CN102722701A - 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

Visual monitoring method and equipment in fingerprint acquisition process

Technical Field

The invention relates to the field of fingerprint identification, in particular to a visual monitoring method and equipment for cheating behaviors of a fingerprint identification system.

Background

Due to the uniqueness and lifetime invariance of fingerprints, methods for identity authentication and identification via fingerprint images have had a long history of development. With the increasing maturity of automatic fingerprint identification technology, at present, the technology is widely applied to criminal investigation, entry and exit management, use control of portable electronic equipment, access control systems of important places and the like.

However, with the great success of fingerprint identification technology, cheating behaviors aiming at fingerprint identification systems are rampant increasingly. Fingerprint recognition systems can be classified into white lists and black lists according to different application occasions. The white list system comprises the use control of the portable electronic equipment, an access control system of an important place and the like. The cheating method for the system mainly includes the steps of embezzlement of fingerprints of legal users, production of fake fingerprints to impersonate the legal users and cheating the system. Blacklist systems include employee crime record checking for sensitive professions, terrorist investigation in entry and exit management, etc. The blacklist fingerprint identification system needs to check whether the person to be checked is on the blacklist.

People on the blacklist tend to cheat in order to avoid their identity being checked out by the system. Because the fingerprint database is very large, in order to increase the search speed, the person to be searched is often required to follow the fingerprints of all the fingers according to a specified sequence. The cheater only needs to disturb the order of the fingers or press the palm print to achieve the purpose of deceiving the system. Because the local part of the palm print is very similar to the fingerprint, and the finger position is difficult to infer from the fingerprint, the fingerprint quality evaluation software of the existing fingerprint identification system cannot automatically detect the cheating behavior.

Currently, this cheating action against the blacklist fingerprint identification system is mainly prevented by manpower, such as having a worker carefully monitor the whole fingerprint collection process. However, this solution has no way to adapt to the increasing application requirements, is expensive depending on human monitoring, and may cause lacuna and fraud.

Therefore, there is a need for a visual monitoring method and apparatus for fingerprint collection to detect cheating actions on the blacklist fingerprint identification system.

Disclosure of Invention

One of the technical problems to be solved by the present invention is to provide a visual monitoring method and apparatus thereof in a fingerprint collection process, which can detect cheating behaviors aiming at a blacklist fingerprint identification system.

In order to solve the technical problem, the invention provides a visual monitoring method in a fingerprint acquisition process, which comprises the following steps: step 10, acquiring an image around the fingerprint acquisition equipment in the fingerprint acquisition process; step 20, extracting a hand region image from the image; step 30, converting the hand area image into a finger-separated hand contour curve; step 40, sequentially inquiring extreme points which satisfy that the radial distance is an extreme value in each point of the hand contour curve separated by the fingers, and arranging the extreme points according to the inquired sequence to generate an extreme point set, wherein the radial distance is the distance from the point on the hand contour curve to the center of the palm; step 50, acquiring an extreme point set which is arranged by taking the finger positions of the hand as a sequence and corresponds to the fingertips and the finger valleys on the basis of the extreme point set, wherein the finger valleys are the end parts of gaps between the fingers; and step 60, determining the finger position of the acquired fingerprint based on the extreme point set corresponding to the finger tip and the finger valley.

According to the visual monitoring method of another aspect of the present invention, in the step 20, a background region image is removed from the image by using a background difference method to obtain a foreground region image; and removing shadow areas from the foreground area image according to the hue, the saturation and the brightness to obtain a hand area image.

The visual monitoring method according to another aspect of the present invention, in the step 30, further includes: and detecting whether the gathered fingers exist in the hand area image, and if not, directly converting the hand area image into a finger-separated hand contour curve.

According to the visual monitoring method of another aspect of the present invention, the presence or absence of the closed fingers is detected by detecting whether or not a straight line having a lower gray value than the average gray value of the hand region image exists in the hand region image.

According to another aspect of the present invention, in the step 30, if there are closed fingers, the following steps are performed: step 311, converting the hand region image into a hand contour curve without separating fingers; step 312, sequentially inquiring points meeting a first preset condition in each point of the hand contour curve with the fingers not separated to obtain a finger gap candidate starting point set; 313, selecting a finger seam starting point meeting a second preset condition from the finger seam candidate starting point set based on the active contour model to determine the finger seam between the closed fingers; and step 314, obtaining a hand contour curve of the finger separation based on the determined finger gaps.

According to another aspect of the present invention, the first predetermined condition is that the radial distance is a minimum value and the radial distance is greater than or equal to a first predetermined threshold; the second preset condition is that the length of the line segment is greater than or equal to a second set threshold value, the line segment extends towards the palm center, the line segment is formed by taking the starting point of each finger seam as the starting point according to the active contour model, and the radial distance is the distance from the point on the hand contour curve with the fingers not separated to the palm center.

According to the visual monitoring method of another aspect of the present invention, in the step 50, the extreme point sets corresponding to the finger tips and the finger valleys arranged in order of the finger positions of the hand are obtained by:

step 501, deleting the wrist extreme points corresponding to the positions of the wrists in the extreme point set, enabling the extreme points after all the wrist extreme points are deleted to be adjacent end to end, enabling non-wrist extreme points adjacent to the wrist extreme points to serve as the first extreme points in the sequence, and sequentially sequencing other extreme points to re-determine the extreme point set;

step 511, judging whether the maximum value points and the minimum value points in the extreme value point set are alternately arranged, if not, entering step 512;

step 512, in the extreme point set, each extreme point is based on the variance of the linear distance and the variance of the index distance of other extreme points, and the variance of the linear distance and the variance of the index distance of all the extreme points in the extreme point set one by one to obtain the variance change value of the linear distance and the variance change value of the index distance of each extreme point, wherein the index distance is a path between the extreme points along the hand contour curve;

step 513, each extreme point in the extreme point set is based on the linear distance variance variation value and the index distance variance variation value of each extreme point one by one to obtain an energy value of each extreme point;

and 514, deleting the extreme point with the maximum energy value from the extreme point set to re-determine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set which is arranged by taking the finger positions of the hand as the sequence and corresponds to the finger tip and the finger valley, ending the operation, otherwise, returning to the step 511.

In the step 512, the straight-line distance variance variation value and the index distance variance variation value of each extreme point are obtained by the following expressions respectively for each extreme point:

<math> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

wherein,L^tis the set of extreme points after the t-th deletion of the extreme point, n_iIs any one extreme point in the set of extreme points, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+1The index distance of (a) is greater than (b), $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+2Linear distance of (d) (L)^t) And var (D (L))^t) Are each d (L)^t) And D (L)^t) The variance of (a) is determined,

indicates the extreme point n_iFrom L^tD (L) after deletion^t) The value of the change in the variance is,

indicates the extreme point n_iFrom L^tD (L) after deletion^t) Variance value of variance.

According to another aspect of the present invention, in step 513, the energy value of each extreme point is obtained by the following expression:

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

<math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

where min denotes minimizing the function, max denotes maximizing the function, N^tIs the t-th deletion poleThe number of extreme points in the set of extreme points following the value point,

indicates the extreme point n_kD (L) after deletion from the set of extreme points^t) The value of the change in the variance is,

indicates the extreme point n_kD (L) after deletion from the set of extreme points^t) Variance, e (i) represents the energy value of the extreme point.

According to another aspect of the visual monitoring method of the present invention, if in step 511, the maximum points and the minimum points in the extreme point set are alternately arranged, step 522 is entered;

step 522, obtaining an index distance variance variation value of each adjacent extreme point pair by each adjacent extreme point pair in the extreme point set, based on the index distance variance of other extreme points and the variance of the index distances of all extreme points in the extreme point set pair by pair, wherein the index distances are paths between the extreme points along the hand contour curve;

523, obtaining a variation value of the linear distance variance of each extreme point in the extreme point set based on the linear distance variances of other extreme points and the variances of the linear distances of all extreme points in the extreme point set one by one;

step 524, obtaining energy values of each adjacent extreme point pair by the adjacent extreme point pairs in the extreme point set, pair by pair based on the variance variation value of the index distance of each adjacent extreme point pair and the variance of the linear distance of each corresponding extreme point in each adjacent extreme point pair;

and 525, deleting the adjacent extreme point pair with the maximum energy value from the extreme point set to re-determine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set which is arranged by taking the finger position of the hand as the sequence and corresponds to the finger tip and the finger valley, ending the operation, otherwise, returning to the step 511.

According to another aspect of the present invention, in the step 522, each adjacent extremum point pair obtains an index distance variance variation value of each adjacent extremum point pair by the following expression:

<math> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <mo>{</mo> <msub> <mi>n</mi> <mi>k</mi> </msub> <mo>,</mo> <msub> <mi>n</mi> <mrow> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>}</mo> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>.</mo> </mrow> </math>

wherein,L^tis the set of extreme points after the t-th deletion of the extreme point, n_k，n_k+1Is any adjacent extreme point pair in the extreme point set, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$ representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+1Index distance of (d), var (L)^t) Is d (L)^t) The variance of (a) is determined,

representing pairs of adjacent extremum points n_k，n_k+1From L^tD (L) after deletion^t) Variance value of variance.

In step 523, according to the visual monitoring method of another aspect of the present invention, the linear distance variance variation value of each extreme point is obtained by the following expression:

<math> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

wherein,L^tis the set of extreme points after the t-th deletion of the extreme point, n_iIs any one extreme point in the set of extreme points, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+2Linear distance of (D), var (L)^t) Represents D (L)^t) The variance of (a) is determined,

indicates the extreme point n_iFrom L^tD (L) after deletion^t) Variance variationThe value is obtained.

According to another aspect of the present invention, in step 524, each of the adjacent pairs of extremum points obtains an energy value of each of the pairs of extremum points by the following expression:

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

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

wherein,

<math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> <mo>,</mo> </mrow> </math> <math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

min denotes minimizing the function, max denotes maximizing the function, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point,

representing pairs of adjacent extremum points n_i，n_i+1From L^tD (L) after deletion^t) The value of the change in the variance is,

and

respectively representing the pairs of extreme points n_i，n_i+1Extreme point n of_iAnd n_i+1D (L) after deletion from the extreme point set respectively^t) The variance value e (i, i +1) represents the adjacent extreme point pair n_i，n_i+1The energy value of (c).

According to another aspect of the visual monitoring method of the present invention, the third preset condition is that the extreme point set is composed of 5 maximum points and 4 minimum points, and the maximum points and the minimum points are alternately arranged.

According to the visual monitoring method of another aspect of the present invention, in the step 60, the digit to which the acquired fingerprint belongs is determined based on the 1 st, 2 nd and 3 rd order extreme points and the 7 th, 8 th and 9 th order extreme points in the 5 th and 4 th minimum extreme point sets.

According to another aspect of the present invention, there is also provided a visual monitoring device in a fingerprint collection process, which determines a finger position of a collected fingerprint by performing the method according to any one of the above.

