CN111639555A - Finger vein image noise accurate extraction and self-adaptive filtering denoising method and device - Google Patents
Finger vein image noise accurate extraction and self-adaptive filtering denoising method and device Download PDFInfo
- Publication number
- CN111639555A CN111639555A CN202010414619.2A CN202010414619A CN111639555A CN 111639555 A CN111639555 A CN 111639555A CN 202010414619 A CN202010414619 A CN 202010414619A CN 111639555 A CN111639555 A CN 111639555A
- Authority
- CN
- China
- Prior art keywords
- noise
- filtering
- finger vein
- response value
- template
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000001914 filtration Methods 0.000 title claims abstract description 92
- 210000003462 vein Anatomy 0.000 title claims abstract description 66
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000000605 extraction Methods 0.000 title claims abstract description 10
- 230000004044 response Effects 0.000 claims abstract description 67
- 238000009826 distribution Methods 0.000 claims abstract description 10
- 238000012216 screening Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000012795 verification Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 4
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 102000001554 Hemoglobins Human genes 0.000 description 1
- 108010054147 Hemoglobins Proteins 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 238000001727 in vivo Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- G06V40/1382—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger
- G06V40/1388—Detecting the live character of the finger, i.e. distinguishing from a fake or cadaver finger using image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Human Computer Interaction (AREA)
- Collating Specific Patterns (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a method and a device for accurately extracting noise of finger vein images and self-adaptive filtering and denoising, wherein the method comprises the following steps: 1) calculating a multidirectional curvature response value of the finger vein image based on a two-dimensional Gaussian template; 2) setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and extracting a noise region image; 3) and filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image. The device comprises a calculation module, a noise region extraction module and a noise region filtering module. The method can realize accurate and effective detection of the noise of the finger vein image and realize accurate and quick filtering of the detected noise areaTherefore, the identification performance of the finger vein image containing noise is effectively improved, and the actual use experience of the finger vein identification system is improved.
Description
Technical Field
The invention belongs to the technical field of biological feature recognition in information security, and particularly relates to a method and a device for accurately extracting noise of a finger vein image and self-adaptively filtering and denoising.
Background
The finger vein recognition technology belongs to the biological feature recognition technology, and is a typical technology in the second generation biological feature recognition technology. The method has the outstanding technical characteristics that the acquired biological characteristics are in vivo, and compared with the traditional biological characteristics such as fingerprints and human faces, the method has higher safety and non-replicability. The characteristic collection depends on the absorption of hemoglobin in blood to near infrared light, the living body identification is fundamentally ensured, and meanwhile, the finger vein lines have uniqueness and long-term invariance, so that the finger vein line is a biological characteristic which is highly safe and has great research value.
Researches find that cracks, peeling areas or dust on an acquisition window on a finger can be reflected in the finger vein image, the gray value of the finger vein image is generally lower than that of a finger vein line area, and the finger vein image belongs to a noise area. When noise exists in the finger vein image, the instability of the noise can cause great problems to the actual use experience of the finger vein recognition system: when the noise of the image is too much, the verification passing rate is reduced, and even the verification cannot be successful. Therefore, when noise exists in the finger vein image, the noise area needs to be detected and filtered for denoising, and the success rate of verification of the image containing the noise is improved.
The finger vein identification method disclosed in patent No. CN110119724A includes the following steps: performing ROI positioning on an image acquired by a high-definition camera; carrying out gray scale normalization processing on different parts of the image; carrying out mean filtering and Gaussian filtering on the image to eliminate noise; judging the center position of the texture by using curvature detection; enhancing the contrast of the gray level image through gamma conversion; judging edge pixel points by using a thinning method based on an index table; fusing, expanding and smoothing the extracted finger vein skeleton; finger vein recognition is carried out through template matching, Hu invariant moment and improved Zernike moment.
In the prior art, the mean filtering and the gaussian filtering are performed on the image to eliminate the noise, and the method for eliminating the noise has the defects that: the noise region is not accurately positioned, and the filtering process is not fast enough.
Disclosure of Invention
The invention aims to overcome the defects that in the prior art, the noise area of a finger vein image is not accurately positioned and the filtering processing speed is not high enough in the finger vein identification process, and provides a method and a device for accurately extracting noise of the finger vein image and adaptively filtering and denoising the noise.
In order to achieve the purpose, the technical scheme provided by the invention is as follows:
the invention relates to a method for accurately extracting noise of a finger vein image and adaptively filtering and denoising, which comprises the following steps of:
1) calculating a multidirectional curvature response value of the finger vein image based on a two-dimensional Gaussian template;
2) setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and a threshold value and extracting a noise region image;
3) and filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image.
Preferably, the specific steps of step 1) include:
1.1) constructing a curvature calculation operator in 8 directions according to a two-dimensional Gaussian function, wherein the two-dimensional Gaussian function is as follows:
wherein x and y represent the length and width of the template, respectively;
1.2) calculating the multidirectional curvature, wherein the calculation of the curvature involves the calculation of a first-order partial derivative, the calculation of a second-order partial derivative and the calculation of a Gaussian function in a straight line LθFirst and second directional derivatives in direction, wherein,
the first partial derivative is calculated by the formula:
the second partial derivative is calculated by the formula:
calculating the Gaussian function in a straight line LθThe calculation formula of the first-order and second-order directional derivatives in the direction is as follows:
in the formula, θ represents an included angle introduced as a certain straight line LθThe included angle with the positive direction of the x axis;
1.3) calculating a multidirectional curvature response value, wherein the calculation formula is as follows:
preferably, the step 2) specifically comprises:
2.1) setting the maximum curvature response value threshold T1According to the maximum curvature response value threshold T1Screening suspected noise points;
2.2) setting a curvature response value threshold S and a quantity threshold T exceeding the curvature response value threshold2According to the rate response threshold S and the number threshold T exceeding the curvature response threshold2Determining a noise area;
2.3) extracting a noise area image.
Preferably, the rate response threshold S is smaller than the maximum curvature response threshold T1。
Preferably, the magnitude of the rate response threshold S is: s ═ T1/2。
Preferably, the step 3) specifically comprises:
3.1) setting a proportion value eta of the noise point, and calculating the size of a maximum median filtering template according to the proportion value eta;
and 3.2) carrying out circle-by-circle median filtering from outside to inside according to the size of the maximum median filtering template to obtain the denoised finger vein image.
Preferably, the specific step of calculating the maximum median filtering template size in step 3.1) includes:
3.1.1) setting the template as a square template, wherein the initial side length of the template is 1;
3.1.2) calculating the proportion of noise points in all points in the size of the current template;
3.1.3) if the proportion of the noise points in all the points in the current template size is less than eta, determining the current template size as the size of the best matched median filtering template, if the proportion of the noise points in all the points in the current template size is greater than or equal to eta, increasing the side length on the basis of the side length of the current template, and returning to the step 3.1.2).
Preferably, the increase of the side length in step 3.1.3) is 1.
Preferably, the specific manner of the circle-by-circle median filtering in step 3.2) from outside to inside is as follows: the median filtering is defined to be performed once in four directions in each circle, namely from top to bottom, from bottom to top, from left to right, and from right to left, wherein only one noise point is processed each time, after the first circle is performed, the outermost noise point of the noise region in the image is filtered, at this time, the outermost noise point is marked as a normal point, the filtering of the next circle is performed, and the size of the maximum median filtering template needs to be recalculated each time the filtering of the circle is performed.
The invention also relates to a device for accurately extracting the noise of the finger vein image and adaptively filtering and denoising, which comprises:
1) the calculation module is used for calculating a multi-direction curvature response value of the finger vein image based on a two-dimensional Gaussian template;
2) a noise region extraction module for setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and extracting a noise region image;
3) and the noise region filtering module is used for filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image.
Compared with the prior art, the technical scheme provided by the invention has the following beneficial effects:
the invention provides a method for accurately extracting noise of a finger vein image and self-adapting filtering and denoising, which comprises the steps of firstly calculating a multidirectional curvature value of the finger vein image based on a two-dimensional Gaussian template, then extracting a noise region according to the curvature value, and filtering noise points through circle-by-circle self-adapting median filtering to realize noise region detection and noise filtering of the finger vein image.
Drawings
FIG. 1 is a flow chart of a method for accurately extracting noise of a vein image and adaptively filtering and denoising;
FIG. 2 is a grayscale image of a finger vein containing noise;
FIG. 3 is a graph of threshold T2Judging a suspected noise area image;
FIG. 4 is an image after adaptive round-by-round median filtering;
FIG. 5 is an example of a gray level image of a finger vein with different noise patterns;
FIG. 6 is a denoised image of a library of different noise images;
fig. 7 is a block diagram of vein image noise accurate extraction and adaptive filtering denoising.
Detailed Description
For further understanding of the present invention, the present invention will be described in detail with reference to examples, which are provided for illustration of the present invention but are not intended to limit the scope of the present invention.
Example one
With reference to fig. 1, the method for accurately extracting noise of finger vein image and adaptively filtering and denoising according to the present invention includes the following steps:
1) for a finger vein grayscale image, as shown in fig. 2, the image size is 320 × 140, it is obvious that there are noise points with low grayscale values and relatively dispersed distributions in the visible image, and the multidirectional curvature response value of the finger vein image is calculated based on a two-dimensional gaussian template, which specifically includes the steps of:
1.1) constructing a curvature calculation operator in 8 directions according to a two-dimensional Gaussian function, wherein the two-dimensional Gaussian function is as follows:
wherein x and y represent the length and width of the template, respectively, the two-dimensional gaussian template selected in this example is a 9 × 9 square template, and x ═ y ═ 9;
1.2) calculating the multidirectional curvature, wherein the calculation of the curvature involves the calculation of a first-order partial derivative, the calculation of a second-order partial derivative and the calculation of a Gaussian function in a straight line LθFirst and second directional derivatives in direction, wherein,
the first partial derivative is calculated by the formula:
the second partial derivative is calculated by the formula:
calculating the Gaussian function in a straight line LθThe calculation formula of the first-order and second-order directional derivatives in the direction is as follows:
in the formula, θ represents an included angle introduced as a certain straight line LθThe angle with the positive direction of the x-axis, in this embodiment, the angle parameter theta is 0 degree, 22.5 degrees, 45 degrees,67.5°,90°,112.5°,135°,157.5°;
1.3) calculating the multidirectional curvature response value KθThe calculation formula is as follows:
2) setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and a threshold value and extracting a noise region image, wherein the method specifically comprises the following steps:
2.1) since the curvature value of the noise region is generally large and since the continuity is poor, there are large values at a plurality of directional angles, a maximum curvature response value threshold T is set first1In this embodiment, T1Is 280 according to the maximum curvature response value threshold value T1Screening suspected noise points, namely when the maximum curvature value of a certain point is greater than T1Then, listing as a suspected noise point;
2.2) setting a rate response threshold S and a number threshold T that exceeds a curvature response threshold2Wherein the magnitude of the rate response threshold S is less than the maximum curvature response threshold T1In this embodiment, S ═ T1A number threshold T exceeding a curvature response threshold25, according to the rate response value threshold S and the number threshold T exceeding the curvature response value threshold2Determining a noise region, namely determining that the curvature response value exceeds 140 and the maximum curvature response value exceeds 280 in at least 5 directions, and considering that the current point is the determined noise region, extracting to obtain a noise region image as shown in fig. 4, as can be seen by comparing fig. 2 and fig. 3, the extracted noise region image can accurately reflect the noise distribution condition in the image, so as to achieve accurate detection.
3) After obtaining an accurate noise region image, filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain a denoised finger vein image, wherein the method specifically comprises the following steps:
3.1) because the size of the median filtering template has direct influence on the filtering effect and efficiency, the noise areas in the finger vein image have the phenomenon of different sizes, and if the whole image uses a larger median filtering template, the filtering time is longer; if a smaller template is used, the filtering effect is poor for some large noisy regions. Thus for a certain noise point, its optimal median filter template size needs to be determined. The median filtering principle is to select the median value after sorting all values in the template to replace the current value, if a certain point is a noise point, the gray value of the point is in the edge area of the sorting of all values in the size of the template, and it is required to ensure that the median value is close to the normal gray range, and it is required to ensure that the proportion of the number of the noise points in the current template cannot be too high, so when the size of the template for median filtering is determined, the proportion value eta of the noise points needs to be set, in this embodiment, the proportion value eta is 30%, that is, it is ensured that the proportion of the noise points in the template currently participating in median filtering does not exceed 30%, the filtering effect is better at this time, and the size of the maximum median filtering template, that
3.1.1) setting the template as a square template, wherein the initial side length of the template is 1;
3.1.2) calculating the proportion of noise points in all points in the size of the current template;
3.1.3) if the proportion of the noise points in all the points in the current template size is smaller than eta, determining that the current template size is the size of the median filtering template with the best match, if the proportion of the noise points in all the points in the current template size is larger than or equal to eta, increasing the side length on the basis of the side length of the current template, wherein the increase of the side length is 1 each time, returning to the step 3.1.2), and obtaining the maximum median filtering template side length of 6 by the method;
3.2) carrying out circle-by-circle median filtering from outside to inside according to the size of the maximum median filtering template to obtain a denoised finger vein image, wherein the specific mode is as follows: performing median filtering once in four directions in each circle, namely performing filtering in the next circle from top to bottom, from bottom to top, from left to right, and from right to left, wherein only one noise point is processed each time, after the first circle is performed, the outermost noise point of a noise area in the image is filtered, and at the moment, the outermost noise point is marked as a normal point, and then performing filtering in the next circle; it should be noted that: every time a circle of filtering is performed, the size of the maximum median filtering template needs to be recalculated, the filtered image is shown in fig. 4, as can be seen from comparison with fig. 2, the noise area in the image is basically filtered out completely, the vein lines become the main area with low gray value in the image, and the vein features are obvious.
In order to analyze the denoising effectiveness of the method, a noise image library with various forms and different types is selected, for example, as shown in fig. 5, images with 3 types of noise in total are used as comparison images, and simultaneously, 1 group of normally acquired image libraries which do not contain noise are also stored and used as template images. The corresponding situation of the 3 noise image libraries after denoising is shown in fig. 6, and it can be seen that the method of the present invention has a better filtering effect for the noise with different degrees and different distribution forms. The comparison test was based on the 4 image libraries described above, each containing 100 fingers, 5 images per finger. The normal image input system is used as a template, the 3 groups of noise image libraries are directly input into the system to be used as a verification image to obtain a verification result without denoising, the 3 groups of noise image libraries are input into the system to be used as a verification image after denoising by the method of the invention to obtain the verification result after denoising, the verification passing rate and the recognition rate are respectively counted, and when the recognition rate is counted, the finger template is shielded and only compared with the remaining 99 finger templates. For each noise image library, there were 1500 results of statistical verification success rate and 1500 results of statistical false recognition rate. A comparison of pre-and post-denoising success rates and false recognition rates was obtained as in table 1.
Table 1: comparison of success rate before and after denoising and false recognition rate
As can be seen from the results in table 1, the finger vein image includes noise, and the recognition rate is reduced to different degrees, and the more the noise is serious, the greater the reduction in the recognition rate is. The method can obviously improve the recognition rate of the noise image after denoising the noise image, the recognition rates of 3 noise image libraries with different degrees are respectively improved by 12.40%, 15.53% and 15.60%, the more serious the noise is, the higher the recognition rate is, and no new recognition is added after denoising in the aspect of the recognition rate, thereby showing that the denoising method can improve the overall recognition performance of the noise image library.
Example two
Referring to fig. 7, the present embodiment relates to a device for accurately extracting noise from finger vein images and adaptively filtering and denoising, including:
1) the calculation module is used for calculating a multi-direction curvature response value of the finger vein image based on a two-dimensional Gaussian template; the computing module is a module for realizing the functions of the step 1) of the embodiment.
2) A noise region extraction module for setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and extracting a noise region image; the noise region extraction module is a module for realizing the function of step 2) of the embodiment.
3) And the noise region filtering module is used for filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image. The noise region filtering module is a module for realizing the function of the step 3) of the embodiment.
The present invention has been described in detail with reference to the embodiments, but the description is only for the preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. All equivalent changes and modifications made within the scope of the present invention shall fall within the scope of the present invention.
Claims (10)
1. A finger vein image noise accurate extraction and self-adaptive filtering denoising method is characterized in that: which comprises the following steps:
1) calculating a multidirectional curvature response value of the finger vein image based on a two-dimensional Gaussian template;
2) setting a maximum curvature response value threshold T1A curvature response value threshold value S and a quantity threshold value exceeding the curvature response value threshold valueT2Determining a noise region according to the curvature response value distribution condition and a threshold value and extracting a noise region image;
3) and filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image.
2. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 1, wherein: the specific steps of the step 1) comprise:
1.1) constructing a curvature calculation operator in 8 directions according to a two-dimensional Gaussian function, wherein the two-dimensional Gaussian function is as follows:
wherein x and y represent the length and width of the template, respectively;
1.2) calculating the multidirectional curvature, wherein the calculation of the curvature involves the calculation of a first-order partial derivative, the calculation of a second-order partial derivative and the calculation of a Gaussian function in a straight line LθFirst and second directional derivatives in direction, wherein,
the first partial derivative is calculated by the formula:
the second partial derivative is calculated by the formula:
calculating the Gaussian function in a straight line LθThe calculation formula of the first-order and second-order directional derivatives in the direction is as follows:
in the formula, θ represents an included angle introduced as a certain straight line LθThe included angle with the positive direction of the x axis;
1.3) calculating a multidirectional curvature response value, wherein the calculation formula is as follows:
3. the method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 1, wherein: the step 2) specifically comprises the following steps:
2.1) setting the maximum curvature response value threshold T1According to the maximum curvature response value threshold T1Screening suspected noise points;
2.2) setting a rate response threshold S and a number threshold T that exceeds a curvature response threshold2According to the rate response threshold S and the number threshold T exceeding the curvature response threshold2Determining a noise area;
2.3) extracting a noise area image.
4. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 3, wherein: the rate response value threshold S is smaller than the maximum curvature response value threshold T1。
5. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 4, wherein: the curvature response value threshold value S has the following size: s ═ T1/2。
6. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 1, wherein: the step 3) specifically comprises the following steps:
3.1) setting a proportion value eta of the noise point, and calculating the size of a maximum median filtering template according to the proportion value eta;
and 3.2) carrying out circle-by-circle median filtering from outside to inside according to the size of the maximum median filtering template to obtain the denoised finger vein image.
7. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 6, wherein: the specific step of calculating the size of the maximum median filtering template in step 3.1) includes:
3.1.1) setting the template as a square template, wherein the initial side length of the template is 1;
3.1.2) calculating the proportion of noise points in all points in the size of the current template;
3.1.3) if the proportion of the noise points in all the points in the current template size is less than eta, determining the current template size as the size of the best matched median filtering template, if the proportion of the noise points in all the points in the current template size is greater than or equal to eta, increasing the side length on the basis of the side length of the current template, and returning to the step 3.1.2).
8. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 7, wherein: the increment of the side length in step 3.1.3) is 1.
9. The method for accurately extracting noise and adaptively filtering and denoising the finger vein image according to claim 6, wherein: the specific way of the circle-by-circle median filtering from outside to inside in the step 3.2) is as follows: the median filtering is defined to be performed once in four directions in each circle, namely from top to bottom, from bottom to top, from left to right, and from right to left, wherein only one noise point is processed each time, after the first circle is performed, the outermost noise point of the noise region in the image is filtered, at this time, the outermost noise point is marked as a normal point, the filtering of the next circle is performed, and the size of the maximum median filtering template needs to be recalculated each time the filtering of the circle is performed.
10. The utility model provides a finger vein image noise accurate extraction and self-adaptation filtering denoising device which characterized in that: it includes:
1) the calculation module is used for calculating a multi-direction curvature response value of the finger vein image based on a two-dimensional Gaussian template;
2) a noise region extraction module for setting a maximum curvature response value threshold T1A curvature response value threshold S and a number threshold T exceeding the curvature response value threshold2Determining a noise region according to the curvature response value distribution condition and extracting a noise region image;
3) and the noise region filtering module is used for filtering the noise region by a circle-by-circle self-adaptive median filtering method to obtain the denoised finger vein image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010414619.2A CN111639555B (en) | 2020-05-15 | 2020-05-15 | Finger vein image noise accurate extraction and adaptive filtering denoising method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010414619.2A CN111639555B (en) | 2020-05-15 | 2020-05-15 | Finger vein image noise accurate extraction and adaptive filtering denoising method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111639555A true CN111639555A (en) | 2020-09-08 |
CN111639555B CN111639555B (en) | 2023-06-20 |
Family
ID=72330834
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010414619.2A Active CN111639555B (en) | 2020-05-15 | 2020-05-15 | Finger vein image noise accurate extraction and adaptive filtering denoising method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111639555B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113790818A (en) * | 2021-08-10 | 2021-12-14 | 中国电子科技集团公司第十三研究所 | Visible light heat reflection temperature measurement method and temperature measurement equipment |
CN115272684A (en) * | 2022-09-29 | 2022-11-01 | 山东圣点世纪科技有限公司 | Method for processing pseudo noise in vein image enhancement process |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20100049407A (en) * | 2008-11-03 | 2010-05-12 | 서울대학교산학협력단 | Method and apparatus for finger vein identification using mean curvature |
US20130051674A1 (en) * | 2010-05-07 | 2013-02-28 | Bart Goossens | Method and device for estimating noise in a reconstructed image |
CN103020958A (en) * | 2012-11-22 | 2013-04-03 | 北京理工大学 | Vessel automatic matching method based on curvature scale space |
JP5951817B1 (en) * | 2015-02-09 | 2016-07-13 | コリセン カンパニー リミテッドKorecen Co.,Ltd. | Finger vein authentication system |
EP3057034A1 (en) * | 2015-02-10 | 2016-08-17 | Korecen Co., Ltd. | Finger vein authentication system |
CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
CN106127151A (en) * | 2016-06-23 | 2016-11-16 | 北京海鑫科金高科技股份有限公司 | Based on the finger vein identification method and the device that improve local binary model |
CN108256456A (en) * | 2018-01-08 | 2018-07-06 | 杭州电子科技大学 | A kind of finger vein identification method based on multiple features Threshold Fusion |
CN108520214A (en) * | 2018-03-28 | 2018-09-11 | 五邑大学 | A kind of finger vein identification method based on multiple dimensioned HOG and SVM |
CN109145792A (en) * | 2018-08-09 | 2019-01-04 | 钧安科技(深圳)有限公司 | Two fingers setting refers to vein identification device and method |
CN110119724A (en) * | 2019-05-16 | 2019-08-13 | 天津科技大学 | A kind of finger vein identification method |
CN110188614A (en) * | 2019-04-30 | 2019-08-30 | 杭州电子科技大学 | It is a kind of based on skin crack segmentation NLM filtering refer to vein denoising method |
CN110751029A (en) * | 2019-09-12 | 2020-02-04 | 南京邮电大学 | Maximum curvature-based adaptive finger vein line extraction method |
CN110852280A (en) * | 2019-11-13 | 2020-02-28 | 圣点世纪科技股份有限公司 | Detection algorithm for mirror foreign matter of finger vein equipment |
-
2020
- 2020-05-15 CN CN202010414619.2A patent/CN111639555B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20100049407A (en) * | 2008-11-03 | 2010-05-12 | 서울대학교산학협력단 | Method and apparatus for finger vein identification using mean curvature |
US20130051674A1 (en) * | 2010-05-07 | 2013-02-28 | Bart Goossens | Method and device for estimating noise in a reconstructed image |
CN103020958A (en) * | 2012-11-22 | 2013-04-03 | 北京理工大学 | Vessel automatic matching method based on curvature scale space |
JP5951817B1 (en) * | 2015-02-09 | 2016-07-13 | コリセン カンパニー リミテッドKorecen Co.,Ltd. | Finger vein authentication system |
EP3057034A1 (en) * | 2015-02-10 | 2016-08-17 | Korecen Co., Ltd. | Finger vein authentication system |
CN106096569A (en) * | 2016-06-20 | 2016-11-09 | 广州中国科学院先进技术研究所 | A kind of finger vein identification method |
CN106127151A (en) * | 2016-06-23 | 2016-11-16 | 北京海鑫科金高科技股份有限公司 | Based on the finger vein identification method and the device that improve local binary model |
CN108256456A (en) * | 2018-01-08 | 2018-07-06 | 杭州电子科技大学 | A kind of finger vein identification method based on multiple features Threshold Fusion |
CN108520214A (en) * | 2018-03-28 | 2018-09-11 | 五邑大学 | A kind of finger vein identification method based on multiple dimensioned HOG and SVM |
CN109145792A (en) * | 2018-08-09 | 2019-01-04 | 钧安科技(深圳)有限公司 | Two fingers setting refers to vein identification device and method |
CN110188614A (en) * | 2019-04-30 | 2019-08-30 | 杭州电子科技大学 | It is a kind of based on skin crack segmentation NLM filtering refer to vein denoising method |
CN110119724A (en) * | 2019-05-16 | 2019-08-13 | 天津科技大学 | A kind of finger vein identification method |
CN110751029A (en) * | 2019-09-12 | 2020-02-04 | 南京邮电大学 | Maximum curvature-based adaptive finger vein line extraction method |
CN110852280A (en) * | 2019-11-13 | 2020-02-28 | 圣点世纪科技股份有限公司 | Detection algorithm for mirror foreign matter of finger vein equipment |
Non-Patent Citations (9)
Title |
---|
LI, JN ET AL: "Finger vein feature extraction based on improved maximum curvature description", 《PROCEEDING OF THE 38TH CHINESE CONTROL CONFERENCE》 * |
LI, JN ET AL: "Finger vein feature extraction based on improved maximum curvature description", 《PROCEEDING OF THE 38TH CHINESE CONTROL CONFERENCE》, 17 October 2019 (2019-10-17), pages 7566 - 7571 * |
NAMBURETE, AIL ET AL: "Computational methods for quantifying in vivo muscle fascicle curvature from ultrasound images", 《JOURNAL OF BIOMECHANICS》 * |
NAMBURETE, AIL ET AL: "Computational methods for quantifying in vivo muscle fascicle curvature from ultrasound images", 《JOURNAL OF BIOMECHANICS》, vol. 44, no. 14, 12 August 2011 (2011-08-12), pages 2538 - 2543 * |
周昌雄;于盛林;祖克举;: "基于L-曲率流滤波器的图像降噪算法", 《光学精密工程》, no. 06, pages 759 - 765 * |
孙晓琳: "指静脉图像模式提取算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
孙晓琳: "指静脉图像模式提取算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 01, 15 January 2013 (2013-01-15), pages 138 - 1698 * |
蓝师伟: "低图像质量下的手指静脉识别算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
蓝师伟: "低图像质量下的手指静脉识别算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 01, 15 January 2020 (2020-01-15), pages 138 - 1234 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113790818A (en) * | 2021-08-10 | 2021-12-14 | 中国电子科技集团公司第十三研究所 | Visible light heat reflection temperature measurement method and temperature measurement equipment |
CN113790818B (en) * | 2021-08-10 | 2023-09-26 | 中国电子科技集团公司第十三研究所 | Visible light heat reflection temperature measurement method and temperature measurement equipment |
CN115272684A (en) * | 2022-09-29 | 2022-11-01 | 山东圣点世纪科技有限公司 | Method for processing pseudo noise in vein image enhancement process |
CN115272684B (en) * | 2022-09-29 | 2022-12-27 | 山东圣点世纪科技有限公司 | Method for processing pseudo noise in vein image enhancement process |
Also Published As
Publication number | Publication date |
---|---|
CN111639555B (en) | 2023-06-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109035163B (en) | Self-adaptive image denoising method based on deep learning | |
CN107767387B (en) | Contour detection method based on variable receptive field scale global modulation | |
CN108009520A (en) | A kind of finger vein identification method and system based on convolution variation self-encoding encoder neutral net | |
CN106548176B (en) | Finger vein image enhancement method based on self-adaptive guide filtering | |
CN107066942A (en) | A kind of living body faces recognition methods and system | |
CN107729820B (en) | Finger vein identification method based on multi-scale HOG | |
CN110543822A (en) | finger vein identification method based on convolutional neural network and supervised discrete hash algorithm | |
CN104091145B (en) | Human body slaps arteries and veins characteristic image acquisition method | |
US11475707B2 (en) | Method for extracting image of face detection and device thereof | |
CN104331683B (en) | A kind of facial expression recognizing method with noise robustness | |
CN104239769A (en) | Identity recognition method and system based on finger vein characteristics | |
CN112597812A (en) | Finger vein identification method and system based on convolutional neural network and SIFT algorithm | |
CN110188614A (en) | It is a kind of based on skin crack segmentation NLM filtering refer to vein denoising method | |
CN106127193B (en) | A kind of facial image recognition method | |
US20180144179A1 (en) | Image processing device, image processing method, and image processing program | |
CN108960185A (en) | Vehicle target detection method and system based on YOLOv2 | |
CN111639555B (en) | Finger vein image noise accurate extraction and adaptive filtering denoising method and device | |
CN114973307B (en) | Finger vein recognition method and system for generating antagonism and cosine ternary loss function | |
CN111223063A (en) | Finger vein image NLM denoising method based on texture features and binuclear function | |
CN115994870B (en) | Image processing method for enhancing denoising | |
CN110046587A (en) | Human face expression feature extracting method based on Gabor difference weight | |
CN107742302B (en) | Contour detection method based on primary visual cortex multi-scale contour fusion | |
CN107784263A (en) | Based on the method for improving the Plane Rotation Face datection for accelerating robust features | |
CN110458773A (en) | A kind of anisotropy parameter method for processing noise based on edge enhancement operator | |
CN107153807A (en) | A kind of non-greedy face identification method of two-dimensional principal component analysis |
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 | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A precise noise extraction and adaptive filtering denoising method and device for finger vein images Granted publication date: 20230620 Pledgee: Bank of China Limited Taiyuan Binzhou sub branch Pledgor: Holy Point Century Technology Co.,Ltd. Registration number: Y2024140000011 |