CN110309738B - Method for labeling OCT fingerprint image - Google Patents
Method for labeling OCT fingerprint image Download PDFInfo
- Publication number
- CN110309738B CN110309738B CN201910522011.9A CN201910522011A CN110309738B CN 110309738 B CN110309738 B CN 110309738B CN 201910522011 A CN201910522011 A CN 201910522011A CN 110309738 B CN110309738 B CN 110309738B
- Authority
- CN
- China
- Prior art keywords
- dimensional
- fingerprint
- fingerprint image
- labeling
- matrix
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
-
- 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
- G06V40/12—Fingerprints or palmprints
- G06V40/13—Sensors therefor
- G06V40/1318—Sensors therefor using electro-optical elements or layers, e.g. electroluminescent sensing
-
- 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
- G06V40/12—Fingerprints or palmprints
- G06V40/1347—Preprocessing; Feature extraction
Abstract
The invention provides a method for labeling an OCT fingerprint image, which comprises the following steps: step S1, reading a preset number of three-dimensional fingerprint images; step S2, preprocessing the three-dimensional fingerprint image to obtain a one-dimensional vector corresponding to the three-dimensional fingerprint image, and then combining the one-dimensional vectors into a two-dimensional fingerprint image according to the sequence of the original three-dimensional fingerprint image; step S3, selecting a labeling area of the two-dimensional fingerprint image to form a two-dimensional fingerprint label; and step S4, mapping the two-dimensional fingerprint label back to the three-dimensional space of the three-dimensional fingerprint image through a mapping algorithm. According to the invention, the three-dimensional fingerprint image is processed into the two-dimensional fingerprint image, so that the two-dimensional fingerprint image can be labeled, and the formed two-dimensional fingerprint label is mapped back to the three-dimensional space of the three-dimensional fingerprint image, so that the label of the OCT fingerprint image is realized, and various characteristics of the fingerprint under the label area can be more accurately seen.
Description
Technical Field
The invention relates to an OCT fingerprint image processing method, in particular to a method for labeling an OCT fingerprint image.
Background
With the maturity of optical coherence tomography, the OCT three-dimensional fingerprint image can be used to analyze the internal information of the fingerprint, and the OCT is an optical coherence tomography, so that various problems faced by the conventional two-dimensional fingerprint image biometric identification can be better avoided, such as various stains on the finger surface, severe damage to the finger epidermis, and the like. However, in the prior art, there is no method for labeling the OCT fingerprint image formed based on the OCT fingerprint imaging technology.
Disclosure of Invention
The invention aims to solve the technical problem of providing a method capable of labeling an OCT fingerprint image.
Therefore, the invention provides a method for labeling an OCT fingerprint image, which comprises the following steps:
step S1, reading a preset number of three-dimensional fingerprint images;
step S2, preprocessing the three-dimensional fingerprint image to obtain a one-dimensional vector corresponding to the three-dimensional fingerprint image, and then combining the one-dimensional vectors into a two-dimensional fingerprint image according to the sequence of the original three-dimensional fingerprint image;
step S3, selecting a labeling area of the two-dimensional fingerprint image to form a two-dimensional fingerprint label;
and step S4, mapping the two-dimensional fingerprint label back to the three-dimensional space of the three-dimensional fingerprint image through a mapping algorithm.
A further development of the invention is that said step S2 comprises the following sub-steps:
step S201, processing each three-dimensional fingerprint image into a grey-white fingerprint picture;
step S202, accumulating and summing the fingerprint gray-white pictures according to the fingerprint length to form a one-dimensional vector corresponding to the fingerprint gray-white pictures;
and S203, combining the one-dimensional vectors corresponding to the grey-white fingerprint picture into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image to obtain a two-dimensional fingerprint image.
The invention is further improved in that 400 OCT three-dimensional fingerprint images are read in step S1; in step S202, each fingerprint gray-white picture is equally divided into 1500 column vectors according to the fingerprint length, the 1500 column vectors are respectively summed up to obtain 1500 values, the 1500 values are sequentially combined into a one-dimensional vector corresponding to the fingerprint gray-white picture, and by analogy, the 400 fingerprint gray-white pictures are processed to obtain 400 corresponding one-dimensional vectors.