One or more embodiments of the present invention may have the following advantages over the prior art:

the visual monitoring method adopts the method that after the texture is collected, the camera acquires the image around the fingerprint collection equipment at the moment, and then the image is analyzed to judge whether the collected texture comes from a certain finger or not and judge which finger the collected texture comes from. When the fingerprint is judged to be a non-finger or an error finger, the collected fingerprint is refused to be accepted, and the influence of cheating on the system is effectively reduced. Compared with the manual monitoring means widely used at present, the method can adapt to the increasing application requirements.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:

FIG. 1 is a flow chart of a method of visual monitoring during fingerprint acquisition according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a hand region segmentation of a visual monitoring method in a fingerprint acquisition process according to an embodiment of the present invention;

FIG. 3(a) is a schematic representation of radial distance minima points on a hand contour image of a hand;

FIG. 3(b) is a schematic view of the radial distance from each point on the hand contour image to the palm center;

FIG. 4(a) is a hand contour image I_cA schematic diagram of (a);

FIG. 4(b) is a diagram illustrating a search curve starting from the minimum point 7;

FIG. 4(c) is a diagram illustrating a search curve starting from the minimum point 8;

FIG. 4(d) is a hand outline image I 'after separating the closed fingers along the finger slits'_cA schematic diagram of (a);

FIG. 5(a) is a schematic diagram of an initial extreme point on a hand contour image;

fig. 5(b) is a schematic diagram of the extreme points finally left on the hand contour map.

Detailed Description

The following detailed description of the embodiments of the present invention will be provided with reference to the drawings and examples, so that how to apply the technical means to solve the technical problems and achieve the technical effects can be fully understood and implemented. It should be noted that, as long as there is no conflict, the embodiments and the features of the embodiments of the present invention may be combined with each other, and the technical solutions formed are within the scope of the present invention.

Additionally, the steps illustrated in the flow charts of the figures may be performed in a computer system such as a set of computer-executable instructions and, although a logical order is illustrated in the flow charts, in some cases, the steps illustrated or described may be performed in an order different than here.

Fig. 1 is a schematic flow chart of a visual monitoring method in a fingerprint acquisition process according to an embodiment of the present invention, and the following describes each step of the present embodiment in detail with reference to fig. 1.

Step S110, acquiring an image I around the fingerprint acquisition equipment in the fingerprint acquisition process, and extracting a hand area image I from the image_h(this step is simply referred to as dividing the hand region image). The image I is a current image acquired by the camera after the fingerprint is collected.

Specifically, after the texture is acquired, an image I of the surroundings of the fingerprint acquisition device at that time is acquired by the camera, and the hand region is extracted by removing the background region from the image I. In this embodiment, since the fingerprint acquisition device is usually placed on a desktop, the background is simple, and there are no moving objects, so that the foreground region image I can be extracted by a simple and fast background difference method_f。

In addition to the simple background, embodiments of the present invention also require that the color of the background not be too similar to the human skin tone.

Fig. 2 is a schematic diagram of a hand area image segmented by a visual monitoring method in a fingerprint acquisition process according to an embodiment of the present invention, and step S110 is specifically implemented by the following steps.

Step S1101, removing the background area image from the image I by using a background difference method to obtain a foreground area image I_f。

Specifically, each pixel of the background region image is simulated in an HSV color space by using a Gaussian model, so that a model M is obtained. Then, the pixel with the overlarge deviation model M is marked as the foreground by adopting a background difference methodThen filtering some noise points by using a morphological operator to obtain a foreground region image I_f。

Step S1102, if the image is in the foreground area image I_fIf the shadow part exists, removing the foreground region image I according to the tone, saturation and brightness_fTo obtain a hand region image I_h。

Because the shadow area of the hand under the influence of illumination is also the foreground area image I_fNeed to be taken from the foreground region image I_fRemoving to obtain hand region image I_h. The shadow area is the area of the hand area image I obtained by removing the shadow area from the foreground area image according to the characteristic that the brightness of some pixels of the background area image is low because the hand blocks a part of light, so that the part of pixels is similar to the background area image in terms of hue and saturation, and the brightness is low_h。

Step S120, detecting a hand region image I_hWhether a closed finger is present.

Specifically, the presence or absence of the closed fingers is detected by detecting whether or not a straight line having a lower gradation value than the average gradation value of the hand region image exists in the hand region image. If the closed fingers exist, the process goes to step S130, otherwise, if the closed fingers do not exist, the hand region image is directly converted into a finger-separated hand contour curve, and the process goes to step S160.

It should be noted that, since the present embodiment relates to detection of finger valleys at edge points of a hand, finger valleys between closed fingers are not in the hand region image I_hThe position of the closed fingers along the finger slit, and thus the need to differentiate the closed fingers. Wherein, the finger valley refers to the terminal part of the gap between the fingers.

When the fingers are closed, the gray level of the finger seam area of the closed fingers is lower than that of the surrounding area, and the finger seam can be generally approximate to a straight line; but in the fully open palmThe gray levels of the respective regions do not differ greatly, and such a straight line does not exist. Therefore if in the hand area image I_hWhen such a straight line is detected, it is considered as a finger slit of the closed finger.

In addition, the finger gap is between the closed fingers, so the distance from the point on the finger gap to the background is mostly larger than the width of the finger, which is one of the limiting conditions of the finger gap.

Step S130, converting the hand region image into a hand contour curve without separating fingers.

