CN100373397C - Pre-processing method for iris image - Google Patents

Abstract

The present invention provides a pre-processing method for iris images, which belongs to the technical field of image treatment and mainly relates to iris personal identification technique in the biometric identification. In the present invention, firstly, the rough location of an inner edge circle center of an iris is realized by binarization, corrosion, expansion and gray projection; then, the inner edge of the iris region is determined by the methods of single gray value comparison and circular curve polymerization; subsequently, the outer edge of the iris region is determined by the methods of gray differential accumulation and the circular curve polymerization; at last, the sharpness value of points and the number of effective pixel points are calculated based on the normalized iris image, and the image quality can be evaluated on clearness and visibility. With the method combining the boundary point search based on gray gradient and the curve fitting provided by the present invention, the location precision and the location speed of irises can be effectively improved. With the quality evaluating method based on the point sharpness and the number of effective points provided by the present invention, the versatility of the traditional quality evaluating algorithm is improved.

Description

A kind of pre-processing method for iris image

Technical field

The invention belongs to technical field of image processing, relate generally to the iris identity recognizing technology in the biological characteristic discriminating.

Background technology

In the current information age, how accurately to identify a people's identity, the protection information security is a crucial social concern that must solve.For this reason, the biological characteristic authentication technique quietly newly rises, and becomes the forward position research topic in information security management field, the present world.The biological characteristic authentication technique be meant utilize human body intrinsic physiological characteristic or behavioural characteristic carry out personal identification and identify.The iris identity recognizing technology is a branch of biological characteristic authentication technique, it is the application of computer image processing technology and mode identification technology in the person identification field, because its high stability and high accuracy have become the popular developing direction that biological characteristic is differentiated in recent years.Iris identity automatic identification technology is widely used at aspects such as bank, public security, airport, networks, has huge economic and realistic meaning.Now it used at border control, taken an overall view of authentication, made a draft of money, information management and building safety management etc., people are broken away from remember the loaded down with trivial details of credit number, account No., identification card number, network entry number.Along with the development of Digital Signal Processing and image processing techniques, the iris identification system reaches its maturity.See document for details: John G.Daugman, " How Iris Recognition Works; " IEEE Transaction on Circuits andSystems for Video Technology, Volume 14, Issue 1, pp.21-30,2004 and document: John G.Daugman, " High Confidence Recognition of Persons by Iris Patterns, " The Proceeding ofIEEE 35 ^ThInternational Carnahan Conference on Security Technology, pp.254-263,2001 is described.

In the iris identity recognizing technology, the iris image pre-service is the key of whole recognition technology, and it comprises Iris Location and iris image quality assessment.Iris Location is the first step of iris recognition, and its execution time and precision will directly influence the speed and the accuracy of whole iris authentication system.In practice, because iris region usually is subjected to blocking of eyelid and eyelashes, Iris Location algorithm accuracy and validity are still waiting further raising.How in the inferior quality iris image that has eyelashes and eyelid occlusion issue, orient iris quickly and accurately, and its border or position are described with mathematical model is the subject matter that we study.See document for details: John G.Daugman, " HighConfidence Visual Recognition of Persons by a Test ofStatistical Independence; " IEEE Transaction onPattern Analysisand Machine Intelligence, volume 15, no.11, pp.1148-1161,1993.The iris image quality assessment is an important link in the automatic iris authentication system, and it has guaranteed to be met in the gatherer process image of quality standard.In the reality, because the focal length problem of collecting device when taking, the rotational problems of moment eyeball, and eyelid and eyelashes usually make the iris image of collection can't carry out follow-up feature extraction to the partial occlusion of iris.In the existed algorithms a kind of effective iris image quality assessment models is not proposed also at present, see document for details: Chen Ji, Hu Guangshu, " Iris Image Quality Evaluation based on Wavelet PacketDecomposition, " Journal ofTsinghua University (Sci﹠amp; Tech), volume 43, no.3, pp.377-380,2003.

The method of present normally used Iris Location has:

(1) goes on foot iris locating methods based on two of shade of gray.It is sought the approximate location of outer rim in the iris, and then utilizes circular detector to carry out fine positioning near this position in the small range by coarse positioning, thereby finds the exact position of outer rim in the iris.But this method search that need iterate in actual applications, operand is bigger, and efficient is not high.See document for details: Li Qingrong, Ma Zheng, " A Iris Location Algorithm, " Journal of UEST of China, volume 31, no.1, pp.7-9.

(2) based on the iris locating method of hough transform.It is by certain operator, extracts the marginal point in the iris image, thereby search is by the position at the maximum circular curve place of marginal point.Its shortcoming is usually can introduce noise in marginal point extracts, and makes that the Iris Location result is inaccurate.See document for details: Richard P.Wildes, " Iris Recognition:an EmergingBiometric Technology, " Proceedings ofthe IEEE, volume85, pp.1348-1363,1997.

At present existing iris method for evaluating quality has:

(1) based on the method for fast fourier transform.It carries out fast two-dimensional fourier transformation to the picture element in two rectangular blocks on the iris region, and then by to the statistics of its high frequency, intermediate frequency and low frequency energy, whether analysis image is clear and exist eyelashes to block.The versatility of this model is not strong, and easily that texture is less clear iris image erroneous judgement is the inferior quality iris image.See document for details: Li Ma, Tieniu Tan, Yunhong Wang, Dexin Zhang, " Personal Identification based on IrisTexture Analysis, " IEEE Transactions on Pattern Analysisand Machine Intelligence, volume.25, no.12, pp.1519-1533.

(2) based on the method for WAVELET PACKET DECOMPOSITION.It is chosen the texture high fdrequency component and distributes the most concentrated sub-band as the feature sub-band, with the criterion of its energy as differentiation picture quality.The shortcoming of this method is can't judge because of eyelashes to block in-problem iris image.See document for details: Chen Ji, Hu Guangshu, " Iris Image Quality Evaluation based on WaveletPacket Decomposition, " Journal of Tsinghua University (Sci﹠amp; Tech), volume43, no.3, pp.377-380,2003.

Above-mentioned iris image Preprocessing Algorithm all has problems to a certain extent, and location algorithm is consuming time more, and is subjected to the interference of eyelashes occlusion issue easily, and stability is not high.The versatility of iris image quality appraisal procedure is not strong.

Summary of the invention

Task of the present invention provides a kind of pre-processing method for iris image, comprises two processes of Iris Location and iris image quality assessment.Iris Location process of the present invention adopts based on the frontier point search of shade of gray and the iris locating method of circular curve match, and it has the advantages that to block accurate positioning under the situation at eyelash; In addition, the present invention has set up the more intense iris image quality assessment models of a cover versatility on the basis of Iris Location.

In order to describe content of the present invention easily, at first some terms are defined.

Definition 1: iris.The center of eyeball is the pupil of black, and the outer intermarginal annular tissue of pupil is iris.It presents the textural characteristics of interlaced similar and spot, filament, striped, crypts.Same individual's iris can change in life hardly the people's, and the iris of different people is different fully.

Definition 2: gray level image.Only comprise monochrome information in the image and without any the image of other colouring informations.

Definition 3: binary-state threshold.Selected gray scale threshold value when image is carried out binaryzation.

Definition 4: binaryzation.The all values of entire image is changed into the process of having only two kinds of values, and generally these two kinds of values are 0 and 1 or 0 and 255.When the value on the image more than or equal to the threshold values of binaryzation the time, the value two-value of this point turns to 1 (or 255); When the value on the image less than the binaryzation threshold values time, the value two-value of this point turns to 0.

Definition 5: mathematical morphology.Go to measure and extract in the image correspondingly-shaped to reach purpose with structural element to graphical analysis and identification with certain form.The fundamental operation of mathematical morphology has 4: expand (or expansion), corrosion (or erosion), unlatching and closed.The operational formula that expands and corrode is: $A\⊕B=\left\{x\right|{\left(\hat{B}\right)}_x\∩A\≠\mathrm{\φ}\}$ With A Θ B={x| (B) _x A}; The operational formula of open operation is: A о B=(A Θ B)  B.The operational formula of closed procedure is: AB=(A  B) Θ B.In the above-mentioned formula, " A " is image collection, and " B " is structural element, and the mapping about initial point, " () are done in " ^ " expression _x" expression translation x, " ∩ " represents to occur simultaneously, and " φ " represents empty set, and "  " tabular form comprises entirely, and "  " is the dilation operation symbol, and " Θ " is the erosion operation symbol, and " о " for opening operational symbol, " " is the closure operation symbol.

Definition 6: Gray Projection.Gray Projection in the two-dimensional space to the one-dimensional space, is divided into horizontal Gray Projection and vertical Gray Projection.Horizontal Gray Projection is meant the gray scale along continuous straight runs in the two dimensional image is added up, is transformed into the one-dimensional space.Transfer function is: $S_h\left(x\right)=\underset{y=1}{\overset{N}{\mathrm{\Σ}}}I(x,y)\·$ Vertical Gray Projection is meant the gray scale in the two dimensional image is vertically added up, is transformed into the one-dimensional space.Transfer function is: $S_v\left(y\right)=\underset{x=1}{\overset{M}{\mathrm{\Σ}}}I(x,y)\·$ S wherein _b(x) the expression horizontal ordinate is the Gray Projection value of x, S _v(y) the expression ordinate is the Gray Projection value of y, and M, N are the width and the height of image, and (x y) is position (x, the gray-scale value of picture element y) to I.

Definition 7: iris inner edge frontier point.Be meant the point that is positioned on pupil outward flange or the iris inward flange.

Definition 8: circle match.The coordinate of known series of points is set up a circular curve equation that can reflect these coordinate points positions.Specifically: equation of a circle is x ²+ y ²+ cx+dy+e=0, c wherein, d and e are the parameters about the radius and the central coordinate of circle point of circular curve, (x y) is the coordinate figure of the point on the circular curve, and the best circular curve with respect to these coordinate points makes error variance and minimum exactly so.Error variance and formula be: ${\mathrm{\ϵ}}^2=\underset{i}{\mathrm{\Σ}}{(x_i^2+y_i^2+{\mathrm{cx}}_i+{\mathrm{dy}}_i+e)}^2,$ Wherein, ε ²Be meant error variance and, (x _i, y _i) be the coordinate of known point.

Definition 9: horizontal first order difference.In the image, the gray-scale value of the back pixel of certain delegation deducts the gray-scale value of front pixel, or the gray-scale value of front pixel deducts the gray-scale value of back pixel, obtains the horizontal first order difference value of this row.Horizontal first order difference can be given prominence to the vertical edge information of image, is convenient to edge extracting.

Definition 10: iris outer rim frontier point.Iris is an annular region, and the point that is positioned on the iris outward flange is called iris outer rim frontier point.

Definition 11: normalization.The iris region of annular is drawn into the identical rectangular area of size, and to eliminate owing to the shooting distance difference, factors such as pupil contraction are to the influence of recognition effect.Concrete computing formula is: $\left\{\begin{array}{c}x(r,\mathrm{\θ})=(1-r)x_p\left(\mathrm{\θ}\right)+{\mathrm{rx}}_i\left(\mathrm{\θ}\right)\\ y(r,\mathrm{\θ})=(1-r)y_p\left(\mathrm{\θ}\right)+{\mathrm{ry}}_i\left(\mathrm{\θ}\right)\end{array}\right.,$ Wherein, r is distributed in interval [0,1], and θ is distributed in interval [0,2 π], and (x _p(θ), y _p(θ)) and (x _i(θ), y _i(θ)) represent iris internal boundary points and outer boundary point on the θ direction respectively.

Define 12. normalization iris images.Original iris image is carried out the rectangular image that normalized obtains afterwards.

Definition 13.8-neighborhood.To a coordinate points is that (it has 4 levels and vertical neighbour's pixel for x, pixel p y), and their coordinate is respectively (x+1, y), (x-1, y), (x, y+1), (x, y-1), and 4 diagonal angle neighbour's pixels, their coordinate points be (x+1, y+1), (x+1, y-1), (x-1, y+1), (x-1, y-1), 8 such pixels are collectively referred to as 8-neighborhood of p, as shown in Figure 1.

Define 14. acutancees.Be used to estimate the operator of Definition of digital picture, its mathematical form is: $f=\underset{i=1}{\overset{m\×n}{\mathrm{\Σ}}}\underset{j=1}{\overset{8}{\mathrm{\Σ}}}|\mathrm{dI}/\mathrm{dx}\backslash /m\×n,$ Wherein dI is the difference of the gray-scale value of the gray-scale value of picture element and this point in certain any eight neighborhoods in the image, and dx is the distance of consecutive point, and m and n are respectively the height and width of image.

Definition 15. effective picture elements.Being meant the picture element that is positioned at iris region in the normalization iris image, mainly is in order to distinguish invalid picture elements such as eyelashes and eyelid.The condition that effective picture element must satisfy is: V _Lash≤ I (x, y)≤V _Eyelid, V wherein _LashAnd V _EyelidBe the thresholding that takes a decision as to whether the picture element that is positioned at eyelashes zone and palpebral region, (x y) is the gray scale of image to I.

Define 16. visibilitys.Be meant the visible level of iris texture in the original iris image, the principal element that influences visibility is eyelid and eyelashes blocking iris region.

Detailed technology scheme of the present invention is:

A kind of pre-processing method for iris image, comprise two processes of Iris Location and iris image quality assessment: described Iris Location process comprises the steps:

The original iris gray level image that step 1. pair camera head is gathered carries out binary conversion treatment, and gray-scale value is greater than threshold value V in the original-gray image _bThe gray-scale value of picture element to compose be 1, less than threshold value V _bThe gray-scale value of picture element to compose be 0.

The bianry image that obtains in the step 2. pair step 1 carries out closure operation in the mathematical morphology with structural element B to it and eliminates little cavity in the bianry image.Described structural element B is one 7 * 7 a matrix, and the value of the element in the intermediate approximation border circular areas is 1, and the value of all the other elements is 0.

Step 3. is calculated the rough center of circle of iris inward flange

At first obtain the level and vertical Gray Projection of image in the calculation procedure 2: the computing formula of horizontal projection is: $S_h\left(x\right)=\underset{y=1}{\overset{N}{\mathrm{\Σ}}}I(x,y),$ The computing formula of vertical Gray Projection is: $S_v\left(y\right)=\underset{x=1}{\overset{M}{\mathrm{\Σ}}}I(x,y)\·$ S wherein _h(x) the expression horizontal ordinate is the Gray Projection value of x, S _v(y) the expression ordinate is the Gray Projection value of y, and M, N are the width and the height of original iris gray level image, and (x y) is position (x, the gray-scale value of picture element y) to I.

Next searches for Gray Projection, gets S _h(x) the horizontal ordinate x during minimum value _oWith get S _v(y) the ordinate y during minimum value _o, with (x _o, y _o) be considered as the rough center of circle of iris inward flange.

Step 4. is calculated the accurate center of circle and the radius of iris inward flange

At first, to coordinate points (x _o, y _o) near several rows carry out the search of iris inward flange point, obtain the coordinate of a series of iris inward flange points.Concrete searching method can be (to be y with the ordinate _oThis behavior example): at ordinate is y _oIn this delegation, with (x _o, y _o) be the center, along continuous straight runs is searched for the point of pixel gray-scale value greater than T left, stops search immediately when searching the pixel gray-scale value greater than T, writes down the coordinate (x of this moment _l, y _o) as the coordinate of an iris inward flange point, carry out along continuous straight runs search to the right by same mode again, obtain another iris inward flange point coordinate (x _r, y _o).

Then, a series of iris inward flange points of gained are justified match, obtain the accurate center of circle (x of iris inward flange _p, y _p) and radius r _p

Step 5. is calculated the outer peripheral accurate center of circle of iris and radius

At first, get coordinate points (x _p, y _p) near several rows, on each row that takes out, carry out the search of iris outer rim point, obtain the coordinate of a series of iris outward flange points.Concrete searching method is following, and (with the ordinate is y _pThis behavior example):

1), coordinates computed point (x _p, y _p) the horizontal first order difference of being expert at.Concrete computing formula is: work as x _pD (x, y during＜x＜N-5 _p)=I (x+5, y _p)-I (x, y _p); As 5＜x≤x _pThe time D (x, y _p)=I (x-5, y _p)-I (x, y _p).Wherein, D (x, y _p) denotation coordination point (x, y _p) horizontal first order difference value, (x, y) (x, gray-scale value y), N are original iris gray level image width to the denotation coordination point to I.

2) be y, at ordinate _pIn the delegation, at interval [x _p+ r _p+ 20, x _p+ r _p+ 100] go up the horizontal first order difference value sum of calculating each point and 20 points in back.Concrete computing formula is: work as x _p+ r _p+ 20＜x＜x _p+ r _p+ 100 o'clock, $(x,y_p)=\underset{i=0}{\overset{20}{\mathrm{\Σ}}}D(x+i,y_p),$ D (x+i, y wherein _p) be coordinate points (x+i, y _p) horizontal first order difference value.And in this interval, find out S (x, y _p) coordinate points of correspondence (x, y when getting maximal value _p) as iris outer rim frontier point.

3), ordinate is y _pIn the delegation, at interval [x _p-r _p-100, x _p-r _p-20] go up the horizontal first order difference value sum of calculating each point and 20 points in front.Concrete computing formula is: work as x _p-r _p-100＜x＜x _p-r _p-20 o'clock, $(x,y_p)=\underset{i=0}{\overset{20}{\mathrm{\Σ}}}D(x-i,y_p),$ D (x-i, y wherein _p) be coordinate points (x-i, y _p) horizontal first order difference value.And in this interval, find out S (x, y _p) coordinate points of correspondence (x, y when getting maximal value _p) as iris outward flange point.

Then, a series of iris outward flange points of gained are justified match, obtain the outer peripheral accurate center of circle (x of iris _i, y _i) and radius r _i

Described iris image quality evaluation process comprises the steps:

The step 6. pair iris image of orienting carries out normalized, obtains size and is the normalization iris image of m * n.Concrete computing formula is: $\left\{\begin{array}{c}x(r,\mathrm{\θ})=(1-r)x_p\left(\mathrm{\θ}\right)+{\mathrm{rx}}_i\left(\mathrm{\θ}\right)\\ y(r,\mathrm{\θ})=(1-r)y_p\left(\mathrm{\θ}\right)+{\mathrm{ry}}_i\left(\mathrm{\θ}\right)\end{array}\right.,$ Wherein, r is distributed in interval [0,1], and θ is distributed in interval [0,2 π], and (x _p(θ), y _p(θ)) and (x _i(θ), y _i(θ)) represent iris inward flange point and outward flange point on the θ direction respectively.

Step 7. is calculated the some acutance of normalization iris image, judges whether image definition satisfies the requirement of iris recognition

At first, the some sharpness value f of calculation procedure 6 gained normalization iris images, concrete computing formula is: $f=\underset{i=1}{\overset{m\×n}{\mathrm{\Σ}}}\underset{j=1}{\overset{8}{\mathrm{\Σ}}}|\mathrm{dI}/\mathrm{dx}\backslash /m\×n,$ Wherein dI is the difference of the gray-scale value of the gray-scale value of picture element and this point in certain any 8-neighborhood in the image, and dx is the distance of consecutive point, and m and n are respectively the height and width of normalization iris image.

Then, with the some sharpness value f of normalization iris image and predefinedly be used to judge whether clearly threshold values V of iris image _fCompare, if f 〉=V _f, think that then the sharpness of image satisfies the requirement of iris authentication system, otherwise, think the requirement of not satisfying iris authentication system.

Step 8: calculate the number of effective picture element in the normalization iris image, judge whether the visibility of iris image satisfies the requirement of iris recognition

At first, the number K of effective picture element in the statistics normalization iris image, concrete computing formula is: $K=\underset{x}{\mathrm{\&Sigma;}}\underset{y}{\mathrm{\&Sigma;}}n(x,y),$ Wherein $n(x,y)=\left\{\begin{array}{cc}1,& V_{\mathrm{lash}}\&le;I(x,y)\&le;V_{\mathrm{eyelid}}\\ 0,& I(x,y)>V_{\mathrm{eyelid}}\mathrm{orI}(x,y)<V_{\mathrm{lash}}\end{array}\right.,,$ V wherein _LashAnd V _EyelidBe the thresholding that takes a decision as to whether the picture element that is positioned at eyelashes zone and palpebral region, (x y) is the gray scale of image to I.

Then, with the number K of effective picture element and predefinedly be used to judge whether iris image exists the threshold values V of eyelid and eyelashes occlusion issue _kCompare, if K 〉=V _k, think that then the visibility of iris image satisfies the requirement of iris authentication system, otherwise, think the requirement of not satisfying iris authentication system.

By above step, just from the original image that contains iris, extract normalized iris image, and judge the requirement whether this image satisfies iris authentication system.

Need to prove:

1. carry out binaryzation in the step 1 and choose a fixing threshold value V _bA fixing threshold value V who chooses _bObtain by a large amount of tests, and select a fixing threshold values to be here because the gray-scale value of the gray-scale value of pupil region and iris region differs very big, even the iris image of taking under different illumination conditions also can guarantee the effect of binaryzation.

2. locate the rough center (x of pupil in the step 3 _o, y _o) be in order to determine to carry out the scope of iris boundary point search.

3. in the step 4 in the iris inward flange point ground searching method, thinking that gray-scale value is exactly the frontier point of pupil greater than the point of threshold value T, is because obviously the increasing progressively of the marginal existence gray-scale value of pupil, when being to be exactly pupil edge greater than a certain value.

4. in the step 5,, be unfavorable for the accurate location of method, must adopt Gaussian function that projection value is carried out smoothing processing because the first order difference horizontal projection curve burr that is obtained by step 4 is a lot.

5. the size of mentioning the normalization iris image in the step 6 is m * n, and value m is the spacing value decision of the θ that got when operating according to normalization, and the spacing value of the r that value n is got when being operated by normalization determines.

6. the picture point sharpness value in the step 7 has mainly characterized the readability of image, and the big more image of some sharpness value f is clear more, and the more little image of f is fuzzy more; Threshold values V _fBe to obtain by a large amount of iris images of same collecting device are tested, this value can the clear and fuzzy iris image of accurate classification.

7. mention V in the step 8 _LashAnd V _Eyelid, it is considered herein that gray-scale value is less than V _LashPicture element be the picture element in eyelashes zone, gray-scale value is greater than V _EyelidBe the picture element of palpebral region, grey value profile is at V _LashAnd V _EyelidBetween be the picture element of iris region; Effectively the big more iris region that is not blocked of picture element number K value is big more, and the eyelid of the more little existence of K value and eyelashes block serious more.

The present invention adopts frontier point search and circle match to combine, and at first realizes the coarse localization in the iris inward flange center of circle by binaryzation, burn into expansion and Gray Projection; Adopt single gray-scale value comparison and circular curve approximating method to determine the inward flange of iris region then, adopt the grey scale difference circular curve approximating method that adds up to determine the outward flange of iris region; According to normalized iris image, calculation level sharpness value and effective pixel points number are from sharpness and two aspect assess image quality of visibility at last.The method that frontier point search and the curve fitting based on shade of gray that adopts the present invention to propose combines can improve Iris Location precision and locating speed effectively; Adopt the quality evaluating method based on an acutance and available point number of the present invention's proposition, improved the versatility of traditional quality assessment algorithm.

Innovation part of the present invention is:

Made full use of the shade of gray information of iris image and the method for curve fitting, carried out curve fitting, thereby the positional information of acquisition iris region reaches the purpose of separating iris by the coordinate that obtains iris outer edge point; And correctly estimated the quality of iris image by the half-tone information of normalization iris image.The present invention at first adopts sciagraphy, to filling to such an extent that the binaryzation iris image carries out level and vertical Gray Projection through hot spot, obtains the rough center of iris inward flange.By scanning to horizontal grey scale curve, find gray-scale value any internal boundary points as iris greater than a certain threshold values, a series of iris inward flange points are justified match, thereby obtain the positional information of iris inward flange.Afterwards,, obtain position coordinates when integrated value is got maximal value as the outward flange point of iris, and then utilize these outward flange points to justify match again by horizontal first order difference being carried out the integration of adjacent 20 points, the outer peripheral positional information of iris.The location that the method for utilizing shade of gray information and circle match to combine is carried out iris region is a characteristic of the present invention, compares with two general step iris locating methods, and it is high 5 percentage points that locating accuracy of the present invention is wanted, and speed improves 60%.During quality evaluation, the present invention adds up the some acutance of normalization iris image and effective picture element number, by comparing with predefined threshold values, thereby iris image quality has been carried out correct evaluation, and versatility is very strong.

Description of drawings

Fig. 1 is 8-neighborhood synoptic diagram of pixel p; Wherein, r is level and vertical next-door neighbour's pixel, and s is diagonal angle neighbour's pixel.

Fig. 2 is the process flow diagram of a kind of pre-processing method for iris image of the present invention.

Embodiment

Adopt method of the present invention, at first use C language and assembly language to write the iris preprocessor; Adopt the original image of CMOS or CCD camera head automatic shooting iris then; Then the iris original image that photographs is input to as source data in the iris preprocessor of DSP embedded system and handles; Through Iris Location and image quality measure, up-to-standard iris image location back output comprises the iris normalized image of enriching texture information.Adopt 2400 to take different illumination conditions good, that comprise different people, the different gray scale iris image of taking posture as source data, locating accuracy is 97.5%, and the location piece image only needs 100ms.

In sum, method of the present invention makes full use of the half-tone information of iris image, in conjunction with the circle fitting method, thereby realizes iris region being provided from the iris original image that is provided rapidly and accurately and making quality evaluation accurately.

Claims (4)

1. a pre-processing method for iris image is characterized in that, comprise two processes of Iris Location and iris image quality assessment: described Iris Location process comprises the steps:

The original iris gray level image that step 1. pair camera head is gathered carries out binary conversion treatment, and gray-scale value is greater than threshold value V in the original-gray image _bThe gray-scale value of picture element to compose be 1, less than threshold value V _bThe gray-scale value of picture element to compose be 0;

The bianry image that obtains in the step 2. pair step 1 carries out closure operation in the mathematical morphology with structural element B to it and eliminates little cavity in the bianry image; Described structural element B is one 7 * 7 a matrix, and the value of the element in the intermediate approximation border circular areas is 1, and the value of all the other elements is 0;

Step 3. is calculated the rough center of circle of iris inward flange

At first obtain the level and vertical Gray Projection of image in the calculation procedure 2: the computing formula of horizontal projection is: $S_h\left(x\right)=\underset{y=1}{\overset{N}{\mathrm{\&Sigma;}}}I(x,y),$ The computing formula of vertical Gray Projection is: $S_v\left(y\right)=\underset{x=1}{\overset{M}{\mathrm{\&Sigma;}}}I(x,y);$ S wherein _h(x) the expression horizontal ordinate is the Gray Projection value of x, S _v(y) the expression ordinate is the Gray Projection value of y, and M, N are the width and the height of original iris gray level image, and (x y) is position (x, the gray-scale value of picture element y) to I;

Next searches for Gray Projection, gets S _h(x) the horizontal ordinate x during minimum value _oWith get S _v(y) the ordinate y during minimum value _o, with (x _o, y _o) be considered as the rough center of circle of iris inward flange;

Step 4. is calculated the accurate center of circle and the radius of iris inward flange

At first, to coordinate points (x _o, y _o) near several rows carry out the search of iris inward flange point, obtain the coordinate of a series of iris inward flange points; Then, a series of iris inward flange points of gained are justified match, obtain the accurate center of circle (x of iris inward flange _p, y _p) and radius r _p

Step 5. is calculated the outer peripheral accurate center of circle of iris and radius

At first, get coordinate points (x _p, y _p) near several rows, on each row that takes out, carry out the search of iris outer rim point, obtain the coordinate of a series of iris outward flange points; Then, a series of iris outward flange points of gained are justified match, obtain the outer peripheral accurate center of circle (x of iris _i, y _i) and radius r _i

Described iris image quality evaluation process comprises the steps:

The step 6. pair iris image of orienting carries out normalized, obtains size and is the normalization iris image of m * n;

Step 7. is calculated the some acutance of normalization iris image, judges requirement that whether image definition satisfy iris recognition at first, the some sharpness value f of calculation procedure 6 gained normalization iris images, and specifically computing formula is: $f=\underset{i=1}{\overset{m\&times;n}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{8}{\mathrm{\&Sigma;}}}|\mathrm{dI}/\mathrm{dx}|/m\&times;n,$ Wherein dI is the difference of the gray-scale value of the gray-scale value of picture element and this point in certain any 8-neighborhood in the image, and dx is the distance of consecutive point, and m and n are respectively the height and width of normalization iris image;

Then, with the some sharpness value f of normalization iris image and predefinedly be used to judge whether clearly threshold values V of iris image _fCompare, if f 〉=V _f, think that then the sharpness of image satisfies the requirement of iris authentication system, otherwise, think the requirement of not satisfying iris authentication system;

Step 8: calculate the number of effective picture element in the normalization iris image, judge whether the visibility of iris image satisfies the requirement of iris recognition

At first, the number K of effective picture element in the statistics normalization iris image, concrete computing formula is: $K=\underset{x}{\mathrm{\&Sigma;}}\underset{y}{\mathrm{\&Sigma;}}n(x,y),$ Wherein $n(x,y)=\left\{\begin{array}{cc}1,& V_{\mathrm{lash}}\&le;I(x,y)\&le;V_{\mathrm{eyelid}}\\ 0,& I(x,y)>V_{\mathrm{eyelid}}\mathrm{orI}(x,y)<V_{\mathrm{lash}}\end{array},,\right.$ V wherein _LashAnd V _EyelidBe the thresholding that takes a decision as to whether the picture element that is positioned at eyelashes zone and palpebral region, (x y) is the gray scale of image to I;

Then, with the number K of effective picture element and predefinedly be used to judge whether iris image exists the threshold values V of eyelid and eyelashes occlusion issue _kCompare, if K 〉=V _k, think that then the visibility of iris image satisfies the requirement of iris authentication system, otherwise, think the requirement of not satisfying iris authentication system.

2. a kind of pre-processing method for iris image according to claim 1 is characterized in that, the searching method of the inward flange point of iris described in the step 4 is: at ordinate is y _oIn this delegation, with (x _o, y _o) be the center, along continuous straight runs is searched for the point of pixel gray-scale value greater than threshold value T left, stops search immediately when searching the pixel gray-scale value greater than threshold value T, writes down the coordinate (x of this moment _l, y _o) as the coordinate of an iris inward flange point, carry out along continuous straight runs search to the right by same mode again, obtain another iris inward flange point coordinate (x _r, y _o); To coordinate points (x _o, y _o) near other several rows carry out above-mentioned search equally, obtain other iris inward flange point.

3. a kind of pre-processing method for iris image according to claim 1 is characterized in that, the searching method of the outer rim point of iris described in the step 5 is: for ordinate is y _pThis delegation,

1), coordinates computed point (x _p, y _p) the horizontal first order difference of being expert at, concrete computing formula is: work as x _pD (x, y during＜x＜N-5 _p)=I (x+5, y _p)-I (x, y _p); As 5＜x≤x _pThe time D (x, y _p)=I (x-5, y _p)-I (x, y _p), wherein, D (x, y _p) denotation coordination point (x, y _p) horizontal first order difference value, I (x, y) (x, gray-scale value y), N are original iris gray level image width to the denotation coordination point;

2), ordinate is y _pIn the delegation, at interval [x _p+ r _p+ 20, x _p+ r _p+ 100] go up the horizontal first order difference value sum of calculating each point and 20 points in back, concrete computing formula is for working as x _p+ r _p+ 20＜x＜x _p+ r _p+ 100 o'clock, $S(x,y_p)=\underset{i=0}{\overset{20}{\mathrm{\&Sigma;}}}D(x+i,y_p),$ D (x+i, y wherein _p) be coordinate points (x+i, y _p) horizontal first order difference value; And in this interval, find out S (x, y _p) coordinate points of correspondence (x, y when getting maximal value _p) as iris outer rim frontier point;

3), ordinate is y _pIn the delegation, at interval [x _p-r _p-100, x _p-r _p-20] go up the horizontal first order difference value sum of calculating each point and 20 points in front, concrete computing formula is: work as x _p-r _p-100＜x＜x _p-r _p-20 o'clock, $S(x,y_p)=\underset{i=0}{\overset{20}{\mathrm{\&Sigma;}}}D(x-i,y_p),$ D (x-i, y wherein _p) be coordinate points (x-i, y _p) horizontal first order difference value; And in this interval, find out S (x, y _p) coordinate points of correspondence (x, y when getting maximal value _p) as iris outward flange point;

For coordinate points (x _p, y _p) near other several rows, carry out above-mentioned search equally, obtain other iris outward flange point.

4. a kind of pre-processing method for iris image according to claim 1 is characterized in that, the concrete computing formula of the normalized of iris image described in the step 6 is: $\left\{\begin{array}{c}x(r,\mathrm{\&theta;})=(1-r)x_p\left(\mathrm{\&theta;}\right)+r{x}_i\left(\mathrm{\&theta;}\right)\\ y(r,\mathrm{\&theta;})=(1-r)y_p\left(\mathrm{\&theta;}\right)+r{y}_i\left(\mathrm{\&theta;}\right)\end{array}\right.,$ Wherein, r is distributed in interval [0,1], and θ is distributed in interval [0,2 π], and (x _p(θ), y _p(θ)) and (x _i(θ), y _i(θ)) represent iris inward flange point and outward flange point on the θ direction respectively.

Images