A further improvement of the present invention is that, in step S202, each numerical value in each one-dimensional vector is further divided by the sum of the vectors where the numerical value is currently located, so as to implement normalization, and obtain 400 normalized one-dimensional vectors.
In step S203, the 400 one-dimensional vectors are combined into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image, so as to obtain a two-dimensional fingerprint image reconstructed from the OCT three-dimensional fingerprint image.
The invention further improves the method and the device, and the method and the device further comprise a step S5, wherein in the step S5, the regenerated 400 two-dimensional fingerprint images and 400 grey-white fingerprint pictures are saved in corresponding folders.
A further development of the invention is that said step S3 comprises the following sub-steps:
step S301, selecting a labeling area of the two-dimensional fingerprint image by means of a mouse or a keyboard;
and step S302, labeling the labeling area through a matrix labeling algorithm or a circle labeling algorithm to obtain a two-dimensional fingerprint label.
In a further improvement of the present invention, in the step S3, the labeling process of the labeled region by the matrix labeling algorithm is as follows: selecting a first matrix point of the two-dimensional fingerprint image through a mouse or a keyboard; then, selecting a second matrix point through the movement of the mouse or the keyboard, wherein the second matrix point is a matrix diagonal coordinate point of the first matrix point, and further respectively obtaining the length and the width of the matrix through a matrix coordinate difference of the first matrix point and the second matrix point; and finally, changing the color of the picture in the matrix to finish the labeling, wherein the two-dimensional image at the moment is a two-dimensional fingerprint label.
In a further improvement of the present invention, in the step S3, the labeling of the labeled region by the circle labeling algorithm is as follows: firstly, selecting a first circle coordinate point and a second circle coordinate point of a two-dimensional fingerprint image through a mouse or a keyboard, and opening a root number for a coordinate square difference of the first circle coordinate point and the second circle coordinate point to obtain a diameter of a circle, so as to obtain a radius of the circle; then calculating the middle point of the two first circle coordinate points and the second circle coordinate point, and taking the middle point as the circle center; and finally, taking the circle center as the center, changing the color of the picture in the circle by taking the radius of the circle to finish the labeling, wherein the two-dimensional image at the moment is a two-dimensional fingerprint label.
The further improvement of the present invention lies in that the step S4 is implemented by mapping the two-dimensional fingerprint label back to the three-dimensional space through a mapping algorithm, and comprises the following steps: and the two-dimensional fingerprint labels are in one-to-one correspondence to the width coordinate of each three-dimensional fingerprint image by utilizing the arrangement sequence of the original three-dimensional fingerprint images, so that the three-dimensional fingerprint structure information with the two-dimensional fingerprint labels is obtained.
Compared with the prior art, the invention has the beneficial effects that: the three-dimensional fingerprint image is processed into the two-dimensional fingerprint image, the two-dimensional fingerprint image can be labeled, and the formed two-dimensional fingerprint label is mapped back to the three-dimensional space of the three-dimensional fingerprint image, so that the OCT fingerprint image can be labeled, various characteristics of fingerprints under a labeling area can be more accurately known from the labeled OCT fingerprint image, the OCT fingerprint image can also be used in various neural networks based on the OCT fingerprint image, and a good data basis is provided for expanding the application of the OCT fingerprint image.
Drawings
FIG. 1 is a schematic workflow diagram of one embodiment of the present invention;
FIG. 2 is a schematic diagram of a three-dimensional fingerprint image according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the working principle of processing and combining three-dimensional fingerprint images into a two-dimensional fingerprint image according to an embodiment of the present invention;
figure 4 is a schematic diagram of a single OCT fingerprint image with labeled regions according to one embodiment of the present invention.
Detailed Description
Preferred embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.