For example, the hand region image in which the fingers are present in close proximity in the hand region image is converted into the hand contour image I in which the fingers are not separated as shown in fig. 4(a)_c。

It should be noted that the finger slits extend outward from the valleys between the closed fingers to the edge of the hand. Since the finger valley is not easily detected, the search is inward from the edge point. In order to speed up the searching process, the embodiment selects only some possible candidate points from the edge points for operation.

Step S140, finding the starting point of the finger gap from the hand contour curve to determine the finger gap between the closed fingers.

Specifically, firstly, points satisfying a first preset condition in each point of a hand contour curve with fingers not separated are sequentially inquired to obtain a finger seam starting point set.

Specifically, the palm center C is set as the center of the largest inscribed circle in the hand region, and the distance from each point on the hand contour curve to the center of the circle is referred to as a radial distance. Since the radial distance at the intersection of the finger slit and the hand contour is usually a minimum point, the first preset condition is set such that the radial distance is equal to or greater than the first set threshold value t_dAnd the point with the minimum value is used as the starting point of the finger gap candidate. The first set threshold may be set to 1.5 to 1.7 times the radius of the inscribed circle, and the embodiment preferably sets t to_dIs set to be 5/3 times of the radius of an inscribed circle, wherein the inscribed circle takes the palm center as the center of a circle and divides the circle by the center of the circleA circle tangent to the edge of the hand region outside the fingers.

FIG. 3(a) is a schematic diagram of the radial distance minimum point on the hand contour image, FIG. 3(b) is a schematic diagram of the radial distance from the contour point to the palm center, and as can be seen from FIG. 3(b), there are 9 minimum points, but only the radial distances from the minimum point 7 and the minimum point 8 to the palm center satisfy a certain threshold t_dSo the minimum value point 7 and the minimum value point 8 become the starting point set of finger slit candidates.

Then, selecting a finger seam starting point satisfying a second preset condition from the finger seam candidate starting point set based on an active contour model (hereinafter referred to as snake model) to determine the finger seam between the closed fingers.

Specifically, after the starting point of the finger slit candidate is obtained, a snake model is used for searching a straight line around the starting point of the finger slit candidate by taking a gray scale image as an energy field. Setting the second preset condition that the length of the line segment is greater than or equal to a second set threshold value t_lAnd extending towards the palm center C, if the condition is met, the candidate point is considered to be a finger joint starting point, and the straight line is taken as a finger joint. The second set threshold may be set to 1.1 to 1.3 times the radius of the inscribed circle, and in this step, t is preferably set to_lSet at 1.2 times the radius of the inscribed circle.

In addition, if a plurality of straight lines satisfying the condition are emitted from the same point, the straight line with the lowest average gray level value is taken as the finger slit.

FIG. 4(a) is a hand contour image I_cIn fig. 4(b), the blue line is the line found from the minimum value point 7, and in fig. 4(c), the blue line is the line found from the minimum value point 8. The line curvature in fig. 4(b) is large, and therefore is discarded. Whereas the line in fig. 4(c) is relatively close to a straight line and the length exceeds the threshold, it can be regarded as a finger-seam.

In step S150, a finger-separated hand contour curve is obtained based on the determined finger seam.

In particular, based on slave stepsThe finger slit obtained in step S140 can be distinguished from the closed finger, and specifically, as shown in fig. 4(d), the hand contour image I 'after the closed finger is separated along the finger slit as shown in fig. 4 (d)'_cSchematic representation of (a).

And step S160, sequentially inquiring the extreme points which satisfy the condition that the radial distance is an extreme value in each point of the hand contour curve separated by the fingers, and arranging the extreme points according to the inquired sequence to generate an extreme point set, wherein the radial distance is the distance from the point on the hand contour image to the palm center.

In general, the radial distance of the contour points located in the fingertip portion is larger than the radial distance of the surrounding contour points, and is a maximum value point, and the contour points located in the finger-valley portion are minimum value points. The algorithm finds out the edge points with the maximum or minimum radial distance, and because of the influence of irregular edges and arm parts, redundant extreme points can be found and need to be removed.

If the edge points meeting the conditions are less than 5 maximum value points and 4 minimum value points, the phenomenon that fingers are mutually shielded exists, the finger positions cannot be accurately estimated, the gesture of the type is rejected by the algorithm, and a warning is sent to the collector. Similarly, if too many edge points satisfy the condition, for example, the maximum threshold of the edge points is set to 13 in this embodiment, which also indicates that the edge is quite irregular, possibly because the user overlaps two hands together to disturb the system, the algorithm will directly reject such gesture, and no operation of removing the redundant extreme points is performed.

In this embodiment, the extreme point is searched for by traversing clockwise from the intersection point of the curve and the centerline of the fingerprint acquisition device. Taking the hand contour image in fig. 4(c) as an example, the extreme point is detected on the image. The detection result is shown in fig. 5(a), and there are 18 extreme points in total, and the extreme points are sorted according to the searched sequence.

In addition, if there are no maxima around the fingerprint acquisition device, indicating that there is no fingertip around, i.e. the acquired texture does not originate from a finger, the system will reject this gesture.

Step S170, acquiring extreme point sets corresponding to the fingertips and the finger valleys respectively, which are arranged in order of the finger positions of the hand, based on the extreme point sets, wherein the finger valleys are the ends of the gaps between the fingers.

The method is specifically obtained through the following substeps:

step 1701, deleting the wrist extreme points corresponding to the wrist positions in the extreme point set, enabling the extreme points after all the wrist extreme points are deleted to be adjacent end to end, enabling non-wrist extreme points adjacent to the wrist extreme points to serve as the first extreme point in the sequence, and sequentially sequencing other extreme points to determine the extreme point set again.

Taking the hand contour image in fig. 4(c) as an example, the extreme point is detected on the image. As shown in fig. 5(a), the solid points are maximum points, and the hollow points are minimum points. n is₇，n₈，n₉When the image is close to the edge of the image, the extreme point is judged as the extreme point on the wrist and is not considered. In step, a point less than 50 pixels from the lower edge of the curve is preferably determined as a wrist extreme point near the edge of the image.

In addition, since a normal valley should be adjacent to two fingertips (maximum points), n after being deleted₆，n₁₀If such a condition is not satisfied, the valley is not included, and thus the deletion may be performed.

N is to be₆，n₇，n₈，n₉，n₁₀After deletion, n is added₁₈And n₁Adjacent end to end, will be n₁₀Adjacent n₁₁As the first extreme point in the sequence, the newly determined set of extreme points is (n)₁₁，n₁₂,n₁₃，n₁₄,n₁₅，n₁₆,n₁₇,n₁₈,n₁，n₂，n₃，n₄，n₅)。

Aiming at removing redundant extreme points, the embodiment of the invention provides an energy function for measuring the quality of the extreme points, the higher the energy is, the worse the quality is, and the most possible combination of the extreme points can be selected according to the point. The contents of the algorithm are explained below.

Step 1702, determining whether the maximum points and the minimum points in the extreme point set are alternately arranged, if not, entering step 1703.

And 1703, obtaining a linear distance variance change value and an index distance variance change value of each extreme point on the basis of the linear distance variance and the index distance variance of other extreme points and the linear distance variance and the index distance variance of all the extreme points in the extreme point set one by one, wherein the index distance is the distance between the extreme points along the hand contour curve.

In step 1704, the extreme points in the extreme point set are based on the variance variation of the linear distance and the variance variation of the index distance of each extreme point one by one to obtain the energy value of each extreme point.

Step 1705, deleting the extreme point with the largest energy value from the extreme point set to re-determine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set corresponding to the finger tip and the finger valley arranged in sequence by taking the finger position of the hand as the sequence, ending the operation, otherwise, returning to step 1702.

Specifically, the extremum point sequence is written as

L^tI.e. the sequence of extreme points after the t-th deletion of the extreme point, n_i(i＝1，…，N^t) Is an extreme point, N^tThe number of the extreme points in the sequence of the extreme points after the t-th deletion of the extreme points is shown. Two extreme points n are defined_iAnd n_jIndex distance d of_i，jDistance of path of two points along the contour, D_i，jIs two points in the image I_cUpper straight line distance. Definition of $d\left(L^t\right)=$ $\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$ $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$ And var (d (L)^t) And var (D (L))^t) Are each d (L)^t) And D (L)^t) The variance of (c).

For normal adjacent fingertips and valleys, each d_i，jRelatively close, each D for two normal adjacent fingertips or two finger valleys_i，jAre all relatively similar, i.e. var (d (L)^t) And var (D (L))^t) ) are relatively small. If a maximum point, d, occurs between two normal maximum points due to an irregular edge_i，jWill become smaller, D_i，jWill also change, causing var (d)(L^t) And var (D (L))^t) Is increased). If the maximum point appears on the arm, D_i，jWill become larger, d_i，jIs not certain, this will cause var (D (L)^t) Is increased).

Note the book <math> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> </mrow> </math> I.e. the extreme point n_iFrom L^tAfter d (L) removal^t) Variance variation, for the same reason, for straight-line distance

Extreme point n_iThe energy of (c) can be calculated by the following expression:

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

<math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

the higher the energy, the higher the extreme point pair var (d (L)^t) And var (D (L))^t) The more influence of) is, the more unstable it is, and should be preferentially deleted. Generally, the maximum points and the minimum points are alternately present. If the two adjacent pixels are both the maximum or minimum point, then the one with the higher energy should be deleted.

If the determination result in the step 1702 is negative, the process proceeds to step 1713, that is, the maximum value points and the minimum value points are alternately arranged adjacently, and the maximum value points are deleted in pairs by the following steps.

Step 1713, obtaining variance values of the variance of the index distances of each adjacent extreme point pair by each adjacent extreme point pair in the extreme point set, based on the variance of the index distances of other extreme points and the variance of the index distances of all extreme points in the extreme point set, wherein the index distances are paths between the extreme points along the hand contour curve.

Step 1714, in the extreme point set, each extreme point is based on the variance of the linear distances of other extreme points and the variance of the linear distances of all the extreme points in the extreme point set one by one to obtain the variance variation value of the linear distance of each extreme point;

step 1715, concentrating each adjacent extreme point pair in the extreme point set, and obtaining the energy value of each adjacent extreme point pair by pair based on the variance of the index distance variance of each adjacent extreme point pair and the variance of the linear distance of each corresponding extreme point in each adjacent extreme point pair;

step 1716, deleting the adjacent extreme point pair with the largest energy value from the extreme point set to redetermine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set which is arranged by taking the finger positions of the hand as the sequence and corresponds to the finger tip and the finger valley, ending the operation, otherwise, returning to the step 1702.

Specifically, the pair of extremum points n_i，n_i+1The energy value of (c) can be calculated by the following expression:

<math> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <mo>{</mo> <msub> <mi>n</mi> <mi>k</mi> </msub> <mo>,</mo> <msub> <mi>n</mi> <mrow> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>}</mo> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>.</mo> </mrow> </math>

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

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

wherein, <math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> <mo>,</mo> </mrow> </math> <math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

min denotes minimizing the function, max denotes maximizing the function, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point,representing pairs of adjacent extremum points n_i，n_i+1From L^tD (L) after deletion^t) The value of the change in the variance is,

andrespectively representing the pairs of extreme points n_i，n_i+1Extreme point n of_iAnd n_i+1D (L) after deletion from the extreme point set respectively^t) The variance value e (i, i +1) represents the adjacent extreme point pair n_i，n_i+1The energy value of (c).

At this time, the extreme point pairs with high energy should be preferentially deleted. According to the algorithm, redundant extreme points can be deleted until the extreme point set meets a third preset condition, and then the operation is finished.

For example, taking the hand contour image in fig. 4(c) as an example, the extreme point is detected on the image. As shown in fig. 5(a), the solid points are maximum points, and the hollow points are minimum points. The extreme points are therefore deleted by an energy function that measures the quality of the extreme points.

Due to (n)₁₁，n₁₂，n₁₃，n₁₄，n₁₅，n₁₆，n₁₇，n₁₈，n₁，n₂，n₃，n₄，n₅) The medium maximum point and the minimum point appear alternately, and therefore, when the extreme point is deleted, the deletion is performed in the form of a point pair. Table 1 shows the energy values calculated according to the energy functions for the pairs of extreme points. Table 2 shows the energy values of other pairs of extreme points after deleting some of the extreme points. Since n is₁Is to separateThe nearest maxima point of the detector, which the algorithm believes corresponds to the finger tip of the finger from which the fingerprint originated, cannot be deleted₁₈，n₁And n₁，n₂The energy value of (c).

TABLE 1

TABLE 2

According to the table 1, the point pair n with the highest energy value is identified₁₄，n₁₅Delete, according to Table 2, n with the highest energy value₄，n₅Deleting the obtained extreme point (n)₁₁，n₁₂，n₁₃，n₁₆，n₁₇，n₁₈，n₁，n₂，n₃) They correspond to fingertips and valleys, respectively, as shown in fig. 5 (b).

And S180, determining the finger position of the acquired fingerprint based on the extreme point set of the finger tip and the finger valley.

Specifically, the digit to which the acquired fingerprint belongs is determined based on the extreme points of the ordered 1 st, 2 nd, and 3 rd bits and the extreme points of the ordered 7 th, 8 th, and 9 th bits from the 5-maximum point and 4-minimum point extreme point sets.

More specifically, since the fingertip corresponding to the maxima closest to the detector is the source of the fingerprint, the position of the fingerprint can be determined from the location of the maxima.For example, in FIG. 5(b), the extreme point n on the arm is used₇，n₈，n₉Dividing for boundaries, if the maximum points are arranged from left to right, then n₁Is the sequence of maximum points left { n }₁₁，n₁₃，n₁₇，n₁，n₃The 4 th of them. Thus, n₁The corresponding fingertip may be the left index finger or the right ring finger.

As for the left and right hand judgment, the distinction is made by using the feature that the straight distance from the index finger to the valley between it and the thumb is significantly longer than that of the thumb.

If the left hand is used, the index finger and the thumb respectively correspond to the second maximum value point and the first maximum value point of the right number. According to the symmetry of the left hand and the right hand, the index finger and the thumb of the right hand respectively correspond to the second maximum value point and the first maximum value point of the left number.

The extreme points may be arranged from left to right, i.e., L ═ n₁，...，n₉}. Note d_left＝D_2,3-D_1,2，d_right＝D_7,8-D_8,9，ρ＝d_left-d_rightThen, if ρ > 0, the description is right hand, otherwise left hand.

It should be noted that this method of determination requires that the finger not be bent too much, otherwise there is no way to project on the two-dimensional image to ensure that such features are still present. Therefore, if d_leftAnd d_rightNot all positive numbers indicate that the finger is bent and not judged properly, and such a gesture will be rejected.

Extreme point result (n) with respect to FIG. 5(b)₁₁，n₁₂，n₁₃，n₁₆，n₁₇，n₁₈，n₁，n₂，n₃) Since ρ < 0 is obtained by calculation, it is judged as a left hand. The maximum point nearest to the detector is n₁Therefore, the finger position is finally determined to be the left index finger. Instructions may then be sent to the detector to receive the fingerprint and add label information to the fingerprint as needed。

In addition, the invention also relates to a visual monitoring device in the fingerprint acquisition process, which determines the finger position of the acquired fingerprint by executing the operation.

According to the embodiment of the invention, after the texture is acquired, the image around the fingerprint acquisition device at the moment is acquired through the camera, and whether the acquired texture comes from a certain finger or not and which finger is judged through analyzing and understanding the image, so that the influence of cheating on the system is effectively reduced. Compared with the manual monitoring means widely used at present, the method can adapt to the increasing application requirements.

Those skilled in the art will appreciate that the modules or steps of the invention described above can be implemented in a general purpose computing device, centralized on a single computing device or distributed across a network of computing devices, and optionally implemented in program code that is executable by a computing device, such that the modules or steps are stored in a memory device and executed by a computing device, fabricated separately into integrated circuit modules, or fabricated as a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.

Although the embodiments of the present invention have been described above, the above descriptions are only for the convenience of understanding the present invention, and are not intended to limit the present invention. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (16)

1. A visual monitoring method in a fingerprint acquisition process is characterized by comprising the following steps:

step 10, acquiring an image around the fingerprint acquisition equipment in the fingerprint acquisition process;

step 20, extracting a hand region image from the image;

step 30, converting the hand area image into a finger-separated hand contour curve;

step 40, sequentially inquiring extreme points which satisfy that the radial distance is an extreme value in each point of the hand contour curve separated by the fingers, and arranging the extreme points according to the inquired sequence to generate an extreme point set, wherein the radial distance is the distance from the point on the hand contour curve to the center of the palm;

step 50, acquiring an extreme point set which is arranged by taking the finger positions of the hand as a sequence and corresponds to the fingertips and the finger valleys on the basis of the extreme point set, wherein the finger valleys are the end parts of gaps between the fingers;

and step 60, determining the finger position of the acquired fingerprint based on the extreme point set corresponding to the finger tip and the finger valley.

2. The method according to claim 1, characterized in that, in said step 20,

removing a background area image from the image by using a background difference method to obtain a foreground area image;

and removing shadow areas from the foreground area image according to the hue, the saturation and the brightness to obtain a hand area image.

3. The method according to claim 1, wherein in the step 30, further comprising:

and detecting whether the gathered fingers exist in the hand area image, and if not, directly converting the hand area image into a finger-separated hand contour curve.

4. The method of claim 3,

and detecting whether the gathered fingers exist or not by detecting whether a straight line with a gray value lower than the average gray value of the hand area image exists or not in the hand area image.

5. The method according to any one of claims 3 or 4, wherein in step 30, if closed fingers are present, the following steps are performed:

step 311, converting the hand region image into a hand contour curve without separating fingers;

step 312, sequentially inquiring points meeting a first preset condition in each point of the hand contour curve with the fingers not separated to obtain a finger gap candidate starting point set;

313, selecting a finger seam starting point meeting a second preset condition from the finger seam candidate starting point set based on the active contour model to determine the finger seam between the closed fingers;

and step 314, obtaining a hand contour curve of the finger separation based on the determined finger gaps.

6. The method of claim 5,

the first preset condition is that the radial distance is a minimum value and the radial distance is greater than or equal to a first set threshold;

the second preset condition is that the length of the line segment is greater than or equal to a second set threshold value, the line segment extends towards the palm center, the line segment is formed by taking the starting point of each finger seam as the starting point according to the active contour model, and the radial distance is the distance from the point on the hand contour curve with the fingers not separated to the palm center.

7. The method according to claim 1, wherein in the step 50, the extreme point set corresponding to the fingertip and the valley in order of the finger position of the hand is obtained by:

step 501, deleting the wrist extreme points corresponding to the positions of the wrists in the extreme point set, enabling the extreme points after all the wrist extreme points are deleted to be adjacent end to end, enabling non-wrist extreme points adjacent to the wrist extreme points to serve as the first extreme points in the sequence, and sequentially sequencing other extreme points to re-determine the extreme point set;

step 511, judging whether the maximum value points and the minimum value points in the extreme value point set are alternately arranged, if not, entering step 512;

step 512, in the extreme point set, each extreme point is based on the variance of the linear distance and the variance of the index distance of other extreme points, and the variance of the linear distance and the variance of the index distance of all the extreme points in the extreme point set one by one to obtain the variance change value of the linear distance and the variance change value of the index distance of each extreme point, wherein the index distance is a path between the extreme points along the hand contour curve;

step 513, each extreme point in the extreme point set is based on the linear distance variance variation value and the index distance variance variation value of each extreme point one by one to obtain an energy value of each extreme point;

and 514, deleting the extreme point with the maximum energy value from the extreme point set to re-determine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set which is arranged by taking the finger positions of the hand as the sequence and corresponds to the finger tip and the finger valley, ending the operation, otherwise, returning to the step 511.

8. The method of claim 7, wherein, in step 512,

the linear distance variance change value and the index distance variance change value of each extreme point are obtained by the following expressions respectively for each extreme point:

<math> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

<math> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

wherein,L^tis the set of extreme points after the t-th deletion of the extreme point, n_iIs any one extreme point in the set of extreme points, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+1The index distance of (a) is greater than (b), $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+2Linear distance of (d) (L)^t) And var (D (L))^t) Are each d (L)^t) And D (L)^t) The variance of (a) is determined,indicates the extreme point n_iFrom L^tD (L) after deletion^t) The value of the change in the variance is,

indicates the extreme point n_iFrom L^tD (L) after deletion^t) Variance value of variance.

9. The method of claim 7, wherein, in step 513,

the energy value of each extreme point is obtained by the following expression respectively:

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

<math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

where min denotes minimizing the function, max denotes maximizing the function, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point,

indicates the extreme point n_kD (L) after deletion from the set of extreme points^t) The value of the change in the variance is,

indicates the extreme point n_kD (L) after deletion from the set of extreme points^t) Variance, e (i) represents the energy value of the extreme point.

10. The method of claim 7, wherein if the maximum points and the minimum points in the extreme point set are alternately arranged in step 511, step 522 is proceeded to;

step 522, obtaining an index distance variance variation value of each adjacent extreme point pair by each adjacent extreme point pair in the extreme point set, based on the index distance variance of other extreme points and the variance of the index distances of all extreme points in the extreme point set pair by pair, wherein the index distances are paths between the extreme points along the hand contour curve;

523, obtaining a variation value of the linear distance variance of each extreme point in the extreme point set based on the linear distance variances of other extreme points and the variances of the linear distances of all extreme points in the extreme point set one by one;

step 524, obtaining energy values of each adjacent extreme point pair by the adjacent extreme point pairs in the extreme point set, pair by pair based on the variance variation value of the index distance of each adjacent extreme point pair and the variance of the linear distance of each corresponding extreme point in each adjacent extreme point pair;

and 525, deleting the adjacent extreme point pair with the maximum energy value from the extreme point set to re-determine the extreme point set, if the extreme point set meets a third preset condition, taking the extreme point set as the extreme point set which is arranged by taking the finger position of the hand as the sequence and corresponds to the finger tip and the finger valley, ending the operation, otherwise, returning to the step 511.

11. The method of claim 10, wherein, in said step 522,

obtaining the variance variation value of the index distance of each adjacent extreme point pair respectively through the following expressions:

<math> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>d</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <mo>{</mo> <msub> <mi>n</mi> <mi>k</mi> </msub> <mo>,</mo> <msub> <mi>n</mi> <mrow> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> </msub> <mo>}</mo> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>.</mo> </mrow> </math>

wherein,

L^tis the set of extreme points after the t-th deletion of the extreme point, n_k，n_k+1Is any adjacent extreme point pair in the extreme point set, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point, $d\left(L^t\right)=\left\{d_{i,i+1}^t\right|i=\mathrm{1,2},...,N^t-1\},$

representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+1Index distance of (d), var (L)^t) Is d (L)^t) The variance of (a) is determined,

representing pairs of adjacent extremum points n_k，n_k+1From L^tD (L) after deletion^t) Variance value of variance.

12. The method of claim 10, wherein, in the step 523,

the linear distance variance variation value of each extreme point is obtained by the following expression:

<math> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>=</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>-</mo> <mi>var</mi> <mrow> <mo>(</mo> <mi>D</mi> <mrow> <mo>(</mo> <msup> <mi>L</mi> <mi>t</mi> </msup> <mo>-</mo> <msub> <mi>n</mi> <mi>i</mi> </msub> <mo>)</mo> </mrow> <mo>)</mo> </mrow> <mo>,</mo> <mrow> <mo>(</mo> <mi>i</mi> <mo>=</mo> <mn>1</mn> <mo>,</mo> <mo>.</mo> <mo>.</mo> <mo>.</mo> <mo>,</mo> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>)</mo> </mrow> </mrow> </math>

wherein,

L^tis the set of extreme points after the t-th deletion of the extreme point, n_iIs any one extreme point in the set of extreme points, N^tIs the t-th deletion poleThe number of extreme points in the set of extreme points following the value point, $D\left(L^t\right)=\left\{D_{i,i+2}^t\right|i=\mathrm{1,2},...,N^t-2\},$ representing n in the extreme point set after the t-th deletion of the extreme point_iAnd n_i+2Linear distance of (D), var (L)^t) Represents D (L)^t) The variance of (a) is determined,indicates the extreme point n_iFrom L^tD (L) after deletion^t) Variance value of variance.

13. The method of claim 10, wherein, in said step 524,

and respectively obtaining the energy value of each extreme point pair by the following expressions:

<math> <mrow> <msubsup> <mi>e</mi> <mn>1</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&delta;</mi> <mrow> <mi>i</mi> <mo>,</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> <mo>-</mo> <mn>1</mn> </mrow> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&delta;</mi> <mrow> <mi>k</mi> <mo>,</mo> <mi>k</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

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

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

wherein, <math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mi>i</mi> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> <mo>,</mo> </mrow> </math> <math> <mrow> <msubsup> <mi>e</mi> <mn>2</mn> <mi>t</mi> </msubsup> <mrow> <mo>(</mo> <mi>i</mi> <mo>+</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>=</mo> <mfrac> <mrow> <msubsup> <mi>&Delta;</mi> <mrow> <mi>i</mi> <mo>+</mo> <mn>1</mn> </mrow> <mi>t</mi> </msubsup> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> <mrow> <msubsup> <mi>max</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> <mo>-</mo> <msubsup> <mi>min</mi> <mrow> <mi>k</mi> <mo>=</mo> <mn>1</mn> </mrow> <msup> <mi>N</mi> <mi>t</mi> </msup> </msubsup> <mrow> <mo>(</mo> <msubsup> <mi>&Delta;</mi> <mi>k</mi> <mi>t</mi> </msubsup> <mo>)</mo> </mrow> </mrow> </mfrac> <mo>.</mo> </mrow> </math>

min denotes minimizing the function, max denotes minimizing the functionTaking the maximum value, N^tIs the number of extreme points in the set of extreme points after the t-th deletion of the extreme point,

representing pairs of adjacent extremum points n_i，n_i+1From L^tD (L) after deletion^t) The value of the change in the variance is,and

respectively representing the pairs of extreme points n_i，n_i+1Extreme point n of_iAnd n_i+1D (L) after deletion from the extreme point set respectively^t) The variance value e (i, i +1) represents the adjacent extreme point pair n_i，n_i+1The energy value of (c).

14. The method according to any one of claims 7 to 13,

the third preset condition is that the extreme point set consists of 5 maximum points and 4 minimum points, and the maximum points and the minimum points are alternately arranged.

15. The method of claim 14, wherein, in step 60,

and determining the digit to which the acquired fingerprint belongs based on the 1 st, 2 nd and 3 rd bit ordered extreme points and the 7 th, 8 th and 9 th bit ordered extreme points in the 5 th maximum value point and 4 th minimum value point extreme point sets.

16. Visual monitoring device during fingerprint acquisition, characterized in that it determines the finger position to which an acquired fingerprint belongs by performing a method according to any one of claims 1 to 15.

Images