Iris image quality evaluation method and system

Abstract

The invention discloses an iris image quality evaluation method and system, and the method comprises the following specific steps: 1, image preprocessing: improving the image quality, and highlighting the iris features; 2, feature region segmentation: detecting iris and pupil boundaries by adopting a Hough circle transformation or ellipse fitting technology, and detecting eyelid and eyelash boundaries by adopting Canny edge detection gradient intensity; 3, multi-dimensional quality evaluation: comprehensively quantifying the image quality through four indexes of available iris region calculation, contrast analysis, geometric characteristic evaluation and image quality evaluation; 4, adjustable weight scoring: generating a comprehensive quality score through a comprehensive quality scoring formula in combination with the adjustable weight of the user; and 5, interactive manual adjustment: the user can manually adjust the boundary of the iris and the pupil, and the evaluation result is updated in real time. The method is suitable for the fields of iris recognition, eye movement tracking, human-computer interaction, medical diagnosis and the like, and has remarkable practicability and flexibility.

Description

Technical Field

[0001] The present invention relates to the technical fields of computer vision, image processing, and biometric recognition, and particularly to an iris image quality assessment method and system thereof. Background Art

[0002] Iris recognition technology has been widely applied in fields such as identity authentication, security monitoring, unlocking of intelligent devices, and financial payment due to its high uniqueness, contactless nature, and security. As a biometric recognition technology, its core relies on the extraction and matching of iris textures, and the quality of iris images directly determines the success rate of feature extraction and the accuracy of recognition. However, in practical applications, due to factors such as acquisition devices, lighting conditions, user cooperation, and ocular interferences (such as eyelids, eyelashes, reflections, etc.), the quality of iris images often varies. Low-quality images (such as blurred, occluded, or insufficient contrast) can lead to feature extraction failures or misidentifications, thereby reducing the reliability of the system.

[0003] Existing iris image quality assessment methods mostly focus on single or a few indicators, such as image sharpness (based on edge sharpness), contrast, or the integrity of the iris region. Although these methods have certain effects in specific scenarios, they have the following limitations:

[0004] Single evaluation dimension: Only focusing on a few indicators, it cannot comprehensively reflect multiple key aspects of image quality;

[0005] Lack of robustness: In complex scenarios (such as strong light, weak light, or eyelid occlusion), traditional methods are prone to failure;

[0006] Lack of flexibility: Unable to adjust the evaluation criteria according to different application requirements, with poor adaptability;

[0007] Weak user interactivity: Lack of intuitive result display and manual intervention mechanism, making it difficult for users to optimize the evaluation results.

[0008] In view of the above problems, the present invention proposes a multi-dimensional, highly robust, and flexibly adjustable iris image quality assessment method and system, which overcomes the limitations of traditional methods through innovative algorithms and technical improvements, and improves the overall performance of the iris recognition system. Summary of the Invention

[0009] The present invention provides a multi-dimensional, highly robust, and flexibly adjustable iris image quality assessment method and system, which overcomes the limitations of traditional methods through innovative algorithms and technical improvements, and improves the overall performance of the iris recognition system to solve the problems raised in the above background art.

[0010] To achieve the above object, the present invention adopts the following technical solutions:

[0011] Iris image quality assessment method, including the following specific steps:

[0012] The first step, image preprocessing: improve the image quality and highlight the iris features. The specific steps are as follows:

[0013] S1, use adaptive histogram equalization to enhance local contrast and improve the problem of uneven illumination;

[0014] S2, apply 5x5 median filtering to remove salt-and-pepper noise;

[0015] S3, use a 5x5 Gaussian blur kernel function to smooth the image and reduce high-frequency noise interference;

[0016] S4, enhance the gray-scale difference between the iris and the surrounding area through linear transformation of the contrast stretching formula to improve feature visibility;

[0017] The second step, feature region segmentation: use Hough circle transform or ellipse fitting technology to detect the iris and pupil boundaries, and use Canny edge detection gradient intensity to detect the eyelid and eyelash boundaries;

[0018] The third step, multi-dimensional quality assessment: comprehensively quantify the image quality by calculating four indicators: available iris area, contrast analysis, geometric property evaluation, and image quality assessment;

[0019] Among them, the available iris area: calculate the proportion of the unobstructed iris area through the available iris area formula, reflecting the proportion of the effective feature area;

[0020] Weber contrast: evaluate the gray-scale difference between the iris-sclera and iris-pupil through the Weber contrast formula;

[0021] Pupil boundary roundness: calculate the pupil boundary roundness through the pupil boundary roundness formula;

[0022] Gray-scale utilization rate: calculate the gray-scale utilization rate through the gray-scale utilization rate entropy value formula;

[0023] The fourth step, adjustable weight scoring: combine the user-adjustable weights, generate a comprehensive quality score through the comprehensive quality scoring formula, and support the results to be exported in graph and Markdown formats for easy analysis and archiving;

[0024] The fifth step, interactive manual adjustment: the user can manually adjust the iris and pupil boundaries to update the evaluation results in real time.

[0025] As a further improvement of this technical solution: the Gaussian blur kernel function formula is:

[0026]

[0027] Among them, G(x, y): the value of the Gaussian kernel at position (x, y);

[0028] σ: the standard deviation of the Gaussian distribution, which controls the degree of blurring, σ = 1.0.

[0029] As a further improvement of this technical solution: the contrast stretching formula is:

[0030]

[0031] Among them, I(x, y): the gray value of the input grayscale image at pixel point (x, y);

[0032] I min ,I max : the minimum and maximum gray values of the input image;

[0033] I stretched (x, y): the stretched gray value, normalized to the range of 0 - 255.

