CN111539258A - Iris image recognition method, iris image recognition apparatus, and storage medium - Google Patents
Iris image recognition method, iris image recognition apparatus, and storage medium Download PDFInfo
- Publication number
- CN111539258A CN111539258A CN202010244416.3A CN202010244416A CN111539258A CN 111539258 A CN111539258 A CN 111539258A CN 202010244416 A CN202010244416 A CN 202010244416A CN 111539258 A CN111539258 A CN 111539258A
- Authority
- CN
- China
- Prior art keywords
- iris image
- processed
- empirical mode
- mode decomposition
- reconstructed
- 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
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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/193—Preprocessing; Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- 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
- G06V40/18—Eye characteristics, e.g. of the iris
- G06V40/197—Matching; Classification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Ophthalmology & Optometry (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Collating Specific Patterns (AREA)
Abstract
The present disclosure relates to an iris image recognition method, an iris image recognition apparatus, and a storage medium, the iris image recognition method including: acquiring an iris image to be processed; decomposing the iris image to be processed based on improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components; selecting a preset number of two-dimensional empirical mode decomposition components from the plurality of two-dimensional empirical mode decomposition components to reconstruct to obtain a reconstructed iris image to be processed; extracting and fusing the distribution characteristics of the reconstructed iris image to be processed; and identifying the distribution characteristics based on the light gradient lifting frame to obtain an identification result. According to the embodiment of the disclosure, the iris image to be processed is decomposed through the improved BEMD, partial decomposition components are extracted to reconstruct the iris image, the characteristics of the reconstructed iris image are extracted and fused, and the identification is carried out by means of the lightgbm algorithm, so that the noise in the iris image and the interference of background factors are effectively eliminated, and the accuracy and the efficiency of iris identification are improved.
Description
Technical Field
The present disclosure relates to the field of image processing technologies, and in particular, to an iris image recognition method, an iris image recognition apparatus, and a storage medium.
Background
Biometric identification is a technology for verifying and identifying identity through inherent physiological characteristics or behavioral characteristics of a human body, wherein iris identification has the advantages of high reliability, good stability and the like, and meanwhile, the biometric identification has attracted extensive attention of academic circles, industrial circles, governments and military troops, and becomes the focus of attention in the field of biometric identification.
When iris recognition is utilized, the quality of an iris image is related to the speed and the precision of recognition, and in the current process of iris recognition, due to the influences of human and environmental factors, for example, the quality of the acquired iris image is poor, the difficulty of iris image recognition is increased, the recognition efficiency is influenced, the recognition precision is reduced, even the iris recognition is failed, and inconvenience is brought to the use of people.
Disclosure of Invention
To overcome the problems in the related art, the present disclosure provides an iris image recognition method, an iris image recognition apparatus, a human eye recognition method, a human eye recognition apparatus, and a storage medium.
According to an aspect of the embodiments of the present disclosure, there is provided an iris image recognition method including: acquiring an iris image to be processed; decomposing the iris image to be processed based on improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components; selecting a preset number of two-dimensional empirical mode decomposition components from the plurality of two-dimensional empirical mode decomposition components to reconstruct to obtain a reconstructed iris image to be processed; extracting and fusing the distribution characteristics of the reconstructed iris image to be processed; and identifying the distribution characteristics based on the light gradient lifting frame to obtain an identification result.
In one embodiment, the decomposition of the iris image to be processed based on the improved two-dimensional empirical mode decomposition comprises: in the improved two-dimensional empirical mode decomposition, the mean value of the envelope surface of the improved two-dimensional empirical mode decomposition is calculated by utilizing a piecewise cubic Hermite polynomial interpolation function.
In one embodiment, the decomposition of the iris image to be processed based on the improved two-dimensional empirical mode decomposition comprises: and carrying out boundary symmetric continuation on the iris image to be processed.
In one embodiment, the extracting and fusing the distribution characteristics of the reconstructed iris image to be processed includes: and extracting the distribution characteristics of the reconstructed iris image to be processed by using a logGabor odd function.
In one embodiment, identifying the distribution features based on a lightweight gradient lift framework comprises: determining the similarity of iris images to be processed by using the Hamming distance; and identifying the Hamming distance based on the light gradient lifting frame.
In one embodiment, the iris image recognition method further includes: preprocessing an iris image to be processed, wherein the preprocessing comprises the steps of; normalization processing is carried out on the iris image to be processed; and blocking the normalized iris image to be processed.
In one embodiment, the method for reconstructing the iris image to be processed comprises the following steps: and enhancing the reconstructed iris image to be processed based on self-adaptive histogram equalization.
According to an aspect of the embodiments of the present disclosure, there is provided an iris image recognition apparatus including: the acquisition module is used for acquiring an iris image to be processed; the decomposition module is used for decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components; the reconstruction module is used for selecting a preset number of two-dimensional empirical mode decomposition components from the plurality of two-dimensional empirical mode decomposition components to reconstruct to obtain a reconstructed iris image to be processed; the extraction module is used for extracting and fusing the distribution characteristics of the reconstructed iris image to be processed; and the identification module is used for identifying the distribution characteristics based on the light gradient lifting frame to obtain an identification result.
In one embodiment, when decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition, the decomposition module is configured to: in the improved two-dimensional empirical mode decomposition, the mean value of the envelope surface of the improved two-dimensional empirical mode decomposition is calculated by utilizing a piecewise cubic Hermite polynomial interpolation function.
In one embodiment, when decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition, the decomposition module is configured to: and carrying out boundary symmetric continuation on the iris image to be processed.
In an embodiment, when extracting and fusing the distribution features of the reconstructed iris image to be processed, the extraction module is configured to: and extracting the distribution characteristics of the reconstructed iris image to be processed by using a log Gabor odd function.
In one embodiment, when identifying the distribution features based on the lightweight gradient boost framework, the identification module is configured to: determining the similarity of iris images to be processed by using the Hamming distance; and identifying the Hamming distance based on the light gradient lifting frame.
In one embodiment, the iris image recognition apparatus further includes: the preprocessing module is used for preprocessing the iris image to be processed, and the preprocessing comprises the following steps: normalization processing is carried out on the iris image to be processed; and blocking the normalized iris image to be processed.
In an embodiment, when obtaining the reconstructed iris image to be processed, the reconstruction module is configured to: and enhancing the reconstructed iris image to be processed based on self-adaptive histogram equalization.
According to still another aspect of the embodiments of the present disclosure, there is provided an iris image recognition apparatus including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to: the iris image recognition method of any one of the preceding claims is performed.
According to still another aspect of embodiments of the present disclosure, there is provided a non-transitory computer-readable storage medium having instructions stored thereon, which, when executed by a processor of a mobile terminal, enable the mobile terminal to perform an iris image recognition method as described in one of the above.
The technical scheme provided by the embodiment of the disclosure can have the following beneficial effects: the iris image to be processed is decomposed based on the improved BEMD, partial decomposition components are extracted to reconstruct the iris image, the characteristics of the reconstructed iris image are extracted and fused, and the recognition is carried out by means of the lightgbm algorithm, so that the noise in the iris image and the interference of background factors are effectively eliminated, and the precision and the efficiency of the iris recognition are improved.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a flowchart illustrating an iris image recognition method according to an exemplary embodiment of the present disclosure.
Fig. 2 is a flowchart illustrating an iris image recognition method according to still another exemplary embodiment of the present disclosure.
Fig. 3 is a block diagram illustrating an iris image recognition apparatus according to an exemplary embodiment of the present disclosure.
Fig. 4 is a block diagram illustrating an iris image recognition apparatus according to an exemplary embodiment of the present disclosure.
FIG. 5 is a block diagram illustrating an apparatus in accordance with an example embodiment.
Detailed Description
Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the invention, as detailed in the appended claims.
Fig. 1 is a flowchart illustrating an iris image recognition method according to an exemplary embodiment of the present disclosure, and as shown in fig. 1, the iris image recognition method includes the following steps.
In step S101, an iris image to be processed is acquired.
The iris image to be processed can be acquired by image acquisition equipment, and can also be acquired by electronic equipment with shooting function such as a mobile phone, a PC, a notebook computer and the like. The number of iris images to be processed may be 1 or more.
In step S102, the iris image to be processed is decomposed based on the improved two-dimensional empirical mode decomposition, and a plurality of two-dimensional empirical mode decomposition components are obtained.
In the field of image processing, an image is used as a visual carrier, and is influenced by external or internal factors in the process of acquisition and transmission so as to generate noise pollution, and the noise reduces the quality of the image, so that important information in the image is lost, and great difficulty is brought to the understanding of the image.
In the process of shooting an iris image, due to the influence of human beings and the environment, the obtained image often has the situations of noise and uneven illumination, wherein the noise belongs to a high-frequency component, and the illumination background belongs to a low-frequency component. The method comprises the steps of decomposing a to-be-processed iris image by using an improved two-dimensional empirical mode decomposition (BEMD) to obtain a plurality of two-dimensional empirical mode decomposition components, separating the noise mode components and the illumination background mode components from the to-be-processed iris image, removing interference factors, and further improving the identification precision and efficiency of the to-be-processed iris image.
The iris image to be processed is partitioned, so that the local features of the iris image to be processed can be conveniently extracted, the local features of the iris image to be processed and the overall features of the iris image to be processed are fused, and the accuracy and the robustness of iris recognition are improved.
For example, I is divided into 4 rows and 8 columns, i.e. 32 sub-blocks, i.e. the iris image I to be processed is represented as:
decomposing the iris image I to be processed by using an improved BEMD algorithm to obtain an intrinsic mode function IMFi1, 2, n-1 and a residual component Rn. Wherein the first modal decomposition component IMF1Corresponding to the noise modal components.
In step S103, a preset number of two-dimensional empirical mode decomposition components are selected from the plurality of two-dimensional empirical mode decomposition components for reconstruction, so as to obtain a reconstructed iris image to be processed.
In order to influence interference factors such as noise, illumination background and the like in the shooting process of the iris image to be processed, the improved BEMD is adopted to decompose the iris image to be processed I to obtain an intrinsic mode function IMFi1, 2, n-1 and a residual component Rn. Taking n to 5 in consideration of final effect and calculation amount, namely, circulating for 4 times, extracting second, third and fourth corresponding decomposition components to reconstruct the iris image, namely
Filtering out noise and background lighting effects.
In step S104, the distribution characteristics of the reconstructed iris image to be processed are extracted and fused.
And extracting and fusing the step-by-step characteristics of the reconstructed iris image so as to more effectively classify and identify the image.
In step S105, the distribution features are identified based on the light gradient lifting frame, and an identification result is obtained.
The Light Gradient Boosting Machine (LightGBM) is a fast, distributed, high-performance Gradient Boosting frame based on a decision tree algorithm, is an improved version of the Gradient Boosting algorithm, and has the main idea that a weak classifier (decision tree) is used for iterative training to obtain an optimal model. The LightGBM model has the advantages of good training effect, difficulty in overfitting and the like, and is used for tasks such as feature selection, search classification and the like. The distribution characteristics are recognized based on lightgbm, and the recognition effect of the iris image to be processed is improved.
According to the iris image recognition method, the to-be-processed iris image is decomposed based on the improved BEMD, partial decomposition components are extracted to reconstruct the iris image, the characteristics of the reconstructed iris image are extracted and fused, and recognition is performed by means of the lightgbm algorithm, so that noise in the iris image and interference of background factors are effectively eliminated, and the accuracy and efficiency of iris recognition are improved.
In one embodiment, in the improved two-dimensional empirical mode decomposition, an envelope surface mean of the improved two-dimensional empirical mode decomposition is calculated using a piecewise cubic Hermite polynomial interpolation function.
The iris image to be processed is R1，1(x，y)＝I(x，y)，x∈[0，M]，y∈[0，N]And M and N are the number of rows and the number of columns of the iris image to be processed.
In the improved BEMD, a cubic spline interpolation method cannot well fit a curved surface, and a segmented cubic Hermite polynomial interpolation function is adopted to reduce the error of curved surface fitting.
1, Ri，j(x, y) I (x, y), finding Ri，j(x, y) all minimum and maximum points, in the modified BEMD, fitting maximum and minimum values respectively by using a piecewise cubic Hermite polynomial interpolation function, i.e. upper and lower envelopes Max composed of maximum and minimum valuesi，j(x, y) and Mini，j(x, y), calculating the envelope mean
Calculating iris image R to be processedi，j(x, y) and envelope mean Mi，jDifference R of (x, y)i，j+1(x，y)，Ri，j+1The calculation method of (x, y) is as follows:
Ri，j+1(x，y)＝Ri，j(x，y)-Mi，j(x，y)
standard deviation SDjIf it is less than the set value, if it is the standard deviation SDjWhen less than the set value, recording Ri，j(x, y) is IMFi(x，y)。
If standard deviation SDjNot less than the set value, taking the iris image to be processed as Ri，j+1(x, y), repeating the step of respectively fitting the maximum value and the minimum value by using the piecewise cubic Hermite polynomial interpolation function, and judging the importance degree of the influence of the characteristics on the final classification result by means of chi-square test until the following conditions are met:
denote the input signal as Ri+1，j(x，y)＝Ri，j(x，y)-IMFi(x, y), j ═ 1, and the above fitting of maxima and minima respectively with piecewise cubic hermitian polynomial interpolation function is repeated until the residual component R remainsn(x, y) is a monotonic function.
In one embodiment, the iris image to be processed is subjected to boundary symmetric continuation.
In the process of decomposing the iris image to be processed by the improved BEMD algorithm, a boundary effect exists, and the influence of the boundary effect is reduced by adopting a method of symmetric continuation of the boundary.
For example, a boundary symmetric extension method may be adopted to extend the upper and lower portions of the iris image to be processed by 16 pixels, and extend the left and right portions by 64 pixels. It can be understood that continuation in the up, down, and left-right directions of the iris image to be processed is symmetrical, further reducing the influence of the boundary effect on the iris image recognition.
In one embodiment, the distribution characteristics of the reconstructed iris image to be processed are extracted by using a log Gabor odd function.
The log Gabor function can be divided into even symmetry and odd symmetry, and compared with the log Gabor function, the logGabor odd function has higher identification precision and reduced calculation amount.
The log Gabor function can be expressed as:
wherein f is0Is the center frequency, theta0Is the angular direction of the filter, σfFor determining radial bandwidth, bandwidth equation Bf＝2(2/ln2)1/2|ln(σf/f0)|，σθFor determining directional bandwidth, bandwidth equation Bθ＝2σθ(2ln2)1/2。
For example, it may be a log Gabor odd function Im (G) in 4 directions and 4 sizesij) And wherein i, j is 1, 2, 3 and 4, and the distribution characteristics of the iris image to be processed are respectively extracted.
For two iris images I to be processed1、I2Each iris image to be processed forms 4 × 4 feature templates using 4 directions, 4 sizes of log Gabor odd functions and phase encoding.
In one embodiment, the similarity of the iris images to be processed is determined by using the Hamming distance, and the Hamming distance is identified based on a light gradient lifting framework.
For two iris images I to be processed1、I2Identification is carried out to determine an iris image I to be processed1、I2Whether the iris images belong to the same human eye or not can utilize 4 directions, 4 sizes of log Gabor odd functions and phase encoding to enable each iris image to be processed to form 4 × 4 characteristic templates.
Calculating two iris images I to be processed under the same function1、I2I.e. the two iris images I to be processed1、I2Overall distance H ofw，Hw＝[Hw1，Hw2，...，Hw16]。
For two iris images I to be processed1、I2The corresponding sub-blocks are respectively acted on I by using log Gabor odd functions of 4 directions and 4 sizes and a phase coding method1，I2Each image forms 4 × 4 × 4 × 8 characteristic submodels, and the corresponding characteristic submodels of the two images under the action of the same function are calculatedHamming distance H between platess，Hs＝[Hs1，Hs2，…，Hs512]。
Fusing two iris images I to be processed1、I2Overall hamming distance HwAnd two iris images I to be processed1、I2Local hamming distance H between corresponding sub-blockssI.e. the metric vector is used to measure the distance H between iris images to be processed ═ Hw，Hs]. The whole hamming distance HwAnd local hamming distance HsThe vector is aggregated into a one-dimensional vector, so that the calculation is simple and convenient, and the identification speed is improved.
And identifying the distribution characteristics based on the LightGBM to obtain an identification result.
Training data by means of a lightgbm classification algorithm and reducing dimensions by combining a univariate feature selection method, and comprehensively selecting the most important 100 feature metrics, namely Hf＝[Hf1，Hf2，...，Hf100]And the accuracy can be effectively improved during identification.
Training data by means of a lightgbm classification algorithm and combining a univariate feature selection method to reduce dimensions, and determining 100 most important features.
Iris image I to be processed1、I2And (5) carrying out recognition to further obtain a recognition result. I.e. the above-mentioned vector HfThe prediction result is y, considering the variable x. When two iris images I to be processed1、I2In the case of an image belonging to the same human eye, y is 1. When two iris images I to be processed1、I2And if the images do not belong to the same human eye, y is 0.
Fig. 2 is a flowchart illustrating an iris image recognition method according to still another exemplary embodiment of the present disclosure, and as shown in fig. 2, the iris image recognition method includes the following steps.
In step S201, an iris image to be processed is acquired.
In step S202, normalization processing is performed on the iris image to be processed, and the iris image to be processed is divided into blocks.
The iris image to be processed is processed with normalization, wherein the normalization refers to the processing and transformation of a series of standards on the iris image to be processed, so that the iris image to be processed is transformed into a fixed standard form.
The normalization processing can be to normalize the circular to-be-processed iris image into a rectangular image, so that not only can unnecessary pixels except the iris be removed, but also the size of the to-be-processed iris image can be compressed, and the subsequent feature extraction and identification operation can be facilitated. And the iris image to be processed after normalization processing is a normalized image of the iris image to be processed. The iris image to be processed is normalized, and the normalized iris image to be processed is obtained, so that the iris recognition precision can be improved, and the recognition result is more accurate.
The normalized iris image to be processed is partitioned, so that the local features of the iris image to be processed can be conveniently extracted, the local features of the iris image to be processed and the overall features of the iris image to be processed are fused, and the accuracy and the robustness of iris recognition are improved.
In step S203, the iris image to be processed is decomposed based on the improved two-dimensional empirical mode decomposition, and a plurality of two-dimensional empirical mode decomposition components are obtained.
In step S204, a preset number of two-dimensional empirical mode decomposition components are selected from the plurality of two-dimensional empirical mode decomposition components to be reconstructed, so as to obtain a reconstructed iris image to be processed.
In step S205, the distribution characteristics of the reconstructed iris image to be processed are extracted and fused.
In step S206, the distribution features are identified based on the light gradient lifting frame, and an identification result is obtained.
In an embodiment, the reconstructed iris image to be processed is enhanced based on adaptive histogram equalization.
And enhancing the contrast of the iris image to be processed by using self-adaptive histogram equalization, and enhancing the reconstructed iris image to be processed to obtain an iris image template to be processed.
Fig. 3 is a block diagram illustrating an iris image recognition apparatus according to an exemplary embodiment of the present disclosure, and as shown in fig. 3, an iris image recognition apparatus 300 includes: an acquisition module 310, a decomposition module 320, a reconstruction module 330, an extraction module 340, and an identification module 350.
An obtaining module 310 is configured to obtain an iris image to be processed.
The decomposition module 320 is configured to decompose the iris image to be processed based on the improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components.
The reconstructing module 330 is configured to select a preset number of two-dimensional empirical mode decomposition components from the multiple two-dimensional empirical mode decomposition components to perform reconstruction, so as to obtain a reconstructed iris image to be processed.
And the extraction module 340 is configured to extract and fuse the distribution characteristics of the reconstructed iris image to be processed.
And the identification module 350 is configured to identify the distribution characteristics based on the light gradient lifting frame to obtain an identification result.
In one embodiment, when decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition, the decomposition module 320 is configured to: in the improved two-dimensional empirical mode decomposition, the mean value of the envelope surface of the improved two-dimensional empirical mode decomposition is calculated by utilizing a piecewise cubic Hermite polynomial interpolation function.
In one embodiment, when decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition, the decomposition module 320 is configured to: and carrying out boundary symmetric continuation on the iris image to be processed.
In an embodiment, when the distribution features of the reconstructed iris image to be processed are extracted and fused, the extraction module 340 is configured to: and extracting the distribution characteristics of the reconstructed iris image to be processed by using a log Gabor odd function.
In one embodiment, when identifying the distribution features based on the lightweight gradient boost framework, the identification module 350 is configured to: determining the similarity of iris images to be processed by using the Hamming distance; and identifying the Hamming distance based on the light gradient lifting frame.
Fig. 4 is a block diagram illustrating an iris image recognition apparatus according to an exemplary embodiment of the present disclosure, and as shown in fig. 4, the iris image recognition apparatus 300 further includes: a pre-processing module 360.
The preprocessing module 360 is used for preprocessing the iris image to be processed, and the preprocessing includes: normalization processing is carried out on the iris image to be processed; and blocking the normalized iris image to be processed.
In an embodiment, when obtaining the reconstructed to-be-processed iris image, the reconstruction module 330 is configured to: and enhancing the reconstructed iris image to be processed based on self-adaptive histogram equalization.
With regard to the apparatus in the above-described embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated here.
Fig. 5 is an illustration of an electronic device 30, in accordance with an example embodiment. The electronic device 30 includes a memory 310, a processor 320, and an Input/Output (I/O) interface 330. The memory 310 is used for storing instructions. And a processor 320 for calling the instructions stored in the memory 310 to execute the iris image recognition method according to the embodiment of the present invention. The processor 320 is connected to the memory 310 and the I/O interface 330, respectively, for example, via a bus system and/or other connection mechanism (not shown). The memory 310 may be used to store programs and data including a program of the iris image recognition method according to an embodiment of the present invention, and the processor 320 executes various functional applications and data processing of the electronic device 30 by executing the program stored in the memory 310.
In an embodiment of the present invention, the processor 320 may be implemented in at least one hardware form of a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Array (PLA), and the processor 320 may be one or a combination of a Central Processing Unit (CPU) or other Processing units with data Processing capability and/or instruction execution capability.
In the embodiment of the present invention, the I/O interface 330 may be used to receive input instructions (e.g., numeric or character information, and generate key signal inputs related to user settings and function control of the electronic device 30, etc.), and may also output various information (e.g., images or sounds, etc.) to the outside. The I/O interface 330 may comprise one or more of a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a mouse, a joystick, a trackball, a microphone, a speaker, a touch panel, and the like.
In some embodiments, the invention provides a computer-readable storage medium having stored thereon computer-executable instructions that, when executed by a processor, perform any of the methods described above.
Although operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in serial order, or that all illustrated operations be performed, to achieve desirable results. In certain environments, multitasking and parallel processing may be advantageous.
The methods and apparatus of the present invention can be accomplished with standard programming techniques with rule based logic or other logic to accomplish the various method steps. It should also be noted that the words "means" and "module," as used herein and in the claims, is intended to encompass implementations using one or more lines of software code, and/or hardware implementations, and/or equipment for receiving inputs.
Any of the steps, operations, or procedures described herein may be performed or implemented using one or more hardware or software modules, alone or in combination with other devices. In one embodiment, the software modules are implemented using a computer program product comprising a computer readable medium containing computer program code, which is executable by a computer processor for performing any or all of the described steps, operations, or procedures.
The foregoing description of the implementation of the invention has been presented for purposes of illustration and description. It is not intended to be exhaustive or to limit the invention to the precise form disclosed, and modifications and variations are possible in light of the above teachings or may be acquired from practice of the invention. The embodiments were chosen and described in order to explain the principles of the invention and its practical application to enable one skilled in the art to utilize the invention in various embodiments and with various modifications as are suited to the particular use contemplated.
Claims (16)
1. An iris image recognition method, comprising:
acquiring an iris image to be processed;
decomposing the iris image to be processed based on improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components;
selecting a preset number of two-dimensional empirical mode decomposition components from the plurality of two-dimensional empirical mode decomposition components to reconstruct to obtain the reconstructed iris image to be processed;
extracting and fusing the reconstructed distribution characteristics of the iris image to be processed;
and identifying the distribution characteristics based on the light gradient lifting frame to obtain an identification result.
2. The iris image recognition method as claimed in claim 1, wherein decomposing the iris image to be processed based on an improved two-dimensional empirical mode decomposition comprises:
in the improved two-dimensional empirical mode decomposition, an envelope surface mean value of the improved two-dimensional empirical mode decomposition is calculated by utilizing a piecewise cubic Hermite polynomial interpolation function.
3. An iris image recognition method according to claim 1 or 2, characterized in that decomposing the iris image to be processed based on improved two-dimensional empirical mode decomposition comprises:
and carrying out boundary symmetric continuation on the iris image to be processed.
4. The iris image recognition method of claim 3, wherein the extracting and fusing the distribution characteristics of the reconstructed iris image to be processed comprises:
and extracting the distribution characteristics of the reconstructed iris image to be processed by using a log Gabor odd function.
5. The iris image recognition method of claim 3, wherein the recognition of the distribution features based on a lightweight gradient boosting framework comprises:
determining the similarity of the iris images to be processed by utilizing the Hamming distance;
and identifying the hamming distance based on a lightweight gradient lifting framework.
6. An iris image recognition method as claimed in claim 4 or 5, wherein the method further comprises:
preprocessing the iris image to be processed, wherein the preprocessing comprises the following steps;
normalizing the iris image to be processed; and
and partitioning the normalized iris image to be processed.
7. The iris image recognition method of claim 6, wherein obtaining the reconstructed iris image to be processed comprises:
and enhancing the reconstructed iris image to be processed based on self-adaptive histogram equalization.
8. An iris image recognition apparatus, comprising:
the acquisition module is used for acquiring an iris image to be processed;
the decomposition module is used for decomposing the iris image to be processed based on improved two-dimensional empirical mode decomposition to obtain a plurality of two-dimensional empirical mode decomposition components;
the reconstruction module is used for selecting a preset number of the two-dimensional empirical mode decomposition components from the plurality of two-dimensional empirical mode decomposition components to reconstruct to obtain the reconstructed iris image to be processed;
the extraction module is used for extracting and fusing the reconstructed distribution characteristics of the iris image to be processed;
and the identification module is used for identifying the distribution characteristics based on the light gradient lifting frame to obtain an identification result.
9. An iris image recognition device as claimed in claim 8, wherein when decomposing the iris image to be processed based on the improved two-dimensional empirical mode decomposition, the decomposition module is configured to:
in the improved two-dimensional empirical mode decomposition, an envelope surface mean value of the improved two-dimensional empirical mode decomposition is calculated by utilizing a piecewise cubic Hermite polynomial interpolation function.
10. An iris image recognition device as claimed in claim 8 or 9, wherein when the to-be-processed iris image is decomposed based on the improved two-dimensional empirical mode decomposition, the decomposition module is configured to:
and carrying out boundary symmetric continuation on the iris image to be processed.
11. The iris image recognition device of claim 10, wherein when the distribution features of the reconstructed iris image to be processed are extracted and fused, the extraction module is configured to:
and extracting the distribution characteristics of the reconstructed iris image to be processed by using a log Gabor odd function.
12. The iris image recognition device as claimed in claim 10, wherein when the distribution feature is recognized based on a lightweight gradient boost framework, the recognition module is configured to:
determining the similarity of the iris images to be processed by utilizing the Hamming distance;
and identifying the hamming distance based on a lightweight gradient lifting framework.
13. An iris image recognition apparatus as claimed in claim 11 or 12, further comprising:
the preprocessing module is used for preprocessing the iris image to be processed, and the preprocessing comprises the following steps:
normalizing the iris image to be processed; and
and partitioning the normalized iris image to be processed.
14. An iris image recognition device of claim 13, wherein when the reconstructed iris image to be processed is obtained, the reconstruction module is configured to:
and enhancing the reconstructed iris image to be processed based on self-adaptive histogram equalization.
15. An iris image recognition apparatus comprising:
a processor;
a memory for storing processor-executable instructions;
wherein the processor is configured to: performing the iris image recognition method of any one of claims 1 to 7.
16. A non-transitory computer-readable storage medium having instructions therein, which when executed by a processor of a mobile terminal, enable the mobile terminal to perform the iris image recognition method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010244416.3A CN111539258B (en) | 2020-03-31 | 2020-03-31 | Iris image recognition method, iris image recognition device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010244416.3A CN111539258B (en) | 2020-03-31 | 2020-03-31 | Iris image recognition method, iris image recognition device, and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111539258A true CN111539258A (en) | 2020-08-14 |
| CN111539258B CN111539258B (en) | 2023-05-12 |
Family
ID=71973307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010244416.3A Active CN111539258B (en) | 2020-03-31 | 2020-03-31 | Iris image recognition method, iris image recognition device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111539258B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015119520A1 (en) * | 2014-02-05 | 2015-08-13 | Walus Michał | The method and decision system of personal characteristics acquisition especially in biometrical authorisation systems |
| CN107578028A (en) * | 2017-09-20 | 2018-01-12 | 广东工业大学 | A kind of face identification method, device, equipment and computer-readable recording medium |
| US20190087726A1 (en) * | 2017-08-30 | 2019-03-21 | The Board Of Regents Of The University Of Texas System | Hypercomplex deep learning methods, architectures, and apparatus for multimodal small, medium, and large-scale data representation, analysis, and applications |
| CN109902780A (en) * | 2019-02-14 | 2019-06-18 | 广州番禺职业技术学院 | Testimony of a witness unification verification terminal and system and method based on multi-modal recognition of face |
-
2020
- 2020-03-31 CN CN202010244416.3A patent/CN111539258B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015119520A1 (en) * | 2014-02-05 | 2015-08-13 | Walus Michał | The method and decision system of personal characteristics acquisition especially in biometrical authorisation systems |
| US20190087726A1 (en) * | 2017-08-30 | 2019-03-21 | The Board Of Regents Of The University Of Texas System | Hypercomplex deep learning methods, architectures, and apparatus for multimodal small, medium, and large-scale data representation, analysis, and applications |
| CN107578028A (en) * | 2017-09-20 | 2018-01-12 | 广东工业大学 | A kind of face identification method, device, equipment and computer-readable recording medium |
| CN109902780A (en) * | 2019-02-14 | 2019-06-18 | 广州番禺职业技术学院 | Testimony of a witness unification verification terminal and system and method based on multi-modal recognition of face |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111539258B (en) | 2023-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Chen et al. | Median filtering forensics based on convolutional neural networks | |
| CN107665364B (en) | Neural network method and apparatus | |
| CN109190531B (en) | Feature extraction and matching and template update for biometric authentication | |
| Qi et al. | Robust feature learning by stacked autoencoder with maximum correntropy criterion | |
| Doegar et al. | Cnn based image forgery detection using pre-trained alexnet model | |
| CN111079639A (en) | Method, device and equipment for constructing garbage image classification model and storage medium | |
| WO2022042123A1 (en) | Image recognition model generation method and apparatus, computer device and storage medium | |
| US20220036167A1 (en) | Sorting method, operation method and operation apparatus for convolutional neural network | |
| CN106650568B (en) | Face recognition method and device | |
| CN110032704B (en) | Data processing method, device, terminal and storage medium | |
| CN111553215A (en) | Personnel association method and device, and graph convolution network training method and device | |
| KR20170109304A (en) | Method for parallel learning of cascade classifier by object recognition | |
| Sha et al. | Graph Laplacian regularization with sparse coding for image restoration and representation | |
| Gan et al. | A PCA-based convolutional network | |
| CN111209940A (en) | Image duplicate removal method and device based on feature point matching | |
| Xu et al. | An efficient and lightweight method for human ear recognition based on MobileNet | |
| JP7121132B2 (en) | Image processing method, apparatus and electronic equipment | |
| CN110659631A (en) | License plate recognition method and terminal equipment | |
| Krupiński et al. | Binarization of degraded document images with generalized Gaussian distribution | |
| CN111539258B (en) | Iris image recognition method, iris image recognition device, and storage medium | |
| CN111488811A (en) | Face recognition method and device, terminal equipment and computer readable medium | |
| Ren et al. | Face and facial expressions recognition and analysis | |
| TWI775038B (en) | Method and device for recognizing character and storage medium | |
| CN113159211B (en) | Method, computing device and computer storage medium for similar image retrieval | |
| Mangayarkarasi et al. | Contour detection based ear recognition for biometric applications |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100081 room 701, floor 7, Fuhai international port, Haidian District, Beijing Applicant after: Beijing wanlihong Technology Co.,Ltd. Address before: 100081 1504, floor 15, Fuhai international port, Daliushu Road, Haidian District, Beijing Applicant before: BEIJING SUPERRED TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |