CN110032936A - The maximum round method for extracting region of printenv hand back vein - Google Patents

The maximum round method for extracting region of printenv hand back vein Download PDF

Info

Publication number
CN110032936A
CN110032936A CN201910174735.9A CN201910174735A CN110032936A CN 110032936 A CN110032936 A CN 110032936A CN 201910174735 A CN201910174735 A CN 201910174735A CN 110032936 A CN110032936 A CN 110032936A
Authority
CN
China
Prior art keywords
circle
maximum
hand
center
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910174735.9A
Other languages
Chinese (zh)
Other versions
CN110032936B (en
Inventor
刘富
姜守坤
侯涛
康冰
李丁园
刘云
王柯
苗岩
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN201910174735.9A priority Critical patent/CN110032936B/en
Publication of CN110032936A publication Critical patent/CN110032936A/en
Application granted granted Critical
Publication of CN110032936B publication Critical patent/CN110032936B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/28—Quantising the image, e.g. histogram thresholding for discrimination between background and foreground patterns
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/14—Vascular patterns

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Image Analysis (AREA)
  • Collating Specific Patterns (AREA)

Abstract

A kind of maximum round method for extracting region of printenv hand back vein, belongs to biometrics identification technology field.Method the purpose of the present invention is carrying out direction correction for the back of the hand maximum border circular areas, thus to the maximum round method for extracting region of the printenv hand back vein of the round extracted region of the maximum of dorsal area and correction.Step of the invention is: 1. use condition random field image segmentation algorithm carries out binary conversion treatment to hand back vein image；2. utilizing the Delaunay Triangulation of MATLAB function delaunay and triangulation coordinates computed sequence；3. finding maximum radius and corresponding central coordinate of circle, the maximum round area image Cmax of the back of the hand is obtained；4. choosing the profile point near little finger, and the distance between calculating and the center of circle, the as vertex at little finger；5. calculating the angle in the direction OA and horizontal direction.The invention has the advantages that the round extracting method of maximum of the invention does not need to carry out parameter and time complexity is low；A kind of method that the present invention provides border circular areas direction correction.