As shown in fig. 1, this embodiment provides a method for labeling an OCT fingerprint image, including the following steps:
step S1, reading a preset number of three-dimensional fingerprint images;
step S2, preprocessing the three-dimensional fingerprint image to obtain a one-dimensional vector corresponding to the three-dimensional fingerprint image, and then combining the one-dimensional vectors into a two-dimensional fingerprint image according to the sequence of the original three-dimensional fingerprint image;
step S3, selecting a labeling area of the two-dimensional fingerprint image to form a two-dimensional fingerprint label;
and step S4, mapping the two-dimensional fingerprint label back to the three-dimensional space of the three-dimensional fingerprint image through a mapping algorithm.
The three-dimensional fingerprint image is an OCT three-dimensional fingerprint image, and can also be called the OCT fingerprint image for short; in this example, the step S1 uses the three-dimensional fingerprint image (3D fingerprint image) of the user collected by OCT (optical coherence tomography) as the read data, and preferably collects and reads at least 400 three-dimensional fingerprint images of the same finger of the same user as the fingerprint image data, that is, the preset number is a value predefined according to actual needs, and is preferably 400.
As shown in fig. 2, the example places 400 three-dimensional fingerprint images in three-dimensional coordinates, wherein the X-axis represents the length of the fingerprint, i.e. the picture length of the three-dimensional fingerprint image, and the numerical value of the X-axis represents the width coordinates of the three-dimensional fingerprint image; the Y-axis represents the number of 400 (or other preset number) three-dimensional fingerprint images, i.e. the Y-axis represents the number of preset numbers; the Z-axis represents the fingerprint depth, i.e. the picture width of the three-dimensional fingerprint image.
The main principle of the embodiment is that 400 (or other preset number) three-dimensional fingerprint images/three-dimensional fingerprint pictures formed by using the OCT technology are subjected to graying on the three-dimensional fingerprint images, then the summation in the z direction is performed according to the fingerprint depth (namely, the picture width taking the z axis as the direction in fig. 2) of the fingerprint in each three-dimensional fingerprint image, normalization is performed after the summation to obtain 400 one-dimensional vectors, and finally the 400 vectors are combined into a two-dimensional matrix according to the sequence of the pictures formed by using the OCT technology, so that the matrix is the two-dimensional fingerprint image reconstructed from the three-dimensional fingerprint images.
The main principle of the marked two-dimensional fingerprint image restoration is that the marked two-dimensional fingerprint image is mapped back to the three-dimensional fingerprint image of the OCT according to the sequence of the original three-dimensional fingerprint image, data processing is concentrated in marked areas in the mapping process, the marked areas are newly converted into fingerprint cross sections, therefore, corresponding three-dimensional structure information of the marked areas can be observed through experiments, and the marking of the OCT fingerprint image is realized through marking the two-dimensional fingerprint image, as shown in figure 4, and a white area in figure 4 is the marked area.
In step S2 of the present embodiment, the three-dimensional fingerprint image is preprocessed according to the fingerprint depth of each three-dimensional fingerprint image to obtain one-dimensional vectors corresponding to the three-dimensional fingerprint image, and then the one-dimensional vectors are combined into a two-dimensional fingerprint image according to the sequence of the original three-dimensional fingerprint image.
Specifically, as shown in fig. 3, step S2 in this example includes the following sub-steps:
step S201, processing each three-dimensional fingerprint image into a grey-white fingerprint picture;
step S202, accumulating and summing the fingerprint gray pictures according to the fingerprint length to form a one-dimensional vector corresponding to the fingerprint gray pictures;
and S203, combining the one-dimensional vectors corresponding to the grey-white fingerprint picture into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image to obtain a two-dimensional fingerprint image.
In fig. 3, in the left half, the Z-axis direction indicates the depth of the fingerprint in each fingerprint image, the Y-axis direction indicates the number of OCT fingerprint images, and the X-axis direction indicates the length of each fingerprint image; in the right half, four two-dimensional fingerprint images represent the results of two-dimensional fingerprint construction with different depths (Z-axis) of 4 fingerprints selected respectively.
In this example, in step S1, 400 OCT three-dimensional fingerprint images are read, and the reading process only needs to read 400 or other preset number of three-dimensional fingerprint images acquired by OCT. In the step S201, each three-dimensional fingerprint image is processed into a fingerprint gray picture according to the fingerprint depth, that is, the RGB color picture is changed into a gray picture, and then accumulation and normalization are realized in subsequent steps.
In step S202 in this example, each fingerprint gray-scale picture is equally divided into 1500 column vectors according to the fingerprint length (i.e., the X axis in fig. 2), the 1500 column vectors are respectively accumulated and summed to obtain 1500 values, the 1500 values are sequentially combined into a one-dimensional vector corresponding to the fingerprint gray-scale picture, and so on, the 400 fingerprint gray-scale pictures are processed to obtain 400 corresponding one-dimensional vectors.
In this example, in step S202, it is preferable to divide each value in each one-dimensional vector by the sum of the vectors where the value is currently located, so as to realize normalization, and obtain 400 normalized one-dimensional vectors. Specifically, each numerical value in each one-dimensional vector is divided by the sum of the one-dimensional vectors in which the numerical value is located, so as to realize normalization, and obtain 400 new one-dimensional vectors after normalization processing.
In step S203 described in this example, 400 one-dimensional vectors are combined into a two-dimensional matrix according to the order of the original three-dimensional fingerprint image, so as to obtain a two-dimensional fingerprint image reconstructed from the OCT three-dimensional fingerprint image.
That is to say, in step S203 of this example, 400 one-dimensional vectors are combined into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image formed by the OCT imaging technology, that is, in step S203, 400 one-dimensional vectors are combined along the y-axis direction in fig. 2 according to the sequence of the original three-dimensional fingerprint image to obtain a two-dimensional matrix, and the two-dimensional matrix is a reconstructed two-dimensional fingerprint image of the three-dimensional fingerprint image.
Certainly, in the step S202 described in this example, the implementation processes of performing summation and normalization on the fingerprint gray-scale images all belong to preferred implementation processes, and the implementation process of combining the two-dimensional matrixes in the step S203 also belongs to preferred implementation processes, in this example, the fingerprint gray-scale images are processed into corresponding one-dimensional vectors and then combined into two-dimensional matrixes, so as to facilitate subsequent data processing; in practical applications, the step S2 can also be implemented in other ways, as long as the three-dimensional fingerprint image is processed into a two-dimensional matrix, so that the labeling can be implemented subsequently.
Step S3 in the present example includes the following substeps:
step S301, selecting a labeling area of the two-dimensional fingerprint image by means of a mouse or a keyboard;
and step S302, labeling the labeling area through a matrix labeling algorithm or a circle labeling algorithm to obtain a two-dimensional fingerprint label.
In step S3, the process of labeling the labeled area by the matrix labeling algorithm is as follows: selecting a first matrix point of the two-dimensional fingerprint image through a mouse or a keyboard; then, selecting a second matrix point through the movement of the mouse or the keyboard, wherein the second matrix point is a matrix diagonal coordinate point of the first matrix point, and further respectively obtaining the length and the width of the matrix through the matrix coordinate difference of the first matrix point and the second matrix point; finally, the color of the picture in the matrix is changed to complete the labeling, for example, the color is changed to black or white, etc., fig. 4 shows that the color of the picture in the matrix is changed to white, and the two-dimensional image at this time is the two-dimensional fingerprint tag, as shown in fig. 4. The matrix is used for labeling, the first matrix point is a first point of a selected matrix labeling area, the second matrix point is a second point of the selected matrix labeling area, and the two points are used as diagonal coordinate points of the matrix, so that matrix labeling can be realized.
In this example, in step S3, the process of labeling the labeled region by using the circle labeling algorithm is as follows: firstly, selecting a first circle coordinate point and a second circle coordinate point of a two-dimensional fingerprint image through a mouse or a keyboard, and opening a root number for a coordinate square difference of the first circle coordinate point and the second circle coordinate point to obtain a diameter of a circle, so as to obtain a radius of the circle (the radius can be obtained by dividing the diameter by 2); then calculating the midpoint of the two first circle coordinate points and the second circle coordinate point, and taking the midpoint as the circle center; and finally, taking the circle center as the center, changing the color of the picture in the circle by using the radius of the circle to finish labeling, such as changing the color into black or white, and the like, wherein the two-dimensional image at the moment is a two-dimensional fingerprint label. The first circle coordinate point refers to a first point for selecting a circular labeling area, the second circle coordinate point refers to a second point for selecting the circular labeling area, the two points obtain the circle center and the radius of the circular labeling area, and circular labeling can be achieved.
In this example, the implementation process of mapping the two-dimensional fingerprint tag back to the three-dimensional space in step S4 through the mapping algorithm is as follows: utilizing the arrangement sequence of the original three-dimensional fingerprint images and the corresponding relationship between the two-dimensional fingerprint images and the original three-dimensional fingerprint images in the step S2, enabling the two-dimensional fingerprint labels to correspond to the width coordinate (namely the X axis in the figure 2) of each three-dimensional fingerprint image one by one, and changing the width coordinate values corresponding to all the three-dimensional fingerprint images in the depth direction of the original three-dimensional fingerprint images by the method so as to obtain the three-dimensional fingerprint structure information with the two-dimensional fingerprint labels
Step S4 described in this example is used to map the two-dimensional fingerprint tag back to the three-dimensional space through a mapping algorithm, and the algorithm may be concentrated on reconstructed labeled regions, which are newly converted into fingerprint cross sections, and the specific implementation process is: the two-dimensional fingerprint labels (including the identification areas) are in one-to-one correspondence to the width coordinate (namely the x axis in fig. 2) of each OCT three-dimensional fingerprint image by utilizing the one-to-one correspondence relationship between the arrangement sequence of the original three-dimensional fingerprint images and the two-dimensional fingerprint images in the y axis direction; and then changing the width coordinate value corresponding to each three-dimensional fingerprint image in the depth direction (namely the y axis in fig. 2) of the OCT original fingerprint image, so that the two-dimensional fingerprint label can be mapped to the three-dimensional space.
The present embodiment further preferably includes step S5, in which step S5, the 400 regenerated two-dimensional fingerprint images and the 400 gray-scale fingerprint images are saved in corresponding folders, so as to facilitate subsequent data processing.
In summary, in this embodiment, the three-dimensional fingerprint image is processed into the two-dimensional fingerprint image, so that the two-dimensional fingerprint image can be labeled, and the formed two-dimensional fingerprint label is mapped back to the three-dimensional space of the three-dimensional fingerprint image, so as to label the OCT fingerprint image, and the labeled OCT fingerprint image can more accurately find various features of the fingerprint under the labeling area, and can also be used in various neural networks based on the OCT fingerprint image, thereby providing a good data base for expanding the application of the OCT fingerprint image.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, numerous simple deductions or substitutions may be made without departing from the spirit of the invention, which shall be deemed to belong to the scope of the invention.
Claims (9)
1. A method for labeling an OCT fingerprint image is characterized by comprising the following steps:
step S1, reading a preset number of three-dimensional fingerprint images;
step S2, preprocessing the three-dimensional fingerprint image to obtain a one-dimensional vector corresponding to the three-dimensional fingerprint image, and then combining the one-dimensional vectors into a two-dimensional fingerprint image according to the sequence of the original three-dimensional fingerprint image;
the step S2 includes the following sub-steps:
step S201, processing each three-dimensional fingerprint image into a grey-white fingerprint picture;
step S202, accumulating and summing the fingerprint gray-white pictures according to the fingerprint length to form a one-dimensional vector corresponding to the fingerprint gray-white pictures;
step S203, combining the one-dimensional vectors corresponding to the grey-white fingerprint picture into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image to obtain a two-dimensional fingerprint image;
step S3, selecting a labeling area of the two-dimensional fingerprint image to form a two-dimensional fingerprint label;
and step S4, mapping the two-dimensional fingerprint label back to the three-dimensional space of the three-dimensional fingerprint image through a mapping algorithm.
2. The method for labeling the OCT fingerprint image of claim 1, wherein 400 OCT three-dimensional fingerprint images are read in step S1; in step S202, each fingerprint gray-white picture is equally divided into 1500 column vectors according to the fingerprint length, the 1500 column vectors are respectively summed up to obtain 1500 values, the 1500 values are sequentially combined into a one-dimensional vector corresponding to the fingerprint gray-white picture, and by analogy, the 400 fingerprint gray-white pictures are processed to obtain 400 corresponding one-dimensional vectors.
3. The method as claimed in claim 2, wherein in step S202, each numerical value in each one-dimensional vector is further divided by a sum of vectors where the numerical value currently exists, so as to achieve normalization, and obtain 400 normalized one-dimensional vectors.
4. The method for labeling the OCT fingerprint image of claim 2, wherein in step S203, the 400 one-dimensional vectors are combined into a two-dimensional matrix according to the sequence of the original three-dimensional fingerprint image, so as to obtain the reconstructed two-dimensional fingerprint image of the OCT three-dimensional fingerprint image.
5. The method for labeling the OCT fingerprint image of claim 2, further comprising a step S5, wherein in the step S5, the regenerated 400 two-dimensional fingerprint images and 400 gray-white fingerprint images are saved in corresponding folders.
6. Method for labeling an OCT fingerprint image according to any one of claims 1 to 4, characterized in that said step S3 comprises the following sub-steps:
step S301, selecting a labeling area of the two-dimensional fingerprint image by means of a mouse or a keyboard;
and step S302, labeling the labeling area through a matrix labeling algorithm or a circle labeling algorithm to obtain a two-dimensional fingerprint label.
7. The method for labeling the OCT fingerprint image of claim 6, wherein in the step S3, the labeling process of the labeling region through the matrix labeling algorithm is as follows: selecting a first matrix point of the two-dimensional fingerprint image through a mouse or a keyboard; then, selecting a second matrix point through the movement of the mouse or the keyboard, wherein the second matrix point is a matrix diagonal coordinate point of the first matrix point, and further respectively obtaining the length and the width of the matrix through a matrix coordinate difference of the first matrix point and the second matrix point; and finally, changing the color of the picture in the matrix to finish the labeling, wherein the two-dimensional image at the moment is a two-dimensional fingerprint label.
8. The method for labeling the OCT fingerprint image of claim 6, wherein in the step S3, the labeling of the labeled region by the circle labeling algorithm is as follows: firstly, selecting a first circle coordinate point and a second circle coordinate point of a two-dimensional fingerprint image through a mouse or a keyboard, and opening a root number for a coordinate square difference of the first circle coordinate point and the second circle coordinate point to obtain a diameter of a circle, so as to obtain a radius of the circle; then calculating the middle point of the two first circle coordinate points and the second circle coordinate point, and taking the middle point as the circle center; and finally, taking the circle center as the center, changing the color of the picture in the circle by using the radius of the circle to finish the labeling, wherein the two-dimensional image at the moment is a two-dimensional fingerprint label.
9. The method for labeling the OCT fingerprint image of any one of claims 1 to 4, wherein the step S4 is implemented by mapping the two-dimensional fingerprint label back to the three-dimensional space through a mapping algorithm by: and corresponding the two-dimensional fingerprint labels to the width coordinate of each three-dimensional fingerprint image one by utilizing the arrangement sequence of the original three-dimensional fingerprint images so as to obtain the three-dimensional fingerprint structure information with the two-dimensional fingerprint labels.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910522011.9A CN110309738B (en) | 2019-06-17 | 2019-06-17 | Method for labeling OCT fingerprint image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910522011.9A CN110309738B (en) | 2019-06-17 | 2019-06-17 | Method for labeling OCT fingerprint image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110309738A CN110309738A (en) | 2019-10-08 |
| CN110309738B true CN110309738B (en) | 2022-09-30 |
Family
ID=68076576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910522011.9A Active CN110309738B (en) | 2019-06-17 | 2019-06-17 | Method for labeling OCT fingerprint image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110309738B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593349A (en) * | 2009-06-26 | 2009-12-02 | 福州华映视讯有限公司 | Bidimensional image is converted to the method for 3-dimensional image |
| CN103810709A (en) * | 2014-02-25 | 2014-05-21 | 南京理工大学 | Vessel-based registration method for eye fundus image and SD-OCT projection image |
| CN106021610A (en) * | 2016-06-28 | 2016-10-12 | 电子科技大学 | Video fingerprint extracting method based on salient region |
| CN106407896A (en) * | 2016-08-29 | 2017-02-15 | 华蓥市高科德电子科技有限公司 | 3D fingerprint information extraction method for handset equipment |
| CN106778663A (en) * | 2016-12-29 | 2017-05-31 | 苏州塞罗尔医学影像科技有限公司 | A kind of fingerprint recognition system |
| CN107146262A (en) * | 2017-04-18 | 2017-09-08 | 清华大学深圳研究生院 | The three-dimensional visualization method and system of a kind of OCT image |
| CN108446633A (en) * | 2018-03-20 | 2018-08-24 | 深圳大学 | A kind of method, system and device of novel finger print automatic anti-fake and In vivo detection |
| CN108664940A (en) * | 2018-05-16 | 2018-10-16 | 山东大学 | A kind of partial fingerprints matching process and system |
| CN108932507A (en) * | 2018-08-06 | 2018-12-04 | 深圳大学 | A kind of automatic anti-fake method and system based on OCT fingerprint image |
| CN109377549A (en) * | 2018-09-29 | 2019-02-22 | 浙江工业大学 | A kind of real-time processing of OCT finger tip data and three-dimensional visualization method |
| CN109376687A (en) * | 2018-11-16 | 2019-02-22 | 深圳市趣创科技有限公司 | Shield lower unlocked by fingerprint discrimination method for improving, device, terminal and storage medium |
| CN109508692A (en) * | 2018-11-30 | 2019-03-22 | 深圳大学 | A kind of gender identification method and system based on 3D fingerprint image |
| CN109657567A (en) * | 2018-11-30 | 2019-04-19 | 深圳大学 | A kind of Weakly supervised characteristic analysis method and system based on 3D fingerprint image |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9892334B2 (en) * | 2015-11-01 | 2018-02-13 | Joshua Noel Hogan | Optical coherence tomography array based subdermal imaging device |
| KR102491855B1 (en) * | 2017-12-11 | 2023-01-26 | 삼성전자주식회사 | 3-dimensional finger print device and electronic device comprising the same |
-
2019
- 2019-06-17 CN CN201910522011.9A patent/CN110309738B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101593349A (en) * | 2009-06-26 | 2009-12-02 | 福州华映视讯有限公司 | Bidimensional image is converted to the method for 3-dimensional image |
| CN103810709A (en) * | 2014-02-25 | 2014-05-21 | 南京理工大学 | Vessel-based registration method for eye fundus image and SD-OCT projection image |
| CN106021610A (en) * | 2016-06-28 | 2016-10-12 | 电子科技大学 | Video fingerprint extracting method based on salient region |
| CN106407896A (en) * | 2016-08-29 | 2017-02-15 | 华蓥市高科德电子科技有限公司 | 3D fingerprint information extraction method for handset equipment |
| CN106778663A (en) * | 2016-12-29 | 2017-05-31 | 苏州塞罗尔医学影像科技有限公司 | A kind of fingerprint recognition system |
| CN107146262A (en) * | 2017-04-18 | 2017-09-08 | 清华大学深圳研究生院 | The three-dimensional visualization method and system of a kind of OCT image |
| CN108446633A (en) * | 2018-03-20 | 2018-08-24 | 深圳大学 | A kind of method, system and device of novel finger print automatic anti-fake and In vivo detection |
| CN108664940A (en) * | 2018-05-16 | 2018-10-16 | 山东大学 | A kind of partial fingerprints matching process and system |
| CN108932507A (en) * | 2018-08-06 | 2018-12-04 | 深圳大学 | A kind of automatic anti-fake method and system based on OCT fingerprint image |
| CN109377549A (en) * | 2018-09-29 | 2019-02-22 | 浙江工业大学 | A kind of real-time processing of OCT finger tip data and three-dimensional visualization method |
| CN109376687A (en) * | 2018-11-16 | 2019-02-22 | 深圳市趣创科技有限公司 | Shield lower unlocked by fingerprint discrimination method for improving, device, terminal and storage medium |
| CN109508692A (en) * | 2018-11-30 | 2019-03-22 | 深圳大学 | A kind of gender identification method and system based on 3D fingerprint image |
| CN109657567A (en) * | 2018-11-30 | 2019-04-19 | 深圳大学 | A kind of Weakly supervised characteristic analysis method and system based on 3D fingerprint image |
Non-Patent Citations (7)
| Title |
|---|
| 3D指纹的获取与重建技术;王恒 等;《加解密技术与图形识别》;20160731;第43、45页 * |
| Automation of Fingerprint Recognition Using OCT Fingerprint Images;Nasibe Akbari 等;《Journal of Signal and Information Processing》;20111109;第3卷(第1期);第117-121页 * |
| Capturing the vital vascular fingerprint with optical coherence tomography;Gangjun Liu 等;《APPLIED OPTICS》;20130801;第52卷(第22期);第5473-5477页 * |
| Egidijus Auksorius 等.Fast subsurface fingerprint imaging with full-field optical coherence tomography system equipped with a silicon camera.《Journal of Biomedical Optics》.2017,第22卷(第9期), * |
| Fast subsurface fingerprint imaging with full-field optical coherence tomography system equipped with a silicon camera;Egidijus Auksorius 等;《Journal of Biomedical Optics》;20170908;第22卷(第9期);第1-21页 * |
| 常用指纹细节特征标注方法的比较研究;李康 等;《刑事技术》;20180815;第43卷(第4期);第272-276页 * |
| 指纹自动识别与人工识别的差异及人机配合策略;王勇 等;《刑事技术》;20180103;第42卷(第6期);第501-503页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110309738A (en) | 2019-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dalla Mura et al. | Classification of hyperspectral images by using extended morphological attribute profiles and independent component analysis | |
| Zhou et al. | Digital image processing: part II | |
| CN112967236B (en) | Image registration method, device, computer equipment and storage medium | |
| JP6554900B2 (en) | Template creation apparatus and template creation method | |
| KR20120047991A (en) | Automatic identification of fingerprint inpainting target areas | |
| JP2013012190A (en) | Method of approximating gabor filter as block-gabor filter, and memory to store data structure for access by application program running on processor | |
| KR101618996B1 (en) | Sampling method and image processing apparatus for estimating homography | |
| JP4232804B2 (en) | Image density conversion method, image enhancement processing apparatus, and program thereof | |
| CN104281856B (en) | For the image pre-processing method and system of brain Medical Images Classification | |
| CN111179433A (en) | Three-dimensional modeling method and device for target object, electronic device and storage medium | |
| US20040179737A1 (en) | Method for contour extraction for object representation | |
| Khaleefah et al. | Review of local binary pattern operators in image feature extraction | |
| CN112950684B (en) | Target feature extraction method, device, equipment and medium based on surface registration | |
| CN110751680A (en) | Image processing method with fast alignment algorithm | |
| CN110309738B (en) | Method for labeling OCT fingerprint image | |
| KR20180034237A (en) | Image processing apparatus, image processing method, storage medium, and program | |
| US20090129699A1 (en) | Image processing system | |
| CN107590829B (en) | Seed point picking method suitable for multi-view dense point cloud data registration | |
| JP5657503B2 (en) | Reading apparatus, method and program | |
| Lindblad et al. | Exact linear time euclidean distance transforms of grid line sampled shapes | |
| Dickscheid et al. | Towards 3D reconstruction of neuronal cell distributions from histological human brain sections | |
| CN112766012B (en) | Two-dimensional code image recognition method and device, electronic equipment and storage medium | |
| Kim et al. | Depth map super-resolution using guided deformable convolution | |
| KR101756959B1 (en) | Image analyze method and apparatus thereby | |
| KR102511821B1 (en) | Methods and apparatus for processing image data for machine vision |
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 |