[0034] As a further improvement of this technical solution: the Canny edge detection gradient intensity formula is:

[0035]

[0036] Among them, G x (x, y),G y (x, y): the gradients of the image in the horizontal and vertical directions, calculated by the Sobel operator;

[0037] G(x, y): the gradient intensity of pixel point (x, y), used for edge detection, and the threshold of G(x, y) is 100 - 200.

[0038] As a further improvement of this technical solution: the formula for the available iris area is:

[0039]

[0040] Among them, A occluded : the pixel area of the occluded area;

[0041] A iris : the total pixel area of the iris;

[0042] R usable : the percentage of the available iris area, which reflects the proportion of the effective features of the iris, and R usable The ideal threshold is set to ≥ 70%.

[0043] As a further improvement of this technical solution: the Weber contrast formula is:

[0044]

[0045] Among them, I iris , I pupil : The median gray value of the iris and pupil regions;

[0046] k: A constant (20 in the text), to avoid the denominator being zero and simulate human eye perception;

[0047] C Weber : Weber contrast, measuring the gray value difference between the iris and the pupil.

[0048] As a further improvement of this technical solution: The pupil boundary roundness formula:

[0049]

[0050] σ dist : The standard deviation of the distance from the edge points to the pupil center;

[0051] μ dist : The average distance from the edge points to the center;

[0052] C circularity : Roundness index, the closer the value is to 1, the closer the boundary is to an ideal circle.

[0053] As a further improvement of this technical solution: The gray - scale utilization entropy value formula is:

[0054]

[0055] p(i): The probability of the gray value i in the iris region;

[0056] H: Information entropy, measuring the richness of the gray - scale distribution of the iris texture.

[0057] As a further improvement of this technical solution: The comprehensive quality score formula is:

[0058]

[0059] Q i : The normalized score of the i - th quality index;

[0060] w i : The weight of the i - th index;

[0061] n: The total number of indexes;

[0062] S total : The comprehensive quality score.

[0063] The present invention also provides a system for an iris image quality assessment method, which includes an image preprocessing unit, a feature region segmentation unit, a multi-dimensional quality assessment unit, an adjustable weight scoring unit, and an interactive manual adjustment unit;

[0064] The image preprocessing unit is used to improve the image quality and highlight the iris features;

[0065] The feature region segmentation unit is used to detect the iris and pupil boundaries by using the Hough circle transform or ellipse fitting technology, and detect the eyelid and eyelash boundaries by using the Canny edge detection gradient intensity;

[0066] The multi-dimensional quality assessment unit is used to comprehensively quantify the image quality by calculating four indicators, namely the available iris region, contrast analysis, geometric property assessment, and image quality assessment, and fusing the four indicators;

[0067] The adjustable weight scoring unit is used to combine the user-adjustable weights, generate a comprehensive quality score through the comprehensive quality scoring formula, and support the export of the results in graph and Markdown formats for easy analysis and archiving;

[0068] The interactive manual adjustment unit is used for the user to manually adjust the iris and pupil boundaries and update the evaluation results in real time.

[0069] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0070] 1. Comprehensiveness: The system adopts multi-dimensional quality assessment indicators, including the available iris region, contrast analysis, geometric property assessment, and image quality assessment, which comprehensively cover the key factors of iris image quality. Compared with traditional methods, the present invention can evaluate iris images more systematically and completely, ensuring the comprehensiveness and reliability of the evaluation results.

[0071] 2. Accuracy: The system introduces an adaptive pupil detection verification mechanism and Weber contrast assessment technology, which greatly improves the accuracy of the evaluation results: Adaptive pupil detection verification mechanism: Utilize the low-gray biological characteristics of the pupil region to verify the detection results, ensure that the detected pupil boundary conforms to the actual situation, and avoid misjudgment. Weber contrast assessment: Based on the human eye perception characteristics, scientifically calculate the gray difference between the iris-pupil and iris-sclera, and provide a contrast assessment that is more in line with visual perception.

[0072] 3. Robustness: Through the boundary detection technology that integrates multiple algorithms, the system can adapt to iris images of different qualities and conditions; it supports multiple algorithms such as Hough circle transform, adaptive threshold segmentation, contour detection and ellipse fitting, and morphological operations, ensuring accurate segmentation of the iris and pupil in complex scenarios such as lighting changes and noise interference. Intelligent eyelid and eyelash detection technology: Through Canny edge detection and morphological operations, it effectively identifies and excludes the interference of eyelids and eyelashes, further improving the robustness of the evaluation.

[0073] 4. Flexibility: The system supports a quality scoring mechanism with adjustable weights. Users can adjust the weights of various indicators according to specific application scenarios (such as security authentication or medical diagnosis) to achieve personalized quality assessment criteria. This flexibility enables the system to meet diverse practical needs.

[0074] 5. Practicality: The system is designed with user-friendly interaction functions, enhancing practicality; Interactive manual adjustment mechanism: Through the graphical user interface (GUI), users can manually correct the boundaries of the iris and pupil, balancing the needs of automatic evaluation and manual intervention; Result visualization and export: It supports the export of evaluation results in Markdown format and provides visual display, facilitating users to analyze, archive, and generate reports.

[0075] 6. Innovation: The system introduces a number of innovative technologies, which are significantly different from existing methods: Entropy-based gray utilization rate evaluation: Quantify the richness of the gray distribution of iris texture through information entropy, providing a new dimension for image quality evaluation. Precise evaluation of pupil boundary roundness: Propose a scientific roundness quantification method through the ratio of the standard deviation to the mean of the distance from edge points to the center. