CN109409237B - Fingerprint feature processing method and device - Google Patents
Fingerprint feature processing method and device Download PDFInfo
- Publication number
- CN109409237B CN109409237B CN201811137203.XA CN201811137203A CN109409237B CN 109409237 B CN109409237 B CN 109409237B CN 201811137203 A CN201811137203 A CN 201811137203A CN 109409237 B CN109409237 B CN 109409237B
- Authority
- CN
- China
- Prior art keywords
- fingerprint
- fingerprint image
- template
- feature
- characteristic
- 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
- 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
-
- 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
-
- 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
- G06V40/1365—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
Abstract
The invention discloses a fingerprint feature processing method and a fingerprint feature processing device, which comprise a template generation step and a fingerprint identification step: the template generation step comprises: repeatedly collecting a first fingerprint image for multiple times; extracting fingerprint features for multiple times from a first fingerprint image repeatedly acquired for multiple times; if the total amount of the multiple fingerprint features exceeds a preset threshold value, generating a feature template; the fingerprint identification step comprises the following steps: acquiring current second fingerprint image data; extracting current fingerprint features in the second fingerprint image data; and comparing the current fingerprint characteristics with the characteristic template to obtain an identification result. The beneficial effects of the implementation of the invention are as follows: the security problem of the fingerprint characteristic data is solved, and the fingerprint characteristic data of the user can be protected; an encryption and decryption method is not adopted, so that the problem of algorithm cracking does not exist; the fingerprint is collected, processed and stored through the fingerprint characteristic processing device, and the terminal such as a mobile phone cannot read the fingerprint characteristic data, so that the privacy data of a user is protected, and the problem of data leakage does not exist.
Description
Technical Field
The present invention relates to the field of fingerprint feature processing technologies, and in particular, to a fingerprint feature processing method and apparatus.
Background
The fingerprint is widely applied to the fields of security authentication and the like due to uniqueness and uniqueness, and the fingerprint module is used as an indispensable acquisition device of fingerprint characteristic data and is often appeared in the related security authentication process.
The existing fingerprint sensor needs to upload fingerprint data to a mobile phone or a fingerprint lock for identification processing, the fingerprint characteristic data can be stolen in the transmission process, and the fingerprint data can be obtained and stored by a background database of a mobile phone/fingerprint lock manufacturer, and the fingerprint data can be used for identity authentication without consent. Therefore, how to enhance the security of the fingerprint feature data becomes an urgent problem to be solved.
Disclosure of Invention
The invention aims to provide a fingerprint feature processing method and a fingerprint feature processing device, which solve the problems that the existing fingerprint sensor needs to upload fingerprint data to a mobile phone or a fingerprint lock for identification processing, the fingerprint feature data can be obtained and stored by a background database of a mobile phone/fingerprint lock manufacturer except that the fingerprint feature data is stolen in the transmission process, and the fingerprint data can be used for identity authentication without agreement.
The technical scheme of the invention is realized as follows:
the invention provides a fingerprint feature processing method, which comprises the steps of template generation and fingerprint identification:
the template generating step includes steps S11-S13:
s11, repeatedly collecting the first fingerprint image for multiple times;
s12, extracting fingerprint characteristics for multiple times from the first fingerprint image repeatedly acquired for multiple times;
s13, if the total amount of the multiple fingerprint features exceeds a preset threshold value, generating a feature template;
the fingerprint recognizing step includes steps S21-S23:
s21, acquiring current second fingerprint image data;
s22, extracting the current fingerprint feature in the second fingerprint image data;
and S23, comparing the current fingerprint features with the feature template to obtain an identification result.
In the fingerprint feature processing method according to the present invention, the step S12 includes: steps S121-S125:
s121, performing noise reduction processing on the first fingerprint image by adopting median filtering;
s122, performing binarization processing on the first fingerprint image subjected to the noise reduction processing;
s123, thinning the first fingerprint image subjected to binarization processing;
s124, extracting characteristic points of the first fingerprint image subjected to thinning processing to obtain fingerprint characteristics;
and S125, repeating the steps S121-S124 for multiple times to acquire the fingerprint characteristics for multiple times.
In the fingerprint feature processing method of the present invention, the step S121 includes steps S1211-S1212:
s1211, traversing all pixel points in the first fingerprint image to obtain current pixel points;
and S1212, replacing the current pixel point by using the median value of eight pixel points of one grid around the current pixel point.
In the fingerprint feature processing method of the present invention, the step S122 includes steps S1221 to S1223:
s1221, converting the first fingerprint image into a histogram, and extracting a gray value meeting a preset condition from the histogram as a pixel threshold;
s1222, setting the value of the pixel point which is larger than the pixel threshold value in the first fingerprint image after the noise reduction processing to be 1;
and S1223, setting the value of the pixel point which is smaller than or equal to the pixel threshold value in the first fingerprint image subjected to noise reduction processing to be 0.
In the fingerprint feature processing method of the present invention, the step S1221 includes steps S12211 to S12213:
s12211, converting the first fingerprint image into a histogram;
s12212, acquiring histogram values corresponding to gray values of 100 to 156, and selecting a minimum value from the acquired histogram values;
and S12213, taking the gray value corresponding to the selected minimum value as a pixel threshold value.
In the fingerprint feature processing method according to the present invention, the step S123 includes steps S1231 to S1233:
s1231, traversing all pixel points in the first fingerprint image to obtain a current pixel point;
s1232, calculating the weighted sum of the values of four pixel points of one adjacent lattice of the current pixel point;
and S1233, judging whether the weighted sum is larger than the pixel threshold value, if so, setting the value of the current pixel point to be 1, and if not, setting the value of the current pixel point to be 0.
In the fingerprint feature processing method of the present invention, the step S124 includes steps S1241 to S1244:
s1241, traversing all pixel points in the first fingerprint image to obtain current pixel points;
s1242, calculating absolute values of differences between the current pixel point and values of eight pixel points of one grid around the current pixel point;
s1243, calculating the sum of all absolute values;
and S1244, if half of the sum is 7 or 3, the current pixel point is a feature point, otherwise, the current pixel point is not a feature point.
In the fingerprint feature processing method of the present invention, step S13 includes steps S131 to S134:
s131, acquiring any two first fingerprint images from the first fingerprint images repeatedly acquired for multiple times;
s132, splicing the two acquired first fingerprint images, and taking the spliced images as an initial template;
s133, respectively calculating the similarity between the first fingerprint images which are repeatedly acquired for multiple times except the spliced two first fingerprint images and the initial template:
Ri＝(Si-Ai)/Si
wherein R isiSimilarity of the ith first fingerprint image of the first fingerprint images repeatedly acquired multiple times except for the spliced two first fingerprint images and the initial template, SiIs the sum of the values of the pixel points of the ith first fingerprint image of the first fingerprint images repeatedly acquired for multiple times except for the spliced two first fingerprint images and the initial template, AiThe difference between the pixel point value of the ith first fingerprint image of the first fingerprint images repeatedly acquired for multiple times except for the spliced two first fingerprint images and the pixel point value of the initial template;
and S134, correcting the initial template according to the similarity to generate a feature template.
In the fingerprint feature processing method of the present invention, step S23 includes steps S231-S235:
s231, calculating a direction field of the second fingerprint image and a direction field of the characteristic template;
s232, respectively rotating the second fingerprint image and the characteristic template according to the direction field of the second fingerprint image and the direction field of the characteristic template so as to enable the directions of the second fingerprint image and the characteristic template to be consistent;
s233, matching the second fingerprint image and the characteristic template to obtain the number of characteristic points matched with the second fingerprint image and the characteristic template;
s234, calculating the matching rate of the second fingerprint image and the feature template:
wherein C is the matching rate of the second fingerprint image and the feature template, N is the number of feature points matched with the second fingerprint image and the feature template, and N is the number of feature points matched with the second fingerprint image and the feature template1Number of feature points of the second fingerprint image, N2R is the similarity between the second fingerprint image and the feature template:
R＝(S-A)/S
s is the sum of the values of the pixel points of the second fingerprint image and the characteristic template, and A is the difference of the values of the pixel points of the second fingerprint image and the characteristic template;
and S235, acquiring an identification result according to the matching rate.
In another aspect, a fingerprint feature processing apparatus is provided, configured to perform the above fingerprint feature processing method, where the fingerprint feature processing apparatus includes:
the acquisition module is used for acquiring a first fingerprint image and a second fingerprint image;
the processing module is used for executing the template generation step and the fingerprint identification step;
the storage module is used for storing the characteristic template; the acquisition module, the processing module and the storage module are connected with each other through a system bus.
Therefore, the invention has the advantages that the safety problem of the fingerprint characteristic data is solved, and the fingerprint characteristic data of the user can be protected; an encryption and decryption method is not adopted, so that the problem that the algorithm is cracked does not exist; fingerprint collection, processing and storage are carried out through the fingerprint feature processing device, and terminals such as a mobile phone or a fingerprint lock cannot read fingerprint feature data, so that privacy data of a user are protected, and the problem of data leakage after decryption does not exist.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
fig. 1 is a flowchart of a fingerprint feature processing method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of the positions of a current pixel and eight pixels of a grid around the current pixel;
FIG. 3 is a schematic diagram of the positions of a current pixel and four pixels of an adjacent grid;
fig. 4 is a block diagram of a fingerprint feature processing apparatus according to an embodiment of the present invention.
Detailed Description
In order to more clearly understand the technical features, objects, and effects of the present invention, embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It is to be understood that the following description is only a specific illustration of the embodiments of the present invention and should not be taken as limiting the scope of the invention.
Referring to fig. 1, fig. 1 is a flowchart of a fingerprint feature processing method according to an embodiment of the present invention, where the fingerprint feature processing method includes a template generation step and a fingerprint identification step:
the template generating step includes steps S11-S13:
s11, repeatedly collecting the first fingerprint image for multiple times; preferably, the first fingerprint image is collected by the collecting module, and the first fingerprint image of the same user is collected for multiple times.
S12, extracting fingerprint characteristics for multiple times from the first fingerprint image repeatedly acquired for multiple times; the step S12 includes: steps S121-S125:
s121, performing noise reduction processing on the first fingerprint image by adopting median filtering; the step S121 includes steps S1211-S1212:
s1211, traversing all the pixel points in the first fingerprint image to obtain the current pixel point.
And S1212, replacing the current pixel point by using the median value of eight pixel points of one grid around the current pixel point. Referring to fig. 2, fig. 2 is a schematic diagram of positions of a current pixel and eight pixels of a surrounding grid, where a dotted pixel in the diagram is the current pixel, and a white pixel is the eight pixels of the surrounding grid. For example: firstly, median filtering is adopted to carry out noise reduction processing on a fingerprint image, all pixel points are traversed during a specific process, and the median of 9 adjacent pixel points around the current pixel point is used for replacing the current pixel point, so that isolated noise points are eliminated. The median value here is calculated using the gray values of the pixels. Specifically, the gray values of 9 pixels are sorted, and the gray value in the 5 th row is the median.
S122, performing binarization processing on the first fingerprint image subjected to the noise reduction processing; the step S122 includes steps S1221 to S1223:
s1221, converting the first fingerprint image into a histogram, and extracting a gray value meeting a preset condition from the histogram as a pixel threshold; the step S1221 includes steps S12211 to S12213:
s12211, converting the first fingerprint image into a histogram; for example: the fingerprint image is subjected to binarization processing, and the threshold value is obtained by obtaining a histogram of the image.
S12212, acquiring histogram values corresponding to gray values of 100 to 156, and selecting a minimum value from the acquired histogram values; the range of the gray value of the pixel point of the image is 0 to 255, after the histogram of the image is solved, the 256 gray values all have corresponding 256 histogram values, so that 56 histogram values corresponding to the 56 gray values with the gray value in the interval of 100 to 156 can be found.
And S12213, taking the gray value corresponding to the selected minimum value as a pixel threshold value. That is, the gray value corresponding to the histogram with the minimum value among the 56 histogram values is selected as the threshold.
And S1222, setting the value of the pixel point which is greater than the pixel threshold value in the first fingerprint image after the noise reduction processing to be 1.
S1223, setting the value of the pixel point less than or equal to the pixel threshold in the first fingerprint image after the noise reduction processing to 0, that is, for the 8-bit fingerprint data, the pixel point greater than the threshold is set to 1, and the pixel point less than or equal to the threshold is set to 0.
S123, thinning the first fingerprint image subjected to binarization processing; the step S123 includes steps S1231 to S1233:
and S1231, traversing all pixel points in the first fingerprint image to obtain the current pixel point.
And S1232, calculating the weighted sum of the values of four pixels of one adjacent lattice of the current pixel. Referring to fig. 3, fig. 3 is a schematic diagram of positions of a current pixel and four pixels of an adjacent cell, where a dotted pixel in the diagram is the current pixel and a white pixel is the four pixels of the adjacent cell. The weighted sum here is preferably a linear weighted sum, which is to be understood as: assuming n parameters x1, x2, x3... xn, with corresponding weight coefficients p1, p2, p3... pn, the weighted sum is: s ═ p1 × 1+ p2 × 2+ p3 × 3+, + pn × xn ═ Σ (pi × xi).
And S1233, judging whether the weighted sum is larger than the pixel threshold value, if so, setting the value of the current pixel point to be 1, and if not, setting the value of the current pixel point to be 0. The fingerprint image after binarization is refined, and the specific method is that all pixel points are traversed, the weighted sum of four adjacent points around the pixel point is calculated, if the weighted sum is greater than a threshold value, the pixel point is 1, otherwise, the pixel point is 0.
S124, extracting characteristic points of the first fingerprint image subjected to thinning processing to obtain fingerprint characteristics; the step S124 includes steps S1241 to S1244:
s1241, traversing all pixel points in the first fingerprint image to obtain current pixel points; the extraction of the feature points is based on the refined fingerprint image data, and all pixel points are traversed.
S1242, calculating absolute values of differences between the current pixel point and values of eight pixel points of one grid around the current pixel point; the value of the pixel point is a binarized value, which is 1 or 0.
And S1243, calculating the sum of all absolute values.
And S1244, if half of the sum is 7 or 3, the current pixel point is a feature point, otherwise, the current pixel point is not a feature point. For example: if the current pixel point pixel value is X and the surrounding 8 pixel point pixel values are Xi, the current pixel point is an endpoint if the current pixel point pixel value is X0-X/2 +. the. II Xi-X/2 (i ranges from 0 to 7); if the calculation result is 3, the current pixel point is a cross point; and the calculation result is of other values, and the current pixel point is not the characteristic point. Both the end points and the cross points are feature points, which are used for better processing of the subsequent images.
And S125, repeating the steps S121-S124 for multiple times to acquire the fingerprint characteristics for multiple times. That is, the first fingerprint images of the same user are acquired multiple times, and steps S121-S124 are performed on each first fingerprint image to acquire fingerprint features multiple times.
S13, if the total amount of the multiple fingerprint features exceeds a preset threshold value, generating a feature template; step S13 includes steps S131-S134:
s131, acquiring any two first fingerprint images from the first fingerprint images repeatedly acquired for multiple times; for example: calculating the direction fields of the two images, rotating the two images according to the direction fields, matching the characteristic points of the two images, and then calculating the similarity of the overlapped partial images, wherein the more the matched characteristic points are, the higher the similarity is. The similarity between the two images is calculated as in step S133.
S132, splicing the two acquired first fingerprint images, and taking the spliced images as an initial template; preferably, when the registered fingerprint image (first fingerprint image) reaches 5 frames, the correlation degree between the images is calculated (the higher the similarity is, the higher the correlation degree is, the correlation degree can be generally represented by adopting the similarity), two images with the highest correlation degree are taken for splicing, and the spliced images are taken as the initial template.
S133, respectively calculating the similarity between the first fingerprint images which are repeatedly acquired for multiple times except the spliced two first fingerprint images and the initial template:
Ri＝(Si-Ai)/Si
wherein R isiSimilarity of the ith first fingerprint image of the first fingerprint images repeatedly acquired multiple times except for the spliced two first fingerprint images and the initial template, SiIs the sum of the values of the pixel points of the ith first fingerprint image of the first fingerprint images repeatedly acquired for multiple times except for the spliced two first fingerprint images and the initial template, AiThe difference between the pixel point value of the ith first fingerprint image of the first fingerprint images repeatedly acquired for multiple times except for the spliced two first fingerprint images and the pixel point value of the initial template;s abovei、AiAll the values of (a) can be obtained by setting corresponding programs. Where i is 1, 2, 3 … … n, n being the number of first fingerprint images acquired repeatedly in addition to the two first fingerprint images being stitched.
And S134, correcting the initial template according to the similarity to generate a feature template. For example: and calculating the similarity of other registered images and the initial template, splicing the image with the highest similarity and the initial template into a new template, sequentially ranking and pushing, and ending the splicing process to generate the feature template when the fingerprint image with the similarity larger than a preset threshold value with the spliced template cannot be found.
The splicing is to select a matching mode with the highest relevance degree for splicing according to the matching conditions of the direction fields and the characteristic points of the two images when the relevance degrees of the two images are calculated.
The fingerprint recognizing step includes steps S21-S23:
and S21, acquiring the current second fingerprint image data.
S22, extracting the current fingerprint feature in the second fingerprint image data; like the step S12 of extracting fingerprint features from the first fingerprint image repeatedly acquired a plurality of times, since the second fingerprint image data is only one, the steps S121-S124 need only be executed once to extract the current fingerprint features in the second fingerprint image data, so the step S22 includes the following steps S221-224:
s221, performing noise reduction processing on the second fingerprint image by adopting median filtering;
s222, performing binarization processing on the second fingerprint image subjected to the noise reduction processing;
s223, thinning the second fingerprint image subjected to the binarization processing;
and S224, extracting the characteristic points of the second fingerprint image subjected to the thinning processing to obtain the fingerprint characteristics. The details of steps S221-224 are similar to steps S121-S124 and are not described herein.
And S23, comparing the current fingerprint features with the feature template to obtain an identification result. Step S23 includes steps S231-S235:
and S231, calculating the direction field of the second fingerprint image and the direction field of the characteristic template.
S232, respectively rotating the second fingerprint image and the characteristic template according to the direction field of the second fingerprint image and the direction field of the characteristic template so as to enable the directions of the second fingerprint image and the characteristic template to be consistent; for example: calculating a direction field of a matched sample image (second fingerprint image) and a direction field of a template image (characteristic template), rotating the images according to the direction fields to enable the directions of the sample and the template to be consistent, then pairing the characteristic points of the sample and the template, and calculating the similarity of the overlapped image areas in each pairing mode.
S233, matching the second fingerprint image and the characteristic template to obtain the number of characteristic points matched with the second fingerprint image and the characteristic template;
s234, calculating the matching rate of the second fingerprint image and the feature template:
wherein C is the matching rate of the second fingerprint image and the feature template, N is the number of feature points matched with the second fingerprint image and the feature template, and N is the number of feature points matched with the second fingerprint image and the feature template1Number of feature points of the second fingerprint image, N2R is the similarity between the second fingerprint image and the feature template:
R＝(S-A)/S
s is the sum of the values of the pixel points of the second fingerprint image and the characteristic template, and A is the difference of the values of the pixel points of the second fingerprint image and the characteristic template;
and S235, acquiring an identification result according to the matching rate. For example: and when the maximum matching rate is greater than a preset threshold value, the identification is passed, otherwise, the identification is not passed, and the threshold value is calculated based on a large amount of experimental data to obtain.
Referring to fig. 4, fig. 4 is a block diagram of a fingerprint feature processing apparatus according to an embodiment of the present invention, the apparatus is configured to execute the fingerprint feature processing method, and the fingerprint feature processing apparatus includes a processing module 2, an acquisition module 1, and a storage module 3. For example, the fingerprint feature processing device is an integrated chip, and the processing module 2, the collecting module 1 and the storage module 3 are all packaged in the integrated chip.
The acquisition module 1 is used for acquiring a first fingerprint image and a second fingerprint image; the acquisition module 1 is responsible for acquiring fingerprint characteristic data and can be a capacitive acquisition device, an ultrasonic acquisition device, an optical acquisition device and the like. Preferably, the acquisition module 1 mainly comprises a Pixel and an ADC (analog to digital converter), wherein the Pixel circuit represents the fingerprint circuit by using a voltage signal, and can utilize the principle of capacitance charge and discharge and the principle of different light reflection distances; the ADC circuit converts the generated voltage signal into image Data and stores the image Data in the Data SRAM. That is, the collecting module is responsible for collecting fingerprint image data (first fingerprint image), and the fingerprint of the user is collected into the image data by the hardware circuit. Similarly, when the fingerprint identification is authenticated, the current fingerprint image data (second fingerprint image) is collected by the collection module 1.
The processing module 2 is used for executing the template generation step and the fingerprint identification step; the fingerprint feature processing device integrates the processing module 2(MCU, DSP, CPU, etc.), and the processing module 2 is used for running an identification algorithm, fingerprint feature data does not need to be transmitted to the outside of the sensor for processing, a reading interface of the fingerprint feature data is fundamentally cut off from hardware, the data safety is protected, and the problem of the fingerprint data safety is thoroughly solved. The processing module is mainly used for running various algorithms of fingerprint identification, including a feature extraction algorithm, a feature synthesis algorithm and an identification comparison algorithm, the processing module can be an MCU or a DSP, and the MCU can adopt processor IP cores of arm company, such as Cortex M3 and M4.
The collected image data is sent to a processing module, and the processing module can operate a feature extraction algorithm to extract features from the fingerprint image data; the characteristic template is formed by combining a certain number of characteristics, if sufficient characteristics are not obtained, the acquisition of fingerprints through the acquisition module 1 is repeated, and more characteristics are obtained by extracting fingerprint characteristics through the processing module 2; after sufficient features are obtained, the processing module 2 runs a feature combination algorithm to generate a feature template. Similarly, during fingerprint identification, the processing module 2 reads the feature template stored during registration from the storage module 3, runs an identification comparison algorithm, and compares the feature extracted at the current time with the feature template to obtain an identification result.
The storage module 3 is used for storing the characteristic template; i.e. the generated feature template will be saved to the storage module 3.
The acquisition module 1, the processing module 2 and the storage module 3 are connected with each other through a system bus. The memory module 3 may be constituted by ROM, Flash, EPPROM, etc. Preferably, the storage module 3 is used to store the feature template, the program Code, the intermediate Data of the algorithm operation, and the fingerprint image Data, wherein the ROM is used to store the program Code, the Code SRAM is used to provide space for operating the program Code, and the Data SRAM is used to store the feature template and the intermediate Data. The system bus is preferably an Arbiter bus.
In order to ensure that the fingerprint feature processing device communicates with the external device, the fingerprint feature processing device further includes an interface module 4, the interface module 4 is responsible for communicating with the external device, and the interface module 4 includes but is not limited to an SPI interface, an I2C interface, a USB interface, and a UART interface. The interface module adopts SPI protocol interface to interact with external terminal, also can include interface protocols such as I2C, USB, UART, the SPI interface is responsible for with the various communication interaction of external terminal. That is, the registration result is notified to the external terminal through the interface template 4, and other processing items are performed by the external terminal. Similarly, in the case of fingerprint recognition, the interface template 4 notifies the external terminal of the recognition result.
In addition, the fingerprint feature processing device further comprises a power management module 5, the power management module 5 preferably comprises Bandgap, LDO25, POR and OSC, and the power management module 5 is responsible for supplying power to all the modules of the whole sensor.
In summary, although the present invention has been described with reference to the preferred embodiments, the above-described preferred embodiments are not intended to limit the present invention, and those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, therefore, the scope of the present invention shall be determined by the appended claims.
Claims (9)
1. A fingerprint feature processing method is characterized by comprising the steps of template generation and fingerprint identification:
the template generating step includes steps S11-S13:
s11, repeatedly collecting the first fingerprint image for multiple times;
s12, extracting fingerprint characteristics for multiple times from the first fingerprint image repeatedly acquired for multiple times;
s13, if the total amount of the multiple fingerprint features exceeds a preset threshold value, generating a feature template;
the fingerprint recognizing step includes steps S21-S23:
s21, acquiring current second fingerprint image data;
s22, extracting the current fingerprint feature in the second fingerprint image data;
s23, comparing the current fingerprint features with the feature template to obtain an identification result;
step S13 includes steps S131-S134:
s131, acquiring any two first fingerprint images from the first fingerprint images repeatedly acquired for multiple times;
s132, splicing the two acquired first fingerprint images, and taking the spliced images as an initial template;
s133, respectively calculating the similarity between the first fingerprint images which are repeatedly acquired for multiple times except the spliced two first fingerprint images and the initial template:
Ri＝(Si-Ai)/Si
wherein R isiSimilarity of the ith first fingerprint image of the first fingerprint images repeatedly acquired multiple times except for the spliced two first fingerprint images and the initial template, SiIs the sum of the values of the pixel points of the ith first fingerprint image of the first fingerprint images repeatedly acquired for multiple times except for the spliced two first fingerprint images and the initial template, AiFor the ith first fingerprint image of the first fingerprint images repeatedly acquired a plurality of times other than the spliced two first fingerprint imagesThe difference between the fingerprint image and the value of the pixel point of the initial template;
and S134, correcting the initial template according to the similarity to generate a feature template.
2. The fingerprint feature processing method according to claim 1, wherein the step S12 includes: steps S121-S125:
s121, performing noise reduction processing on the first fingerprint image by adopting median filtering;
s122, performing binarization processing on the first fingerprint image subjected to the noise reduction processing;
s123, thinning the first fingerprint image subjected to binarization processing;
s124, extracting characteristic points of the first fingerprint image subjected to thinning processing to obtain fingerprint characteristics;
and S125, repeating the steps S121-S124 for multiple times to acquire the fingerprint characteristics for multiple times.
3. The fingerprint feature processing method according to claim 2, wherein the step S121 comprises steps S1211-S1212:
s1211, traversing all pixel points in the first fingerprint image to obtain current pixel points;
and S1212, replacing the current pixel point by using the median value of eight pixel points of one grid around the current pixel point.
4. The fingerprint feature processing method according to claim 2, wherein the step S122 comprises steps S1221 to S1223:
s1221, converting the first fingerprint image into a histogram, and extracting a gray value meeting a preset condition from the histogram as a pixel threshold;
s1222, setting the value of the pixel point which is larger than the pixel threshold value in the first fingerprint image after the noise reduction processing to be 1;
and S1223, setting the value of the pixel point which is smaller than or equal to the pixel threshold value in the first fingerprint image subjected to noise reduction processing to be 0.
5. The fingerprint feature processing method according to claim 4, wherein the step S1221 comprises steps S12211 to S12213:
s12211, converting the first fingerprint image into a histogram;
s12212, acquiring histogram values corresponding to gray values of 100 to 156, and selecting a minimum value from the acquired histogram values;
and S12213, taking the gray value corresponding to the selected minimum value as a pixel threshold value.
6. The fingerprint feature processing method according to claim 4, wherein the step S123 comprises steps S1231-S1233:
s1231, traversing all pixel points in the first fingerprint image to obtain a current pixel point;
s1232, calculating the weighted sum of the values of four pixel points of one adjacent lattice of the current pixel point;
and S1233, judging whether the weighted sum is larger than the pixel threshold value, if so, setting the value of the current pixel point to be 1, and if not, setting the value of the current pixel point to be 0.
7. The fingerprint feature processing method according to claim 4, wherein the step S124 comprises steps S1241-S1244:
s1241, traversing all pixel points in the first fingerprint image to obtain current pixel points;
s1242, calculating absolute values of differences between the current pixel point and values of eight pixel points of one grid around the current pixel point;
s1243, calculating the sum of all absolute values;
and S1244, if half of the sum is 7 or 3, the current pixel point is a feature point, otherwise, the current pixel point is not a feature point.
8. The fingerprint feature processing method according to claim 1, wherein step S23 includes steps S231-S235:
s231, calculating a direction field of the second fingerprint image and a direction field of the characteristic template;
s232, respectively rotating the second fingerprint image and the characteristic template according to the direction field of the second fingerprint image and the direction field of the characteristic template so as to enable the directions of the second fingerprint image and the characteristic template to be consistent;
s233, matching the second fingerprint image and the characteristic template to obtain the number of characteristic points matched with the second fingerprint image and the characteristic template;
s234, calculating the matching rate of the second fingerprint image and the feature template:
wherein C is the matching rate of the second fingerprint image and the feature template, N is the number of feature points matched with the second fingerprint image and the feature template, and N is the number of feature points matched with the second fingerprint image and the feature templateiNumber of feature points of the second fingerprint image, N2R is the similarity between the second fingerprint image and the feature template:
R＝(S-A)/S
s is the sum of the values of the pixel points of the second fingerprint image and the characteristic template, and A is the difference of the values of the pixel points of the second fingerprint image and the characteristic template;
and S235, acquiring an identification result according to the matching rate.
9. A fingerprint feature processing apparatus for performing the fingerprint feature processing method according to any one of claims 1 to 8, the fingerprint feature processing apparatus comprising:
the acquisition module is used for acquiring a first fingerprint image and a second fingerprint image;
the processing module is used for executing the template generation step and the fingerprint identification step;
the storage module is used for storing the characteristic template; the acquisition module, the processing module and the storage module are connected with each other through a system bus.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811137203.XA CN109409237B (en) | 2018-09-28 | 2018-09-28 | Fingerprint feature processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811137203.XA CN109409237B (en) | 2018-09-28 | 2018-09-28 | Fingerprint feature processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109409237A CN109409237A (en) | 2019-03-01 |
| CN109409237B true CN109409237B (en) | 2022-03-29 |
Family
ID=65466492
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811137203.XA Active CN109409237B (en) | 2018-09-28 | 2018-09-28 | Fingerprint feature processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109409237B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112837222A (en) * | 2021-01-25 | 2021-05-25 | 深圳市奔凯安全技术股份有限公司 | Fingerprint image splicing method and device, storage medium and electronic equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104281841A (en) * | 2014-09-30 | 2015-01-14 | 深圳市汇顶科技股份有限公司 | Fingerprint identification system and fingerprint processing method and device thereof |
| CN106778461A (en) * | 2016-03-17 | 2017-05-31 | 深圳信炜科技有限公司 | Finger prints processing method, finger prints processing device, fingerprint recognition system and electronic equipment |
| CN108121946A (en) * | 2017-11-15 | 2018-06-05 | 大唐微电子技术有限公司 | A kind of Pre-processing Method for Fingerprint Image and device |
-
2018
- 2018-09-28 CN CN201811137203.XA patent/CN109409237B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104281841A (en) * | 2014-09-30 | 2015-01-14 | 深圳市汇顶科技股份有限公司 | Fingerprint identification system and fingerprint processing method and device thereof |
| CN106778461A (en) * | 2016-03-17 | 2017-05-31 | 深圳信炜科技有限公司 | Finger prints processing method, finger prints processing device, fingerprint recognition system and electronic equipment |
| CN108121946A (en) * | 2017-11-15 | 2018-06-05 | 大唐微电子技术有限公司 | A kind of Pre-processing Method for Fingerprint Image and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109409237A (en) | 2019-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Benzaoui et al. | Ear biometric recognition using local texture descriptors | |
| Cvejic et al. | Region-based multimodal image fusion using ICA bases | |
| KR101183391B1 (en) | Image comparison by metric embeddings | |
| KR101637229B1 (en) | Apparatus and method for extracting feature point based on SIFT, and face recognition system using thereof | |
| CN111814574A (en) | Face living body detection system, terminal and storage medium applying double-branch three-dimensional convolution model | |
| CN104751040A (en) | Fingerprint detection method based on intelligent mobile information equipment | |
| WO2019047248A1 (en) | Feature extraction method and device for hyperspectral remotely sensed image | |
| CN112487886A (en) | Method and device for identifying face with shielding, storage medium and terminal | |
| CN112668544B (en) | Pedestrian re-identification method based on hard sample confusion and feature robustness enhancement | |
| RU2628201C1 (en) | Method of adaptive quantization for encoding iris image | |
| Panella et al. | A smartphone-based application using machine learning for gesture recognition: Using feature extraction and template matching via Hu image moments to recognize gestures | |
| Krish et al. | Pre‐registration of latent fingerprints based on orientation field | |
| Mursi et al. | An improved SIFT-PCA-based copy-move image forgery detection method | |
| Bruno et al. | Palmprint principal lines extraction | |
| CN109409237B (en) | Fingerprint feature processing method and device | |
| Dwaich et al. | Signature texture features extraction using GLCM approach in android studio | |
| Abdo et al. | Iris recognition based on histogram equalization and discrete cosine transform | |
| Patil et al. | Expression invariant face recognition using semidecimated DWT, Patch-LDSMT, feature and score level fusion | |
| US20200380288A1 (en) | Proposal region filter for digital image processing | |
| Narang et al. | Robust face recognition method based on SIFT features using Levenberg-Marquardt Backpropagation neural networks | |
| Walker et al. | Locating salient facial features using image invariants | |
| Alim et al. | The most discriminant subbands for face recognition: A novel information-theoretic framework | |
| CN111079551A (en) | Finger vein identification method and device based on singular value decomposition and storage medium | |
| Mehrotra et al. | An efficient dual stage approach for iris feature extraction using interest point pairing | |
| Kanchana et al. | Quadtree decomposition for palm print feature representation in palmprint recognition system |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20211208 Address after: 518000 Guangdong city of Shenzhen province Nanshan District South Road No. 10 four Applicant after: Wang Haibin Address before: 518000 room 003, floor 9, Beike building, No. 18, Keyuan Road, high tech Industrial Park, Yuehai street, Nanshan District, Shenzhen, Guangdong Applicant before: SHENZHEN HAOHAIXIN TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |