TW201937406A - Non-directional fingerprint and palmprint identification method and non-directional fingerprint and palmprint data creation method - Google Patents

Abstract

The invention relates to a non-directional finger palmprint identification method and a non-directional finger palmprint data creation method. The finger palmprint recognition method senses a user's hand fingerprint and palm print by a finger palm pattern recognition device and generates a fingerprint and The palm print image is transmitted to a processor, and the palm palm image is subjected to a recognition calculation process to obtain a fingerprint and a palm print of a specific finger, and the fingerprint and the palm print of the specific finger are crossed according to the at least one reference line. The identification process generates a complex intersection point, and then the processor calculates the distance between the intersection points to generate a plurality of feature points, and converts the feature points into the digitized feature data and a feature database. A stored profile is compared to produce a comparison result.

Description

Non-directional finger palmprint identification method and non-directional finger palmprint data creation method

The invention relates to a non-directional finger palmprint identification method and a non-directional finger palmprint data establishing method, in particular to a hand fingerprint and a palm print which can achieve non-directional recognition and achieve the palm finger recognition speed. effect.

Biometric technology has always been a key element in the development of safety technology, and common biometric technologies such as retina, iris, face and other identification techniques often seen in movies, as well as fingerprints that are already common ( Fingerprint), palm print, voice pattern recognition, etc., and the most common ones are uniqueness, convenience and the most specific fingerprint identification technology. In order to improve the security of hand fingerprint identification, palm pattern recognition is added, and the conventional hand fingerprint and palm recognition are aimed at one-third or two-thirds of the hands of the actual user through a camera. The user's palm must be in a display on the camera, and the user's one-third or two-thirds of the hand's full palm contour must be constantly adjusted to exactly match the display. After the virtual full palm outline is displayed, the camera will start to capture the full palm image of the actual user and transmit it to a processor, which first performs image recognition on the fingerprint and palm print image of the entire palm of the shot. After processing, the processor compares the hand fingerprint and the palm print image after the identification processing with the fingerprint and palm print corresponding to the user in a database, and if the comparison result is yes, the representative is in the corresponding database. If the comparison result is not, it means that it is not the user in the database, so as to effectively achieve the purpose of identity identification. Although it is known that the camera can be recognized by the non-contact capture of the user's finger and palm (ie, the whole palm) image, it is safer than single fingerprint recognition, but it extends another problem, that is, the conventional processor is in the identification calculation. The method is complicated and takes a long time. For example, the user needs to constantly adjust the distance of the mobile camera to the far or the left or the left and right direction, so that the actual user's full palm contour conforms to the virtual full palm contour in the display screen. The camera will start shooting, so that the input data is long and the input comparison time is also long, and the processor identification calculation is also quite complicated, and the cost is also increased. In addition, since the virtual full palm outline displayed on the display screen has a certain angle (for example, 45 degrees), each finger of the user's hand is also placed corresponding to the contour of the virtual full palm contour. The position of a finger is such that the user can only recognize it in a single fixed angle direction, and the user often encounters it in actual use, because the opening angle position of one of the fingers of the whole palm and the corresponding display screen In the virtual full palm outline, the position of one finger opening angle is a little bit different, which has caused the processor to recognize the error and judge the non-user himself. However, the user is not convenient in use and has long recognition time and identification efficiency. Poor question. In addition, it is known that the way to identify the user's full palm through the camera is easily deceived by the hacker's fake photo of the hand to decipher the camera, making the data easy to leak, resulting in poor security.

