CN104228767A - Palm print authentication-based car starting method - Google Patents

Abstract

The invention relates to the field of cars, in particular to a palm print authentication-based car starting method. The palm print authentication-based car starting method provided by the invention comprises the following steps: 1, performing collection on palm prints of a hand by using a palm print collection instrument, then, transmitting collected palm print information to a palm print processing and recognition unit, and performing preprocessing on the palm print information by the palm print processing and recognition unit; 2, performing classification authentication on the palm print information by using a field feature line segment classifier. The palm print authentication-based car starting method disclosed by the invention has the beneficial effects that discriminant authentication is performed on the palm print information by using comparatively popular nearest feature line classification, and thus, a car owner can realize car starting only by inputting the own palm print information and passing authentication, and other people cannot start the car; the problems of extension inaccuracy and computational complexity of a nearest feature line are also improved.

Description

A kind of automobile starting method based on palmprint authentication

Technical field

The present invention relates to automotive field, particularly relate to a kind of automobile starting method based on palmprint authentication.

Background technology

Vehicle startup system is a subsystem very crucial in the middle of current onboard system, receives very large concern in a lot of car manufactures and R&D institution.Vehicle startup system is mainly used in antitheft and convenient driving.

Automobile starting technology main at present mainly contains: enter start up system by an automobile key by automobile starting or based on the automobile no-key of RFID.Based on the vehicle startup system of key, often because leave behind key, or key can be lost and allows chaufeur have a headache very much.Although the automobile no-key start up system based on RFID only needs band mobile phone just passable, avoid the multiple carry-on object of key handset, but mobile phone is also likely omitted or is lost.

Segregator occupies very important position in pattern recognition system.Nearest neighbor classifier (NN) and nearest center segregator (NM) are two more famous segregators.As the expansion of nearest neighbor classifier, nearest feature line (NFL) segregator is suggested.Although nearest feature line segregator successfully improves classification performance relative to nearest neighbor classifier, NFL still comes with some shortcomings, and such as, it exists the coarse problem of expansion and larger computation complexity problem.

In order to solve computation complexity problem, some nearest feature line segregators improved are suggested, such as nearest neighbor line classifier, nearest mid point segregator, based on the nearest neighbor classifier etc. at center.In order to improve expansion inaccuracy problem, some segregators strengthened are suggested, such as nearest eigencenter segregator, the nearest feature line segregator of restriction, the shortest Eigenvector segregator etc.

Summary of the invention

For the defect existed in prior art or deficiency, technical matters to be solved by this invention is: provide a kind of automobile starting method based on palmprint authentication, this method can realize the startup of automobile in the situation (or other authenticating items) not with key, chaufeur is thoroughly broken away from and with the worry of a lot of article, will also improve expansion inaccuracy and the computation complexity problem of nearest feature line.

The technical scheme that the present invention takes, for providing a kind of automobile starting method based on palmprint authentication, comprises the following steps

Step 1: use the palmmprint of palmmprint acquisition instrument to staff to gather, then the palmprint information be collected is passed to palmmprint process and recognizer, palmmprint process and recognizer carry out pretreatment to palmprint information;

Step 2: use the Eigenvector segregator in field to carry out classification certification to palmprint information; The method of calculating of the Eigenvector segregator in described field is as follows:

Suppose that each class has two samples at least, defining a recurrence unique point tolerance is:

$d_{\mathrm{NFLR}}(y,y_i^c{y}_j^c)=\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|---\left(1\right)$

In formula: with represent two master samples, y represents test sample book, represent and return unique point, can be calculated as:

$y_r^{\mathrm{ij},c}={\mathrm{Aw}}_{\mathrm{ij}}^c---\left(2\right)$

In formula: be linear regression coeffficient, linear regression coeffficient can by formula (3) solution square

Difference obtains:

$w_{\mathrm{ij}}^c={\left(A^{T}A\right)}^{-1}{A}^{T}y---\left(3\right)$

In formula: if A ^ta is unusual, by A ^ta+0.01I replaces A ^ta, can return unique point by formula (1) and formula (2) be expressed as:

$y_r^{\mathrm{ij},c}=A{\left(A^{T}A\right)}^{-1}{A}^{T}y=p_{\mathrm{ij}}^{c}y---\left(4\right)$

Calculate the Eigenvector distance of neighborhood: first check two master samples with whether in the hypersphere of definition, then carry out the radius calculation of Eigenvector in field, radius is the aviation value that test sample book arrives all sample distances of all classes, and it is expressed as:

$d_{\mathrm{avg}}=\frac{{\mathrm{\Σ}}_{c=1}^M{\mathrm{\Σ}}_{i=1}^{N_c}\left|\right|y-y_i^c\left|\right|}{{\mathrm{\Σ}}_{c=1}^m{N}_c}---\left(5\right)$

In formula: y represents test sample book, also represent the center of the hypersphere of definition;

The Eigenvector distance of neighborhood is expressed as:

$d_{\mathrm{NFLS}}(y,\stackrel{\‾}{y_i^c{y}_j^c})=\underset{\min (l_i^c,l_j^c),y\∈{\mathrm{\Θ}}_3^{\mathrm{II}}}{\overset{\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|,y\∈{\mathrm{\Θ}}_1^{\mathrm{II}}}{\frac{\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|+\min (l_i^c,l_j^c)}{2},y\∈{\mathrm{\Θ}}_2^{\mathrm{II}}}}---\left(6\right)$

In formula: refer to with situation all in hypersphere; refer to with only has a situation in hypersphere; refer to with situation all outside hypersphere.

As a further improvement on the present invention, two samples in described step 2 with all in hypersphere, so Eigenvector distance in field equal the recurrence unique point distance of definition in formula (1).

As a further improvement on the present invention, two samples in described step 2 with only has one in hypersphere, so the Eigenvector distance in field equal d ₁and d ₂aviation value, wherein d ₁equal the recurrence unique point distance of definition in formula (1), d ₂refer to smaller that in the spacing of test sample book and two master samples.

As a further improvement on the present invention, two samples in step 2 are stated described in with all outside hypersphere, so Eigenvector distance in field equal smaller that in the spacing of test sample book and two master samples.

The invention has the beneficial effects as follows: 1. inside automobile, add an easy palmmprint acquisition instrument, substantially can not affect design original in automobile, retain traditional attractive in appearance.2. present invention uses popular nearest tagsort and differentiation certification is carried out to palmprint information.3., because everyone the different volume in palmprint information room, so car owner only needs to input the palmprint information of oneself and just can realize automobile starting by certification, and other people can not start automobile.

Accompanying drawing explanation

Fig. 1 is that the Eigenvector of the automobile starting method neighborhood that the present invention is based on palmprint authentication measures the situation schematic diagram of two master samples all in hypersphere;

Fig. 2 is that the Eigenvector of the automobile starting method neighborhood that the present invention is based on palmprint authentication is measured two master samples and only had a situation schematic diagram in hypersphere;

Fig. 3 is that the Eigenvector of the automobile starting method neighborhood that the present invention is based on palmprint authentication measures the situation schematic diagram of two master samples all outside hypersphere.

Detailed description of the invention

Illustrate below in conjunction with accompanying drawing and detailed description of the invention the present invention is further described.

The present invention is based on the vehicle startup system of palmprint authentication by adding an easy palmmprint acquisition instrument inside automobile, thus by gathering the palmmprint of staff, this Measures compare simply can not be made troubles to chaufeur.After having gathered heart palmprint information, this new vehicle startup system can use the Eigenvector segregator of the neighborhood of proposition to carry out palmprint information certification, to judge whether being that car owner is at startup automobile.

Chaufeur uses the palmmprint of an easy palmmprint acquisition instrument to staff to gather, and then the palmprint information be collected is passed to palmmprint process and recognizer.After palmmprint process and recognizer receive the palmmprint signal that collector sheet transmits, its can carry out some pretreatments to palmmprint signal, is then that the Eigenvector segregator of the neighborhood proposed with us carries out classification certification to it.

As shown in Figure 1 to Figure 3, in order to improve expansion inaccuracy and the computation complexity problem of nearest feature line, we propose the Eigenvector segregator of neighborhood.First a new distance metric is suggested, and it is recalled and returns unique point to measure.With nearest eigencenter segregator, the nearest feature line segregator of restriction, the shortest Eigenvector segregator is similar, returns characteristic measure and supposes that each class has two samples at least.Return unique point tolerance to be defined as

$d_{\mathrm{NFLR}}(y,y_i^c,y_j^c)=\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|---\left(1\right)$

Wherein return unique point can by calculating by

$y_r^{\mathrm{ij},c}={\mathrm{Aw}}_{\mathrm{ij}}^c---\left(2\right)$

Wherein be linear regression coeffficient, it can be obtained by solution least square error,

$w_{\mathrm{ij}}^c={\left(A^{T}A\right)}^{-1}{A}^{T}y---\left(3\right)$

If A ^ta is unusual, A ^ta will by A ^ta+0.01I replaces. and by formula (1) and (2), I can know recurrence unique point

$y_r^{\mathrm{ij},c}=A{\left(A^{T}A\right)}^{-1}{A}^{T}y=p_{\mathrm{ij}}^{c}y---\left(4\right)$

Wherein the unique point returned can be obtained by the linear combination of two master samples, if it is by due to other unique point. and we precalculate and store the calculated amount so returning unique point will be lower.

The nearest feature line section of neighborhood does not directly calculate the distance between test sample book and characteristic curve.The Eigenvector of neighborhood first checks two master samples with whether in the hypersphere of definition.As shown in Figure 1, the center of the hypersphere of definition is test sample y, and radius is the aviation value that test sample book arrives all sample distances of all classes, it can calculate by

$d_{\mathrm{avg}}=\frac{{\mathrm{\Σ}}_{c=1}^M{\mathrm{\Σ}}_{i=1}^{N_c}\left|\right|y-y_i^c\left|\right|}{{\mathrm{\Σ}}_{c=1}^m{N}_c}---\left(5\right)$

If with all in hypersphere, as shown in Figure 1, so Eigenvector distance of neighborhood, equal formula (1) if the recurrence unique point distance of middle definition. with only have one in hypersphere, as shown in Figure 2, so Eigenvector distance of neighborhood, equal d ₁and d ₂aviation value, wherein d ₁equal the recurrence unique point distance of definition in formula (1), d ₂if refer to smaller that in the spacing of test sample book and two master samples. with all outside hypersphere, as shown in Figure 3, so Eigenvector distance of neighborhood, equal smaller that in the spacing of test sample book and two master samples. in summary, the Eigenvector distance of neighborhood can be expressed as

$d_{\mathrm{NFLS}}(y,\stackrel{\‾}{y_i^c{y}_j^c})=\left\{\begin{array}{c}\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|,y\∈{\mathrm{\Θ}}_1^{\mathrm{II}}\\ \frac{\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|+\min (l_i^c,l_j^c)}{2},y\∈{\mathrm{\Θ}}_2^{\mathrm{II}}\\ \min (l_i^c,l_j^c),y\∈{\mathrm{\Θ}}_3^{\mathrm{II}}\end{array}\right.---\left(6\right)$

Wherein: refer to with situation all in hypersphere; refer to with only has a situation in hypersphere; refer to with situation all outside hypersphere.

Above content is in conjunction with concrete preferred implementation further description made for the present invention, can not assert that specific embodiment of the invention is confined to these explanations.For general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, some simple deduction or replace can also be made, all should be considered as belonging to protection scope of the present invention.

Claims (4)

1., based on an automobile starting method for palmprint authentication, it is characterized in that: comprise the following steps

Step 1: use the palmmprint of palmmprint acquisition instrument to staff to gather, then the palmprint information be collected is passed to palmmprint process and recognizer, palmmprint process and recognizer carry out pretreatment to palmprint information;

Step 2: use the Eigenvector segregator in field to carry out classification certification to palmprint information; The method of calculating of the Eigenvector segregator in described field is as follows:

Suppose that each class has two samples at least, defining a recurrence unique point tolerance is:

$d_{\mathrm{NFLR}}(y,y_i^c,y_j^c)=\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|---\left(1\right)$

In formula: with represent two master samples, y represents test sample book, represent and return unique point, can be calculated as:

$y_r^{\mathrm{ij},c}={\mathrm{Aw}}_{\mathrm{ij}}^c---\left(2\right)$

In formula: be linear regression coeffficient, linear regression coeffficient can by formula (3) solution square

Difference obtains:

$w_{\mathrm{ij}}^c={\left(A^{T}A\right)}^{-1}{A}^{T}y---\left(3\right)$

In formula: if A ^ta is unusual, by A ^ta+0.01I replaces A ^ta, by formula (1) and formula

(2) unique point can be returned be expressed as:

$y_r^{\mathrm{ij},c}=A{\left(A^{T}A\right)}^{-1}{A}^{T}y=p_{\mathrm{ij}}^{c}y---\left(4\right)$

Calculate the Eigenvector distance of neighborhood: first check two master samples with whether in the hypersphere of definition, then carry out the radius calculation of Eigenvector in field, radius is the aviation value that test sample book arrives all sample distances of all classes, and it is expressed as:

$d_{\mathrm{avg}}=\frac{{\mathrm{\Σ}}_{c=1}^M{\mathrm{\Σ}}_{i=1}^{N_c}\left|\right|y-y_i^c\left|\right|}{{\mathrm{\Σ}}_{c=1}^m{N}_c}---\left(5\right)$

In formula: y represents test sample book, also represent the center of the hypersphere of definition;

The Eigenvector distance of neighborhood is expressed as:

$d_{\mathrm{NFLS}}(y,\stackrel{\‾}{y_i^c{y}_j^c})=\underset{\min (l_i^c,l_j^c),y\∈{\mathrm{\Θ}}_3^{\mathrm{II}}}{\overset{\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|,y\∈{\mathrm{\Θ}}_1^{\mathrm{II}}}{\frac{\left|\right|y-y_r^{\mathrm{ij},c}\left|\right|+\min (l_i^c,l_j^c)}{2},y\∈{\mathrm{\Θ}}_2^{\mathrm{II}}}}---\left(6\right)$

In formula: refer to with situation all in hypersphere; refer to with only has a situation in hypersphere; refer to with situation all outside hypersphere.

2. the automobile starting method based on palmprint authentication according to claim 1, is characterized in that: two samples in described step 2 with all in hypersphere, so Eigenvector distance in field equal the recurrence unique point distance of definition in formula (1).

3. the automobile starting method based on palmprint authentication according to claim 1, is characterized in that: two samples in described step 2 with only has one in hypersphere, so the Eigenvector distance in field equal d ₁and d ₂aviation value, wherein d ₁equal the recurrence unique point distance of definition in formula (1), d ₂refer to smaller that in the spacing of test sample book and two master samples.

4. the automobile starting method based on palmprint authentication according to claim 1, is characterized in that: described in state two samples in step 2 with all outside hypersphere, so Eigenvector distance in field equal smaller that in the spacing of test sample book and two master samples.