CN102722701B - Visual monitoring method and device in fingerprint collection process - Google Patents

Abstract

The invention discloses a visual monitoring method in a fingerprint collection process. The visual monitoring method comprises the following steps: obtaining an image of surrounds of a fingerprint collection device in the fingerprint collection process; extracting a hand area image from the image; converting the hand area image to a finger-separated hand contour curve; sequentially querying extreme points in each point in the finger-separated hand contour curve as the extreme points meet that the radial distance is an extremal value, arraying the extreme points according to querying sequence to generate an extreme point set, wherein the radial distance is the distance from the points of the hand contour curve to the centre of the palm; obtaining the extreme point set which is arranged according to the extreme point set by taking finger positions of a hand as an sequence and corresponds to finger tips and finger valleys; and determining finger positions of collected fingerprints according to the extreme point set corresponding to the finger tips and the finger valleys. According to the visual monitoring method, the finger positions of collected fingerprints are judged by analyzing the images of the collected fingerprints, and the influence of cheating behaviors on a system is effectively reduced.

Description

Vision monitoring method in fingerprint collecting process and equipment thereof

Technical field

The present invention relates to fingerprint recognition field, relate in particular to a kind of vision monitoring method and equipment thereof of the cheating for fingerprint recognition system.

Background technology

Due to uniqueness and the lifelong unchangeability of fingerprint, the method for carrying out authentication and identification by fingerprint image has had very long developing history.Along with auto Fingerprint Identification System is increasingly mature, at present, this technology is widely used in the use control of criminal investigation, entry-exit management, mancarried electronic aid, the gate control system of important place etc.

But, along with a large amount of successful Application of fingerprint identification technology, for the cheating of fingerprint recognition system, be also becoming increasingly rampant.According to the difference of application scenario, fingerprint recognition system can be divided into white list and blacklist two classes.White list system comprises the use control of mancarried electronic aid, the gate control system of important place etc.Cheat method for this type systematic is mainly the fingerprint of usurping validated user, makes pseudo-fingerprint to pretend to be validated user, fraud system.Blacklist system comprises responsive professional employee's previous conviction inspection, the investigation of the terrorist in entry-exit management etc.Blacklist fingerprint recognition system need to check that personnel to be checked are whether on blacklist.

Man on the blacklist person, tend to cheating and by system, found its identity to avoid.Because fingerprint database is very huge, in order to improve seek rate, often require person to be checked order in accordance with regulations by the fingerprint of each finger.As long as cribber upsets finger order, or presses palmmprint, just can reach the object of fraud system.Because the part of palmmprint is closely similar with fingerprint, and from fingerprint itself, be difficult to infer and refer to position, the fingerprint quality evaluation software of existing fingerprint recognition system cannot detect this cheating automatically.

At present, be mainly to rely on manpower to prevent this cheating for blacklist fingerprint recognition system, such as allowing staff carefully monitor whole fingerprint collecting process.But this solution has no idea to adapt to the application requirements day by day increasing, rely on manpower monitoring cost larger, and may occur slack and corrupt practice.

Therefore, in the urgent need to vision monitoring method and the equipment thereof in a kind of fingerprint collecting process, to detect the cheating for blacklist fingerprint recognition system.

Summary of the invention

One of technical matters to be solved by this invention is need to provide a kind of can detect for vision monitoring method and equipment thereof in the fingerprint collecting process of the cheating of blacklist fingerprint recognition system.

In order to solve the problems of the technologies described above, the invention provides the vision monitoring method in a kind of fingerprint collecting process, the method comprises: step 10, obtain fingerprint collecting device image around in fingerprint collecting process; Step 20 is extracted hand region image from described image; Step 30, is converted to the separated hand contour curve of finger by described hand region image; Step 40, inquire about successively in the each point of the separated hand contour curve of described finger and meet the extreme point that radial distance is extreme value, described extreme point is arranged to generate extreme point set according to the sequencing inquiring, and wherein said radial distance is that point on described hand contour curve is to the distance in the centre of the palm; Step 50, based on described extreme point set obtain take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, wherein said finger paddy for finger with point between the end in gap; Step 60, based on described and finger tip with refer to that extreme point set that paddy is corresponding determines the finger position under the fingerprint collecting,

Wherein, in described step 50, the finger position that obtains as follows take hand as tactic, with finger tip with refer to the extreme point set that paddy is corresponding:

Step 501, wrist extreme point corresponding with wrist location in described extreme point set is deleted, the extreme point head and the tail of deleting after all wrist extreme points are adjacent, extreme point using the non-wrist extreme point adjacent with described wrist extreme point as sequence first, other extreme points sort to redefine extreme point set successively;

Step 511, judgement is each maximum point and each minimum point alternative arrangement whether in described extreme point set, if the determination result is NO, enters step 512;

Step 512, in described extreme point set described in each extreme point one by one the variance of air line distance of all extreme points and the variance of index distance in the air line distance variance based on other extreme points and index distance variance and described extreme point set, to obtain air line distance variance changing value and the index distance variance changing value of extreme point described in each, wherein said index distance is along the path of described hand contour curve between extreme point;

Step 513, in described extreme point set described in each extreme point one by one the air line distance variance changing value based on extreme point described in each and index distance variance changing value, to obtain the energy value of extreme point described in each;

Step 514, the extreme point of described energy value maximum is deleted to redefine extreme point set from extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, using this extreme point set as take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, end operation, otherwise return to described step 511, the described the 3rd is pre-conditionedly comprised of 5 maximum points and 4 minimum points for described extreme point set, and maximum point and minimum point are arranged alternately with each other.

Vision monitoring method according to a further aspect of the invention, in described step 20, utilizes background subtraction method from described image, to remove background area image to obtain foreground region image; According to tone, saturation degree and brightness, from described foreground region image, remove shadow region to obtain hand region image.

Vision monitoring method according to a further aspect of the invention, in described step 30, also comprise: detect in described hand region image whether have the finger closing up, if do not exist, directly described hand region image is converted to the separated hand contour curve of finger.

Vision monitoring method according to a further aspect of the invention, by detect in described hand region image, whether exist gray-scale value lower than the straight line of the average gray value of described hand region image, detect whether there is the finger closing up.

Vision monitoring method according to a further aspect of the invention, in described step 30, if while there is the finger closing up, carry out following steps: step 311, is converted to the unsegregated hand contour curve of finger by described hand region image; Step 312, inquires about successively in the each point of the unsegregated hand contour curve of described finger and meets the first pre-conditioned point, to obtain the set of webs candidate starting point; Step 313 is selected to meet the second pre-conditioned webs starting point and is determined the webs between the finger closing up from the set of described webs candidate starting point based on active contour model; Step 314, the webs based on definite obtain pointing separated hand contour curve, wherein, described first pre-conditioned for radial distance be that minimal value and radial distance are for being more than or equal to the first setting threshold; Described second is pre-conditionedly more than or equal to the second setting threshold and described line segment extends to the centre of the palm for the length of line segment, described line segment is according to active contour model, take each webs starting point as the formed line segment of initial point, wherein, described radial distance is that point on the unsegregated hand contour curve of described finger is to the distance in the centre of the palm.

Vision monitoring method according to a further aspect of the invention, in described step 512, described in each, extreme point obtains respectively air line distance variance changing value and the index distance variance changing value of extreme point described in each by following expression:

${\mathrm{\δ}}_i^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-n_i)\right),(i=1,...,N^t)$

${\mathrm{\Δ}}_i^t=\mathrm{var}\left(D\left(L^t\right)\right)-\mathrm{var}\left(D(L^t-n_i)\right),(i=1,...,N^t)$

Wherein,

, L ^tto delete extreme point extreme point set afterwards, n for the t time _ithe arbitrary extreme point in extreme point set, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+1index distance, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+2air line distance, var (d (L ^t)) and var (D (L ^t)) be respectively d (L ^t) and D (L ^t) variance, expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance,

expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance.

Vision monitoring method according to a further aspect of the invention, in described step 513, described in each, extreme point obtains respectively the energy value of extreme point described in each by following expression:

$e_1^t\left(i\right)=\frac{{\mathrm{\δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)}.$

$e_2^t\left(i\right)=\frac{{\mathrm{\Δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.$

$e\left(i\right)={\left(e_1^t\left(i\right)\right)}^2+{\left(e_2^t\left(i\right)\right)}^2.$

Wherein, min represents to make function to get minimum value, and max represents to make function to get maximal value, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time,

expression is by extreme point n _kd (L after deleting from extreme point set ^t) changing value of variance, expression is by extreme point n _kd (L after deleting from extreme point set ^t) changing value of variance, e (i) represents the energy value of extreme point.

Vision monitoring method according to a further aspect of the invention, if in step 511, in described extreme point set, each maximum point and each minimum point are alternative arrangements, enter step 522;

Step 522, in described extreme point set each adjacent extreme point to, by the variance of the index distance of all extreme points in the index distance variance to based on other extreme points and described extreme point set, to obtain the right index distance variance changing value of adjacent extreme point described in each, wherein said index distance is along the path of described hand contour curve between extreme point;

Step 523, in described extreme point set described in each extreme point one by one the variance of the air line distance of all extreme points in the air line distance variance based on other extreme points and described extreme point set, to obtain the air line distance variance changing value of extreme point described in each;

Step 524, described extreme point concentrate each adjacent extreme point to, by the variance of the air line distance of each the corresponding extreme point to based on the right index distance variance changing value of each adjacent extreme point and described each adjacent extreme point centering, to obtain the right energy value of adjacent extreme point described in each;

Step 525, by the adjacent extreme point of described energy value maximum to deleting from extreme point set to redefine extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, using this extreme point set as take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, end operation, otherwise return to described step 511.

Vision monitoring method according to a further aspect of the invention, in described step 522, each adjacent extreme point is to obtaining the right index distance variance changing value of adjacent extreme point described in each by following expression respectively:

${\mathrm{\δ}}_{k,k+1}^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-\{n_k,n_{k+1}\left\}\right)\right).$

Wherein,

l ^tto delete extreme point extreme point set afterwards, n for the t time _k, n _k+1the arbitrary adjacent extreme point pair in extreme point set, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+1index distance, var (d (L ^t)) be d (L ^t) variance,

represent that adjacent extreme point is to n _k, n _k+1from L ^td (L after middle deletion ^t) changing value of variance.

Vision monitoring method according to a further aspect of the invention, in described step 523, described in each, extreme point obtains the air line distance variance changing value of extreme point described in each by following expression:

${\mathrm{\Δ}}_i^t=\mathrm{var}\left(D\left(L^t\right)\right)-\mathrm{var}\left(D(L^t-n_i)\right),(i=1,...,N^t)$

Wherein,

l ^tto delete extreme point extreme point set afterwards, n for the t time _ithe arbitrary extreme point in extreme point set, ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+2air line distance, var (D (L ^t)) expression D (L ^t) variance,

expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance.