Description

The maximum round method for extracting region of printenv hand back vein
Technical field
The invention belongs to biometrics identification technology fields.
Background technique
In the information age, the occasion for needing to carry out authentication is more and more, by the biography of password, key, magnetic card etc. System authentication mode has been unable to meet the demand of people.Biometrics identification technology is gradually replacing conventional authentication mode, has become For an important branch of field of identity authentication.Wherein Palm-dorsal vein recognition because its own antifalsification height, internal feature etc. it is excellent Point becomes an important component of biometrics identification technology.In order to reduce the data processing amount in identification process and mention High accuracy of identification, needs in identification process, carries out region of interesting extraction.Conventional method is all to extract rectangular area, and justify The region area that shape region includes is bigger, and information is richer.The time complexity of existing border circular areas extracting method is relatively high, It is existing the back of the hand border circular areas extracting method below:
1) maximum inscribed circle region is searched for directly in dorsal area, using each pixel as the center of circle, calculates the center of circle to the back of the hand The distance of edges of regions, minimum range are the corresponding radius in the current center of circle, and the maximum value in all radiuses is maximum border circular areas Radius, the corresponding center of circle is the center of circle of maximum border circular areas, and the time complexity of the algorithm is that wherein N is dorsal area Number of pixels, M be the back of the hand wire-frame image vegetarian refreshments length；
2) mass center for calculating dorsal area calculates the distance for arriving the back of the hand edge, minimum range is both correspondence using mass center as the center of circle Radius, the time complexity of the algorithm is N, and N is the area of dorsal area；
3) mass center for calculating dorsal area, in the region of center of mass selection fixed size as center of circle region, in the area The center of circle is scanned in domain, and calculates the distance for arriving the back of the hand edge, finds maximum radius and the corresponding center of circle, the time of the algorithm Complexity is N × M, and wherein n is the area in fixed size region, and m is the length of the back of the hand wire-frame image vegetarian refreshments, when the value of n compares When big, the complexity of algorithm is very high, and the round maximum closer to dorsal area of the maximum of extraction is round, when the value of n is smaller When, the complexity reduction of algorithm, but the center of circle extracted will appear deviation dorsal area maximum round the case where corresponding to the center of circle, circle Shape difference in areas is situation described in (2) when n takes 1, is situation described in (1) when n takes N away from also bigger；
4) using Wei Nuotu be calculated it is some may be the center of circle vertex, calculate vertex backgear exterior feature in one's hands distance obtain accordingly The center of circle and radius, the time complexity of the algorithm be M^2, M is the number of the back of the hand wire-frame image vegetarian refreshments；
First three algorithm is influenced by parameter selection, and for different images, the circle that fixed parameter not can guarantee extraction is whole The maximum of a dorsal area is round.4th kind of algorithm needs to be traversed for entire the back of the hand profile and all vertex Wei Nuotu, and the time is complicated Degree is high.
Summary of the invention
Method the purpose of the present invention is carrying out direction correction for the back of the hand maximum border circular areas, thus to dorsal area Maximum circle extracted region and the maximum round method for extracting region of the printenv hand back vein of correction.
Step of the invention is:
1. use condition random field image segmentation algorithm carries out binary conversion treatment to hand back vein image, dorsal area is obtained Bianry image BW, in bianry image searching value be 0.5 contour, contour is the profile diagram of dorsal area, from figure As the first left profile point of bottom, entire profile diagram is traversed according to clock-wise order, and by the XY coordinate of all profile points It is stored in variable X according to traversal order, obtains an orderly profile point coordinate sequence X2×N, N is the back of the hand profile point number, profile point Coordinate is (X (1, i), X (2, i)), i=1,2 ..., N；
2. utilizing MATLAB function delaunay and triangulation coordinates computed sequence X2×NDelaunay triangle Subdivision obtains triangulation network TRM×6, M is the number of triangulation network intermediate cam shape, and the vertex of triangulation network intermediate cam shape is the back of the hand profile point, Vertex (A, B, C are vertex number) coordinate is respectively
(TR (i, 1), TR (i, 2)), (TR (i, 3), TR (i, 4)), (TR (i, 5), TR (i, 6)), wherein i=1,2, M, according to Distance between two points formula dis=sqrt (xx1-xx2)2+(yy1-yy2)2), the side length of all M triangles is calculated, is denoted as disM×3, according to the cosine law (cos thetaA=(b2+c2-a2The angle of)/2bc, thetaA for vertex A, a, b, c are respectively Side BC, AC, the length of AB calculate the angle of all triangular apex, judge triangle class according to the angle at Atria angle Type deletes obtuse angle and right angled triangle, only retains acute triangle, obtain acute triangle apex coordinate set Acuten×6, so Acute triangle unfaithful intention coordinate and circumradius are calculated afterwards thetaAiFor the angle of i-th of triangular apex A, aiFor the length of i-th triangle edges BC, the n center of circle [(x is obtained1, y1), (x2, y2) ..., (xn, yn)] and radius [r1, r2..., rn], n is acute triangle number, and the center of circle and radius one are a pair of It answers；
