CN109815791B - Blood vessel-based identity recognition method and device - Google Patents
Blood vessel-based identity recognition method and device Download PDFInfo
- Publication number
- CN109815791B CN109815791B CN201811526968.2A CN201811526968A CN109815791B CN 109815791 B CN109815791 B CN 109815791B CN 201811526968 A CN201811526968 A CN 201811526968A CN 109815791 B CN109815791 B CN 109815791B
- Authority
- CN
- China
- Prior art keywords
- sub
- block
- similarity
- vein image
- descriptor
- 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
Abstract
The embodiment of the invention provides an identity recognition method and device based on blood vessels. The blood vessel-based identity recognition method comprises the following steps: acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of the preset central pixel, and extracting a descriptor of each sub-block; according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in the vein image library; determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library; wherein the descriptor is used for describing the characteristics of the sub-block. According to the blood vessel-based identity recognition method and device provided by the embodiment of the invention, the descriptor of each sub-block of the vein image to be recognized is obtained, and the identity information corresponding to the vein image to be recognized is determined according to the descriptor of each sub-block, so that the robustness is better, and the recognition efficiency and accuracy can be improved when non-rigid deformation exists.
Description
Technical Field
The embodiment of the invention relates to the technical field of image processing, in particular to an identity recognition method and device based on blood vessels.
Background
With the increasing requirements for property security and personal identity authentication, biometric identification technology has been developed rapidly in security. Among them, a common method in the blood vessel-based identification method is a vein identification method. The hand vein recognition method is of interest to both the industrial circle and the academic circle for the following two reasons. First, the hand vessels stabilize and the vascular pattern varies from person to person. Second, the hand veins are below the skin and are therefore difficult to tamper with. However, the hand posture is not constant in the acquisition process, so that the vein image has obvious deformation, and therefore, the robust and rapid hand blood vessel-based identity recognition method is of great importance.
The existing blood vessel-based identity recognition method mainly comprises a texture-based recognition method. Most of texture-based blood vessel-based identity recognition methods are based on texture information of images to extract pixel features and use a global matching method to perform feature matching, but the existing methods have the following two defects: firstly, the pixel-based features are single and not robust enough; secondly, the global matching method cannot effectively process non-rigid deformation in the image to complete the matching of the image. Due to the defects, the identification accuracy of the existing blood vessel-based identity identification method is not high.
Disclosure of Invention
In view of the problems in the prior art, embodiments of the present invention provide a blood vessel-based identification method and apparatus that overcome the above problems or at least partially solve the above problems.
In a first aspect, an embodiment of the present invention provides an identity recognition method based on a blood vessel, including:
acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of a preset central pixel, and extracting a descriptor of each sub-block;
according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in a vein image library;
determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in a vein image library;
wherein the descriptor is used for describing the characteristics of the sub-block.
In a second aspect, an embodiment of the present invention provides a blood vessel-based identity recognition apparatus, including:
the descriptor acquisition module is used for acquiring a plurality of sub-blocks of the vein image to be identified according to the preset interval of the central pixel and extracting a descriptor of each sub-block;
the similarity obtaining module is used for obtaining the similarity between the vein image to be identified and each sample image in a vein image library according to the descriptor of each sub-block;
the identity recognition module is used for determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library;
wherein the descriptor is used for describing the characteristics of the sub-block.
In a third aspect, an embodiment of the present invention provides an electronic device, including:
at least one processor; and
at least one memory communicatively coupled to the processor, wherein:
the memory stores program instructions executable by the processor to invoke the method of vessel-based identification provided by any of the various possible implementations of the first aspect.
In a fourth aspect, embodiments of the present invention provide a non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform a blood vessel based identification method as provided in any one of the various possible implementations of the first aspect.
According to the blood vessel-based identity recognition method and device provided by the embodiment of the invention, the plurality of sub-blocks of the vein image to be recognized are obtained, the descriptor of each sub-block is extracted, the similarity between the vein image to be recognized and the sample image is obtained according to the descriptor of each sub-block, the identity information corresponding to the vein image to be recognized is determined, the robustness is better, and the recognition efficiency and accuracy can be improved when non-rigid deformation exists.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a blood vessel-based identification method according to an embodiment of the present invention;
FIG. 2 is a functional block diagram of a blood vessel based identification apparatus provided in accordance with an embodiment of the present invention;
fig. 3 is a block diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The embodiments of the present invention, and all other embodiments obtained by a person of ordinary skill in the art without any inventive step, are within the scope of the present invention.
In order to overcome the above problems in the prior art, embodiments of the present invention provide a blood vessel-based identity recognition method, which includes establishing a descriptor of a regional subblock having a distinguishing force, finding a local optimal match according to the descriptor, determining a similarity of an image according to the local optimal match, and performing blood vessel-based identity recognition according to the similarity of the image to realize vein-based identity recognition.
In the embodiments of the present invention, the identification based on the blood vessel is vein identification.
Fig. 1 is a schematic flow chart of a blood vessel-based identification method according to an embodiment of the present invention. As shown in fig. 1, a blood vessel-based identification method includes: s101, acquiring a plurality of sub-blocks of the vein image to be identified according to a preset interval of central pixels, and extracting a descriptor of each sub-block; wherein the descriptor is used for describing the characteristics of the sub-block.
It should be noted that the vein image to be identified and the sample image in the vein image library in the implementation of the present invention are both vein images of the back of the hand. The sample image refers to a vein image of each user collected in advance, and the identity information of the user is a label of the sample image. All sample images constitute a vein image library.
It should be noted that the size of the vein image to be recognized is the same as the size of each sample image in the vein image library, and is a preset size. For example, the preset size is 81 × 81 pixels. It can be understood that, if the size of the vein image to be recognized is not the preset size, the size of the vein image to be recognized is adjusted to the preset size.
And after the vein image to be identified is obtained, obtaining a plurality of sub-blocks from the vein image to be identified. And the sub-block refers to an image of one area in the vein image to be identified, and is a part of the vein image to be identified.
Specifically, the position of the central pixel of each sub-block may be determined according to a preset interval of the central pixel, so as to divide the vein image to be identified into a plurality of portions, and the sub-blocks may be obtained according to each of the divided portions.
It should be noted that the size of the vein image to be recognized, i.e., the preset size, is an integral multiple of the interval of the preset central pixels. For example, the size of the vein image to be recognized is 81 × 81 pixels, and the interval of the preset center pixels may be 3 pixels, thereby determining the positions of 9 × 9 center pixels.
And acquiring a plurality of sub-blocks of the vein image to be identified, and extracting the characteristics of each sub-block to serve as a descriptor of the sub-block. And the characteristics of the sub-blocks are obtained according to the characteristics of each pixel contained in the sub-blocks. The characteristics of the pixels can be obtained according to the gray-scale values of the pixels.
And S102, acquiring the similarity between the vein image to be identified and each sample image in the vein image library according to the descriptor of each sub-block.
In order to perform blood vessel-based identification, the similarity between the vein image to be identified and each sample image in the vein image library needs to be acquired.
For the vein image to be identified and any sample image, the similarity of the vein image to be identified and any sample image is obtained based on the descriptor of each sub-block of the vein image to be identified. Specifically, based on the descriptor of each sub-block of the vein image to be identified, region matching is performed on the sample image, and the similarity between the sub-block and the region matched with the sub-block is obtained. And obtaining the similarity of the vein image to be identified and the sample image according to the similarity of each sub-block and the region matched with the sub-block.
And S103, determining the identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library.
After the similarity between the vein image to be recognized and each sample image in the vein image library is obtained, the sample image matched with the vein image to be recognized in the vein image library can be determined based on the similarity, the identity information represented by the label of the sample image is determined as the identity information corresponding to the vein image to be recognized, and therefore the vein image to be recognized is corresponding to the identity information of the user, namely the user corresponding to the vein image to be recognized is determined.
The embodiment of the present invention is not limited in particular to a specific method for determining a sample image in a vein image library that matches a vein image to be identified. For example, the sample image with the largest similarity may be determined as the image matching the vein image to be identified.
According to the embodiment of the invention, the similarity between the vein image to be recognized and the sample image is obtained according to the descriptor of each sub-block by obtaining the plurality of sub-blocks of the vein image to be recognized and extracting the descriptor of each sub-block, and the identity information corresponding to the vein image to be recognized is determined, so that the method and the device have better robustness, and the recognition efficiency and accuracy can be improved when non-rigid deformation exists.
Based on the content of the above embodiments, the specific step of extracting the descriptor of each sub-block includes: for each sub-block, acquiring a feature vector of each pixel contained in the sub-block; and normalizing and averaging the feature vectors of the pixels contained in the sub-blocks to obtain the descriptors of the sub-blocks.
Specifically, since the sub-block includes a plurality of pixels, a feature vector of each pixel included in the sub-block is extracted first. The feature vector of the pixel can be obtained according to the gray value of the pixel.
The feature vectors of the pixels included in the sub-block may be averaged to obtain the descriptor of the sub-block. Since there is a significant difference between the pixels contained in the blood vessel region and the non-blood vessel region, normalization can be performed when averaging the feature vectors of the pixels contained in the sub-block. Normalization can be performed by introducing a normalization constant or a normalization parameter. Normalization is a dimensionless processing means, which changes the absolute value of the feature vector into a certain relative value relation, and can simplify the calculation and reduce the magnitude.
The descriptor of the sub-block can be obtained by normalizing and averaging the feature vectors of the pixels included in the sub-block by the following calculation formula.
Wherein R isp(i) A value representing the ith element of the descriptor of the sub-block; FVj(i) A value indicating the ith element of the feature vector of the jth pixel included in the sub-block; μ denotes a normalization constant; m represents the dimension of the feature vector of the pixel (the number of elements included); n represents the number of pixels included in the sub-block.
The values of m and n can be chosen according to the actual implementation, for example, m is 8 and n is 9.
It is understood that the feature vector of the pixel includes m elements, and the (m +1) th element of the descriptor of the sub-block can be constructed according to the values of the elements of the feature vectors, so that the descriptor of the sub-block includes (m +1) elements.
According to the embodiment of the invention, the descriptor of the sub-block is obtained by normalizing and averaging the feature vectors of the pixels, so that the feature of the sub-block has better robustness, the robustness of the blood vessel-based identity recognition method can be improved, the calculation can be simplified, and the time consumption is reduced.
Based on the content of the foregoing embodiments, the specific step of obtaining the feature vector of each pixel included in the sub-block includes: and for each pixel contained in the sub-block, filtering the vein image to be identified through an eight-direction filter, and forming a feature vector of the pixel according to the response of the pixel in eight directions.
Specifically, the feature vector of a pixel is composed of responses of the pixel in eight directions when the vein image to be recognized is filtered by an eight-direction filter.
The directional filter is widely used for image processing such as image compression, enhancement, edge detection, denoising and the like.
It should be noted that after the subblocks of the vein image to be identified are obtained, the vein image to be identified is filtered through the eight-direction filter, and the response of each pixel in the eight directions is obtained, so that the feature vector of each pixel and the descriptor of each subblock are obtained; or filtering the vein image to be identified by an eight-direction filter to obtain the characteristic vector of each pixel, and then determining the position of each subblock and the pixels contained in the subblock according to the interval of a preset central pixel so as to obtain the descriptors of each subblock.
Since the filter response is almost 0 in the non-blood vessel region, normalizing the feature vector can reduce the similarity between the blood vessel region and the non-blood vessel region, thereby effectively distinguishing the blood vessel region from the non-blood vessel region.
According to the embodiment of the invention, the vein image to be recognized is filtered through the eight-direction filter, the characteristic vector of the pixel is obtained, the blood vessel region and the non-blood vessel region can be effectively distinguished, and thus the robustness and the accuracy of the blood vessel-based identity recognition method can be improved.
Based on the content of each embodiment, the specific step of obtaining the similarity between the vein image to be identified and each sample image in the vein image library according to the descriptor of each sub-block includes: for each sample image in the vein image library, determining a region corresponding to each sub-block in the sample image; acquiring a similarity graph of each sub-block according to the descriptor of each sub-block and the feature vector of each pixel contained in the region corresponding to the sub-block; and according to the similarity of each sub-block, obtaining the similarity of the vein image to be identified and the sample image.
Specifically, when the similarity between the vein image to be identified and each sample image in the vein image library is obtained, for each sub-block of the vein image to be identified, an area corresponding to the sub-block in the sample image is determined.
The position of the sub-block in the vein image to be identified corresponds to the position of the region corresponding to the sub-block in the sample image. The position of the central pixel of the sub-block in the vein image to be identified is the same as the position of the central pixel of the region corresponding to the sub-block in the sample image. The side length of the region corresponding to the sub-block is a preset value and is larger than the preset interval of the central pixels. The region corresponding to the sub-block may be a square region surrounded by the side length with the center pixel of the sub-block as the center.
For each sub-block of the vein image to be identified, a similarity map of the sub-block can be obtained according to the descriptor of the sub-block and the descriptors of the sub-blocks contained in the region corresponding to the sub-block. And the similarity graph of the sub-blocks is used for describing the similarity of the sub-blocks and the corresponding regions of the sub-blocks. The sub-block included in the corresponding region refers to a sub-block whose central pixel is located in the corresponding region.
It is understood that the feature vector of any pixel of the sample image is filtered by the eight-direction filter, and then the pixel is in eight directions. The method for determining the position of each sub-block in the sample image is the same as the method for determining the position of each sub-block in the image to be identified, and is not described herein again. The method for obtaining the descriptor of any sub-block in the sample image is the same as the method for obtaining the descriptor of any sub-block in the image to be identified, and is not described herein again.
Because the similarity of the sub-block describes the similarity of the sub-block and the corresponding region of the sub-block, after the similarity of each sub-block is obtained, the similarity of the vein image to be identified and the sample image can be obtained according to the similarity of each sub-block.
According to the embodiment of the invention, the similarity between the vein image to be identified and the sample image is obtained according to the similarity between each subblock and the corresponding region of the subblock, the similarity of the subblock is calculated to ensure that the subblock finds the local optimal matching, and the identification efficiency and accuracy can be improved when non-rigid deformation exists.
Based on the content of the foregoing embodiments, the specific step of obtaining the similarity graph of each sub-block according to the descriptor of each sub-block and the descriptors of the sub-blocks included in the region corresponding to the sub-block includes: for each sub-block, according to the descriptor of the sub-block and the descriptors of the sub-blocks contained in the region corresponding to the sub-block, obtaining the similarity between the sub-block and the sub-blocks contained in the region corresponding to the sub-block; and respectively taking the similarity between the sub-block and each sub-block contained in the region corresponding to the sub-block as the value of the pixel corresponding to each sub-block contained in the region corresponding to the sub-block in the similarity graph of the sub-block.
It is understood that the size of the similarity map of sub-blocks corresponds to the number of sub-blocks included in the region to which the sub-blocks correspond.
Any pixel A in the similar graph of the sub-block corresponds to a sub-block contained in the region corresponding to the sub-block. The position (coordinates) of the pixel a in the similarity map of the sub-block corresponds to the position of the sub-block B in the above-described area.
The coordinate of the central pixel of the sub-block B' in the image to be recognized is p, and specifically, p can be selected from pxAnd pyRespectively, the abscissa and the ordinate. In any sample image, the coordinate of the central pixel of the region corresponding to the sub-block B' is p, and the side length of the corresponding region is 2L, so that the coordinate of the central pixel of any sub-block included in the region corresponding to the sub-block in the sample image can be represented as p + (Δ p)x,Δpy) In particular, can be represented by px+ΔpxAnd py+ΔpyRespectively representing the abscissa and the ordinate of the central pixel of any sub-block contained in the region corresponding to the sub-block in the sample image. (Δ p)x,Δpy) Indicates the amount of shift, and also indicates the coordinate of the center pixel in the sample image as p + (Δ p)x,Δpy) The sub-block of (B) corresponds to the coordinates of the pixels in the similarity map of B'. Δ pxAnd Δ pyThe value ranges from-L to L.
For sub-block B' and the central pixel coordinate in the sample image is p + (Δ p)x,Δpy) A sub-block of (2), a calculation of similarity between the twoThe formula is as follows:
wherein, Cp(Δpx,Δpy) Indicates the position of (Δ p) in the similar diagramx,Δpy) The value of the pixel of (a);
the sub-block B' in the image to be identified, which represents the coordinate of the central pixel as p, and the coordinate of the central pixel in the sample image as p + (delta p)x,Δpy) Similarity between the sub-blocks of (a); rpA descriptor of a sub-block B' in the image to be identified, which represents the coordinate p of the central pixel; rp(i) Represents RpThe value of the ith element of (a);
the coordinate representing the center pixel in the sample image is p + (Δ p)x,Δpy) A descriptor of the sub-block of (1);
the coordinate representing the center pixel in the sample image is p + (Δ p)x,Δpy) The value of the ith element of the descriptor of the sub-block of (1).
It can be understood that, the calculation process of the similarity between the sub-block in the image to be identified and the sub-block in the sample image is to perform point multiplication on descriptors of the sub-block and the sub-block according to the same element position and then perform averaging. The calculation process is similar to convolution.
After the values of the pixels in the similarity map of the sub-block are obtained, the similarity map of the sub-block can be obtained.
According to the method and the device, the similarity graph of the sub-block is obtained according to the descriptor of the sub-block in the image to be identified and the descriptor of each sub-block contained in the region corresponding to the sub-block in the sample image, the similarity of the sub-block and the region corresponding to the sub-block can be reflected more accurately, and therefore a more accurate identification result based on the blood vessel can be obtained.
Based on the content of the above embodiments, the specific step of obtaining the similarity between the vein image to be identified and the sample image according to the similarity map of each sub-block includes: after the similarity images of each sub-block are gathered, pooling is carried out according to a preset window and a preset step length until the similarity between the vein image to be identified and the sample image is obtained.
Specifically, after the similar image of each sub-block is obtained, the similar images of the sub-blocks are gathered to obtain the similar image of the vein image to be identified.
When the similar images of the sub-blocks are aggregated, the aggregation can be performed according to the positions of the sub-blocks. For example, the relative positional relationship of the similarity map of each sub-block is the same as the relative positional relationship of each sub-block.
And after the similar images of each sub-block are converged to obtain the similar image of the vein image to be identified, pooling the similar images of the vein image to be identified according to a preset window and step length until a numerical value is obtained, and taking the numerical value as the similarity between the vein image to be identified and the sample image.
It should be noted that the size number of the vein image to be identified is an integral multiple of the side length and the step length of the preset window, so that a value is obtained through iterative pooling. For example, the size of the vein image to be recognized is 81 × 81 pixels, the preset window is 3 × 3, and the preset step size is 3.
And (4) iterative pooling, namely continuously pooling the result of the last pooling. That is, the object of the current pooling is the result of the previous pooling.
Preferably, the similarity maps of the vein images to be identified are pooled using maximal pooling.
The embodiment of the invention obtains the similarity between the vein image to be identified and the sample image by performing the maximum pooling on the similar images of the collected sub-blocks, can ensure the maximum similarity response value of the local sub-blocks, realizes the local optimal matching of the sub-blocks in the collecting process, and can obtain a more accurate identification result based on the blood vessel.
Based on the content of the above embodiments, the specific step of obtaining the plurality of sub-blocks of the vein image to be recognized according to the preset interval of the central pixel includes: determining the position of the central pixel of each sub-block in the vein image to be identified according to the preset interval of the central pixel; and taking the central pixel of each sub-block as a circle center and the interval as a circular area determined by the diameter as the sub-block.
Specifically, when a plurality of sub-blocks of the vein image to be identified are obtained, the position of the central pixel of each sub-block in the vein image to be identified is determined according to the preset interval of the central pixel.
After the position of the central pixel of each sub-block is determined, a plurality of square areas can be determined as original sub-areas by taking the central pixel of the sub-block as the center and the interval of the preset central pixel as the side length.
The original sub-region may be directly used as a sub-block, or an inscribed circle region of the original sub-region may be used as a sub-block.
When the inscribed circle region of the original sub-region is taken as a sub-block, the specific steps of obtaining the sub-block are as follows: and for the central pixel of each sub-block, determining a circular area by taking the central pixel of the sub-block as a circle center and the preset interval of the central pixels as diameters, and taking the circular area as the sub-block.
Accordingly, since the other pixels included in the circular region are not the actual pixels in the vein image to be identified except for the central pixel, the feature vectors of the other pixels included in the circular region are obtained by interpolation according to the feature vectors of the pixels included in the original sub-region corresponding to the circular region except for the central pixel.
The embodiment of the invention takes the circular area with the central pixel as the circle center and the interval as the diameter determination as the circular sub-block, can reflect the filtering response of the blood vessel in each direction by utilizing the non-deformation of the rotation of the circle, and has better characteristic acquisition continuity, thereby obtaining more accurate identification result based on the blood vessel.
Fig. 2 is a functional block diagram of a blood vessel-based identification apparatus according to an embodiment of the present invention. Based on the content of the foregoing embodiments, as shown in fig. 2, the blood vessel-based identification apparatus is a vein identification apparatus, and includes a descriptor obtaining module 201, a similarity obtaining module 202, and an identification module 203, where:
a descriptor obtaining module 201, configured to obtain a plurality of sub-blocks of the vein image to be identified according to a preset interval of the central pixel, and extract a descriptor of each sub-block;
the similarity obtaining module 202 is configured to obtain, according to the descriptor of each sub-block, a similarity between the vein image to be identified and each sample image in the vein image library;
the identity recognition module 203 is used for determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library;
wherein the descriptor is used for describing the characteristics of the sub-block.
Specifically, the descriptor obtaining module 201 may determine the position of the central pixel of each sub-block according to a preset interval of the central pixel, so as to divide the vein image to be identified into a plurality of portions, and obtain the sub-blocks according to each divided portion; and for each sub-block, extracting the features of the sub-block as a descriptor of the sub-block. And the characteristics of the sub-blocks are obtained according to the characteristics of each pixel contained in the sub-blocks. The characteristics of the pixels can be obtained according to the gray-scale values of the pixels.
And the similarity obtaining module 202 is used for obtaining the similarity between the vein image to be identified and any sample image based on the descriptor of each sub-block of the vein image to be identified.
The identity recognition module 203 may determine, based on the similarity, a sample image in the vein image library that matches the vein image to be recognized, and determine the identity information represented by the label of the sample image as the identity information corresponding to the vein image to be recognized, so as to correspond the vein image to be recognized to the identity information of the user, that is, determine the user corresponding to the vein image to be recognized.
The specific method and process for implementing the corresponding function of each module included in the blood vessel-based identity recognition device provided by the embodiment of the present invention are described in the above embodiments of the blood vessel-based identity recognition method, and are not described herein again.
The blood vessel-based identification device is used in the blood vessel-based identification method of the foregoing embodiments. Therefore, the description and definition in the blood vessel-based identification method in the foregoing embodiments can be used for understanding the execution modules in the embodiments of the present invention.
According to the embodiment of the invention, the similarity between the vein image to be recognized and the sample image is obtained according to the descriptor of each sub-block by obtaining the plurality of sub-blocks of the vein image to be recognized and extracting the descriptor of each sub-block, and the identity information corresponding to the vein image to be recognized is determined, so that the method and the device have better robustness, and the recognition efficiency and accuracy can be improved when non-rigid deformation exists.
Fig. 3 is a block diagram of an electronic device according to an embodiment of the present invention. Based on the content of the above embodiment, as shown in fig. 3, the electronic device may include: a processor (processor)301, a memory (memory)302, and a bus 303; wherein, the processor 301 and the memory 302 complete the communication with each other through the bus 303; the processor 301 is configured to invoke computer program instructions stored in the memory 302 and executable on the processor 301 to perform the methods provided by the various method embodiments described above, including, for example: acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of the preset central pixel, and extracting a descriptor of each sub-block; according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in the vein image library; determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library; wherein the descriptor is used for describing the characteristics of the sub-block.
Another embodiment of the present invention discloses a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the methods provided by the above-mentioned method embodiments, for example, including: acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of the preset central pixel, and extracting a descriptor of each sub-block; according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in the vein image library; determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library; wherein the descriptor is used for describing the characteristics of the sub-block.
Furthermore, the logic instructions in the memory 302 may be implemented in software functional units and stored in a computer readable storage medium when sold or used as a stand-alone product. Based on such understanding, the technical solutions of the embodiments of the present invention may be essentially implemented or make a contribution to the prior art, or may be implemented in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the methods described in the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
Another embodiment of the present invention provides a non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the methods provided by the above method embodiments, for example, including: acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of the preset central pixel, and extracting a descriptor of each sub-block; according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in the vein image library; determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library; wherein the descriptor is used for describing the characteristics of the sub-block.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. It is understood that the above-described technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method of the above-described embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (8)
1. A blood vessel-based identity recognition method is characterized by comprising the following steps:
acquiring a plurality of sub-blocks of the vein image to be identified according to the interval of a preset central pixel, and extracting a descriptor of each sub-block;
according to the descriptor of each sub-block, obtaining the similarity between the vein image to be identified and each sample image in a vein image library;
determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in a vein image library;
wherein the descriptor is used for describing the characteristics of the sub-block;
the specific steps of obtaining the similarity between the vein image to be identified and each sample image in the vein image library according to the descriptor of each sub-block comprise:
for each sample image in a vein image library, determining a region corresponding to each sub-block in the sample image;
obtaining a similar graph of each sub-block according to the descriptor of each sub-block and the descriptor of each sub-block contained in the region corresponding to the sub-block;
according to the similarity graph of each sub-block, obtaining the similarity between the vein image to be identified and the sample image;
the specific step of obtaining the similarity graph of each sub-block according to the descriptor of each sub-block and the descriptors of the sub-blocks included in the region corresponding to the sub-block includes:
for each sub-block, according to the descriptor of the sub-block and the descriptor of each sub-block contained in the region corresponding to the sub-block, obtaining the similarity between the sub-block and each sub-block contained in the region corresponding to the sub-block;
and respectively taking the similarity between the sub-block and each sub-block contained in the region corresponding to the sub-block as the value of the pixel corresponding to each sub-block contained in the region corresponding to the sub-block in the similarity map of the sub-block.
2. The blood vessel-based identification method according to claim 1, wherein the specific step of extracting the descriptor of each sub-block comprises:
for each sub-block, acquiring a feature vector of each pixel contained in the sub-block;
and normalizing and averaging the feature vectors of the pixels contained in the sub-block to obtain the descriptor of the sub-block.
3. The blood vessel-based identification method according to claim 2, wherein the step of obtaining the feature vector of each pixel included in the sub-block comprises:
and for each pixel contained in the sub-block, filtering the vein image to be identified through an eight-direction filter, and forming a feature vector of the pixel according to the response of the pixel in eight directions.
4. The blood vessel-based identity recognition method according to claim 1, wherein the specific step of obtaining the similarity between the vein image to be recognized and the sample image according to the similarity map of each sub-block comprises:
after the similarity graphs of each sub-block are gathered, pooling is carried out according to a preset window and a preset step length until the similarity between the vein image to be identified and the sample image is obtained.
5. The blood vessel-based identity recognition method according to any one of claims 1 to 4, wherein the specific step of obtaining the plurality of sub-blocks of the vein image to be recognized according to the preset interval of the central pixel comprises:
determining the position of the central pixel of each sub-block in the vein image to be identified according to the interval of the preset central pixel;
and taking the central pixel of each sub-block as a circle center, and taking the interval as a circular area determined by the diameter as the sub-block.
6. A blood vessel based identification device, comprising:
the descriptor acquisition module is used for acquiring a plurality of sub-blocks of the vein image to be identified according to the preset interval of the central pixel and extracting a descriptor of each sub-block;
the similarity obtaining module is used for obtaining the similarity between the vein image to be identified and each sample image in a vein image library according to the descriptor of each sub-block;
the identity recognition module is used for determining identity information corresponding to the vein image to be recognized according to the similarity between the vein image to be recognized and each sample image in the vein image library;
wherein the descriptor is used for describing the characteristics of the sub-block;
the similarity obtaining module is specifically configured to:
for each sample image in a vein image library, determining a region corresponding to each sub-block in the sample image;
obtaining a similar graph of each sub-block according to the descriptor of each sub-block and the descriptor of each sub-block contained in the region corresponding to the sub-block;
according to the similarity graph of each sub-block, obtaining the similarity between the vein image to be identified and the sample image;
the specific step of obtaining the similarity graph of each sub-block according to the descriptor of each sub-block and the descriptors of the sub-blocks included in the region corresponding to the sub-block includes:
for each sub-block, according to the descriptor of the sub-block and the descriptor of each sub-block contained in the region corresponding to the sub-block, obtaining the similarity between the sub-block and each sub-block contained in the region corresponding to the sub-block;
and respectively taking the similarity between the sub-block and each sub-block contained in the region corresponding to the sub-block as the value of the pixel corresponding to each sub-block contained in the region corresponding to the sub-block in the similarity map of the sub-block.
7. An electronic device, comprising:
at least one processor; and
at least one memory communicatively coupled to the processor, wherein:
the memory stores program instructions executable by the processor, the processor invoking the program instructions to perform the method of any of claims 1 to 5.
8. A non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526968.2A CN109815791B (en) | 2018-12-13 | 2018-12-13 | Blood vessel-based identity recognition method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811526968.2A CN109815791B (en) | 2018-12-13 | 2018-12-13 | Blood vessel-based identity recognition method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109815791A CN109815791A (en) | 2019-05-28 |
| CN109815791B true CN109815791B (en) | 2021-02-19 |
Family
ID=66601566
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811526968.2A Active CN109815791B (en) | 2018-12-13 | 2018-12-13 | Blood vessel-based identity recognition method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109815791B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110192843B (en) * | 2019-05-31 | 2022-04-15 | Oppo广东移动通信有限公司 | Information pushing method and related product |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074021A (en) * | 2011-01-07 | 2011-05-25 | 北京理工大学 | Covariance matching-based kernel tracking method |
| CN105760841A (en) * | 2016-02-22 | 2016-07-13 | 桂林航天工业学院 | Identify recognition method and identify recognition system |
| CN108509927A (en) * | 2018-04-09 | 2018-09-07 | 中国民航大学 | A kind of finger venous image recognition methods based on Local Symmetric graph structure |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102063727B (en) * | 2011-01-09 | 2012-10-03 | 北京理工大学 | Covariance matching-based active contour tracking method |
| US9390327B2 (en) * | 2013-09-16 | 2016-07-12 | Eyeverify, Llc | Feature extraction and matching for biometric authentication |
| CN108596126B (en) * | 2018-04-28 | 2021-09-14 | 中国民航大学 | Finger vein image identification method based on improved LGS weighted coding |
-
2018
- 2018-12-13 CN CN201811526968.2A patent/CN109815791B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102074021A (en) * | 2011-01-07 | 2011-05-25 | 北京理工大学 | Covariance matching-based kernel tracking method |
| CN105760841A (en) * | 2016-02-22 | 2016-07-13 | 桂林航天工业学院 | Identify recognition method and identify recognition system |
| CN108509927A (en) * | 2018-04-09 | 2018-09-07 | 中国民航大学 | A kind of finger venous image recognition methods based on Local Symmetric graph structure |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109815791A (en) | 2019-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106981077B (en) | Infrared image and visible light image registration method based on DCE and LSS | |
| CN110334762B (en) | Feature matching method based on quad tree combined with ORB and SIFT | |
| CN109919960B (en) | Image continuous edge detection method based on multi-scale Gabor filter | |
| US9092697B2 (en) | Image recognition system and method for identifying similarities in different images | |
| CN107577979B (en) | Method and device for quickly identifying DataMatrix type two-dimensional code and electronic equipment | |
| CN107545223B (en) | Image recognition method and electronic equipment | |
| WO2018176514A1 (en) | Fingerprint registration method and device | |
| WO2017070923A1 (en) | Human face recognition method and apparatus | |
| CN108269274A (en) | Method for registering images based on Fourier transformation and Hough transform | |
| CN111709426B (en) | Diatom recognition method based on contour and texture | |
| CN111260564A (en) | Image processing method and device and computer storage medium | |
| CN105139013A (en) | Object recognition method integrating shape features and interest points | |
| CN109815791B (en) | Blood vessel-based identity recognition method and device | |
| CN104268550A (en) | Feature extraction method and device | |
| CN106778844B (en) | Method and system for matching cracks in tunnel | |
| Niigaki et al. | Circular object detection based on separability and uniformity of feature distributions using Bhattacharyya coefficient | |
| CN111178398A (en) | Method, system, storage medium and device for detecting tampering of image information of identity card | |
| CN115690104A (en) | Wafer crack detection method and device and storage medium | |
| CN111967460B (en) | Text detection method and device, electronic equipment and computer storage medium | |
| CN110188601B (en) | Airport remote sensing image detection method based on learning | |
| CN111753723B (en) | Fingerprint identification method and device based on density calibration | |
| CN113705660A (en) | Target identification method and related equipment | |
| CN110287943B (en) | Image object recognition method and device, electronic equipment and storage medium | |
| CN109213515B (en) | Multi-platform lower buried point normalization method and device and electronic equipment | |
| CN112733670A (en) | Fingerprint feature extraction method and device, electronic equipment and storage medium |
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 |