Vision monitoring method according to a further aspect of the invention, in described step 524, described in each, adjacent extreme point is to obtaining the right energy value of extreme point described in each by following expression respectively:

$e_1^t(i,i+1)=\frac{{\mathrm{\δ}}_{i,i+1}^t-{\min }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k.k+1}^t\right)}{{\max }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k,k+1}^t\right)-{\min }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k,k+1}^t\right)}.$

$e_2^t(i,i+1)=(e_2^t\left(i\right)+e_2^t(i+1))/2.$

$e(i,i+1)={\left(e_1^t(i,i+1)\right)}^2+{\left(e_2^t(i,i+1)\right)}^2$

Wherein, $e_2^t\left(i\right)=\frac{{\mathrm{\Δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.,$ $e_2^t(i+1)=\frac{{\mathrm{\Δ}}_{i+1}^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.$

Min represents to make function to get minimum value, and max represents to make function to get maximal value, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time,

represent that adjacent extreme point is to n _i, n _i+1from L ^td (L after middle deletion ^t) changing value of variance,

with represent extreme point n respectively _i, n _i+1extreme point n _iand n _i+1d (L after deleting from extreme point set respectively ^t) changing value of variance, e (i, i+1) represents that adjacent extreme point is to n _i, n _i+1energy value.

Vision monitoring method according to a further aspect of the invention, in described step 60, the extreme point that the extreme point by the 1st, 2 and 3 of the sequences based in 5 maximum points and the set of 4 minimum point extreme points and sequence are the 7th, 8 and 9 is determined the finger position under the fingerprint collecting.

According to a further aspect in the invention, also provide the vision monitoring equipment in a kind of fingerprint collecting process, the vision monitoring equipment in described fingerprint collecting process is determined the finger position under the fingerprint collecting by carrying out according to the method described in above-mentioned any one.

Compared with prior art, one or more embodiment of the present invention can have the following advantages by tool:

Vision monitoring method of the present invention has adopted after collecting texture and has obtained now fingerprint collecting equipment image around by camera, then by the analysis of this image being judged to whether the texture collecting derives from certain finger and judgement is the method for which finger.When being judged as non-finger, or be judged as wrong finger, the fingerprint that refusal acceptance is collected, this will reduce the impact of cheating on system effectively.Than now widely used manual monitoring means, this method can adapt to the application requirements day by day increasing.

Other features and advantages of the present invention will be set forth in the following description, and, partly from instructions, become apparent, or understand by implementing the present invention.Object of the present invention and other advantages can be realized and be obtained by specifically noted structure in instructions, claims and accompanying drawing.

Accompanying drawing explanation

Accompanying drawing is used to provide a further understanding of the present invention, and forms a part for instructions,, jointly for explaining the present invention, is not construed as limiting the invention with embodiments of the invention.In the accompanying drawings:

Fig. 1 is according to the schematic flow sheet of the vision monitoring method in the fingerprint collecting process of the embodiment of the present invention;

Fig. 2 is according to the schematic diagram that hand region is cut apart of the vision monitoring method in the fingerprint collecting process of the embodiment of the present invention;

The schematic diagram of Fig. 3 (a) radial distance minimum point on the hand contour images of hand;

On Fig. 3 (b) hand contour images, each point is to the radial distance schematic diagram in the centre of the palm;

Fig. 4 (a) is hand contour images I _cschematic diagram;

Fig. 4 (b) is from the initial curve synoptic diagram searching out of minimum point 7;

Fig. 4 (c) is from the initial curve synoptic diagram searching out of minimum point 8;

Fig. 4 (d) will close up finger along webs after separating hand contour images I ' _cschematic diagram;

The schematic diagram of the initial extreme point of Fig. 5 (a) on hand contour images;

The schematic diagram of the extreme point that Fig. 5 (b) finally stays on the profile diagram of hand.

Embodiment

Below with reference to drawings and Examples, describe embodiments of the present invention in detail, to the present invention, how application technology means solve technical matters whereby, and the implementation procedure of reaching technique effect can fully understand and implement according to this.It should be noted that, only otherwise form conflict, each embodiment in the present invention and each feature in each embodiment can mutually combine, and formed technical scheme is all within protection scope of the present invention.

In addition, in the step shown in the process flow diagram of accompanying drawing, can in the computer system such as one group of computer executable instructions, carry out, and, although there is shown logical order in flow process, but in some cases, can carry out shown or described step with the order being different from herein.

Fig. 1, according to the schematic flow sheet of the vision monitoring method in the fingerprint collecting process of the embodiment of the present invention, describes each step of the present embodiment in detail below with reference to Fig. 1.

Step S110, obtains fingerprint collecting device image I around in fingerprint collecting process, extracts hand region image I from this image _h(this step is referred to as hand area image is cut apart).Wherein image I is for gathering the present image obtaining by camera after fingerprint.

Particularly, after collecting texture, by camera, obtain now fingerprint collecting equipment image I around, by removing background area so that hand region is extracted from image I.In the present embodiment, because fingerprint collecting equipment is normally positioned on desktop, background is comparatively simple, and there is no what mobile object, therefore can adopt simple and quick background subtraction method to extract foreground region image I _f.

Except background simply, the embodiment of the present invention also requires the color of background and the human body complexion can not be too similar.

Fig. 2 is that step S110 specifically realizes by following steps according to the pictorial diagram that hand area image is cut apart of the vision monitoring method in the fingerprint collecting process of the embodiment of the present invention.

Step S1101, utilizes background subtraction method from image I, to remove background area image to obtain foreground region image I _f.

Particularly, utilize Gauss model to simulate in hsv color space each pixel of background area image, obtain model M.Then adopt background subtraction method, by departing from the excessive pixel of model M, be labeled as prospect, then use some noises of morphological operator filtering, obtain foreground region image I _f.

Step S1102, if at foreground region image I _fin containing shaded portion, according to tone, saturation degree and brightness, remove foreground region image I _fin dash area to obtain hand region image I _h.

The shadow region of setting about due to the impact in illumination is also foreground region image I _fa part, need to be by it from foreground area image I _fremoval obtains hand region image I _h.Shadow region is because hand has blocked a part of light, and cause the brightness step-down of some pixels of background area image, so can be according to this part pixel more similar to background area image aspect tone and saturation degree, and the lower feature of brightness ratio removes to obtain hand region image I by shadow region from foreground region image _h.

Step S120, detects hand region image I _hin whether there is the finger closing up.

Particularly, by detecting in hand region image, whether exist gray-scale value to detect whether there is the finger closing up lower than the straight line of the average gray value of hand region image.If there is the finger closing up, enter step S130, otherwise, there is not the finger closing up, directly hand region image be converted to the separated hand contour curve of finger and enter step S160.

It should be noted that, because the present embodiment relates to, in the marginal point of hand, refer to that paddy detects, and finger paddy between the finger closing up is not in hand region image I _hedge, therefore the finger closing up need to be distinguished the finger closing up the position along webs.Wherein, refer to that paddy refers to the terminal region in gap between finger and finger.

Due to when finger closes up, the gray scale in webs region of closing up finger is all lower than region around, and webs generally can be approximated to be straight line; And in the palm opening completely, regional gray scale difference is little, there is not such straight line.If therefore in hand region image I _hin such straight line detected, just think that it is the webs that close up finger.

In addition, webs are to close up between finger, so the point on webs is greater than finger width to the distance majority of background, and this is also one of restrictive condition of webs.

Step S130, is converted to the unsegregated hand contour curve of finger by hand region image.

For example, will in hand region image, exist the hand region image of the finger closing up to be converted to the unsegregated hand contour images of finger I as shown in Figure 4 (a) _c.

It should be noted that, webs are from close up finger paddy between the finger edge in one's hands that stretches out always.Owing to referring to that paddy is not easy to detect, therefore from marginal point, start inside searching.In order to accelerate searching process, the present embodiment has only been chosen some possible candidate points and has been operated from marginal point.

Step S140 finds the starting point of webs to determine the webs between the finger closing up from hand contour curve.

Particularly, first, in the each point of the unsegregated hand contour curve of inquiry finger, meet the first pre-conditioned point successively, to obtain the set of webs starting point.

Particularly, centre of the palm C is set as to the center of circle of maximum inscribed circle in hand region, each point on hand contour curve is called radial distance to the distance in the center of circle.Conventionally, the radial distance of the intersection point of webs and hand profile is minimum point, so the first pre-conditioned being set as is more than or equal to the first setting threshold t by radial distance _dand be minimizing as webs candidate starting point.The first setting threshold can be set as 1.5 to 1.7 times of inscribed circle radius, and the present embodiment is preferably by t _dvalue be set to 5/3 times of inscribed circle radius, wherein, incircle is for take the centre of the palm circle tangent with the edge of hand region except finger as the center of circle.

Fig. 3 (a) is the schematic diagram of radial distance minimum point on hand contour images, Fig. 3 (b) is that point is to the radial distance schematic diagram in the centre of the palm, from Fig. 3 (b), can find out, minimum point one has 9, but only has minimum point 7 and minimum point 8 to be greater than certain threshold value t to the radial distance in the centre of the palm is satisfied _drequirement, so minimum point 7 and minimum point 8 become the set of webs candidate starting point.

Then, based on active contour model (hereinafter to be referred as snake model), from the set of webs candidate starting point, select to meet the second pre-conditioned webs starting point and determine the webs between the finger closing up.

Particularly, after obtaining webs candidate starting point, utilize snake model to using gray-scale map as energy field, at surrounding's searching straight line of webs candidate starting point.The second pre-conditioned length that is set as line segment is more than or equal to the second setting threshold t _l, and to centre of the palm C along stretching, if meet this condition, think that this candidate point is a webs starting point, this straight line is as webs.The second setting threshold can be set to 1.1 to 1.3 times of inscribed circle radius, in this step, preferably, t _lbe set to 1.2 times of inscribed circle radius.

In addition, if having many straight lines that satisfy condition from same point, will using the minimum straight line of average gray value as webs.

Fig. 4 (a) is hand contour images I _cschematic diagram, the blue line in Fig. 4 (b) be from minimum point 7 set off in searchs to line, the blue line in Fig. 4 (c) be from minimum point 8 set off in searchs to line.Because the line curvature in Fig. 4 (b) is larger, so given up.And line in Fig. 4 (c) relatively approaches straight line and length surpasses threshold value, can be used as webs.

Step S150, the webs based on definite obtain pointing separated hand contour curve.

Particularly, the webs based on obtaining from step S140 can distinguish the finger closing up, concrete as shown in Fig. 4 (d), as Fig. 4 (d) will close up finger along webs after separating hand contour images I ' _cschematic diagram.

Step S160, inquire about in the each point of pointing separated hand contour curve and meet the extreme point that radial distance is extreme value successively, extreme point is arranged to generate extreme point set according to the sequencing inquiring, and wherein radial distance is that point on hand contour images is to the distance in the centre of the palm.

It should be noted that, the general radial distance of point that is positioned at pad of finger is all larger than the radial distance of point around, is maximum point, is positioned at and refers to that the point at paddy position is generally minimum point.Algorithm is that very big or minimum marginal point is all found out by radial distance, owing to being subject to the impact of broken edge and arm segment, may have unnecessary extreme point, needs to remove.

If the marginal point satisfying condition is less than 5 maximum points and 4 minimum points, illustrate that finger has the phenomenon of blocking mutually to exist, cannot accurately estimate to refer to position, algorithm will be refused the gesture of this class, and gives a warning to collector.Same, if the marginal point satisfying condition is too many, for example marginal point max-thresholds is set to 13 in the present embodiment, also illustrate that edge is very irregular, likely because of user, two heavy-handed EVAC (Evacuation Network Computer Model) that stack are caused, algorithm also will directly be refused the gesture of this class, the operation of not removing unnecessary extreme point.

In the present embodiment, from the intersection point of curve and fingerprint collecting equipment center line, start to travel through in the direction of the clock, find extreme point.The hand contour images of take in Fig. 4 (c) is example, on this image, detects extreme point.Testing result, as shown in Fig. 5 (a), has 18 extreme points, according to the sequencing sequence finding.

In addition, if there is no maximum point around at fingerprint collecting equipment, around explanation, there is no so finger tip, the texture namely collecting is not to derive from certain finger, and system will be refused this gesture.

Step S170, based on extreme point set obtain take hand finger position as tactic, with finger tip with refer to respectively corresponding extreme point set of paddy, its middle finger paddy for finger with point between the end in gap.

Specifically by following sub-step, obtain:

Step 1701, wrist extreme point corresponding with wrist location in extreme point set is deleted, the extreme point head and the tail of deleting after all wrist extreme points are adjacent, extreme point using the non-wrist extreme point adjacent with wrist extreme point as sequence first, other extreme points sort to redefine extreme point set successively.

The hand contour images of take in Fig. 4 (c) is example, on this image, detects extreme point.Testing result is as shown in Fig. 5 (a), and solid dot is maximum point, and hollow dots is minimum point.N ₇, n ₈, n ₉approach image border, above-mentioned extreme point is judged as the extreme point in wrist, does not therefore consider.In step, the point that preferably distance of distance Curve lower limb is less than to 50 pixels is judged to be the wrist extreme point that approaches image border.

In addition, owing to referring to that normally paddy should be adjacent with two finger tips (maximum point), after deleted, n ₆, n ₁₀do not meet such condition, can not refer to paddy, therefore can delete yet.

By n ₆, n ₇, n ₈, n ₉, n ₁₀after deletion, by n ₁₈with n ₁head and the tail adjacent, will with n ₁₀adjacent n ₁₁as the extreme point of sequence first, the extreme point set after redefining is (n ₁₁, n ₁₂, n ₁₃, n ₁₄, n ₁₅, n ₁₆, n ₁₇, n ₁₈, n ₁, n ₂, n ₃, n ₄, n ₅).

For removing unnecessary extreme point, the embodiment of the present invention has proposed an energy function of weighing extreme point quality, and the explanation quality that energy is higher is poorer, can select a most possible extreme point combination according to this point.The following describes the content of this algorithm.

Step 1702, judgement is each maximum point and each minimum point alternative arrangement whether in extreme point set, if the determination result is NO, enters step 1703.

Step 1703, in extreme point set each extreme point one by one the variance of air line distance of all extreme points and the variance of index distance in the air line distance variance based on other extreme points and index distance variance and extreme point set, to obtain air line distance variance changing value and the index distance variance changing value of each extreme point, wherein index distance is along the distance of hand contour curve between extreme point.

Step 1704, in extreme point set each extreme point one by one the air line distance variance changing value based on each extreme point and index distance variance changing value, to obtain the energy value of each extreme point.

Step 1705, the extreme point of energy value maximum is deleted to redefine extreme point set from extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, this extreme point set is to tactic with finger tip with refer to the extreme point set that paddy is corresponding as take the finger position of hand, end operation, otherwise return to step 1702.

Particularly, extreme point sequence is designated as

l ^tdelete extreme point extreme point sequence afterwards, n for the t time _i(i=1 ..., N ^t) be extreme point, N ^tit is the number of deleting extreme point in extreme point extreme point sequence afterwards for the t time.Define two extreme point n _iand n _jindex apart from d _i,jbe two points along the path distance of profile, _i,jthat two points are in image I _con air line distance.Definition $d\left(L^t\right)=$ $\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$ $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$ And var (d (L ^t)), and var (D (L ^t)) be respectively d (L ^t) and D (L ^t) variance.

For normal adjacent finger tip and finger paddy, each d _i,jmore close, for normal adjacent two finger tips or two finger paddy, each D _i,jall more close, var (d (L namely ^t)) and var (D (L ^t)) smaller.If there is a maximum point causing due to broken edge in the middle of two normal maximum points, d _i,jwill diminish, D _i,jalso can change, cause var (d (L ^t)) and var (D (L ^t)) increase.If appeared on arm D by maximum point _i,jwill become large, d _i,hvariation not sure, this will make var (D (L ^t)) increase.

Note ${\mathrm{\δ}}_i^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-n_i)\right),$ Be about to extreme point n _ifrom L ^td (L after middle removal ^t) variation of variance, in like manner, for air line distance, have ${\mathrm{\Δ}}_i^t=\mathrm{var}\left(D\left(L^t\right)\right)-\mathrm{var}(D(L^t-$ $n_i)).$

Extreme point n _ienergy can calculate by following expression:

$e_1^t\left(i\right)=\frac{{\mathrm{\δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\δ}}_k^t\right)}.$

$e_2^t\left(i\right)=\frac{{\mathrm{\Δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.$

$e\left(i\right)={\left(e_1^t\left(i\right)\right)}^2+{\left(e_2^t\left(i\right)\right)}^2.$

What energy was higher just illustrates that extreme point is to var (d (L ^t)) and var (D (L ^t)) impact larger, more unstable, should be preferentially deleted.In general, maximum point and minimum point are alternately to occur.If adjacent two are both maximum point or minimum point, high that of energy should be deleted so.

If the determination result is NO in step 1702, enter step 1713, each maximum point alternative arrangement adjacent with each minimum point namely, extreme point is deleted in couples by following steps.

Step 1713, in extreme point set each adjacent extreme point to, by the variance of the index distance of all extreme points in the index distance variance to based on other extreme points and extreme point set, to obtain the right index distance variance changing value of each adjacent extreme point, wherein index distance is along the path of hand contour curve between extreme point.

Step 1714, in extreme point set described in each extreme point one by one the variance of the air line distance of all extreme points in the air line distance variance based on other extreme points and extreme point set, to obtain the air line distance variance changing value of each extreme point;

Step 1715, extreme point concentrate each adjacent extreme point to, by the variance of the air line distance of each the corresponding extreme point to based on the right index distance variance changing value of each adjacent extreme point and each adjacent extreme point centering, to obtain the right energy value of each adjacent extreme point;

Step 1716, by the adjacent extreme point of energy value maximum to deleting from extreme point set to redefine extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, using this extreme point set as take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, end operation, otherwise return to described step 1702.

Particularly, extreme point is to n _i, n _i+1energy value can calculate by following expression:

${\mathrm{\δ}}_{k,k+1}^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-\{n_k,n_{k+1}\left\}\right)\right).$

$e_1^t(i,i+1)=\frac{{\mathrm{\δ}}_{i,i+1}^t-{\min }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k.k+1}^t\right)}{{\max }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k,k+1}^t\right)-{\min }_{k=1}^{N^t-1}\left({\mathrm{\δ}}_{k,k+1}^t\right)}.$

$e_2^t(i,i+1)=(e_2^t\left(i\right)+e_2^t(i+1))/2.$

$e(i,i+1)={\left(e_1^t(i,i+1)\right)}^2+{\left(e_2^t(i,i+1)\right)}^2.$

Wherein, $e_2^t\left(i\right)=\frac{{\mathrm{\Δ}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.,$ $e_2^t(i+1)=\frac{{\mathrm{\Δ}}_{i+1}^t-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\Δ}}_k^t\right)}.$

Min represents to make function to get minimum value, and max represents to make function to get maximal value, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time,

represent that adjacent extreme point is to n _i, n _i+1from L ^td (L after middle deletion ^t) changing value of variance,

with

represent extreme point n respectively _i, n _i+1extreme point n _iand n _i+1d (L after deleting from extreme point set respectively ^t) changing value of variance, e (i, i+1) represents that adjacent extreme point is to n _i, n _i+1energy value.

At this moment, the extreme point that energy is high is to just should be preferentially deleted.According to such algorithm, unnecessary extreme point can be deleted, until extreme point set meets the 3rd pre-conditioned rear end operation, in the present embodiment, preferably, the 3rd is pre-conditionedly comprised of 5 maximum points and 4 minimum points for extreme point set, and maximum point and minimum point are arranged alternately with each other, maximum point and minimum point be corresponding finger tip and finger paddy respectively.

For example, the hand contour images of take in Fig. 4 (c) is example, on this image, detects extreme point.Testing result is as shown in Fig. 5 (a), and solid dot is maximum point, and hollow dots is minimum point.Therefore by weighing the energy function of extreme point quality, delete extreme point.

Due to (n ₁₁, n ₁₂, n ₁₃, n ₁₄, n ₁₅, n ₁₆, n ₁₇, n ₁₈, n ₁, n ₂, n ₃, n ₄, n ₅) middle maximum point and alternately appearance of minimum point, while therefore deleting extreme point, to put right form, delete.Table 1 is according to the energy value calculating about the right energy function of extreme point.Table 2 is for deleting the right energy value of other extreme points after some extreme point.Due to n ₁be from the nearest maximum point of detecting device, algorithm thinks that it, can not be deleted corresponding to the finger tip of fingerprint source finger. therefore not calculation level to n ₁₈, n ₁and n ₁, n ₂energy value.

Table 1

Table 2

According to shown in table 1, by the highest point of energy value to n ₁₄, n ₁₅delete, according to shown in table 2, by the highest n of energy value ₄, n ₅delete the extreme point (n obtaining ₁₁, n ₁₂, n ₁₃, n ₁₆, n ₁₇, n ₁₈, n ₁, n ₂, n ₃), they are corresponding finger tip and finger paddy respectively, as shown in Fig. 5 (b).

S180, the finger position under the fingerprint collecting is determined in the extreme point set based on finger tip and finger paddy.

Particularly, the extreme point of the 7th, 8 and 9 of the extreme point by the 1st, 2 and 3 of the sequences based in 5 maximum points and the set of 4 minimum point extreme points and sequences is determined the finger position under the fingerprint collecting.

More specifically, due to from finger tip corresponding to the nearest maximum point of the detecting device source of fingerprint just, therefore by the position of maximum point, just can determine and refer to.For example, in Fig. 5 (b), with the extreme point n on arm ₇, n ₈, n ₉for boundary divides, if maximum point is arranged from left to right, n ₁the maximum point sequence { n staying ₁₁, n ₁₃, n ₁₇, n ₁, n ₃in the 4th.Like this, n ₁corresponding finger tip is likely left index finger or right ring finger.

As for right-hand man's judgement, by having utilized forefinger to distinguish to the air line distance of the finger paddy between it and thumb the feature than the significant length of thumb more a lot.

If left hand, forefinger right several second and first maximum points corresponding to thumb difference.According to right-hand man's symmetry, the forefinger of the right hand left several second and first maximum points corresponding to thumb difference.

Extreme point can be arranged from left to right, i.e. L={n ₁..., n ₉.Note d _ieft=D _2,3-D _1,2, d _right=D _7,8-D _8,9, ρ=d _left-d _right, so, if there is ρ > 0, explanation is the right hand, otherwise is left hand.

It should be noted that, this determination methods requires too bending of finger, otherwise be projected on two dimensional image, has no idea to guarantee that such feature also exists.Therefore, if d _leftand d _rightbe not positive number entirely, illustrate that finger is crooked, cannot normally judge, such posture will be rejected.

Extreme point result (n about Fig. 5 (b) ₁₁, n ₁₂, n ₁₃, n ₁₆, n ₁₇, n ₁₈, n ₁, n ₂, n ₃), by calculating ρ < 0, so be judged as left hand.From the nearest maximum point of detecting device, be n ₁so final definite finger position is left index finger.Can send instruction to receive this fingerprint to detecting device as required afterwards, and this fingerprint is increased to label information.

In addition, the invention still further relates to the vision monitoring equipment in a kind of fingerprint collecting process, the vision monitoring equipment of this fingerprint collecting process is determined the finger position under the fingerprint collecting by carrying out above operation.

The embodiment of the present invention is obtained now fingerprint capturer image around by camera after collecting texture, again by the analysis and understanding of this image being judged to whether the texture collecting derives from certain finger, and be which finger, effectively reduce the impact of cheating on system.Than now widely used manual monitoring means, this method can adapt to the application requirements day by day increasing.

Those skilled in the art should be understood that, above-mentioned each module of the present invention or each step can realize with general calculation element, they can concentrate on single calculation element, or be distributed on the network that a plurality of calculation elements form, alternatively, they can be realized with the executable program code of calculation element, thereby, they can be stored in memory storage and be carried out by calculation element, or they are made into respectively to each integrated circuit modules, or a plurality of modules in them or step are made into single integrated circuit module realize.Like this, the present invention is not restricted to any specific hardware and software combination.

Although the disclosed embodiment of the present invention as above, the embodiment that described content just adopts for the ease of understanding the present invention, not in order to limit the present invention.Technician in any the technical field of the invention; do not departing under the prerequisite of the disclosed spirit and scope of the present invention; can do any modification and variation what implement in form and in details; but scope of patent protection of the present invention, still must be as the criterion with the scope that appending claims was defined.

Claims (12)

1. the vision monitoring method in fingerprint collecting process, is characterized in that, comprising:

Step 10, obtains fingerprint collecting device image around in fingerprint collecting process;

Step 20 is extracted hand region image from described image;

Step 30, is converted to the separated hand contour curve of finger by described hand region image;

Step 40, inquire about successively in the each point of the separated hand contour curve of described finger and meet the extreme point that radial distance is extreme value, described extreme point is arranged to generate extreme point set according to the sequencing inquiring, and wherein said radial distance is that point on described hand contour curve is to the distance in the centre of the palm;

Step 50, based on described extreme point set obtain take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, wherein said finger paddy for finger with point between the end in gap;

Step 60, based on described and finger tip with refer to that extreme point set that paddy is corresponding determines the finger position under the fingerprint collecting,

Wherein, in described step 50, the finger position that obtains as follows take hand as tactic, with finger tip with refer to the extreme point set that paddy is corresponding:

Step 501, wrist extreme point corresponding with wrist location in described extreme point set is deleted, the extreme point head and the tail of deleting after all wrist extreme points are adjacent, extreme point using the non-wrist extreme point adjacent with described wrist extreme point as sequence first, other extreme points sort to redefine extreme point set successively;

Step 511, judgement is each maximum point and each minimum point alternative arrangement whether in described extreme point set, if the determination result is NO, enters step 512;

Step 512, in described extreme point set described in each extreme point one by one the variance of air line distance of all extreme points and the variance of index distance in the air line distance variance based on other extreme points and index distance variance and described extreme point set, to obtain air line distance variance changing value and the index distance variance changing value of extreme point described in each, wherein said index distance is along the path of described hand contour curve between extreme point;

Step 513, in described extreme point set described in each extreme point one by one the air line distance variance changing value based on extreme point described in each and index distance variance changing value, to obtain the energy value of extreme point described in each;

Step 514, the extreme point of described energy value maximum is deleted to redefine extreme point set from extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, using this extreme point set as take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, end operation, otherwise return to described step 511, the described the 3rd is pre-conditionedly comprised of 5 maximum points and 4 minimum points for described extreme point set, and maximum point and minimum point are arranged alternately with each other.

2. method according to claim 1, is characterized in that, in described step 20,

Utilize background subtraction method from described image, to remove background area image to obtain foreground region image;

According to tone, saturation degree and brightness, from described foreground region image, remove shadow region to obtain hand region image.

3. method according to claim 1, is characterized in that, in described step 30, also comprises:

Detect in described hand region image whether have the finger closing up, if do not exist, directly described hand region image is converted to the separated hand contour curve of finger.

4. method according to claim 3, is characterized in that,

By detect in described hand region image, whether exist gray-scale value lower than the straight line of the average gray value of described hand region image, detect whether there is the finger closing up.

5. according to the method described in claim 3 or claim 4 any one, it is characterized in that, in described step 30, if while there is the finger closing up, carry out following steps:

Step 311, is converted to the unsegregated hand contour curve of finger by described hand region image;

Step 312, inquires about successively in the each point of the unsegregated hand contour curve of described finger and meets the first pre-conditioned point, to obtain the set of webs candidate starting point;

Step 313 is selected to meet the second pre-conditioned webs starting point and is determined the webs between the finger closing up from the set of described webs candidate starting point based on active contour model;

Step 314, the webs based on definite obtain pointing separated hand contour curve,

Wherein, described first pre-conditioned for radial distance be that minimal value and radial distance are for being more than or equal to the first setting threshold; Described second is pre-conditionedly more than or equal to the second setting threshold and described line segment extends to the centre of the palm for the length of line segment, described line segment is according to active contour model, take each webs starting point as the formed line segment of initial point, wherein, described radial distance is that point on the unsegregated hand contour curve of described finger is to the distance in the centre of the palm.

6. method according to claim 1, is characterized in that, in described step 512,

Described in each, extreme point obtains respectively air line distance variance changing value and the index distance variance changing value of extreme point described in each by following expression:

${\mathrm{\&delta;}}_i^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-n_i)\right),(i=1,...,N^t)$

${\mathrm{\&Delta;}}_i^t=\mathrm{var}\left(D\left(L^t\right)\right)-\mathrm{var}\left(D(L^t-n_i)\right),(i=1,...,N^t)$

Wherein, , L ^tto delete extreme point extreme point set afterwards, n for the t time _ithe arbitrary extreme point in extreme point set, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+1index distance, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+2air line distance, var (d (L ^t)) and var (D (L ^t)) be respectively d (L ^t) and D (L ^t) variance,

expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance,

expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance.

7. method according to claim 1, is characterized in that, in described step 513,

Described in each, extreme point obtains respectively the energy value of extreme point described in each by following expression:

$e_1^t\left(i\right)=\frac{{\mathrm{\&delta;}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\&delta;}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\&delta;}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\&delta;}}_k^t\right)}.$

$e_2^t\left(i\right)=\frac{{\mathrm{\&Delta;}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}.$

$e\left(i\right)={\left(e_1^t\left(i\right)\right)}^2+{\left(e_2^t\left(i\right)\right)}^2.$

Wherein, min represents to make function to get minimum value, and max represents to make function to get maximal value, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time,

expression is by extreme point n _kd (L after deleting from extreme point set ^t) changing value of variance,

expression is by extreme point n _kd (L after deleting from extreme point set ^t) changing value of variance, e (i) represents the energy value of extreme point.

8. method according to claim 1, is characterized in that, if in step 511, in described extreme point set, each maximum point and each minimum point are alternative arrangements, enter step 522;

Step 522, in described extreme point set each adjacent extreme point to, by the variance of the index distance of all extreme points in the index distance variance to based on other extreme points and described extreme point set, to obtain the right index distance variance changing value of adjacent extreme point described in each, wherein said index distance is along the path of described hand contour curve between extreme point;

Step 523, in described extreme point set described in each extreme point one by one the variance of the air line distance of all extreme points in the air line distance variance based on other extreme points and described extreme point set, to obtain the air line distance variance changing value of extreme point described in each;

Step 524, described extreme point concentrate each adjacent extreme point to, by the variance of the air line distance of each the corresponding extreme point to based on the right index distance variance changing value of each adjacent extreme point and described each adjacent extreme point centering, to obtain the right energy value of adjacent extreme point described in each;

Step 525, by the adjacent extreme point of described energy value maximum to deleting from extreme point set to redefine extreme point set, if it is the 3rd pre-conditioned that extreme point set meets, using this extreme point set as take hand finger position as tactic, with finger tip with refer to the extreme point set that paddy is corresponding, end operation, otherwise return to described step 511.

9. method according to claim 8, is characterized in that, in described step 522,

Each adjacent extreme point is to obtaining the right index distance variance changing value of adjacent extreme point described in each by following expression respectively:

${\mathrm{\&delta;}}_{k,k+1}^t=\mathrm{var}\left(d\left(L^t\right)\right)-\mathrm{var}\left(d(L^t-\{n_k,n_{k+1}\left\}\right)\right).$

Wherein,

l ^tto delete extreme point extreme point set afterwards, n for the t time _k, n _k+1the arbitrary adjacent extreme point pair in extreme point set, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$ represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+1index distance, var (d (L ^t)) be d (L ^t) variance,

represent that adjacent extreme point is to n _k, n _k+1from L ^td (L after middle deletion ^t) changing value of variance.

10. method according to claim 8, is characterized in that, in described step 523,

Described in each, extreme point obtains the air line distance variance changing value of extreme point described in each by following expression:

${\mathrm{\&Delta;}}_i^t=\mathrm{var}\left(D\left(L^t\right)\right)-\mathrm{var}\left(D(L^t-n_i)\right),(i=1,...,N^t)$

Wherein,

l ^tto delete extreme point extreme point set afterwards, n for the t time _ithe arbitrary extreme point in extreme point set, ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

represent to delete for the t time n in extreme point extreme point set afterwards _iand n _i+2air line distance, var (D (L ^t)) expression D (L ^t) variance,

expression is by extreme point n _ifrom L ^td (L after middle deletion ^t) changing value of variance.

11. methods according to claim 8, is characterized in that, in described step 524,

Described in each, adjacent extreme point is to obtaining the right energy value of extreme point described in each by following expression respectively:

$e_1^t(i,i+1)=\frac{{\mathrm{\&delta;}}_{i,i+1}^t-{\min }_{k=1}^{N^t-1}\left({\mathrm{\&delta;}}_{k.k+1}^t\right)}{{\max }_{k=1}^{N^t-1}\left({\mathrm{\&delta;}}_{k,k+1}^t\right)-{\min }_{k=1}^{N^t-1}\left({\mathrm{\&delta;}}_{k,k+1}^t\right)}.$

$e_2^t(i,i+1)=(e_2^t\left(i\right)+e_2^t(i+1))/2.$

$e(i,i+1)={\left(e_1^t(i,i+1)\right)}^2+{\left(e_2^t(i,i+1)\right)}^2$

Wherein, $e_2^t\left(i\right)=\frac{{\mathrm{\&Delta;}}_i^t-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}.,$ $e_2^t(i+1)=\frac{{\mathrm{\&Delta;}}_{i+1}^t-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}{{\max }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)-{\min }_{k=1}^{N^t}\left({\mathrm{\&Delta;}}_k^t\right)}.$

Min represents to make function to get minimum value, and max represents to make function to get maximal value, N ^tthe number of deleting extreme point in extreme point extreme point set afterwards for the t time,

represent that adjacent extreme point is to n _i, n _i+1from L ^td (L after middle deletion ^t) changing value of variance,

with

represent extreme point n respectively _i, n _i+1extreme point n _iand n _i+1d (L after deleting from extreme point set respectively ^t) changing value of variance, e (i, i+1) represents that adjacent extreme point is to n _i, n _i+1energy value.

12. methods according to claim 1, is characterized in that, in described step 60,

The extreme point that extreme point by the 1st, 2 and 3 of the sequences based in 5 maximum points and the set of 4 minimum point extreme points and sequence are the 7th, 8 and 9 is determined the finger position under the fingerprint collecting.

Images