3. finding maximum radius and corresponding central coordinate of circle, maximum radius is maximum round zone radius rc, right The center of circle answered is the center of circle (x of maximum border circular areasc, yc), with the center of circle (xc, yc) and radius rcThe border circular areas surrounded is hand Maximum inscribed circle in backgear exterior feature, corresponding circular image is the back of the hand border circular areas, and it is corresponding then to extract greatest circle region Hand Vein Image obtains the maximum round area image Cmax of the back of the hand；
4. choosing the profile point near little finger
[mux1, muy1), (mux2, muy2) ..., (muxl, muyl)], and calculate and the center of circle (xc, yc) the distance between Dis (i) =sqrt (muxi-xc)2+(muyi-yc)2), i=1,2 ..., l adjust the distance Dis progress smoothly using MATLAB function smooth Processing, finds profile point (x corresponding apart from maximum valued, yd), the as vertex at little finger；5. calculating the direction OA and level side To angle(xc, yc) be greatest circle central coordinate of circle, (xd, yd) little finger vertex seat Mark, rotates 90- θ around picture centre for dorsal area image Cmax counterclockwise, makes 90 degree of angle of the direction OA and horizontal direction, Obtain stable the back of the hand maximum border circular areas Circle.
The invention has the advantages that the round extracting method of maximum of the invention does not need to carry out parameter and time complexity is low； A kind of method that the present invention provides border circular areas direction correction.
Detailed description of the invention
Fig. 1 is the maximum round area schematic of the back of the hand；
Fig. 2 is the maximum round extracted region flow chart of the back of the hand；
Fig. 3 is Hand Vein Image；
Fig. 4 is the back of the hand profile diagram；
Fig. 5 is acute triangle in the triangulation network；
Fig. 6 is the center of circle and little finger region distance curve；
Fig. 7 is the horizontal sextant angle schematic diagram before the correction of direction；
Fig. 8 is the horizontal sextant angle schematic diagram after the correction of direction；
Fig. 9 is the back of the hand maximum border circular areas before the correction of direction；
Figure 10 is the back of the hand maximum border circular areas after the correction of direction；
Figure 11 is the relational graph of the center of circle and triangle, wherein (a) is acute triangle, is (b) right angled triangle, is (c) obtuse angle Triangle.
Specific embodiment
Step of the invention is:
1. use condition random field image segmentation algorithm carries out binary conversion treatment to hand back vein image, the two of dorsal area are obtained It is worth image BW, the contour that searching value is 0.5 in bianry image, contour is the profile diagram of dorsal area, from image bottom The first left profile point in portion traverses entire profile diagram according to clock-wise order, and by the XY coordinate of all profile points according to Traversal order is stored in variable X, obtains an orderly profile point coordinate sequence X2×N, N is the back of the hand profile point number, profile point coordinate For (X (1, i), X (2, i)), i=1,2 ..., N；
2. utilizing MATLAB function delaunay and triangulation coordinates computed sequence X2×NDelaunay triangle cut open Point, obtain triangulation network TRM×6, M is the number of triangulation network intermediate cam shape, and the vertex of triangulation network intermediate cam shape is the back of the hand profile point, top Point (A, B, C are vertex number) coordinate is respectively (TR (i, 1), TR (i, 2)), (TR (i, 3), TR (i, 4)), (TR (i, 5), TR (i, 6)), wherein i=1,2 ..., M, according to distance between two points formula dis=sqrt (xx1-xx2)2+(yy1-yy2)2), it calculates The side length of all M triangles, is denoted as disM×3, according to the cosine law (cos thetaA=(b2+c2-a2)/2bcthetaA is The angle of vertex A, a, b, c are respectively side BC, and the length of AC, AB calculate the angle of all triangular apex, according to triangle The angle at three angles judges triangular type, deletes obtuse angle and right angled triangle, only retains acute triangle, obtain acute angle triangle Shape apex coordinate set Acuten×6, then calculate acute triangle unfaithful intention coordinate and circumradiusthetaAiFor the angle of i-th of triangular apex A, aiIt is i-th The length of triangle edges BC obtains the n center of circle [(x1, y1), (x2, y2) ..., (xn, yn)] and radius [r1, r2..., rn], n For acute triangle number, the center of circle and radius are corresponded；
3. finding maximum radius and corresponding central coordinate of circle, maximum radius is maximum round zone radius rc, corresponding The center of circle is the center of circle (x of maximum border circular areasc, yc), with the center of circle (xc, yc) and radius rcThe border circular areas surrounded is the back of the hand wheel Maximum inscribed circle in exterior feature, corresponding circular image are the back of the hand border circular areas, then extract the corresponding the back of the hand in greatest circle region Image obtains the maximum round area image Cmax of the back of the hand；
4. choosing the profile point [(mux near little finger1, muy1), (mux2, muy2) ..., (muxl, muyl)], and calculate with The center of circle (xc, yc) the distance between Dis (i)=sqrt ((muxi-xc)2+(muyl-yc)2), i=1,2 ..., l use MATLAB The function smooth Dis that adjusts the distance is smoothed, and finds profile point (x corresponding apart from maximum valued, yd), as little finger The vertex at place；
5. calculating the angle in the direction OA and horizontal direction(xc, yc) sat for the center of circle of greatest circle Mark, (xd, yd) little finger vertex coordinate, dorsal area image Cmax is rotated into 90- θ around picture centre counterclockwise, makes the side OA It is 90 degree to get to stable the back of the hand maximum border circular areas Circle to the angle with horizontal direction.
The present invention is described in further details below:
The China and foreign countries Fig. 1 lateral curvature line is the back of the hand profile, and circle is the maximum border circular areas in the dorsal area, and five-pointed star is the greatest circle The center of circle of shape, triangle are the acute triangle that the back of the hand profile obtains after triangulation, the radius of the triangle circumscribed circle Be it is maximum in all acute triangles, circle is the circumscribed circle of the acute triangle, and the center of circle is circumcentre of a triangle, and Fig. 6 is The distance Curve of little finger adjacent contours and the center of circle, wherein five-pointed star be little finger at the farthest point of circle center distance, i.e., small thumb Vertex at finger, the O point in Fig. 7 are the center of circle, and A point is at little finger and the maximum of points of circle center distance, horizontal sextant angle are OA and water The angle of flat axis, the direction correction in Fig. 8 is that OA is rotated to vertical direction.
In view of above-mentioned the maximum round extracted region process the problem of, the present invention proposes that a kind of printenv the back of the hand is quiet The maximum round method for extracting region of arteries and veins, specifically includes the following steps:
1, Hand Vein Image binaryzation and contours extract: carrying out binarization operation to Hand Vein Image, divide the image into prospect and background, Wherein prospect is hand region, carries out Morphological scale-space to bianry image, then extracts the profile of hand region；
2, it calculates the triangulated graph of profile: calculating the triangulated graph of the back of the hand profile, obtain all vertex of a triangle (triangles The node of subdivision), and calculate the side length of triangle；
3, acute triangle detection is with maximum round extracted region: detecting, mentions to all triangles that triangulation obtains Acute triangle therein is taken, and calculates the circumradius of acute triangle, the maximum circumscribed circle of radius is dorsal area Maximum border circular areas, triangle excenter are the circular center of circle；
4, vertex is detected at little finger: the center of circle is calculated at a distance from little finger side profile, at little finger at maximum value Vertex；
5, border circular areas direction is corrected: being reference with vertex line at the center of circle and little finger, being rotated up to border circular areas makes line In vertical direction.
Binarization method described in step 1 uses condition random field image segmentation algorithm, can preferably extract the back of the hand Region avoids extracting the situation that mistake occurs in dorsal area under complex environment, and Morphological scale-space is deposited for removing in dorsal area Isolated point, hole, thus extract a complete dorsal area；Contours extract is by finding gray threshold in bianry image For the coordinate of the point of 0.5 place contour, Hand Vein Image profile is obtained；
Triangulated graph described in step 2, which calculates, uses Delaunay algorithm, obtains multiple the back of the hand profiles extracted with step 1 Point is the triangle on vertex, and dorsal area is divided into multiple regions, is not overlapped each other between the region of each triangle；
The detection of acute triangle described in step 3 is detected using the property that three angles of acute triangle are all acute angles, The angle at Atria angle is wherein calculated using the cosine law；
The round extracted region of the maximum is to calculate radius and the center of circle of the acute triangle circumscribed circle detected, and radius is maximum Circle be that maximum in dorsal area is round, the radius and the corresponding center of circle are radius and the center of circle of maximum border circular areas, with The border circular areas that the radius and the center of circle are surrounded is the back of the hand maximum border circular areas；
Vertex detection is by choosing profile point near little finger and calculating it between the center of circle at little finger described in step 4 Distance, apart from maximum point be little finger at vertex, corresponding coordinate be little finger vertex coordinate；
The correction of border circular areas direction described in step 5 is to choose reference line to correct for direction, to guarantee to same the back of the hand When different gestures, the collected Hand Vein Image of different time carry out border circular areas extraction, there is robustness, direction is corrected with circle For the line on vertex as reference, being rotated up to border circular areas makes line be in vertical side at the little finger extracted in the heart and step 4 To this is using in the case of clenching fist, and root is opposite with circle center line connecting stable on the outside of little finger, thus using it to maximum circle The correction of domain direction；
As a further improvement of the present invention, the gray threshold in the contours extract can select between 0-1 it is any its His numerical value；
As a further improvement of the present invention, the described acute triangle detection can select triangle wherein both sides quadratic sum is small In third side square property detected；
As a further improvement of the present invention, border circular areas direction correction can revolve the center of circle and little finger vertex line Go to any equal angular.
A, (two class of parameter setting is corresponding dorsal area image and back respectively to use condition random field image segmentation algorithm Scape image, condition random field image segmentation are general-purpose algorithms) binary conversion treatment is carried out to hand back vein image, obtain dorsal area (dorsal area is white (value for 1) to the bianry image BW of (there are the regions of hand in image), and background area is that (value is black 0)), the contour that searching value is 0.5 in bianry image, contour are that (dorsal area is contour for the profile diagram of dorsal area Above line, below contour, the enclosed graphic zone that contour surrounds is dorsal area for background area), from image base First left profile point (Far Left profile point at wrist), traverse entire profile diagram according to clock-wise order, and will own The XY coordinate of profile point is stored in variable X according to traversal order, obtains an orderly profile point coordinate sequence X2×N, N is the back of the hand wheel Exterior feature point number, profile point coordinate is (X (1, i), X (2, i)), i=1,2 ..., N；
B, MATLAB function delaunay and triangulation coordinates computed sequence X is utilized2×NDelaunay triangle cut open Point, obtain triangulation network TRM×6, M is the number of triangulation network intermediate cam shape, and the vertex of triangulation network intermediate cam shape is the back of the hand profile point, top Point (A, B, C are vertex number) coordinate is respectively (TR (i, 1), TR (i, 2)), (TR (i, 3), TR (i, 4)), (TR (i, 5), TR (i, 6)), wherein i=1,2 ..., M, according to distance between two points formula dis=sqrt ((xx1-xx2)2+(yy1-yy2)2), wherein (xx1, yy1) and (xx2, yy2) be corresponding two apex coordinates in Atria side (when calculating the distance between vertex AB, (xx1, yy1) and (xx2, yy2) respectively correspond (TR (i, 1),TR (i, 2)) and (TR (i, 3),TR (i, 4))), calculate all M three Angular side length, is denoted as disM×3(disI, 1, disI, 2And disI, 3Indicate that i-th of three sides of a triangle is long, i=1,2, L, M), according to the cosine law ((cos thetaA=(b2+c2-a2(thetaA is the angle of vertex A to)/2bc, and a, b, c is respectively side The length of BC, AC, AB)) angle that calculates all triangular apex, triangle class is judged according to the angle at Atria angle Type deletes obtuse angle and right angled triangle, only retains acute triangle, obtain acute triangle apex coordinate set, wherein (Acute (i, 3), Acute (i, 4)), Acute (i, 5), Acute (i, 6)), i=1,2 ..., n is the vertex of acute triangle Then coordinate calculates acute triangle unfaithful intention (the circumscribed circle center of circle) coordinate and circumradiusI= 1,2 ..., n (thetaAiIt is the length of i-th of triangle edges BC for the angle of i-th of triangular apex A), obtain the n center of circle [(x1, y1), (x2, y2) ..., (xn, yn)] and radius [r1, r2..., rn] (n is acute triangle number, the center of circle and radius one One is corresponding)；
C, all radiuses are traversed, maximum radius and corresponding central coordinate of circle are found.Maximum radius is maximum border circular areas Radius rc, the corresponding center of circle is the center of circle (x of maximum border circular areasc, yc).With the center of circle (xc, yc) and radius rcThe circle surrounded Shape region is the maximum inscribed circle in the back of the hand profile, and corresponding circular image is the back of the hand border circular areas, then extracts greatest circle The corresponding Hand Vein Image in region obtains the maximum round area image Cmax of the back of the hand:
D, the profile point [(mux near little finger is chosen1, muy1), (mux2, muy2) ..., (muxl, muyl)], and calculate with The center of circle (xc, yc) the distance between Dis (i)=sqrt ((muxi-xc)2+(muyi-yc)2), i=1,2 ..., l use MATLAB The function smooth Dis that adjusts the distance is smoothed, and is traversed all distances later, is found profile point corresponding apart from maximum value (xd, yd), the as vertex at little finger；
E, the angle in the direction OA and horizontal direction is calculated(xc, yc) be greatest circle central coordinate of circle, (xd, yd) little finger vertex coordinate, dorsal area image Cmax is rotated into 90- θ around picture centre counterclockwise, make the direction OA with The angle of horizontal direction is 90 degree to get to stable the back of the hand maximum border circular areas Circle.
ROI is extracted
By during four steps, needing to carry out secondary O (N^2) distance of N* (N-3) (N is essentially profile point number) known to analysis It calculates, ROI extraction time is slightly long.Therefore we improve for this part, and the maximum inscribed circle region of extraction meets following Condition:
1. the radius of maximum inscribed circle is the maximum circle of radius in the back of the hand contour area；
2. the corresponding triangle of circumscribed circle is acute triangle；Unfaithful intention must be in triangle interior, therefore must be in the back of the hand contoured interior. It ensure that the center of circle is located inside dorsal area simultaneously.Figure 11 is the relationship of triangle and unfaithful intention.
Specific extraction step:
1. extracting the back of the hand profile；
2. calculating the triangulation and unfaithful intention of profile；
3. judging triangle classification, extract acute triangle part (three angles are all acute angles)；
4. finding maximum circumradius, positioning centre according to formula (1)-(3).Abc is triangle side length, and p is semi-perimeter, s For triangle area；
5. angle adjusts.It is found by analysis, after extracting maximum inscribed circle, little finger joint has apart from the center of circle most Big value point (assuming that the center of circle of same the back of the hand and maximum of points immobilize), ROI is rotated to unified direction with this by us.
" L " in above formula represents ellipsis " ... ".
According to above-mentioned steps, hand back vein image is verified.
The maximum round extracted region of the back of the hand is carried out using the flow chart in Fig. 2, step a institute is carried out to the Hand Vein Image of Fig. 3 The contours extract stated, obtained profile sequence are as shown in Figure 4.
Fig. 4 triangulation for carrying out step b is calculated, and extracts the acute triangle in the triangulation network, as shown in Figure 5 (figure In there are two triangle, small does not see), the corresponding circumradius length of triangle is 90 and 1.17 respectively.
By step c, the center of circle that the center of circle that radius is 90 is maximum border circular areas is obtained, the o point in Fig. 7 is circle The center of circle.Circle in Fig. 7 is the maximum border circular areas in the back of the hand profile, and Fig. 9 is the round Hand Vein Image extracted Cmax。
Shown in solid, the distance between profile point and the center of circle by step d, in the profile point such as Fig. 7 of extraction near A point Curve is as shown in fig. 6, wherein the star point in Fig. 6 is point apart from maximum value, the A point in corresponding diagram 7.
By step e, the angle of OA and horizontal direction, angle as shown in Figure 7, then by image around the center of circle inverse time are calculated Needle rotation, obtains postrotational border circular areas Circle, as shown in Figure 10, the corresponding rotation schematic diagram in the position Fig. 8.
The region of interesting extraction that 6 different images are carried out to the above method regard image 1 as reference picture, calculates The angular deviation of other images and image 1, concrete outcome are as follows:

Claims (1)

1. a kind of maximum round method for extracting region of printenv hand back vein, it is characterised in that:
1. use condition random field image segmentation algorithm carries out binary conversion treatment to hand back vein image, the two of dorsal area are obtained It is worth image BW, the contour that searching value is 0.5 in bianry image, contour is the profile diagram of dorsal area, from image bottom The first left profile point in portion traverses entire profile diagram according to clock-wise order, and by the XY coordinate of all profile points according to Traversal order is stored in variable X, obtains an orderly profile point coordinate sequence, N is the back of the hand profile point number, and profile point is sat It is designated as；
2. utilizing MATLAB function delaunay and triangulation coordinates computed sequenceDelaunay Triangulation, Obtain the triangulation network, M is the number of triangulation network intermediate cam shape, and the vertex of triangulation network intermediate cam shape is the back of the hand profile point, vertex (A, B, C are vertex number) coordinate is respectively, According to distance between two points formula, calculate the side of all M triangles It is long, it is denoted as, according to the cosine law, thetaA is the angle of vertex A Degree, a, b, c are respectively side BC, and the length of AC, AB calculate the angle of all triangular apex, according to the angle at Atria angle Degree judges triangular type, deletes obtuse angle and right angled triangle, only retains acute triangle, obtain acute triangle apex coordinate Set, then calculate acute triangle unfaithful intention coordinate and circumradius,For the angle of i-th of triangular apex A,For the length of i-th of triangle edges BC, the n center of circle is obtainedAnd radius, n be acute triangle number, the center of circle and Radius corresponds；
3. finding maximum radius and corresponding central coordinate of circle, maximum radius is maximum round zone radius, corresponding The center of circle is the center of circle of maximum border circular areas, with the center of circleAnd radiusThe border circular areas surrounded is hand Maximum inscribed circle in backgear exterior feature, corresponding circular image is the back of the hand border circular areas, and it is corresponding then to extract greatest circle region Hand Vein Image obtains the maximum round area image Cmax of the back of the hand；
4. choosing the profile point near little finger, And calculate and the center of circleThe distance between, It is smoothed using the MATLAB function smooth Dis that adjusts the distance, finds profile point corresponding apart from maximum value, Vertex as at little finger；
5. calculating the angle in the direction OA and horizontal direction,It is sat for the center of circle of greatest circle Mark,Dorsal area image Cmax is rotated 90- θ around picture centre counterclockwise, makes the side OA by the coordinate on little finger vertex It is 90 degree to get to stable the back of the hand maximum border circular areas Circle to the angle with horizontal direction.
CN201910174735.9A 2019-03-08 2019-03-08 Method for extracting maximum circular area of non-parameter hand back vein Active CN110032936B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910174735.9A CN110032936B (en) 2019-03-08 2019-03-08 Method for extracting maximum circular area of non-parameter hand back vein

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910174735.9A CN110032936B (en) 2019-03-08 2019-03-08 Method for extracting maximum circular area of non-parameter hand back vein

Publications (2)

Publication Number Publication Date
CN110032936A true CN110032936A (en) 2019-07-19
CN110032936B CN110032936B (en) 2022-08-09

Family

ID=67235150

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910174735.9A Active CN110032936B (en) 2019-03-08 2019-03-08 Method for extracting maximum circular area of non-parameter hand back vein

Country Status (1)

Country Link
CN (1) CN110032936B (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111881733A (en) * 2020-06-17 2020-11-03 艾普工华科技（武汉）有限公司 Worker operation step specification visual identification judgment and guidance method and system
CN114454266A (en) * 2022-02-14 2022-05-10 三菱电机自动化(中国)有限公司 Log cutting device, method and computer readable medium
US11490986B2 (en) * 2019-10-11 2022-11-08 Beyeonics Surgical Ltd. System and method for improved electronic assisted medical procedures
CN115359249A (en) * 2022-10-21 2022-11-18 山东圣点世纪科技有限公司 Palm image ROI region extraction method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090171627A1 (en) * 2007-12-30 2009-07-02 Olson Eric S System and Method for Surface Reconstruction from an Unstructured Point Set
CN102682279A (en) * 2011-02-01 2012-09-19 星友科技股份有限公司 High-speed fingerprint feature comparison system and method implemented by classified triangles
US20140307067A1 (en) * 2006-12-28 2014-10-16 David Byron Douglas Method and apparatus for three dimensional viewing of images
CN106919941A (en) * 2017-04-26 2017-07-04 华南理工大学 A kind of three-dimensional finger vein identification method and system
CN107729883A (en) * 2017-11-22 2018-02-23 吉林大学 A kind of vein image area-of-interest exacting method

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140307067A1 (en) * 2006-12-28 2014-10-16 David Byron Douglas Method and apparatus for three dimensional viewing of images
US20090171627A1 (en) * 2007-12-30 2009-07-02 Olson Eric S System and Method for Surface Reconstruction from an Unstructured Point Set
CN102682279A (en) * 2011-02-01 2012-09-19 星友科技股份有限公司 High-speed fingerprint feature comparison system and method implemented by classified triangles
CN106919941A (en) * 2017-04-26 2017-07-04 华南理工大学 A kind of three-dimensional finger vein identification method and system
CN107729883A (en) * 2017-11-22 2018-02-23 吉林大学 A kind of vein image area-of-interest exacting method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
WANG SHENGHUA ETC: ""a new han shape positioning algorithm based on Voronoi diagram"", <PROCEEDINGS OF THE 35TH CHINESE CONTROL CONFERENCE> *
王升华: ""基于DSP的嵌入式手形识别系统算法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *
魏鹏飞: ""手指静脉图像分割方法研究"", 《中国优秀博硕士学位论文全文数据库(硕士)》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11490986B2 (en) * 2019-10-11 2022-11-08 Beyeonics Surgical Ltd. System and method for improved electronic assisted medical procedures
US11918424B2 (en) 2019-10-11 2024-03-05 Beyeonics Surgical Ltd. System and method for improved electronic assisted medical procedures
CN111881733A (en) * 2020-06-17 2020-11-03 艾普工华科技（武汉）有限公司 Worker operation step specification visual identification judgment and guidance method and system
CN111881733B (en) * 2020-06-17 2023-07-21 艾普工华科技（武汉）有限公司 Method and system for judging and guiding worker operation step standard visual identification
CN114454266A (en) * 2022-02-14 2022-05-10 三菱电机自动化(中国)有限公司 Log cutting device, method and computer readable medium
CN114454266B (en) * 2022-02-14 2023-09-08 三菱电机自动化(中国)有限公司 Log cutting device, method and computer readable medium
CN115359249A (en) * 2022-10-21 2022-11-18 山东圣点世纪科技有限公司 Palm image ROI region extraction method and system

Also Published As

Publication number Publication date
CN110032936B (en) 2022-08-09

Similar Documents

Publication Publication Date Title
CN110032936A (en) The maximum round method for extracting region of printenv hand back vein
Kanhangad et al. Contactless and pose invariant biometric identification using hand surface
CN105046252B (en) A kind of RMB prefix code recognition methods
Zuo et al. On a methodology for robust segmentation of nonideal iris images
CN100382751C (en) Canthus and pupil location method based on VPP and improved SUSAN
CN103310196B (en) The finger vein identification method of area-of-interest and direction element
CN107103323A (en) A kind of target identification method based on image outline feature
CN103136525B (en) A kind of special-shaped Extended target high-precision locating method utilizing Generalized Hough Transform
CN109903313A (en) Real-time pose tracking method based on target three-dimensional model
CN105354866A (en) Polygon contour similarity detection method
CN105893981B (en) A kind of human face posture antidote
CN108491786A (en) A kind of method for detecting human face based on hierarchical network and Cluster merging
CN106096569A (en) A kind of finger vein identification method
CN108010082A (en) A kind of method of geometric match
CN108805915A (en) A kind of close-range image provincial characteristics matching process of anti-visual angle change
CN107025449A (en) A kind of inclination image linear feature matching process of unchanged view angle regional area constraint
CN112837263B (en) Identity card information positioning method under complex background
Haala et al. Recognition of road and river patterns by relational matching
CN109147040A (en) Human body dot cloud hole method for repairing and mending based on template
CN110472521A (en) A kind of Pupil diameter calibration method and system
CN109711404A (en) Recognition methods, device and the computer readable storage medium of seal ballot paper full-filling
CN1204531C (en) Human eye location method based on GaborEge model
CN110458167A (en) A kind of metalwork surface curvature line of text antidote
US6735343B2 (en) Polygon finder and pruned tree geometric match method
CN114783010B (en) Extraction method of interest region of palm print image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant