CN106156784A - A kind of characteristic recognition method and electronic equipment - Google Patents
A kind of characteristic recognition method and electronic equipment Download PDFInfo
- Publication number
- CN106156784A CN106156784A CN201510144573.6A CN201510144573A CN106156784A CN 106156784 A CN106156784 A CN 106156784A CN 201510144573 A CN201510144573 A CN 201510144573A CN 106156784 A CN106156784 A CN 106156784A
- Authority
- CN
- China
- Prior art keywords
- result
- comparison result
- characteristic information
- weight
- comparison
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Character Discrimination (AREA)
Abstract
The open a kind of characteristic recognition method of the present invention and electronic equipment.Described method includes: obtain characteristic information to be identified；Use the first recognizer to be compared with existing characteristic information by described characteristic information, obtain the first comparison result；Use the second recognizer to be compared with existing characteristic information by described characteristic information, obtain the second comparison result；Carry out computing based on described first comparison result and described second comparison result, obtain the 3rd comparison result；Judge whether described global alignment result meets first pre-conditioned, obtain the first judged result；When described first judged result represent described global alignment result meet first pre-conditioned time, determine that described characteristic information matches with existing characteristic information.Use method or the electronic equipment of the present invention, it is possible to use different types of recognizer carries out feature identification, thus has the feature of various features recognizer concurrently.
Description
Technical field
The present invention relates to sensing detection field, particularly relate to a kind of characteristic recognition method and electronic equipment.
Background technology
Along with the development of electronic product, the function that electronic equipment is had also gets more and more.At present, special
Levy identification function be applied in increasing electronic equipment.
As a example by fingerprint recognition, prior art, there is the multiple algorithm that can be used for fingerprint recognition.But it is single
Certain algorithm be generally of obvious shortcoming.Such as, the algorithm that accuracy of identification is high, usual computing
Journey is complicated, and recognition time is longer；The algorithm that recognition speed is fast, the generally recognized precision is relatively low.
Therefore, a kind of characteristic recognition method is needed badly, in order to the feature of comprehensive various features recognizer.
Summary of the invention
It is an object of the invention to provide a kind of characteristic recognition method and electronic equipment, it is possible to use various features is known
Other algorithm, has the feature of various features recognizer concurrently.
For achieving the above object, the invention provides following scheme:
A kind of characteristic recognition method, including:
Obtain characteristic information to be identified；
Use the first recognizer to be compared with existing characteristic information by described characteristic information, obtain the first ratio
To result；
Use the second recognizer to be compared with existing characteristic information by described characteristic information, obtain the second ratio
To result；
Carry out computing based on described first comparison result and described second comparison result, obtain the 3rd comparison knot
Really；
Judge whether described global alignment result meets first pre-conditioned, obtain the first judged result；
When described first judged result represent described global alignment result meet first pre-conditioned time, determine institute
State characteristic information to match with existing characteristic information.
Optionally, described carry out computing based on described first comparison result and described second comparison result, obtain
3rd comparison result, specifically includes:
Use the first parameter and the first comparison result to carry out computing, obtain the first operation result；
Use the second parameter and the second comparison result to carry out computing, obtain the second operation result；
Described first operation result and described second operation result are carried out sum operation.
Optionally, described use the first parameter and the first comparison result carry out computing, obtain the first operation result,
Specifically include:
Use the first weight to be weighted with the first comparison result, obtain the first weighted results；
Described use the second parameter and the second comparison result carry out computing, obtain the second operation result, specifically wrap
Include:
Use the second weight to be weighted with the second comparison result, obtain the second weighted results；
Described described first operation result and described second operation result are carried out sum operation, specifically include:
Described first weighted results and described second weighted results are carried out sum operation, obtains weighted sum knot
Really；
Accordingly, described to judge whether described global alignment result meets first pre-conditioned, specifically includes:
Judge that whether described weighted sum result is more than or equal to the first predetermined threshold value.
Optionally, described carry out computing based on described first comparison result and described second comparison result before,
Also include:
Identify current applied environment information；
Described first weight and described second weight is determined based on described applied environment information.
Optionally, described determine described first weight and described second weight based on described applied environment information,
Specifically include:
The preferred misclassification rate that current environment is corresponding is determined based on described applied environment information；
Obtain the first misclassification rate of described first recognizer；
Obtain the second misclassification rate of described second recognizer；
Described first misclassification rate, the second misclassification rate are compared with described environment misclassification rate, it is thus achieved that the 4th ratio
To result；
The first weight and the second weight is determined based on described 4th comparison result；
Wherein, if the absolute value of the difference of described first misclassification rate and described environment misclassification rate is less than described second
The absolute value of the difference of misclassification rate and described environment misclassification rate, it is determined that described first weight more than described the
Two weights；
If described first misclassification rate is known more than described second with the absolute value of the difference of described environment misclassification rate by mistake
The absolute value of the difference of rate and described environment misclassification rate, it is determined that described first weight less than described second power
Weight.
Optionally, described first comparison result is specially the first matching degree numerical value, described second comparison result tool
Body is the second matching degree numerical value, described transports with described second comparison result based on described first comparison result
Before calculation, also include:
Relatively described first matching degree numerical value and the size of described second matching degree numerical value；
When described first matching degree numerical value is less than described second matching degree numerical value, for described first comparison result
Distribute the first weight, distribute the second weight for described second comparison result；
When described first matching degree numerical value is more than described second matching degree numerical value, for described first comparison result
Distribute the second weight, distribute the first weight for described second comparison result；
Described first weight is more than described second weight.
Optionally, described carry out computing based on described first comparison result and described second comparison result, obtain
3rd comparison result, specifically includes:
Judge described first comparison result indicate whether described in described first recognizer characteristic information with
Existing characteristic information comparison success, obtains the second judged result；
Judge described second comparison result indicate whether described in described second recognizer characteristic information with
Existing characteristic information comparison success, obtains the 3rd judged result；
Accordingly, described to judge whether described global alignment result meets first pre-conditioned, specifically includes:
Judge that described second judged result the most all represents certainly with described 3rd judged result.
A kind of electronic equipment, including:
Characteristic acquisition unit, for obtaining characteristic information to be identified；
First recognition unit, for using the first recognizer to be entered with existing characteristic information by described characteristic information
Row comparison, obtains the first comparison result；
Second recognition unit, for using the second recognizer to be entered with existing characteristic information by described characteristic information
Row comparison, obtains the second comparison result；
Arithmetic element, for carrying out computing based on described first comparison result and described second comparison result,
To the 3rd comparison result；
First judging unit, is used for judging whether described global alignment result meets first pre-conditioned, obtains
First judged result；
Determine unit, for representing that described global alignment result meets first and presets when described first judged result
During condition, determine that described characteristic information matches with existing characteristic information.
Optionally, described arithmetic element, specifically include:
First operator unit, for using the first parameter and the first comparison result to carry out computing, obtains first
Operation result；
Second operator unit, for using the second parameter and the second comparison result to carry out computing, obtains second
Operation result；
Summation subelement, for carrying out summation behaviour to described first operation result and described second operation result
Make.
Optionally, described first operator unit, specifically include:
First weighting subelement, for using the first weight to be weighted with the first comparison result, obtains first
Weighted results；
Described second operator unit, specifically includes:
Second weighting subelement, for using the second weight to be weighted with the second comparison result, obtains second
Weighted results；
Described summation subelement, specifically includes:
Weighted results summation subelement, for carrying out with described second weighted results described first weighted results
Sum operation, obtains weighted sum result；
Accordingly, described first judging unit, specifically include:
First judgment sub-unit, is used for judging whether described weighted sum result presets threshold more than or equal to first
Value.
The specific embodiment provided according to the present invention, the invention discloses techniques below effect:
Characteristic recognition method in the embodiment of the present invention and electronic equipment, by using the first recognizer by institute
State characteristic information to compare with existing characteristic information, obtain the first comparison result；Use the second recognizer
Described characteristic information is compared with existing characteristic information, obtains the second comparison result；Based on described first
Comparison result and described second comparison result carry out computing, obtain the 3rd comparison result；Judge described comprehensive ratio
Whether result is met first pre-conditioned；When described global alignment result meet first pre-conditioned time, really
Fixed described characteristic information matches with existing characteristic information；Different types of recognizer can be utilized to carry out spy
Levy identification, thus have the feature of various features recognizer concurrently.
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 enforcement
In example, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only
Some embodiments of the present invention, for those of ordinary skill in the art, are not paying creative work
On the premise of, it is also possible to other accompanying drawing is obtained according to these accompanying drawings.
Fig. 1 is the flow chart of inventive feature recognition methods embodiment 1；
Fig. 2 is the flow chart of inventive feature recognition methods embodiment 2；
Fig. 3 is the flow chart of inventive feature recognition methods embodiment 3；
Fig. 4 is the flow chart of inventive feature recognition methods embodiment 4；
Fig. 5 is the flow chart of inventive feature recognition methods embodiment 5；
Fig. 6 is the structure chart of the electronic equipment embodiment of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clearly
Chu, be fully described by, it is clear that described embodiment be only a part of embodiment of the present invention rather than
Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creation
The every other embodiment obtained under property work premise, broadly falls into the scope of protection of the invention.
Understandable for enabling the above-mentioned purpose of the present invention, feature and advantage to become apparent from, below in conjunction with the accompanying drawings and
The present invention is further detailed explanation for detailed description of the invention.
Characteristic recognition method in the present invention, can apply to the identification of multiple types another characteristic.Such as, may be used
To be applied to fingerprint recognition, Application on Voiceprint Recognition, pupil identification, image recognition etc..Feature in the present invention is known
The executive agent of other method, can be the various electronic equipments with feature identification function, such as, mobile phone,
Panel computer, notebook computer or various types of safety device etc..
Fig. 1 is the flow chart of inventive feature recognition methods embodiment 1.As it is shown in figure 1, the method can
To include:
Step 101: obtain characteristic information to be identified；
Described characteristic information can be various types of characteristic information.For example, it may be finger print information, vocal print
Information, image information etc..
Corresponding sensor can be first used to obtain characteristic information to be identified.The executive agent of this method is the most right
The characteristic information that sensor gets obtains.
Step 102: use the first recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the first comparison result；
Described existing characteristic information is the characteristic information prestored.Described existing characteristic information can be to electricity
Subset has the characteristic information of the user of specified permission.
Described first recognizer can be various types of recognizer.For example, it may be to characteristic information
The algorithm that is identified of entirety, it is also possible to be that the part in characteristic information with obvious feature is carried out
The algorithm identified.
Described first comparison result can have the various form of expression.Such as, described first comparison result can be
Numeric form, or described first comparison result can also be the character style representing positive or negative.
Step 103: use the second recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the second comparison result；
Described second recognizer is different algorithms from described first recognizer.
Described second comparison result can also have the various form of expression.Such as, described second comparison result is permissible
It is numeric form, or described second comparison result can also be the character style representing positive or negative.
Step 104: carry out computing based on described first comparison result and described second comparison result, obtains the
Three comparison results；
The form of described computing includes multiple.Such as, when described first comparison result and described second comparison knot
When fruit is numeric form, described calculating process can be can be to described first comparison result and described second
Comparison result is weighted averagely；It agree when described first comparison result and described second comparison result are expression
During the character style determined or negate, described computing can be logical operations.
Step 105: judge whether described global alignment result meets first pre-conditioned, obtain the first judgement
Result；
Described first pre-conditioned can set according to the actual requirements.And described first pre-conditioned with described combine
The concrete manifestation form closing comparison result is relevant.If described global alignment result is numeric form, then described
First pre-conditioned can be predetermined threshold value.If described global alignment result is to represent the word of positive or negative
During symbol form, the most described first pre-conditioned can be described global alignment result be yes.
Step 106: to represent that described global alignment result meets first pre-conditioned when described first judged result
Time, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, by using the first recognizer to be carried out with existing characteristic information by described characteristic information
Comparison, obtains the first comparison result；Use the second recognizer by described characteristic information and existing characteristic information
Compare, obtain the second comparison result；Enter with described second comparison result based on described first comparison result
Row operation, obtains the 3rd comparison result；Judge whether described global alignment result meets first pre-conditioned；
When described global alignment result meet first pre-conditioned time, determine described characteristic information and existing characteristic information
Match；Different types of recognizer can be utilized to carry out feature identification, thus have various features identification concurrently
The feature of algorithm.
In actual application, described carry out computing based on described first comparison result and described second comparison result,
Obtain the 3rd comparison result, specifically may comprise steps of:
Use the first parameter and the first comparison result to carry out computing, obtain the first operation result；
Use the second parameter and the second comparison result to carry out computing, obtain the second operation result；
Described first operation result and described second operation result are carried out sum operation.
In above-mentioned steps, described first comparison result and described second comparison result can be numeric form.
Described use the first parameter and the first comparison result carry out computing, obtain the first operation result, specifically may be used
To include:
Use the first weight to be weighted with the first comparison result, obtain the first weighted results.
Described use the second parameter and the second comparison result carry out computing, obtain the second operation result, specifically may be used
To include:
Use the second weight to be weighted with the second comparison result, obtain the second weighted results.
Described described first operation result and described second operation result are carried out sum operation, specifically can wrap
Include:
Described first weighted results and described second weighted results are carried out sum operation, obtains weighted sum knot
Really；
Accordingly, described to judge whether described global alignment result meets first pre-conditioned, specifically includes:
Judge that whether described weighted sum result is more than or equal to the first predetermined threshold value.
Fig. 2 is the flow chart of inventive feature recognition methods embodiment 2.As in figure 2 it is shown, the method can
To include:
Step 201: obtain characteristic information to be identified；
Step 202: use the first recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the first comparison result；
Step 203: use the second recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the second comparison result；
Step 204: use the first weight to be weighted with the first comparison result, obtain the first weighted results；
Step 205: use the second weight to be weighted with the second comparison result, obtain the second weighted results；
Step 206: described first weighted results and described second weighted results are carried out sum operation, obtains
Weighted sum result；
Step 207: judge that described weighted sum result, whether more than or equal to the first predetermined threshold value, obtains the
One judged result；
Step 208: when described first judged result represents that described weighted sum result is pre-more than or equal to first
If during threshold value, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, described first comparison result can be the comparison using the first recognizer comparison to obtain
Mark, described second comparison result can be the alignment score using the second recognizer comparison to obtain.Assume
The full marks of the comparison result of described first recognizer are 100, and coupling mark is 60, the most described first comparison
Result reach 60 points or more than, will judge that in described first recognizer characteristic information to be identified is with existing
Characteristic information matches.The full marks assuming the comparison result of the second recognizer are 100, and coupling mark is also
60.If described first comparison result is 50 points, described second comparison result is 70 points, described first power
Being heavily 0.5, described second weight is also 0.5.Then weighted sum result is 60 points.Assume that first presets threshold
Value is 60 points.Then after weighted sum, originally cannot the spy to be identified that the match is successful in the first recognizer
Reference ceases, after using the method for the present embodiment, by the match is successful.The i.e. method of the present embodiment is above-mentioned in use
During the weight assumed, by make the Stringency of coupling between the first recognizer and the second recognizer it
Between so that matching precision is moderate.
In actual application, after described first weight and described second weight can pre-set, it is used for multiple times,
Electronic equipment can also be used to determine described first weight and described second power voluntarily before weighting every time
Weight.Concrete, use electronic equipment to determine described first weight and described second weight voluntarily, can use
In the following manner:
Before carrying out computing based on described first comparison result with described second comparison result,
Identify current applied environment information；
Described first weight and described second weight is determined based on described applied environment information.
Fig. 3 is the flow chart of inventive feature recognition methods embodiment 3.As it is shown on figure 3, the method can
To include:
Step 301: obtain characteristic information to be identified；
Step 302: use the first recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the first comparison result；
Step 303: use the second recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the second comparison result；
Step 304: identify current applied environment information；
Step 305: determine the preferred misclassification rate that current environment is corresponding based on described applied environment information；
Step 306: obtain the first misclassification rate of described first recognizer；
Step 307: obtain the second misclassification rate of described second recognizer；
Step 308: described first misclassification rate, the second misclassification rate are compared with described environment misclassification rate,
Obtain the 4th comparison result；
Step 309: determine the first weight and the second weight based on described 4th comparison result；
Wherein, if the absolute value of the difference of described first misclassification rate and described environment misclassification rate is less than described second
The absolute value of the difference of misclassification rate and described environment misclassification rate, it is determined that described first weight more than described the
Two weights；
If described first misclassification rate is known more than described second with the absolute value of the difference of described environment misclassification rate by mistake
The absolute value of the difference of rate and described environment misclassification rate, it is determined that described first weight less than described second power
Weight.
Step 310: use the first weight to be weighted with the first comparison result, obtain the first weighted results；
Step 311: use the second weight to be weighted with the second comparison result, obtain the second weighted results；
Step 312: described first weighted results and described second weighted results are carried out sum operation, obtains
Weighted sum result；
Step 313: judge that described weighted sum result, whether more than or equal to the first predetermined threshold value, obtains the
One judged result；
Step 314: when described first judged result represents that described weighted sum result is pre-more than or equal to first
If during threshold value, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, by determining the first weight and the second weight based on described 4th comparison result；Wherein,
If the absolute value of the difference of described first misclassification rate and described environment misclassification rate less than described second misclassification rate with
The absolute value of the difference of described environment misclassification rate, it is determined that described first weight more than described second weight；
If the absolute value of the difference of described first misclassification rate and described environment misclassification rate more than described second misclassification rate with
The absolute value of the difference of described environment misclassification rate, it is determined that described first weight less than described second weight.
Can be the weight that algorithm immediate with preferred misclassification rate distribution is bigger, so that the mistake of the algorithm after Zong He
Knowledge rate is closer to preferred misclassification rate.
In actual application, it is also possible to adopt and determine described first weight and described second weight in other ways.
Fig. 4 is the flow chart of inventive feature recognition methods embodiment 4.As shown in Figure 4, the method can
To include:
Step 401: obtain characteristic information to be identified；
Step 402: use the first recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the first matching degree numerical value；
Step 403: use the second recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the second matching degree numerical value；
Step 404: relatively described first matching degree numerical value and the size of described second matching degree numerical value；
Step 405: when described first matching degree numerical value is less than described second matching degree numerical value, for described the
One comparison result distributes the first weight, distributes the second weight for described second comparison result；
Step 406: when described first matching degree numerical value is more than described second matching degree numerical value, for described the
One comparison result distributes the second weight, distributes the first weight for described second comparison result；
Described first weight is more than described second weight.
Step 407: use the first weight to be weighted with the first comparison result, obtain the first weighted results；
Step 408: use the second weight to be weighted with the second comparison result, obtain the second weighted results；
Step 409: described first weighted results and described second weighted results are carried out sum operation, obtains
Weighted sum result；
Step 410: judge that described weighted sum result, whether more than or equal to the first predetermined threshold value, obtains the
One judged result；
Step 411: when described first judged result represents that described weighted sum result is pre-more than or equal to first
If during threshold value, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, big by relatively described first matching degree numerical value and described second matching degree numerical value
Little；When described first matching degree numerical value is less than described second matching degree numerical value, for described first comparison result
Distribute the first weight, distribute the second weight for described second comparison result；When described first matching degree numerical value is big
When described second matching degree numerical value, distribute the second weight for described first comparison result, for described second ratio
Result is distributed the first weight；Can be to identify that the algorithm that Stringency is higher distributes bigger weight, make to combine
The identification Stringency of the characteristic recognition method after conjunction improves.
Fig. 5 is the flow chart of inventive feature recognition methods embodiment 5.As it is shown in figure 5, the method can
To include:
Step 501: obtain characteristic information to be identified；
Step 502: use the first recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the first comparison result；
Step 503: use the second recognizer to be compared with existing characteristic information by described characteristic information,
Obtain the second comparison result；
Step 504: judge that described first comparison result indicates whether described in described first recognizer special
Reference breath and existing characteristic information comparison success, obtain the second judged result；
Step 505: judge that described second comparison result indicates whether described in described second recognizer special
Reference breath and existing characteristic information comparison success, obtain the 3rd judged result；
Step 506: judge that described second judged result the most all represents with described 3rd judged result certainly,
Obtain the first judged result.
Step 507: when described first judged result represents that described second judged result judges knot with the described 3rd
When fruit all represents certainly, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, described carry out computing based on described first comparison result and described second comparison result,
Obtain the 3rd comparison result, specifically include: judge described first comparison result to indicate whether described first and know
Characteristic information described in other algorithm and existing characteristic information comparison success, obtain the second judged result；Judge institute
State the second comparison result and indicate whether characteristic information and existing feature letter described in described second recognizer
Breath comparison success, obtains the 3rd judged result；It is described that to judge whether described global alignment result meets first pre-
If condition, specifically include: judge that described second judged result the most all represents with described 3rd judged result and agree
Fixed；Therefore, in the present embodiment, comprehensive after the first misclassification rate and second that misclassification rate is the first recognizer
The product of the second misclassification rate of recognizer.And the first misclassification rate and the second misclassification rate are respectively less than or equal to percentage
5, so the characteristic recognition method in the present embodiment has the lowest misclassification rate.
The invention also discloses a kind of electronic equipment.Described electronic equipment can be various to have feature identification merit
The electronic equipment of energy, such as, mobile phone, panel computer, notebook computer or various types of safety device
Etc..
Fig. 6 is the structure chart of the electronic equipment embodiment of the present invention.As shown in Figure 6, this electronic equipment is permissible
Including:
Characteristic acquisition unit 601, for obtaining characteristic information to be identified；
First recognition unit 602, for using the first recognizer by described characteristic information and existing feature letter
Breath is compared, and obtains the first comparison result；
Second recognition unit 603, for using the second recognizer by described characteristic information and existing feature letter
Breath is compared, and obtains the second comparison result；
Arithmetic element 604, for carrying out computing based on described first comparison result and described second comparison result,
Obtain the 3rd comparison result；
First judging unit 605, is used for judging whether described global alignment result meets first pre-conditioned,
Obtain the first judged result；
Determine unit 606, for representing that described global alignment result meets first when described first judged result
Time pre-conditioned, determine that described characteristic information matches with existing characteristic information.
In the present embodiment, by using the first recognizer to be carried out with existing characteristic information by described characteristic information
Comparison, obtains the first comparison result；Use the second recognizer by described characteristic information and existing characteristic information
Compare, obtain the second comparison result；Enter with described second comparison result based on described first comparison result
Row operation, obtains the 3rd comparison result；Judge whether described global alignment result meets first pre-conditioned；
When described global alignment result meet first pre-conditioned time, determine described characteristic information and existing characteristic information
Match；Different types of recognizer can be utilized to carry out feature identification, thus have various features identification concurrently
The feature of algorithm.
In actual application, described arithmetic element 604, specifically include:
First operator unit, for using the first parameter and the first comparison result to carry out computing, obtains first
Operation result；
Second operator unit, for using the second parameter and the second comparison result to carry out computing, obtains second
Operation result；
Summation subelement, for carrying out summation behaviour to described first operation result and described second operation result
Make.
In actual application, described first operator unit, specifically include:
First weighting subelement, for using the first weight to be weighted with the first comparison result, obtains first
Weighted results；
Described second operator unit, specifically includes:
Second weighting subelement, for using the second weight to be weighted with the second comparison result, obtains second
Weighted results；
Described summation subelement, specifically includes:
Weighted results summation subelement, for carrying out with described second weighted results described first weighted results
Sum operation, obtains weighted sum result；
Accordingly, described first judging unit, specifically include:
First judgment sub-unit, is used for judging whether described weighted sum result presets threshold more than or equal to first
Value.
In actual application, this electronic equipment can also include:
Applied environment information identificating unit, for based on described first comparison result and described second comparison knot
Before fruit carries out computing, identify current applied environment information；
Weight determining unit, for determining described first weight and described second based on described applied environment information
Weight.
In actual application, described weight determining unit, specifically may include that
Preferably misclassification rate determines subelement, for determining that current environment is corresponding based on described applied environment information
Preferably misclassification rate；
First misclassification rate obtains subelement, for obtaining the first misclassification rate of described first recognizer；
Second misclassification rate obtains subelement, for obtaining the second misclassification rate of described second recognizer；
Misclassification rate comparer unit, for knowing described first misclassification rate, the second misclassification rate with described environment by mistake
Rate is compared, it is thus achieved that the 4th comparison result；
Weight determines subelement, for determining the first weight and the second weight based on described 4th comparison result；
Wherein, if the absolute value of the difference of described first misclassification rate and described environment misclassification rate is less than described second
The absolute value of the difference of misclassification rate and described environment misclassification rate, it is determined that described first weight more than described the
Two weights；
If described first misclassification rate is known more than described second with the absolute value of the difference of described environment misclassification rate by mistake
The absolute value of the difference of rate and described environment misclassification rate, it is determined that described first weight less than described second power
Weight.
In actual application, described first comparison result is specially the first matching degree numerical value, described second comparison knot
Fruit is specially the second matching degree numerical value, and this electronic equipment can also include:
Matching degree numerical value comparing unit, for based on described first comparison result and described second comparison result
Before carrying out computing, relatively described first matching degree numerical value and the size of described second matching degree numerical value；
First allocation unit, is used for when described first matching degree numerical value is less than described second matching degree numerical value,
Distribute the first weight for described first comparison result, distribute the second weight for described second comparison result；
Second allocation unit, is used for when described first matching degree numerical value is more than described second matching degree numerical value,
Distribute the second weight for described first comparison result, distribute the first weight for described second comparison result；
Described first weight is more than described second weight.
In actual application, described arithmetic element 604, specifically may include that
Second judgment sub-unit, is used for judging described first comparison result to indicate whether described first and identifies calculation
Characteristic information described in method and existing characteristic information comparison success, obtain the second judged result；
3rd judgment sub-unit, is used for judging described second comparison result to indicate whether described second and identifies calculation
Characteristic information described in method and existing characteristic information comparison success, obtain the 3rd judged result；
Accordingly, described first judging unit 605, specifically may include that
Result judgment sub-unit, is used for judging that described second judged result is the most equal with described 3rd judged result
Represent certainly.
In this specification, each embodiment uses the mode gone forward one by one to describe, and what each embodiment stressed is
With the difference of other embodiments, between each embodiment, identical similar portion sees mutually.For
For electronic equipment disclosed in embodiment, owing to it corresponds to the method disclosed in Example, so describe
Fairly simple, relevant part sees method part and illustrates.
Principle and the embodiment of the present invention are set forth by specific case used herein, above enforcement
The explanation of example is only intended to help to understand method and the core concept thereof of the present invention；Simultaneously for this area
Those skilled in the art, according to the thought of the present invention, the most all can change
Part.In sum, this specification content should not be construed as limitation of the present invention.
Claims (10)
1. a characteristic recognition method, it is characterised in that including:
Obtain characteristic information to be identified；
Use the first recognizer to be compared with existing characteristic information by described characteristic information, obtain the first ratio
To result；
Use the second recognizer to be compared with existing characteristic information by described characteristic information, obtain the second ratio
To result；
Carry out computing based on described first comparison result and described second comparison result, obtain the 3rd comparison knot
Really；
Judge whether described global alignment result meets first pre-conditioned, obtain the first judged result；
When described first judged result represent described global alignment result meet first pre-conditioned time, determine institute
State characteristic information to match with existing characteristic information.
Method the most according to claim 1, it is characterised in that described based on described first comparison knot
Fruit carries out computing with described second comparison result, obtains the 3rd comparison result, specifically includes:
Use the first parameter and the first comparison result to carry out computing, obtain the first operation result；
Use the second parameter and the second comparison result to carry out computing, obtain the second operation result；
Described first operation result and described second operation result are carried out sum operation.
Method the most according to claim 2, it is characterised in that described use the first parameter and first
Comparison result carries out computing, obtains the first operation result, specifically includes:
Use the first weight to be weighted with the first comparison result, obtain the first weighted results；
Described use the second parameter and the second comparison result carry out computing, obtain the second operation result, specifically wrap
Include:
Use the second weight to be weighted with the second comparison result, obtain the second weighted results；
Described described first operation result and described second operation result are carried out sum operation, specifically include:
Described first weighted results and described second weighted results are carried out sum operation, obtains weighted sum knot
Really；
Accordingly, described to judge whether described global alignment result meets first pre-conditioned, specifically includes:
Judge that whether described weighted sum result is more than or equal to the first predetermined threshold value.
Method the most according to claim 3, it is characterised in that described based on described first comparison knot
Before fruit carries out computing with described second comparison result, also include:
Identify current applied environment information；
Described first weight and described second weight is determined based on described applied environment information.
Method the most according to claim 4, it is characterised in that described believe based on described applied environment
Breath determines described first weight and described second weight, specifically includes:
The preferred misclassification rate that current environment is corresponding is determined based on described applied environment information；
Obtain the first misclassification rate of described first recognizer；
Obtain the second misclassification rate of described second recognizer；
Described first misclassification rate, the second misclassification rate are compared with described environment misclassification rate, it is thus achieved that the 4th ratio
To result；
The first weight and the second weight is determined based on described 4th comparison result；
Wherein, if the absolute value of the difference of described first misclassification rate and described environment misclassification rate is less than described second
The absolute value of the difference of misclassification rate and described environment misclassification rate, it is determined that described first weight more than described the
Two weights；
If described first misclassification rate is known more than described second with the absolute value of the difference of described environment misclassification rate by mistake
The absolute value of the difference of rate and described environment misclassification rate, it is determined that described first weight less than described second power
Weight.
Method the most according to claim 3, it is characterised in that described first comparison result is specially
First matching degree numerical value, described second comparison result is specially the second matching degree numerical value, described based on described
Before one comparison result and described second comparison result carry out computing, also include:
Relatively described first matching degree numerical value and the size of described second matching degree numerical value；
When described first matching degree numerical value is less than described second matching degree numerical value, for described first comparison result
Distribute the first weight, distribute the second weight for described second comparison result；
When described first matching degree numerical value is more than described second matching degree numerical value, for described first comparison result
Distribute the second weight, distribute the first weight for described second comparison result；
Described first weight is more than described second weight.
Method the most according to claim 1, it is characterised in that described based on described first comparison knot
Fruit carries out computing with described second comparison result, obtains the 3rd comparison result, specifically includes:
Judge described first comparison result indicate whether described in described first recognizer characteristic information with
Existing characteristic information comparison success, obtains the second judged result；
Judge described second comparison result indicate whether described in described second recognizer characteristic information with
Existing characteristic information comparison success, obtains the 3rd judged result；
Accordingly, described to judge whether described global alignment result meets first pre-conditioned, specifically includes:
Judge that described second judged result the most all represents certainly with described 3rd judged result.
8. an electronic equipment, it is characterised in that including:
Characteristic acquisition unit, for obtaining characteristic information to be identified；
First recognition unit, for using the first recognizer to be entered with existing characteristic information by described characteristic information
Row comparison, obtains the first comparison result；
Second recognition unit, for using the second recognizer to be entered with existing characteristic information by described characteristic information
Row comparison, obtains the second comparison result；
Arithmetic element, for carrying out computing based on described first comparison result and described second comparison result,
To the 3rd comparison result；
First judging unit, is used for judging whether described global alignment result meets first pre-conditioned, obtains
First judged result；
Determine unit, for representing that described global alignment result meets first and presets when described first judged result
During condition, determine that described characteristic information matches with existing characteristic information.
Electronic equipment the most according to claim 8, it is characterised in that described arithmetic element, specifically
Including:
First operator unit, for using the first parameter and the first comparison result to carry out computing, obtains first
Operation result；
Second operator unit, for using the second parameter and the second comparison result to carry out computing, obtains second
Operation result；
Summation subelement, for carrying out summation behaviour to described first operation result and described second operation result
Make.
Electronic equipment the most according to claim 9, it is characterised in that described first operator unit,
Specifically include:
First weighting subelement, for using the first weight to be weighted with the first comparison result, obtains first
Weighted results；
Described second operator unit, specifically includes:
Second weighting subelement, for using the second weight to be weighted with the second comparison result, obtains second
Weighted results；
Described summation subelement, specifically includes:
Weighted results summation subelement, for carrying out with described second weighted results described first weighted results
Sum operation, obtains weighted sum result；
Accordingly, described first judging unit, specifically include:
First judgment sub-unit, is used for judging whether described weighted sum result presets threshold more than or equal to first
Value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510144573.6A CN106156784B (en) | 2015-03-30 | 2015-03-30 | A kind of characteristic recognition method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510144573.6A CN106156784B (en) | 2015-03-30 | 2015-03-30 | A kind of characteristic recognition method and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106156784A true CN106156784A (en) | 2016-11-23 |
| CN106156784B CN106156784B (en) | 2019-10-29 |
Family
ID=57340660
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510144573.6A Active CN106156784B (en) | 2015-03-30 | 2015-03-30 | A kind of characteristic recognition method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106156784B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111128144A (en) * | 2019-10-16 | 2020-05-08 | 国网浙江省电力有限公司金华供电公司 | Voice power grid dispatching system and method |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101630367A (en) * | 2009-07-31 | 2010-01-20 | 北京科技大学 | Rejection method for identifying handwritten character based on multiple classifiers |
| CN101859368A (en) * | 2009-04-09 | 2010-10-13 | 普诚科技股份有限公司 | Image identification device and method |
| CN102479328A (en) * | 2010-11-26 | 2012-05-30 | 神盾股份有限公司 | Identity verification device and method based on biological characteristics |
| CN102799899A (en) * | 2012-06-29 | 2012-11-28 | 北京理工大学 | Special audio event layered and generalized identification method based on SVM (Support Vector Machine) and GMM (Gaussian Mixture Model) |
| CN102982349A (en) * | 2012-11-09 | 2013-03-20 | 深圳市捷顺科技实业股份有限公司 | Image recognition method and device |
| CN103593648A (en) * | 2013-10-22 | 2014-02-19 | 上海交通大学 | Face recognition method for open environment |
| CN103853959A (en) * | 2012-12-05 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Permission control device and method |
| KR20140137168A (en) * | 2013-05-22 | 2014-12-02 | 재단법인대구경북과학기술원 | Image Recognition Calculating Apparatus and the Method |
| CN104392245A (en) * | 2014-12-15 | 2015-03-04 | 长春理工大学 | Multi-sensor fusion road surface type identification method and device |
-
2015
- 2015-03-30 CN CN201510144573.6A patent/CN106156784B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101859368A (en) * | 2009-04-09 | 2010-10-13 | 普诚科技股份有限公司 | Image identification device and method |
| CN101630367A (en) * | 2009-07-31 | 2010-01-20 | 北京科技大学 | Rejection method for identifying handwritten character based on multiple classifiers |
| CN102479328A (en) * | 2010-11-26 | 2012-05-30 | 神盾股份有限公司 | Identity verification device and method based on biological characteristics |
| CN102799899A (en) * | 2012-06-29 | 2012-11-28 | 北京理工大学 | Special audio event layered and generalized identification method based on SVM (Support Vector Machine) and GMM (Gaussian Mixture Model) |
| CN102982349A (en) * | 2012-11-09 | 2013-03-20 | 深圳市捷顺科技实业股份有限公司 | Image recognition method and device |
| CN103853959A (en) * | 2012-12-05 | 2014-06-11 | 腾讯科技（深圳）有限公司 | Permission control device and method |
| KR20140137168A (en) * | 2013-05-22 | 2014-12-02 | 재단법인대구경북과학기술원 | Image Recognition Calculating Apparatus and the Method |
| CN103593648A (en) * | 2013-10-22 | 2014-02-19 | 上海交通大学 | Face recognition method for open environment |
| CN104392245A (en) * | 2014-12-15 | 2015-03-04 | 长春理工大学 | Multi-sensor fusion road surface type identification method and device |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111128144A (en) * | 2019-10-16 | 2020-05-08 | 国网浙江省电力有限公司金华供电公司 | Voice power grid dispatching system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106156784B (en) | 2019-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7046481B2 (en) | Battery management method and battery management device | |
| CN106716297B (en) | A kind of method, apparatus and touch screen terminal of fingerprint recognition | |
| CN103533546B (en) | Implicit user verification and privacy protection method based on multi-dimensional behavior characteristics | |
| CN105975823A (en) | Verification method and apparatus used for distinguishing man and machine | |
| CN104991687B (en) | A kind of method and system for obtaining touch-screen equipment curve manipulation track | |
| CN110706261A (en) | Vehicle violation detection method and device, computer equipment and storage medium | |
| CN110852450B (en) | Method and device for identifying countermeasure sample to protect model security | |
| CN107256561A (en) | Method for tracking target and device | |
| CN109344257A (en) | Text emotion recognition methods and device, electronic equipment, storage medium | |
| CN106326913A (en) | Money laundering account determination method and device | |
| CN111191671B (en) | Electrical appliance waveform detection method, system, electronic equipment and storage medium | |
| US11721229B2 (en) | Question correction method, device, electronic equipment and storage medium for oral calculation questions | |
| CN107179831A (en) | Start method, device, storage medium and the terminal of application | |
| CN106663421B (en) | Sound recognition system and sound identification method | |
| CN109189892A (en) | A kind of recommended method and device based on article review | |
| CN106156784A (en) | A kind of characteristic recognition method and electronic equipment | |
| CN105809018A (en) | Sliding track verification method and terminal | |
| CN114325166A (en) | Method, device and equipment for evaluating unlocking and locking functions of vehicle | |
| CN101876985A (en) | WEB text sentiment theme recognizing method based on mixed model | |
| CN111976530B (en) | Method for controlling a safety system of a charging station for charging an electric vehicle | |
| Wangsuk et al. | Trajectory mining for keystroke dynamics authentication | |
| CN107077617A (en) | fingerprint extraction method and device | |
| CN105824998B (en) | A kind of development blasting construction intelligent design system and method | |
| CN115752543A (en) | Method for calibrating steering wheel hands-off detection function and electronic equipment | |
| CN109743200B (en) | Resource feature-based cloud computing platform computing task cost prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |