CN112381809A - Iris image definition evaluation method based on relative entropy - Google Patents

Iris image definition evaluation method based on relative entropy Download PDF

Info

Publication number
CN112381809A
CN112381809A CN202011299777.4A CN202011299777A CN112381809A CN 112381809 A CN112381809 A CN 112381809A CN 202011299777 A CN202011299777 A CN 202011299777A CN 112381809 A CN112381809 A CN 112381809A
Authority
CN
China
Prior art keywords
image
iris
vector
relative entropy
energy
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.)
Pending
Application number
CN202011299777.4A
Other languages
Chinese (zh)
Inventor
李海青
张新会
侯广琦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin Zhongke Hongxing Technology Co ltd
Original Assignee
Tianjin Zhongke Hongxing Technology Co ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tianjin Zhongke Hongxing Technology Co ltd filed Critical Tianjin Zhongke Hongxing Technology Co ltd
Priority to CN202011299777.4A priority Critical patent/CN112381809A/en
Publication of CN112381809A publication Critical patent/CN112381809A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/20—Special algorithmic details
    • G06T2207/20024—Filtering details
    • G06T2207/20032—Median filtering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30041—Eye; Retina; Ophthalmic
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Geometry (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention provides an image quality evaluation method and system, comprising the following steps: firstly, determining a candidate area image to be evaluated according to an original iris image, and obtaining a local maximum gradient energy image through local maximum gradient calculation so as to obtain an energy vector of the local maximum gradient energy image. And further performing low-pass filtering on the candidate area image to be evaluated to obtain a reference image, and performing statistics to obtain an energy vector of the reference image. And finally, quantifying the definition into an evaluation score to visually represent the quality of the original iris image to be evaluated. The method provided by the invention does not need to acquire a reference image in advance, does not need to repeatedly adjust complicated parameters, has strong operability, has an evaluation result similar to human eye perception, and has a very wide application prospect in an iris recognition system.

Description

Iris image definition evaluation method based on relative entropy
Technical Field
The embodiment of the invention relates to the technical field of image processing, in particular to an iris image definition evaluation method based on relative entropy.
Background
Iris recognition is an important means of biological feature recognition, and is gradually applied to scenes such as terrorism prevention, finance, social security and the like which need to accurately identify personal identity by means of uniqueness, high anti-counterfeiting performance and high stability. The quality evaluation is the first link of all biological characteristic identification processes, and the accurate evaluation of the characteristic quality is an important guarantee for improving the identification rate and reducing the recognition rate. The iris quality evaluation mainly comprises five items of content, namely definition, iris-sclera contrast, iris-pupil contrast, gray level utilization rate and effective iris area ratio. Besides the definition, other indexes have definite formulas for quantitative calculation.
The iris definition evaluation means widely adopted at present is ISO/IEC 29794-6: 2015 (E). The high-frequency component proportion in the iris image is obtained through statistics and quantification so as to represent the definition of the iris image, but the method is very sensitive to the image brightness and the individual iris difference in practical research, and the unclear image can not be screened out by well determining the threshold value in practical use. In the face of new technical challenges, such as remote iris recognition, advancing iris recognition and the like, a problem to be solved urgently is to quickly and reasonably evaluate the image quality of an iris image to be recognized and use the evaluation as a key judgment basis for judging whether subsequent recognition operation is carried out, so that the iris recognition result has higher accuracy, robustness, instantaneity and safety, and the image quality evaluation method has very important practical application value.
In view of the above, there is a need for a method for evaluating the sharpness of an iris image quickly and accurately, and the method can evaluate the consistency of different image brightness and different individual irises.
Disclosure of Invention
The invention provides an iris image definition evaluation method based on relative entropy. The method does not need to manually select statistical characteristics, can effectively deal with the distorted images of unknown distortion types, and obtains more accurate and effective image quality evaluation results.
In order to achieve the above object, the present application provides the following technical features:
a method for evaluating the sharpness of an iris image based on relative entropy comprises the following steps:
firstly, determining an image B in a candidate area to be evaluated according to an iris center point and an iris radius in an original iris image A;
secondly, performing local maximum gradient calculation on the image B to obtain a local maximum gradient energy image C;
thirdly, obtaining a vector P of energy distribution of the original iris image A by traversing and counting pixel value distribution in the image C;
fourthly, converting the image B by adopting a low-pass filtering method to obtain a reference image E;
fifthly, repeating the second step and the third step to obtain a vector Q of the energy distribution of the reference image E;
sixthly, calculating the relative entropy difference L of the vector P and the vector Q by adopting a cross entropy quantitative method;
wherein, calculating the difference index of the relative entropy difference L is as follows:
Figure BSA0000225193070000021
and converting the relative entropy difference L between the vector P and the vector Q to obtain a definition evaluation score S of the original iris image A.
Preferably, the iris center point determination method is coarse positioning by a discrete circular dynamic contour line method.
Preferably, the image sub-blocks selected by the iris region of the original iris image and the reference image are square.
Preferably, the image quality is considered to be too poor for an image in which the center of the iris cannot be located.
Further, after the iris center coordinate and the iris radius are determined, the candidate area coordinate range to be evaluated is as follows:
x∈(Px-∝r，Px+∝r)，y∈(Py-∝r，Py+∝r)
wherein the coordinate (P)x，Py) And taking r as the iris center, taking alpha as a multiple, and taking the value range of the multiple alpha as (1.2, 1.3) in order to ensure the integrity and the effectiveness of the iris image.
Further, an image B is obtained according to the coordinate range of the candidate area to be evaluated, and then gradient map calculation is carried out on the image B.
Preferably, a Sobel operator (Sobel operator) is used to perform convolution operation approximation on the image B to obtain gradient values of the image in the x and y directions.
Preferably, the Sobel operator used comprises two sets of 3X 3 vectors, which are convolved with the image to obtain the approximate gradients in the X and Y directions, respectively.
Further, a gradient map image C of the image B is calculated and obtained, as shown in formula 2:
Figure BSA0000225193070000023
wherein, lambda is the gradient fusion coefficient of the X direction and the Y direction, GxFor X-direction convolution templates, GyAnd the template is a convolution template in the Y direction, B is an image B in the candidate area to be evaluated, and C is a gradient map corresponding to the image B.
Further, a maximum gradient energy map image D of the image C is calculated and obtained, as shown in formula 3:
Figure BSA0000225193070000022
wherein said image C represents a gradient map of the image of the sharpness to be evaluated, RxThe image D is an image block with the length of the side being R and the x being the center, and represents a maximum gradient energy graph.
Further, histogram information of the maximum gradient energy map is calculated as an energy vector.
Preferably, the method of calculating the histogram information of the maximum gradient energy map as the energy vector is a traversal method.
Specifically, the number of times of occurrence of the gray value t in the image D is counted by traversing the image D, and histogram information of the image D is obtained as an energy vector p (t).
Further, a reference image E of the image B is constructed.
Preferably, the reference image E is obtained by low-pass filtering, including but not limited to mean filtering, gaussian filtering, median filtering, and the like.
Preferably, the filter kernel specifications of the mean filtering method include, but are not limited to [ 3T 3], [ 5T 5], and [ 7T 7 ].
According to the above steps, the energy vector q (t) of the reference image E can be obtained in the same manner, where t is 0, 1.
Further, the difference L between the two energy vectors can be obtained by quantitatively evaluating the difference in high frequency information between the reference image and the original image.
Specifically, the distance between the vector P and the vector Q is quantitatively calculated by using a cross entropy formula, and is recorded as L, as shown in formula 4:
Figure BSA0000225193070000031
wherein L is the distance between the energy vector P and the energy vector Q.
And finally, quantifying the definition into an evaluation score to visually represent the quality of the iris image to be evaluated.
Specifically, the energy vector distance L is normalized to obtain a sharpness evaluation S, and a calculation formula is shown in formula 5:
Figure BSA0000225193070000032
where L represents the distance between the energy vector P and the energy vector Q, and K represents the normalized center value and represents the normalization coefficient.
An embodiment of the present invention further provides an image quality evaluation device, including:
at least one central processor, at least one memory, a communication interface and a bus;
the central processing unit, the memory and the communication interface complete mutual communication through the bus; the communication interface is used for information transmission between the evaluation equipment and the communication equipment of the display; the memory stores program instructions executable by the central processing unit, and the central processing unit calls the program instructions to execute the image quality evaluation method.
Embodiments of the present invention also provide a non-transitory computer-readable storage medium storing computer instructions, which cause the computer to execute the image evaluation method.
Compared with the prior art, the technical scheme provided by the invention has the following main advantages that:
(1) the invention provides a no-reference iris definition evaluation algorithm, which does not need to acquire a reference image in advance, has strong operability and has very wide application prospect in an iris recognition system.
(2) According to the method, the energy difference between the original image and the constructed reference image is calculated through the cross entropy, the robustness of the algorithm to sensor change, environment change and individual change is improved, the evaluation result is close to human eye perception, and the method has good generalization.
(3) The quality evaluation algorithm has few parameters, does not need repeated and complicated parameter adjustment, quantizes the output iris definition to a numerical value between 0 and 100, and is convenient for users to understand and use.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a schematic flow chart of an iris image sharpness evaluation method based on relative entropy according to the present invention;
FIG. 2 is an exemplary diagram of an original iris image of an iris image sharpness evaluation method based on relative entropy according to the present invention;
FIG. 3 is an exemplary diagram of an image of a candidate region to be evaluated according to the iris image sharpness evaluation method based on relative entropy;
FIG. 4 is an exemplary diagram of a gradient image corresponding to a candidate region image to be evaluated in the iris image sharpness evaluation method based on relative entropy;
FIG. 5 is an exemplary diagram of a local maximum gradient image of an iris image sharpness evaluation method based on relative entropy according to the present invention;
fig. 6 is an exemplary evaluation result of the iris image sharpness evaluation method based on the relative entropy according to the present invention.
In the figure:
(a) original iris image evaluated as 18 according to the method of the present invention (b) original iris image evaluated as 51 according to the method of the present invention
(c) Original iris image evaluated as 73 according to the method of the invention (d) original iris image evaluated as 93 according to the method of the invention
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the following description will clearly describe the embodiments of the present invention in a complete description with reference to the accompanying drawings. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In the prior art, although a common image evaluation algorithm is relatively wide, for iris images, a relatively small target area and a relatively large individual iris difference are caused, so that in an actual iris image evaluation process, the iris image difference acquired among different iris sensors is large, the iris image difference acquired under different light source intensities of the same sensor is also large, an unclear image cannot be screened out well by determining a threshold value, and the iris image evaluation cannot be performed.
In view of the above factors, the present invention provides a method for evaluating sharpness of an iris image based on relative entropy, which specifically includes:
firstly, determining a candidate region; specifically, first, iris positioning is performed on an iris image, the center coordinate of a pupil is positioned, and the radius of the iris with the center of the pupil as the center of a circle and the outer edge of the iris as the circle is measured.
And intercepting a rectangular area with the pupil as the center and alpha times of the iris radius as the side length to intercept an image B with definition to be evaluated.
Secondly, calculating a local maximum gradient energy map; specifically, a gradient map C in the X direction and the Y direction is obtained by adopting a convolution approximation of a Sobel operator and an image B, and a local maximum gradient energy map D is generated by counting the maximum value in the image C in a blocking mode.
Thirdly, calculating an energy vector; the image D is traversed to obtain histogram information P as an energy vector.
Fourthly, constructing a reference image; after processing the image B with low-pass filtering, a reference image is obtained, and the low-pass filtering includes, but is not limited to, mean filtering, gaussian filtering, median filtering, and the like. And counting the energy vector Q according to the second, third and fourth steps.
Fifthly, measuring the energy vector difference; and (5) quantitatively calculating the distance between the vector P and the vector Q by using the cross entropy.
Sixthly, quantizing definition; and converting the energy vector difference to obtain a definition evaluation score.
Fig. 1 is a schematic flow chart of an iris image sharpness evaluation method based on relative entropy, as shown in fig. 1.
Firstly, a candidate area needs to be determined, iris rough positioning is carried out on an iris image A to be evaluated, and an iris center coordinate (P) is determinedx，Py) And an iris radius r.
The iris rough positioning algorithm includes, but is not limited to, a discrete circular dynamic contour line method, a gray threshold segmentation method, a circular Hough transform, a point Hough transform and the like.
Intercept with point (P)x，Py) And taking a rectangular area with the radius of alpha times of the iris as the side length as a center as a definition image B to be evaluated.
The coordinate range of the image B to be evaluated is as follows:
x∈(Px-αr，Px+αr)，y∈(Py-αr，Py+αr)。
in the ideal process of intercepting the iris image, the iris should be uniform and round, the intercepted image should be a square with the side length equal to the diameter of the iris, and the iris should be internally tangent to the square image. However, in actual operation, the above requirements cannot be completely met, and therefore, in order to ensure the integrity and the effectiveness of the iris image, the value range of the multiple α is preferably (1.2, 1.3).
After the candidate region is obtained, a local maximum gradient energy map is calculated.
Preferably, a Sobel operator (Sobel operator) is used to perform convolution operation approximation on the image B to obtain gradient values of the image in the x and y directions.
Specifically, the sobel operator is as follows:
Figure BSA0000225193070000061
the Sobel operator includes two groups of 3X 3 vectors, which are convolved with the image respectively to obtain the gradient approximate values in the X direction and the Y direction.
And calculating and obtaining a gradient map C of the image B according to the formula 2.
Figure BSA0000225193070000062
Wherein B denotes an image of the sharpness to be evaluated, GxFor X-direction convolution templates, GyIs convolution template in Y direction, lambda is gradient fusion coefficient between X direction and Y direction, and C is gradient map corresponding to B.
The maximum gradient energy map D of map C is calculated and obtained according to equation 3.
Figure BSA0000225193070000063
Where C denotes the gradient map of the image to be evaluated for sharpness, RxAn image block with a side length R with x as the center is shown, and D represents the maximum gradient energy map.
Through the process, the maximum gradient energy map of the image to be detected can be obtained, and the energy vector of the maximum gradient energy map D is further calculated.
Specifically, the number of times of occurrence of the gray-scale value t in the image D is counted by traversing the image D, so as to obtain histogram information p (t) of the image D, where t is 0, 1.
Further, a reference image E is further constructed.
Specifically, after processing the image B with low-pass filtering, a reference image E is obtained.
Preferably, the low-pass filtering includes, but is not limited to, mean filtering, gaussian filtering, median filtering, and the like.
Of these, it is worth noting that the size of the filter kernel is suggested to be 3 x 3 or 5 x 5.
It should be noted here that the clear iris image has abundant texture and more high-frequency components, the blurred iris image has unclear texture and less high-frequency components. The reference image is obtained by low-pass filtering, which loses the high-frequency information in the original image. And the sharper the iris image is, the more high frequency information is lost.
Furthermore, the definition of the iris image can be obtained by quantitatively evaluating the difference of high-frequency information between the reference image and the original image.
According to the first, second and third steps, the energy vector q (t) of the reference image E can be obtained in the same way, where t is 0, 1.
Further, a difference between the two energy vectors is measured.
Specifically, the distance between the vector P and the vector Q is quantitatively calculated by using a cross entropy formula, which is denoted as L, and the calculation formula is as follows:
Figure BSA0000225193070000064
wherein, L is the distance between the energy vector P and the energy vector Q, and experiments show that when n is 220, the measurement result can most reflect the iris definition.
And finally, quantifying the definition into an evaluation score to visually represent the quality of the iris image to be evaluated.
The energy vector distance L is normalized according to equation 5 to obtain a sharpness estimate S.
Figure BSA0000225193070000071
Where L represents the distance between the energy vector P and the energy vector Q, K represents the normalized center value, and m represents the normalization coefficient.
The invention relates to an iris image definition evaluation method based on relative entropy, which is described in detail by combining the technical scheme as follows:
fig. 2 is an exemplary diagram of an original iris image.
Further, an image in the candidate region to be evaluated is obtained by interception.
Fig. 3 is an exemplary diagram of an original candidate area image to be evaluated.
Further, local maximum gradient energy map is obtained by performing local maximum gradient calculation on the candidate region image to be evaluated.
Fig. 4 is an exemplary diagram of a local maximum gradient energy map corresponding to a candidate region image to be evaluated of an original iris image.
Further, by traversing the statistical local maximum gradient energy map,
fig. 5 is an exemplary diagram of a local maximum gradient image of an original iris image.
Fig. 6 is an exemplary evaluation result of the iris image sharpness evaluation method based on the relative entropy, where when K is 0.06, m is 2, and α is 1.2, the following sharpness evaluation scores are obtained:
FIG. (a) is an original iris image evaluated as 18 by the method of the present invention;
FIG. (b) shows an original iris image evaluated as 51 by the method of the present invention;
FIG. (c) shows an original iris image evaluated as 73 by the method of the present invention;
fig. (d) shows the original iris image evaluated as 93 by the method of the present invention.
The iris image difference collected between different iris sensors is large, and the iris image difference collected under different light source intensities of the same sensor is also large. The method can quickly and efficiently quantitatively evaluate the definition of the iris without being influenced by the model of the sensor and the intensity of the light source, and simultaneously can perform non-reference quantitative evaluation on the definition of the large iris database image acquired by various acquisition equipment, thereby enhancing the robustness and the practicability of the image evaluation method.
The embodiments described above are presented to enable a person having ordinary skill in the art to make and use the invention. It will be readily apparent to those skilled in the art that various modifications to these examples can be made, and the generic principles described herein can be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the above embodiments, and those skilled in the art should make improvements and modifications within the scope of the present invention based on the disclosure of the present invention.

Claims (10)

1. An iris image definition evaluation method based on relative entropy is characterized by comprising the following steps:
firstly, determining an image B in a candidate area to be evaluated according to an iris center point and an iris radius in an original iris image A;
secondly, calculating the local maximum gradient of the image B to obtain a local maximum gradient energy map C;
thirdly, obtaining a vector P of energy distribution of the original iris image A by traversing and counting pixel value distribution in the image C;
fourthly, converting the image B by adopting a low-pass filtering method to obtain a reference image E;
fifthly, repeating the step 2 and the step 3 to obtain a vector Q of the energy distribution of the reference image E; sixthly, calculating the relative entropy difference L of the vector P and the vector Q by adopting a cross entropy quantitative method;
wherein a difference indicator of the relative entropy difference L is calculated:
Figure FSA0000225193060000011
and converting the relative entropy difference L between the vector P and the vector Q to obtain a definition evaluation score S of the original iris image A.
2. The method for evaluating the sharpness of an iris image based on relative entropy of claim 1, wherein the iris center point determining method comprises: firstly, the original iris image A is roughly positioned, and the central coordinate (P) of the iris is determined by a discrete circular dynamic contour line methodx，Py) And iris radius r, in said iris center coordinate (P)x，Py) Taking the rectangular area with the radius alpha times of the iris r as the side length as the center, and taking the rectangular area as the image B in the candidate area to be evaluated, wherein the coordinate range of the image B in the candidate area to be evaluated is as follows:
x∈(Px-∝r，Px+∝r)，y∈(Py-∝r，Py+∝r)。
3. the method for evaluating sharpness of an iris image based on relative entropy of claim 1, wherein the gradient map calculation for the image B further comprises:
and performing convolution operation approximation on the image B by using a Sobel operator to obtain gradient values of the image in the x direction and the y direction. The Sobel operator includes two groups of 3X 3 vectors, which are convolved with the image respectively to obtain the gradient approximate values in the X direction and the Y direction.
Calculating and obtaining a gradient map image C of the image B according to the formula 2:
Figure FSA0000225193060000012
wherein image B represents the image to be evaluated for sharpness, GxFor X-direction convolution templates, GyThe image C is a gradient map corresponding to the image B.
Calculating and obtaining a maximum gradient energy map image D of the image C according to formula 3:
Figure FSA0000225193060000013
wherein said image C represents a gradient map of the image to be evaluated for sharpness, RxRepresenting an image block with a side length R with x as the center, and the image D representing a maximum gradient energy map.
4. A method for evaluating iris image clarity based on relative entropy as claimed in claim 1, wherein the evaluation candidate region range is the iris center coordinate (P)x，Py) The center is alpha times of the iris radius r, the iris radius r is a rectangular area with side length, and the value range of the multiple alpha is (1.2, 1.3).
5. The method for evaluating the sharpness of an iris image based on relative entropy of claim 1, wherein the method for calculating the energy vector P is to traverse the image D to count the number of times the gray value t appears in the image D, and obtain histogram information of the image D as the energy vector.
6. A method for evaluating the sharpness of an iris image based on relative entropy as claimed in claim 1, wherein the sub-blocks of the image selected in the iris region of the original iris image and the iris region of the reference image are square.
7. The method for evaluating the sharpness of an iris image based on relative entropy of claim 1, wherein the filtering kernel specification of the mean filtering method includes but is not limited to [3 a 3], [5 a 5], and [7 a 7 ].
8. Energy calculation for image B as claimed in claim 3, characterized in that the image quality is considered too poor for images in which the center of the iris cannot be located.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the steps of a relative entropy based iris image sharpness evaluation method according to any one of claims 1 to 8.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the steps of the image fusion quality evaluation method according to any one of claims 1 to 7.
CN202011299777.4A 2020-11-19 2020-11-19 Iris image definition evaluation method based on relative entropy Pending CN112381809A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011299777.4A CN112381809A (en) 2020-11-19 2020-11-19 Iris image definition evaluation method based on relative entropy

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011299777.4A CN112381809A (en) 2020-11-19 2020-11-19 Iris image definition evaluation method based on relative entropy

Publications (1)

Publication Number Publication Date
CN112381809A true CN112381809A (en) 2021-02-19

Family

ID=74585195

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011299777.4A Pending CN112381809A (en) 2020-11-19 2020-11-19 Iris image definition evaluation method based on relative entropy

Country Status (1)

Country Link
CN (1) CN112381809A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113160088A (en) * 2021-04-30 2021-07-23 河南大学 Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113160088A (en) * 2021-04-30 2021-07-23 河南大学 Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy
CN113160088B (en) * 2021-04-30 2022-08-12 河南大学 Speckle interference phase image filtering evaluation method based on Sobel operator and image entropy

Similar Documents

Publication Publication Date Title
CN116168026B (en) Water quality detection method and system based on computer vision
CN108363970B (en) Method and system for identifying fish species
EP1991948B1 (en) An iris recognition system having image quality metrics
Wilson et al. A new metric for grey-scale image comparison
CN103034838B (en) A kind of special vehicle instrument type identification based on characteristics of image and scaling method
CN115018844B (en) Plastic film quality evaluation method based on artificial intelligence
CN111507426B (en) Non-reference image quality grading evaluation method and device based on visual fusion characteristics
Garg et al. Survey on multi-focus image fusion algorithms
CN113837974B (en) NSST domain power equipment infrared image enhancement method based on improved BEEPS filtering algorithm
CN111160284A (en) Method, system, equipment and storage medium for evaluating quality of face photo
CN107273812B (en) A kind of living body iris method for anti-counterfeit for authentication
CN112001904A (en) Comprehensive evaluation module and evaluation method for remote sensing image quality definition
CN116993731B (en) Shield tunneling machine tool bit defect detection method based on image
CN115272303B (en) Textile fabric defect degree evaluation method, device and system based on Gaussian blur
van Zwanenberg et al. Edge detection techniques for quantifying spatial imaging system performance and image quality
Thamaraimanalan et al. Multi biometric authentication using SVM and ANN classifiers
CN112381809A (en) Iris image definition evaluation method based on relative entropy
CN109376782B (en) Support vector machine cataract classification method and device based on eye image features
CN111445435B (en) Multi-block wavelet transform-based reference-free image quality evaluation method
CN108416770B (en) Image quality evaluation method based on visual saliency
CN116129195A (en) Image quality evaluation device, image quality evaluation method, electronic device, and storage medium
CN111062954A (en) Infrared image segmentation method, device and equipment based on difference information statistics
Guan et al. A new metric for latent fingerprint image preprocessing
Kiruthiga et al. Smoothening of iris images and pupil segmentation using fractional derivative and wavelet transform
CN112529845A (en) Image quality value determination method, image quality value determination device, storage medium, and electronic device

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