2.
Default credibility refers to when matching result is more than this numerical value, then it is assumed that matching result is credible, such as: combine Fig. 7 a
With Fig. 7 b, wherein, Fig. 7 a is the second template image one schematic diagram, and Fig. 7 b is image one schematic diagram to be identified, it is assumed that preset
Credibility a is 0.8, and the second template image mates with image to be identified, when the second template image slides into a certain position,
If matching result is less than this default credibility, then it is considered as incredible result and directly abandons, if matching result is pre-higher than this
If credibility, being then considered as segmented result, (x y) and matching result, is more than as obtained 7 matching results for position that record is corresponding
Preset the value of credibility, respectively mark in figure: 1., 2., 3., 4., 5., 6., 7. corresponding position;
1., 2., 3., 4., 5., 6., 7. corresponding matching result is as follows:
①{0.9790031909942627},②{0.9757121205329895},
③{0.9693324289321899},④{0.82920902967453},
⑤{0.8287180066108704},⑥{0.8188462257385254},
⑦{0.8054645657539368}
Assume to set and preset error threshold as b=0.03, it can be seen that maximum match result is C=
0.9790031909942627, then first threshold is C-b=0.9490031909942627, therefore, from 1., 2., 3., 4., 5.,
6. the second candidate image that, 7. middle acquisition matching result is more than first threshold is the image that position 1., 2., 3. is corresponding, as
It is effectively matched result.
Wherein, (x, determination mode y) is as follows:
In conjunction with Fig. 8 a and Fig. 8 b, Fig. 8 a is the another kind of schematic diagram of the second template image, and Fig. 8 b is the another of image to be identified
A kind of schematic diagram, the second template image size is M*N, and the size of image to be identified is K*L.
Slip initial coordinate: the second template image upper left corner overlaps with the image upper left corner to be identified, (x y) corresponds to coordinate
(0,0);
Sliding process coordinate: often slide laterally a pixel, abscissa+1, one pixel of every longitudinal sliding motion, vertical coordinate+1;
Slip end coordinate: the second template image lower right corner overlaps with the image lower right corner to be identified, (x y) corresponds to coordinate
(K-M, L-N).
Wherein, matching way includes but not limited to following several:
1, normalized difference quadratic sum coupling
Matching result m is standardized as [0,1].Best matching result is 0, and matching result the poorest then m value is the biggest.Presetting can
Reliability a, when m < during a, it is judged that the credibility of Corresponding matching result is enough, adds segmented result set.
Set and preset error threshold b (such as b=0.03), take out best matching result C in segmented result set.By standby
Select matching result in results set > result of C+b all deletes, thus, obtain the second candidate image.
2, standard relevant matches
Matching result m is standardized as [0,1].Worst matching result is 0, and matching result the best then m value is the biggest.Presetting can
Reliability a, as m > a time, it is judged that the credibility of Corresponding matching result is enough, adds segmented result set.
Set and preset error threshold b (such as b=0.03), take out best matching result C in segmented result set.By standby
Select matching result in results set < result of C-b is all deleted, thus, obtain the second candidate image.
3, standard relevant matches
Matching result m is standardized as [-1,1].This span is converted into [0,1], sets and preset credibility a, work as m
> a time, it is judged that the credibility of Corresponding matching result is enough, adds segmented result set.
Set and preset error threshold b (such as b=0.03), take out best matching result C in segmented result set.By standby
< result of C-b is all deleted, and obtains the second candidate image to select matching result in results set.
S602: combine coordinate and the coordinate of the second template image of N number of second candidate image, determine and the second template image
The target image matched.
I.e. determine the second candidate image that in the coordinate of N number of second candidate image, the coordinate of distance the second template image is nearest
For the target image matched with the second template image.
The present embodiment, carries out screening for the first time by default credibility, obtains matching result more than the figure presetting credibility
Picture, recycles the intensity of credibility between effective result, arranges default error threshold, according to default error threshold and maximum
Join result and first threshold is set, filter out the from matching result more than presetting the image of credibility further according to first threshold
Two candidate images, in conjunction with coordinate and the coordinate of the second template image of the second candidate image, determine and the second template image phase
The target image of coupling, thus, improve the accuracy of images match.
The image matching method of the various embodiments described above of the present invention can be applicable in mobile phone test scene, gets and the second mould
After the target image of plate images match, according to the operation that the second template image is corresponding, target image is carried out identical operation,
Such as: click, slip etc., thus, it is achieved automatization's mobile phone test.
It is appreciated that the image matching method of the various embodiments described above of the present invention can also be applied in other scene, right
This, the present invention does not limits.
Fig. 9 is the structural representation of image matching apparatus embodiment of the present invention, and the device of the present embodiment includes processing module
901, scaling module 902 and matching module 903, wherein, processing module 901 is for the first resolution according to image to be identified
Rate and the second resolution of the first template image, determine scale coefficient;Scaling module 902 is used for using scale coefficient to
One template image zooms in and out process, obtains the second template image, and the resolution of the second template image is identical with first resolution;
Matching module 903, for using the second template image to mate with image to be identified, obtains and second from image to be identified
The target image that template image matches.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment one, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, processing module 901 is specifically for obtaining scaling pattern;In conjunction with scaling mould
Formula, according to first resolution and the second resolution of the first template image of image to be identified, determines scale coefficient.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment two, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, scaling pattern is first mode;Processing module 901 is specifically for basisDetermine scale coefficient, wherein, [RWP1,RHP1] it is first resolution, [RWP2,RHP2] it is
Two resolution, [DW, DH] is the design resolution of image to be identified, RWP1For the width of second resolution, RHP1For second resolution
Height, RWP2For the width of first resolution, RHP2For the height of first resolution, DW is the width of design resolution, and DH differentiates for design
The height of rate;Scaling module 902 is specifically for according to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] to the first template image
Zoom in and out process, [RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the resolution of the second template image
Width, RHP2'It it is the height of the resolution of the second template image.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment three, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, scaling pattern is the second pattern;
Processing module 901 specifically for
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is first resolution,
[RWP2,RHP2] it is second resolution, [DW, DH] is the design resolution of image to be identified, RWP1For the width of second resolution,
RHP1For the height of second resolution, RWP2For the width of first resolution, RHP2For the height of first resolution, DW is design resolution
Width, DH is the height of design resolution;
Scaling module 902 is specifically for according to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] to the first Prototype drawing
As zooming in and out process, [RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the resolution of the second template image
Width, RHP2'It it is the height of the resolution of the second template image.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment four, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, scaling pattern is the 3rd pattern;
Processing module 901 specifically for
According toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is first resolution, [RWP2,RHP2] it is second resolution, RWP1For second resolution
Width, RHP1For the height of second resolution, RWP2For the width of first resolution, RHP2Height for first resolution;
Scaling module 902 is specifically for according to [RWP2',RHP2']=[αWRWP2,αHRHP2] the first template image is entered
Row scaling processes, [RWP2',RHP2'] it is the resolution of the second template image, RWP2'It is the width of the resolution of the second template image,
RHP2'It it is the height of the resolution of the second template image.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment five, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, scaling module 902 is corresponding specifically for using at least two scaling pattern
Scale coefficient respectively the first template image is zoomed in and out process, obtain at least two the second template image;
Matching module 903 mates with image to be identified respectively specifically for employing at least two the second template image,
The first candidate image that at least two the second template image matches is obtained from image to be identified;From at least two the second template
The first candidate image that image matches determining, the candidate image that matching result is the highest is target image.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment six, in fact
Existing principle is similar with technique effect, and here is omitted.
In the embodiment shown in fig. 9, matching module 903 is specifically for using the second template image to enter in image to be identified
Row coupling, according to best matching result and default error threshold, obtains N number of second candidate image, and N is the integer more than or equal to 2,
First threshold is the difference of maximum match result and default error threshold;Coordinate and the second mould in conjunction with N number of second candidate image
The coordinate of plate image, determines the target image matched with the second template image.
The device of the present embodiment, corresponding can be used for performing the technical scheme of embodiment of the method shown in embodiment seven, in fact
Existing principle is similar with technique effect, and here is omitted.
One of ordinary skill in the art will appreciate that: all or part of step realizing above-mentioned each method embodiment can be led to
The hardware crossing programmed instruction relevant completes.Aforesaid program can be stored in a computer read/write memory medium.This journey
Sequence upon execution, performs to include the step of above-mentioned each method embodiment;And aforesaid storage medium includes: ROM, RAM, magnetic disc or
The various media that can store program code such as person's CD.
Last it is noted that various embodiments above is only in order to illustrate technical scheme, it is not intended to limit;To the greatest extent
The present invention has been described in detail by pipe with reference to foregoing embodiments, it will be understood by those within the art that: it depends on
So the technical scheme described in foregoing embodiments can be modified, or the most some or all of technical characteristic is entered
Row equivalent;And these amendments or replacement, do not make the essence of appropriate technical solution depart from various embodiments of the present invention technology
The scope of scheme.
Claims (14)
1. an image matching method, it is characterised in that including:
First resolution according to image to be identified and the second resolution of the first template image, determine scale coefficient;
Use described scale coefficient that described first template image is zoomed in and out process, obtain the second template image, described second
The resolution of template image is identical with described first resolution;
Use described second template image to mate with described image to be identified, obtain with described from described image to be identified
The target image that second template image matches.
Method the most according to claim 1, it is characterised in that the described first resolution according to image to be identified and first
The second resolution of template image, determines scale coefficient, including:
Obtain scaling pattern;
In conjunction with described scaling pattern, according to first resolution and second resolution of the first template image of image to be identified
Rate, determines scale coefficient.
Method the most according to claim 2, it is characterised in that described scaling pattern is first mode;
Scaling pattern described in described combination, according to the first resolution of image to be identified and second point of the first template image
Resolution, determines scale coefficient, including:
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution,
[RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described second
The width of resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2Differentiate for described first
The height of rate, DW is the width of described design resolution, and DH is the height of described design resolution;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image, including:
According to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] described first template image is zoomed in and out process, [RWP2',
RHP2'] it is the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'For described
The height of the resolution of the second template image.
Method the most according to claim 2, it is characterised in that described scaling pattern is the second pattern;
Scaling pattern described in described combination, according to the first resolution of image to be identified and second point of the first template image
Resolution, determines scale coefficient, including:
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution,
[RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described second
The width of resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2Differentiate for described first
The height of rate, DW is the width of described design resolution, and DH is the height of described design resolution;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image, including:
According to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] described first template image is zoomed in and out process, [RWP2',
RHP2'] it is the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'For described
The height of the resolution of the second template image.
Method the most according to claim 2, it is characterised in that described scaling pattern is the 3rd pattern;
Scaling pattern described in described combination, according to the first resolution of image to be identified and second point of the first template image
Resolution, determines scale coefficient, including:
According toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, RWP1For described second point
The width of resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first resolution
Height;
The described scale coefficient of described employing zooms in and out process to described first template image, obtains the second template image, including:
According to [RWP2',RHP2']=[αWRWP2,αHRHP2] described first template image is zoomed in and out process, [RWP2',RHP2']
For the resolution of described second template image, RWP2'For the width of the resolution of described second template image, RHP2'For described second
The height of the resolution of template image.
Method the most according to claim 1, it is characterised in that the described scale coefficient of described employing is to described first Prototype drawing
As zooming in and out process, obtain the second template image, including:
The scale coefficient using at least two scaling pattern corresponding zooms in and out process to described first template image respectively,
Obtain at least two the second template image;
Described second template image of described employing mates with described image to be identified, from described image to be identified obtain with
The target image that described second template image matches, including:
Described at least two the second template image is used to mate with described image to be identified respectively, from described image to be identified
The first candidate image that middle acquisition described at least two the second template image matches;
First candidate's figure that matching result is the highest is determined from the candidate image that described at least two the second template image matches
Picture is described target image.
7. according to the method described in any one of Claims 1 to 5, it is characterised in that described second template image of described employing with
Described image to be identified mates, and obtains the target figure matched with described second template image from described image to be identified
Picture, including:
Described second template image is used to mate in described image to be identified, according to best matching result and default error
Threshold value, obtains N number of second candidate image, and described N is the integer more than or equal to 2;
In conjunction with coordinate and the coordinate of described second template image of described N number of second candidate image, determine and described second template
The target image that image matches.
8. an image matching apparatus, it is characterised in that including:
Processing module, for the first resolution according to image to be identified and the second resolution of the first template image, determines chi
Degree coefficient;
Scaling module, is used for using described scale coefficient that described first template image is zoomed in and out process, obtains second
Template image, the resolution of described second template image is identical with described first resolution;
Matching module, is used for using described second template image to mate with described image to be identified, from described figure to be identified
The target image matched with described second template image is obtained in Xiang.
Device the most according to claim 8, it is characterised in that described processing module is specifically for obtaining scaling mould
Formula;In conjunction with described scaling pattern, according to first resolution and the second resolution of the first template image of image to be identified,
Determine scale coefficient.
Device the most according to claim 9, it is characterised in that described scaling pattern is first mode;
Described processing module is specifically for basisDetermine scale coefficient, wherein, [RWP1,
RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, [DW, DH] is the design of described image to be identified
Resolution, RWP1For the width of described second resolution, RHP1For the height of described second resolution, RWP2For described first resolution
Width, RHP2For the height of described first resolution, DW is the width of described design resolution, and DH is the height of described design resolution;
Described scaling module is specifically for according to [RWP2',RHP2']=[αMINRWP2,αMINRHP2] to described first Prototype drawing
As zooming in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'For described second template image
The width of resolution, RHP2'Height for the resolution of described second template image.
11. devices according to claim 9, it is characterised in that described scaling pattern is the second pattern;
Described processing module specifically for
According toDetermine scale coefficient, wherein, [RWP1,RHP1] it is described first resolution,
[RWP2,RHP2] it is described second resolution, [DW, DH] is the design resolution of described image to be identified, RWP1For described second
The width of resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2Differentiate for described first
The height of rate, DW is the width of described design resolution, and DH is the height of described design resolution;
Described scaling module is specifically for according to [RWP2',RHP2']=[αMAXRWP2,αMAXRHP2] to described first Prototype drawing
As zooming in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'For described second template image
The width of resolution, RHP2'Height for the resolution of described second template image.
12. devices according to claim 9, it is characterised in that described scaling pattern is the 3rd pattern;
Described processing module specifically for
According toDetermine wide scale coefficient;
According toDetermine high scale coefficient;
Wherein, [RWP1,RHP1] it is described first resolution, [RWP2,RHP2] it is described second resolution, RWP1For described second point
The width of resolution, RHP1For the height of described second resolution, RWP2For the width of described first resolution, RHP2For described first resolution
Height;
Described scaling module is specifically for according to [RWP2',RHP2']=[αWRWP2,αHRHP2] to described first template image
Zoom in and out process, [RWP2',RHP2'] it is the resolution of described second template image, RWP2'Dividing for described second template image
The width of resolution, RHP2'Height for the resolution of described second template image.
13. devices according to claim 8, it is characterised in that described scaling module is specifically for using at least two
Scale coefficient corresponding to kind of scaling pattern zooms in and out process to described first template image respectively, obtains at least two the
Two template images;
Described matching module is carried out with described image to be identified respectively specifically for using described at least two the second template image
Coupling, obtains the first candidate image that described at least two the second template image matches from described image to be identified;From institute
State and the first candidate image that at least two the second template image matches determining, the candidate image that matching result is the highest is described
Target image.
14. the device according to Claim 8~described in 12 any one, it is characterised in that described matching module is specifically for using
Described second template image mates in described image to be identified, according to best matching result and default error threshold, obtains
Taking N number of second candidate image, described N is the integer more than or equal to 2;In conjunction with the coordinate of described N number of second candidate image and described
The coordinate of the second template image, determines the target image matched with described second template image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610436897.1A CN106096659A (en) | 2016-06-16 | 2016-06-16 | Image matching method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610436897.1A CN106096659A (en) | 2016-06-16 | 2016-06-16 | Image matching method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106096659A true CN106096659A (en) | 2016-11-09 |
Family
ID=57235645
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610436897.1A Pending CN106096659A (en) | 2016-06-16 | 2016-06-16 | Image matching method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106096659A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109313708A (en) * | 2017-12-22 | 2019-02-05 | 深圳配天智能技术研究院有限公司 | Image matching method and vision system |
CN111149101A (en) * | 2019-12-27 | 2020-05-12 | 威创集团股份有限公司 | Target pattern searching method and computer readable storage medium |
CN111368915A (en) * | 2020-03-04 | 2020-07-03 | 广东博智林机器人有限公司 | Drawing verification method, device, equipment and storage medium |
CN111797933A (en) * | 2020-07-10 | 2020-10-20 | 百度在线网络技术(北京)有限公司 | Template matching method and device, electronic equipment and storage medium |
CN112488177A (en) * | 2020-11-26 | 2021-03-12 | 金蝶软件(中国)有限公司 | Image matching method and related equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533466A (en) * | 2009-04-09 | 2009-09-16 | 南京壹进制信息技术有限公司 | Image processing method for positioning eyes |
CN101840567A (en) * | 2010-04-23 | 2010-09-22 | 清华大学 | Image reducing method for maintaining image topology structure |
CN102521795A (en) * | 2011-12-15 | 2012-06-27 | 中国科学院自动化研究所 | Cross matching fingerprint image scaling method based on global ridge distance |
CN105511725A (en) * | 2015-12-09 | 2016-04-20 | 网易(杭州)网络有限公司 | Method and device for displaying controls in interface |
-
2016
- 2016-06-16 CN CN201610436897.1A patent/CN106096659A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101533466A (en) * | 2009-04-09 | 2009-09-16 | 南京壹进制信息技术有限公司 | Image processing method for positioning eyes |
CN101840567A (en) * | 2010-04-23 | 2010-09-22 | 清华大学 | Image reducing method for maintaining image topology structure |
CN102521795A (en) * | 2011-12-15 | 2012-06-27 | 中国科学院自动化研究所 | Cross matching fingerprint image scaling method based on global ridge distance |
CN105511725A (en) * | 2015-12-09 | 2016-04-20 | 网易(杭州)网络有限公司 | Method and device for displaying controls in interface |
Non-Patent Citations (1)
Title |
---|
那彦: "《基于多分辨分析理论的图像融合方法》", 31 May 2007, 西安电子科技大学出版社 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109313708A (en) * | 2017-12-22 | 2019-02-05 | 深圳配天智能技术研究院有限公司 | Image matching method and vision system |
CN111149101A (en) * | 2019-12-27 | 2020-05-12 | 威创集团股份有限公司 | Target pattern searching method and computer readable storage medium |
WO2021128243A1 (en) * | 2019-12-27 | 2021-07-01 | 威创集团股份有限公司 | Target pattern lookup method and computer-readable storage medium |
CN111149101B (en) * | 2019-12-27 | 2023-08-18 | 威创集团股份有限公司 | Target pattern searching method and computer readable storage medium |
CN111368915A (en) * | 2020-03-04 | 2020-07-03 | 广东博智林机器人有限公司 | Drawing verification method, device, equipment and storage medium |
CN111797933A (en) * | 2020-07-10 | 2020-10-20 | 百度在线网络技术(北京)有限公司 | Template matching method and device, electronic equipment and storage medium |
CN112488177A (en) * | 2020-11-26 | 2021-03-12 | 金蝶软件(中国)有限公司 | Image matching method and related equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106096659A (en) | Image matching method and device | |
US9998651B2 (en) | Image processing apparatus and image processing method | |
EP3226172B1 (en) | Fingerprint registration method, device and terminal equipment | |
EP2042976B1 (en) | Image processing method | |
RU2421775C2 (en) | Information processing device and method, and computer programme | |
CN109389038A (en) | A kind of detection method of information, device and equipment | |
JP2007305097A (en) | Motion detection method for one-dimensional fingerprint detection module | |
CN107729790A (en) | Quick Response Code localization method and device | |
US20110199502A1 (en) | Image pickup apparatus | |
US20180025247A1 (en) | Information processing apparatus, information processing method, and program | |
CN105094675A (en) | Man-machine interaction method and touch screen wearable device | |
US20150154761A1 (en) | Scene scan | |
US20100080490A1 (en) | Apparatus and method for processing image | |
CN103608761A (en) | Input device, input method and recording medium | |
US20150154736A1 (en) | Linking Together Scene Scans | |
CN105955626A (en) | Photographing search method and device | |
CN109615171A (en) | Characteristic threshold value determines that method and device, problem objects determine method and device | |
US10895919B2 (en) | Gesture control method and apparatus for display screen | |
CN106062608A (en) | Imaging device and focusing control method | |
KR100780057B1 (en) | Device for detecting gradual shot conversion in moving picture and method for operating the device | |
CN107682691B (en) | A kind of method, terminal and the computer readable storage medium of camera focus calibration | |
CN110008802A (en) | Selection target face and face recognition comparison method, device from multiple faces | |
WO2020181522A1 (en) | Defective pixel detection method, image processing chip, and electronic device | |
CN114495041A (en) | Method, device, equipment and medium for measuring distance between vehicle and target object | |
JP2008211534A (en) | Face detecting device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161109 |