The purpose of the present invention is to provide a non-directional effect by using a finger palm identification device for the hand fingerprint of a user's hand and the palm print (ie, the whole palm) to be directly contactlessly identified to achieve the effect of identity recognition. Finger palmprint identification method and method of establishing non-directional finger palmprint data. One of the purposes of this creation is to provide a non-directional finger palmprint identification method and a non-directional finger palmprint data creation method which can improve hand fingerprint and palmprint recognition speed and achieve cost saving. One of the purposes of this creation is to provide a non-directional finger palmprint identification method and a non-directional finger palmprint data creation method which are effective in shortening the recognition time and safety. In order to achieve the above purpose, the present invention provides a non-directional finger palm pattern recognition method, which firstly uses a finger palm pattern recognition device to sense and scan a user's hand fingerprint and palm print, and generates a hand fingerprint and a palm print image to be transmitted to a processor. The processor performs image processing according to the fingerprint and the palm print image to generate a processed palm print image, and then the processor performs a recognition calculation process on the palm print image to obtain two fingers of the palm print image. a valley between the two, and the processor extracts the finger palm image according to the valley points, wherein a certain finger is between two adjacent valley points (ie, a specific finger between two adjacent valley points), and Selecting at least one reference line on the specific finger to extend to the bottom end of the palm, and continuing the processor to perform a recognition calculation process on the palm print image to obtain the fingerprint of the specific finger and the palm print, and according to the at least one reference line Performing cross-identification processing on the fingerprint and palmprint of the specific finger, generating a plurality of intersections, and then the processor intersects the intersections The distance between the two is calculated to generate a plurality of feature points, and the feature points are converted into digitalized feature data and compared with a feature data stored in a feature database to generate a comparison result; Through the design of the method of the invention, the effect of non-directional identification and the speed of hand fingerprint and palmprint recognition can be improved and the cost can be saved, and the effect of shortening the recognition comparison time can be effectively achieved. The invention further provides a non-directional finger palmprint data establishing method, which firstly uses a finger palm pattern recognition device to sense and scan a user's hand fingerprint and palm print, and generates a hand fingerprint and a palm print image to be transmitted to a processor. The image processing is performed according to the fingerprint and the palm print image to generate a processed palm print image, and then the processor performs a recognition calculation process on the palm print image to obtain a finger between the two fingers of the palm print image. a valley point, and the processor recognizes the finger palmprint image according to the valley points, wherein a specific finger is between two adjacent valley points (ie, a specific finger is taken between two adjacent valley points), and the specific finger is Selecting at least one reference line to extend to the bottom end of the palm, and continuing the processor to perform a recognition calculation process on the palm print image to obtain the fingerprint of the specific finger and the palm print, and according to the at least one reference line and the specific finger Hand fingerprint and palmprint are cross-identified to generate complex intersections, and then the processor distances the intersections from each other Performing calculus processing to generate complex feature points, and converting the feature points into digitized feature data and storing them in a feature database for reading use; the design of the method of the present invention enables non-directional The effect of sexual recognition and the improvement of hand fingerprint and palmprint recognition speed and cost savings, and effectively achieve the effect of shortening the recognition comparison time.

The above object of the present invention, as well as its structural and functional features, will be described in accordance with the preferred embodiments of the drawings. The invention relates to a non-directional finger palmprint identification method and a non-directional finger palmprint data establishing method. 1 is a block diagram of a finger recognition device according to a first embodiment of the present invention; FIG. 3A is a schematic diagram of a finger palmprint identification method according to a first embodiment of the present invention; The method selects a valley point, a reference line, and an intersection point between the two fingers in the finger palm image; and FIG. 3B is a schematic diagram of the intersection point and the feature point of the palm palmprint image selection method according to the first embodiment of the present invention; 4A is a schematic view showing the operation of the finger palm marking device of the present invention; FIG. 5A is a schematic view showing the operation of the hand of the user of the present invention on the finger palm identifying device; FIG. 5B is a view of the use of the present invention; The hand of the person performs an operation diagram in the other direction on the palm print recognition device. As shown in the figure, the finger palmprint identification method of the present invention is applied to a finger palm pattern recognition device 1. The finger palm pattern recognition device 1 can be disposed on each product (such as a door, an automobile interior panel or a lift) during actual implementation of the present invention. The machine (not shown) or an electronic product (such as an LED TV, a notebook or a tablet; not shown), and the palm print recognition device 1 of the present embodiment is a flat finger palm recognition device 1. The finger palm identification device 1 has a plurality of sensors 11 and a contact surface 12, the contact surface 12 is in a plane, and is disposed on one side surface of the finger palm recognition device 1 for the user 4 In the first embodiment of the hand 41 (such as the left hand or the right hand), the sensor is placed as a capacitive finger palm sensor. The sensor 11 can be selected from a transparent material. a conductive material such as Indium Tin Oxides (ITO) or Antimony Tin Oxide (ATO), and the sensors 11 are disposed under the contact surface 12 on the palm print identifying device 1 And electrically connected to a corresponding processor 2, wherein the sensors 11 are configured to sense a hand fingerprint and a palm print (ie, a full palm) of the hand 41 of the user 4 directly contacting the contact surface 12. To generate a fingerprint and palm print image and transmit it to the processor 2. The entire palm of the hand 41 of the user 4 can be contactlessly contacted on the corresponding contact surface 12 (as shown in Figs. 5A, 5B). In addition, in the actual implementation of the present invention, the sensors 11 can be directly designed to be formed in a region of each product or the electronic product itself, for example, the sensors 11 are directly formed on the glass door itself, such as by etching. The position of the intermediate portion is directly formed at the left region of the display surface of the LED television such as etching, and the position of the region through the respective products or the electronic device itself is used as a substrate formed by the equal sensor 11, so that the finger can be referred to according to requirements. The palm print recognition device 1 is provided on each desired product or electronic device. In an implementation, as shown in FIG. 4B, FIG. 4B is a schematic diagram of another implementation of the palmprint recognition device of the present invention. As shown in the figure, the finger palm identification device 1 can be designed as a flexible curved palm finger recognition device 1 and the contact surface 12 has a curved surface for the user's hand 41 (such as the left or right hand). Place the entire palm with the affixed contacts. The processor 2 is a central processing unit 2 (CPU) or a microcontroller (MCU). The processor 2 in this embodiment is disposed outside the finger palm recognition device 1 as a separate component, and the processor 2 is powered. The character palm identification device 1 and a feature database 7 corresponding to the connection are in the memory (such as flash memory or random access memory or hard disk or solid state drive). ) for reading or storing the feature data of the corresponding user 4. In one embodiment, the processor 2 is modified to be an element in the fingerprint recognition device 1. The finger palm recognition device 1 includes a processor 2 and the sensors 11. Referring to FIG. 2, FIG. 3A and FIG. 3B, the non-directional finger palmprint identification method of the embodiment of the present invention includes the following steps: S1: sensing a user's hand fingerprint and palm print by using a finger palm pattern recognition device, and generating The fingerprint and the palm print image are transmitted to a processor; the complex sensor 11 (such as the capacitive palm print sensor 11) of the palm print recognition device 1 senses that the scan directly contacts the user on the contact surface 12 The hand fingerprint and the palm print of the hand 41 of the 4 (ie, the left hand or the palm of the right hand), and the hand fingerprint and the palm print image are transmitted to the processor 2. S2: the processor performs image processing according to the fingerprint and the palm print image to generate a processed palm print image; the processor 2 performs image processing according to the fingerprint and the palm print image, for example, the fingerprint and the palm print The image is subjected to noise removal to back (such as image noise and background removal), Gaussian smoothing (such as processing the edges of the image; or Gaussian blur), intelligent enhancement (such as enhanced image contours), binarization ( The image processing such as converting the grayscale image into a binary image and thinning is performed to generate the processed palmprint image 5. S3: the processor performs a recognition calculation process on the palm print image to obtain a valley between the two fingers of the palm print image, and the processor extracts the palm print image according to the valley points, wherein the two phases are Between the adjacent valley points is a specific finger, and at least one reference line is selected on the specific finger to extend to the bottom end of the palm; the processor 2 performs a recognition calculation process on the palm print image 5 to obtain the palm print image 5 a valley point between the two fingers 51, the valley points having a first valley point 541, a second valley point 542, a third valley point 543, and a fourth valley point 544, and the valley points are The embodiment shows that the finger 51 of the first valley point 541 in the hand 41 (left or right hand) is between the thumb and the finger 51 as the index finger, and the finger 51 of the second valley point 542 located in the hand 41 is the index finger and the finger 51. Between the middle fingers, the third valley point 543 is located between the finger 51 of the hand 41 is the middle finger and the finger 51 is the ring finger, and the fourth valley point is located between the finger 51 of the hand 41 is the ring finger and the finger 51 is the little finger. But it is not limited to this. And the processor 2 extracts the second and third valley points 542 and 543 of the finger palm print image 5 according to the valley points to identify the specific finger 51' (for example, the middle finger of the hand 41), and the specific finger The reference line L1 is selected from 51' to extend from the middle of the top end of the specific finger 51' to the bottom end of the palm 53. The reference line L1 is shown in the present embodiment as a reference line L1 indicating that the bottom end of the palm 53 of the hand 41 extends to the middle of the specific finger 51' of the hand 41 in the finger palm print image 5. S4: the processor performs a recognition calculation process on the palm print image to obtain the fingerprint of the specific finger and the palm print, and performs cross recognition processing on the fingerprint and the palm print of the specific finger according to the at least one reference line. Generating a complex intersection; the processor 2 performs a recognition calculation process on the palm print image 5 to obtain the hand fingerprint 511 of the specific finger 51' and the palm print 531 of the hand 41 in the palm print image 5, and the processing The hand 2 of the specific finger 51' intersecting the reference line L1 on the reference line L1 in the finger palm print image 5 intersects with the complex finger streak and the complex main line of the palm print 531. Identification processing to generate such intersections. The knuckles of the specific finger 51' are shown in the embodiment as a first knuckle 5121, a second knuckle 5122, and a third knuckle 5123, and the main lines of the palm 531 The line includes a first main line 5131, a second main line 5312, and a third main line 5313. The first main line 5311 is located below the valley corresponding to the adjacent line. The two main lines 5312 are located between the first main line 5311 and the third main line 5313 as the smart line, and the third main line 5313 is located below the corresponding main main line 5312 as the life line. In addition, the intersections include a fingerprint intersection 551, a first knuckle intersection 552, a second knuckle intersection 553, a third knuckle intersection 554, and a first main line intersection. Point 555, a second main line intersection 556 and a third main line intersection 557, the first knuckle intersection 552 is located at the hand fingerprint intersection 551 and the second knuckle on the specific finger 51' Between the intersections 553, the second knuckle intersection 553 is located between the third knuckle intersection 554 and the first knuckle intersection 552, the third knuckle intersection 554 is adjacent The second and third valleys are between 542 and 543. The first main line intersection 555 is located on the palm 53 and corresponds to the upper third knuckle intersection 554. The second main line intersection 556 is located at the first main line intersection 555 and the Between the third main line intersections 557, the third main line intersection 557 is located below the corresponding second main line intersection 556. S5: The processor performs a processing process on the distance between the intersections to generate a plurality of feature points, and converts the feature points into the digitized feature data and a feature data stored in a feature database. The alignment is processed to produce a comparison result. The processor 2 performs a calculation process on the distance between the intersections, for example, the first fingerprint and the first phalanx intersection are defined between the hand fingerprint intersection 551 and the first knuckle intersection 552. Points 552, 553 define a second distance therebetween, second and third finger joint intersections 553, 554 define a third distance therebetween, and the third knuckle intersection 554 intersects with the first main line A fourth distance is defined between the points 555, the first and second main line intersections 555, 556 define a fifth distance between each other, and the second and third main line intersections 556, 557 define a sixth between each other. The distances are such that the first, second, third, fourth, fifth, and sixth distances correspond to the first, second, third, fourth, fifth, and sixth feature points 561, 562, respectively. 563, 564, 565, 566, and then the processor 2 converts the feature points into a digitized form such as a binary digit, an octal digit, a decimal digit, a hexadecimal digit, or other hexadecimal (eg, hexadecimal) Or the ninth digit of the feature data, and stored in the feature database 7 The matching characteristic data is subjected to comparison processing to generate the comparison result. If the comparison result is consistent, the identity of the user 4 being identified is correct, and if the comparison result is non-conformity, the representative is identifying the user. 4 identity is incorrect. Therefore, through the design of the finger palmprint identification method of the present invention, the entire palm of the hand 41 of the user 4 can be directly placed on the finger palm recognition device 1 by 360 degrees without directivity to achieve the rapid recognition of the palm pattern. And it is only necessary to confirm that the specific finger 51' (such as the middle finger) is pressed on the contact surface 12 of the fingerprint recognition device 1, and the remaining fingers 51 and the palm 53 can be freely pressed on the contact surface 12 without direction, so as to effectively achieve the use convenience. Good sex. In addition, the method of the present invention can effectively improve the recognition degree (or the identification accuracy) and effectively improve the safety, and can effectively achieve cost saving and shorten the identification comparison time. Referring to FIG. 6 and referring to the figures of FIGS. 1 , 3A and 3B , the method for establishing the non-directional finger palm data of the embodiment of the present invention is applied to the finger palm identification device 1 of the present invention. When the user 4 establishes the entire palm of the hand 41 (ie, the hand fingerprint and the palm print) into the feature database 7, the initial user 4 can be established by the finger palm data creation method of the embodiment of the present invention. The fingerprint and palm print of the palm of the hand establish the feature data corresponding to the user 4 and store it in the feature database 7. The flow steps S1-S4 of the finger palmprint data creation method of the present invention are the same as the flow steps S1-S4 of the finger palmprint identification method of the first embodiment, and therefore are not described again, and the finger palmprint data of the present embodiment is established. In step 5 of the method, the processor performs a processing process on the distance between the intersections to generate a plurality of feature points corresponding to the equidistance (ie, the first, second, third, fourth, fifth, and sixth distances respectively correspond to each other) First, second, third, fourth, fifth, and sixth feature points 561, 562, 563, 564, 565, 566), and the processor 2 converts the feature points into digitized feature data and the first step The process step S5 of the finger palmprint identification method is the same. In this embodiment, the process mainly includes: the processor 2 stores the digitized feature data into a feature database 7 for storage or reading. The processor 2 converts the feature points into a digitized form such as binary digits, octal digits, decimal digits, hexadecimal digits, or other hexadecimal (eg, hexadecimal or hexadecimal) digits. Then, the processor 2 stores the digitized feature data into the feature database 7 for storage or reading, thereby completing the hand fingerprint 511 and the palmprint 531 of the hand 41 of the initial user 4. Feature data is established. Therefore, through the design of the method for establishing the palmprint data of the present invention, the entire palm of the hand 41 of the user 4 can be directly contacted by the 360 degree non-directionality and placed on the finger palm identification device 1 to achieve full palm pattern recognition and establishment of data. Effect, and only need to confirm that the specific finger 51' (such as the middle finger of the left palm or the right palm) is pressed on the contact surface 12 of the palm print recognition device 1, and the remaining fingers 51 and the palm 53 can freely press the contact without the directionality. On the face 12, and the remaining fingers 51 can be combined or opened, it can be used without being limited by the angle at which the finger 51 is opened, so that it is effective to achieve ease of use. In addition, the method of the present invention can effectively improve the recognition degree (or the identification accuracy) and effectively improve the safety, and can effectively achieve cost saving and shorten the identification comparison time. Please refer to FIG. 7A for a finger palmprint identification method according to a second embodiment of the present invention. FIG. 7B is a schematic diagram of a finger palmprint identification method for a finger palmprint image according to a second embodiment of the present invention; The intersection point and the feature point schematic diagram are selected; FIG. 8 is a schematic flow chart of the finger palmprint identification method according to the second embodiment of the present invention. As shown in the figure, the method for recognizing the palm print of the second embodiment of the present invention is substantially the same as the method for recognizing the palm print of the first embodiment. Therefore, the description of the first embodiment is not repeated here. A reference line L1 on the finger 51' is modified to be a plurality of reference lines L1, which are the flow steps S1-S2 of the finger palmprint identification method of the present embodiment and the flow steps S1-S2 of the finger palmprint identification method of the first embodiment. The same is not repeated here, and the difference is that the finger palmprint recognition method of the embodiment includes steps 3, 4, and 5: S3: the processor performs a recognition calculation process on the palm print image to obtain the palm print image of the finger. a valley between the two fingers, and the processor extracts the finger palm image according to the valley points, wherein a second finger is between a plurality of adjacent valley points, and a plurality of reference lines are selected on the specific finger. Going to the bottom end of the palm; the processor 2 performs a recognition calculation process on the palm print image 5 to obtain a valley point between the two fingers 51 of the finger palm print image 5, wherein the valley points have one a first valley point 541, a second valley point 542, a third valley point 543, and a fourth valley point 544, and the valley points in the embodiment are the same as the valley points of the first embodiment. Therefore, it is not repeated. And the processor 2 extracts the second and third valley points 542 and 543 of the finger palm print image 5 according to the valley points to identify the specific finger 51' (for example, the middle finger of the hand 41), and the specific finger The reference line L1 is selected from 51' to extend from the top end of one of the side edges of the specific finger 51' to the bottom end of the palm 53. The reference line L1 is shown in the present embodiment as two reference lines L1, and the bottom ends of the palms 53 of the hand 41 extend to the both sides of the specific finger 51' of the hand 41 in the palm print image 5. S4: the processor performs a recognition calculation process on the palm print image to obtain the fingerprint of the specific finger and the palm print, and according to the reference line and the fingerprint of the specific finger and the complex finger knuckle and the palm print The complex main line has a cross-point identification process to generate a complex intersection; the processor 2 then performs a recognition calculation process on the palm print image 5 to obtain a hand fingerprint 511 of the specific finger 51' in the finger palm print image 5. And a palm print 531 of the hand 41, and the processor 2 according to the finger fingerprint 511 of the specific finger 51' and the upper knuckle of the finger fingerprint image 5 intersecting the reference line L1 on the reference line L1 The pattern and the plurality of main lines of the palm print 531 are subjected to intersection recognition processing to generate the intersections. The finger knuckles of the specific finger 51' are the same as those of the specific finger 51' of the first embodiment, and therefore will not be described again. In addition, the intersections include two opposite hand fingerprint intersections 551, two opposing first knuckle intersections 552, two opposing second knuckle intersections 553, and two opposing third knuckle intersections. 554, two opposite first main line intersections 555, two opposite second main line intersections 556, and two opposite third main line intersections 557, the first knuckle intersection 552 is located at the specific Between the hand fingerprint intersection 551 on the finger 51 and the second knuckle intersection 553, the second knuckle intersection 553 is located at the third knuckle intersection 554 and intersect with the first knuckle Between points 552, the third knuckle intersection 554 is located adjacent the second and third valley points 542, 543. The first main line intersection 555 is located on the palm 53 and corresponds to the upper third knuckle intersection 554. The second main line intersection 556 is located at the first main line intersection 555 and the Between the third main line intersections 557, the third main line intersection 557 is located below the corresponding second main line intersection 556. S5: The processor performs a processing process on the distance between the intersections to generate a plurality of feature points, and converts the feature points into the digitized feature data and a feature data stored in a feature database. The alignment is processed to produce a comparison result. The processor 2 performs a calculation process on the distance between the intersections, for example, a first distance between the two opposite hand fingerprint intersections 551 and the corresponding two first knuckle intersections 552. The two opposing first and second finger joint intersections 552, 553 define a second distance between each other, and the two opposite second and third finger joint intersections 553, 554 define a third distance therebetween. A second third knuckle intersection 553 and a corresponding first main line intersection 555 define a fourth distance, and the two opposite first and second main line intersections 555, 556 are mutually Defining a fifth distance and each of the two opposing second and third main line intersections 556, 557 define a sixth distance from each other to generate the feature points corresponding to the equidistance, that is, the two opposite The first, second, third, fourth, fifth, and sixth distances correspond to the first, second, third, fourth, fifth, and sixth feature points 561, 562, 563, 564, 565, and 566, and then the processor 2 The feature points are converted into digital forms such as binary digits, octal digits, Character data of hexadecimal digits, hexadecimal digits or other hexadecimal digits (such as hexadecimal or hexadecimal digits), and then compared with the feature data stored in the feature database 7 to generate the ratio For the result, if the comparison result is consistent, the user 4 who is being identified is correct in identity, and if the comparison result is non-conform, the user 4 who is being identified is not in the correct identity. In an embodiment, in the foregoing step S5, the processor 2 performs the distance between the intersections and the width defined by the plurality of reference lines L1 indicated on the two sides of the specific finger 51'. Calculating processing to generate the feature points corresponding to the equidistant distances and the width feature points corresponding to the widths, and then the processor 2 converts the feature points and the width feature points into the digitized feature data, and the feature The feature data stored in the database 7 is compared and processed to generate the comparison result. The method of the present invention can effectively improve the recognition degree (or the identification accuracy) and effectively improve the security. In addition, referring to FIG. 9 and referring to the 7A and 7B drawings, the flow steps S1-S4 of the non-directional finger palm data establishing method of the second embodiment of the present invention and the finger of the foregoing second embodiment are used. The process steps S1-S4 of the palmprint identification method are the same, and therefore will not be described again here, but in step 5 of the finger palmprint data establishing method of the embodiment, the processor 2 performs calculation processing on the distance between the intersections. To generate a plurality of feature points corresponding to the equidistant distances (ie, the first, second, third, fourth, fifth, and sixth distances of the respective two corresponding first, second, third, fourth, fifth, and sixth feature points 561, 562, 563, 564, 565, 566), and the step of converting the feature points into digitalized feature data by the processor 2 is the same as the flow step S5 of the finger palmprint identification method of the second embodiment, and the main embodiment is After the step is changed, the processor 2 stores the digitized feature data into a feature database 7 for storage or reading, thereby completing the hand of the initial user 4's hand 41. Fingerprint 511 and palmprint 531 feature data construction . Therefore, through the design of the palm palmprint identification method of the present invention, the effect of quickly recognizing the non-directional full palm pattern is achieved, and the convenience of use is achieved, the cost is saved, and the recognition comparison time is shortened.

1‧‧‧ finger palmprint identification device

11‧‧‧ Sensor

12‧‧‧Contact surface

2‧‧‧ Processor

4‧‧‧Users

41‧‧‧Hands

5‧‧‧ Finger Palmprint Image

51‧‧‧ fingers

51'‧‧‧Special fingers

511‧‧‧Hand fingerprint

5121, 5122, 5123‧‧‧ first, second and third finger joint

53‧‧‧ palm

531‧‧‧ palm print

5311, 5312, 5313‧‧‧ first, second and third main lines

541, 542, 543, 544‧‧‧ first, second, third, and fourth valley points

L1‧‧‧ baseline

551‧‧‧Hand fingerprint intersection

552, 553, 554‧‧‧ first, second and third finger joint intersection

555, 556, 557‧‧‧ first, second and third main line intersections

561, 562, 563, 564, 565, 566‧‧‧ first, second, third, fourth, fifth and sixth feature points

7‧‧‧Characteristic database

Figure 1 is a block diagram showing the use of the fingerprint identification device of the present invention. FIG. 2 is a schematic flow chart of a finger palmprint identification method according to a first embodiment of the present invention. FIG. 3A is a schematic diagram of a finger palm pattern recognition method according to a first embodiment of the present invention for selecting a valley point, a reference line, and an intersection point between two fingers in a finger palm image. FIG. 3B is a schematic diagram showing the intersection of the finger palmprint image and the feature point of the finger palmprint identification method according to the first embodiment of the present invention. Fig. 4A is a schematic view showing the operation of the palmprint recognition device of the present invention. FIG. 4B is a schematic view showing another embodiment of the operation of the finger palm recognition device of the present invention. Fig. 5A is a schematic view showing the operation of the hand of the user of the present invention in one direction on the finger palm identification device. Fig. 5B is a schematic view showing the operation of the user's hand in the other direction on the palm print recognition device of the present invention. Figure 6 is a flow chart showing the method for establishing finger palmprint data according to the first embodiment of the present invention. FIG. 7A is a schematic diagram showing the intersection point and feature points of the finger palmprint image recognition method according to the second embodiment of the present invention. FIG. 7B is a schematic diagram showing the intersection point and feature points of the finger palmprint image recognition method according to the second embodiment of the present invention. FIG. 8 is a schematic flow chart of a finger palmprint identification method according to a second embodiment of the present invention. Figure 9 is a flow chart showing the method for establishing finger palmprint data according to the second embodiment of the present invention.

Claims (24)

A non-directional finger palmprint identification method includes: sensing a user's hand fingerprint and palm print by using a finger palm recognition device, and generating a fingerprint and palm print image to a processor; the processor according to the hand The fingerprint and the palm print image are processed to generate a processed palm print image; the processor performs a recognition calculation process on the palm print image to obtain a valley between the two fingers of the palm print image, and the The processor extracts the finger palm image according to the valley identification, wherein a specific finger is between two adjacent valley points, and at least one reference line is selected on the specific finger to extend to the bottom end of the palm; the processor is the finger The palmprint image is subjected to an identification calculation process to obtain the fingerprint of the specific finger and the palmprint, and cross-identify the fingerprint and the palmprint of the specific finger according to the at least one reference line to generate a plurality of intersections; the processor Calculating the distance between the intersections to generate complex feature points and converting the feature points Digitized feature data for comparison with a process wherein a data library stored feature data to generate a comparison result. The non-directional finger palmprint identification method according to claim 1, wherein the finger palm pattern recognition device has a plurality of sensors and a contact surface disposed on one side of the finger palm identification device. For the user's hand to place the affixing contact, the sensors are disposed on the finger palm identification device and electrically connected to the processor, the sensors are used to sense the scanning contact at the The hand fingerprint and the palm print of the user's hand on the contact surface generate the fingerprint and the palm print image. The non-directional finger palmprint identification method according to claim 2, wherein the user's hand can be contactlessly contacted on the contact surface. The non-directional finger palmprint identification method according to claim 1, wherein the valley points have a first valley point, a second valley point, a third valley point, and a fourth valley point, The processor extracts the second and third valley points adjacent to the palm print image according to the valley points as the specific finger, and selects the reference line on the specific finger to extend from the middle of the top end of the specific finger to the The bottom of the palm. The non-directional finger palmprint identification method according to claim 4, wherein the processor performs the fingerprint according to the reference line and the fingerprint and the complex finger pattern of the specific finger and the plurality of main lines of the palm print. The intersection identification process produces the intersections. The non-directional finger palmprint identification method according to claim 5, wherein the intersection includes a fingerprint intersection, a first knuckle intersection, a second knuckle intersection, and a first a third knuckle intersection and a first main line intersection, the first knuckle intersection being located between the hand fingerprint intersection and the second knuckle intersection on the specific finger, the second The knuckle intersection is located between the third knuckle intersection and the first knuckle intersection, the first main line intersection is located on the palm and corresponding to the third knuckle intersection . The non-directional finger palmprint identification method according to claim 6, wherein the intersections comprise a second main line intersection and a third main line intersection, the second main line crossing The point is between the intersection of the first main line and the intersection of the third main line. The non-directional finger palmprint identification method described in claim 1, wherein the feature data is a binary digit, an octal digit, a decimal digit or a hexadecimal digit. The non-directional finger palmprint identification method according to the first aspect of the invention, wherein the palm print pattern recognition device is a flat finger palm recognition device or a flexible curved finger palm recognition device. The non-directional finger palmprint identification method according to claim 2, wherein the contact surface is a plane or a curved surface. The non-directional finger palmprint identification method according to claim 1, wherein the valley points have a first valley point, a second valley point, a third valley point, and a fourth valley point, The processor extracts the second and third valley points adjacent to the finger palm image according to the valley points as the specific finger, and selects a plurality of reference lines on the specific finger to extend from a top end of one side of the specific finger To the bottom of the palm. The non-directional finger palmprint identification method according to claim 11, wherein the processor according to the reference line and the finger fingerprint of the specific finger and the complex finger streak and the complex main line of the palm print Cross-point identification processing is performed to generate the intersections. A method for establishing a non-directional finger palmprint data, comprising: sensing a user's hand fingerprint and palm print by using a finger palm pattern recognition device, and generating a fingerprint and a palm print image to be transmitted to a processor; The fingerprint of the hand and the image of the palm print are processed to generate a processed palm print image; the processor performs a recognition calculation process on the palm print image to obtain a valley between the two fingers of the palm print image, and The processor extracts the finger palm image according to the valley identification, wherein a specific finger is between two adjacent valley points, and at least one reference line is selected on the specific finger to extend to the bottom end of the palm; the processor The finger palm image is subjected to a recognition calculation process to obtain the fingerprint of the specific finger and the palm print, and the intersection recognition process is performed according to the at least one reference line and the fingerprint of the specific finger and the palm print to generate a plurality of intersections; Calculating the distance between the intersections to generate complex feature points and combining the features Conversion bit number of the stored feature data and feature data to a storage compartment for reading or use. The method for establishing a non-directional finger palmprint data according to claim 13 , wherein the palm print pattern recognition device has a plurality of sensors and a contact surface disposed on a side of the finger palm identification device The sensor is disposed on the fingerprint sensing device and is electrically connected to the processor. The sensors are used to sense the scanning contact. The hand fingerprint and the palm print of the user's hand on the contact surface generate the hand fingerprint and the palm print image. The non-directional finger palmprint data establishing method according to claim 14, wherein the user's hand can be contactlessly contacted on the contact surface. The method for establishing a non-directional finger palmprint data according to claim 13 , wherein the valley points have a first valley point, a second valley point, a third valley point, and a fourth valley point. The processor extracts the second and third valley points adjacent to the palm print image according to the valley points as the specific finger, and selects the reference line on the specific finger to extend from the middle of the top end of the specific finger to The bottom of the palm. The non-directional finger palmprint data establishing method according to claim 16, wherein the processor according to the reference line and the finger fingerprint of the specific finger and the complex finger streak and the complex main line of the palm print Cross-point identification processing is performed to generate the intersections. The method for establishing a non-directional finger palmprint data according to claim 17, wherein the intersection includes a fingerprint intersection, a first knuckle intersection, a second knuckle intersection, and a first a third knuckle intersection and a first main line intersection, the first knuckle intersection being located between the hand fingerprint intersection and the second knuckle intersection on the specific finger, the second The knuckle intersection is located between the third knuckle intersection and the first knuckle intersection, the first main line intersection is located on the palm and corresponding to the third knuckle intersection . The method for establishing a non-directional finger palmprint data according to claim 18, wherein the intersections comprise a second main line intersection and a third main line intersection, the second main line The intersection is located between the intersection of the first main line and the intersection of the third main line. The method for establishing a non-directional finger palmprint data according to claim 13 of the patent application scope, wherein the feature data is a characteristic data of a binary digit, an octal digit, a decimal digit or a hexadecimal digit. The method for establishing a non-directional finger palmprint data according to claim 13 wherein the finger palmprint identifying device is a flat finger palm identifying device or a flexible curved finger palm identifying device. The method for establishing a non-directional finger palmprint data according to claim 14, wherein the contact surface is a plane or a curved surface. The method for establishing a non-directional finger palmprint data according to claim 13 , wherein the valley points have a first valley point, a second valley point, a third valley point, and a fourth valley point. The processor extracts the second and third valley points of the finger palm image according to the valley points for the specific finger, and selects a plurality of reference lines from the top of the two sides of the specific finger on the specific finger. Extend to the bottom of the palm. The method for establishing a non-directional finger palmprint data according to claim 23, wherein the processor according to the reference line and the finger fingerprint of the specific finger and the complex finger streak and the plurality of main lines of the palm print Lines are identified by intersections to produce such intersections.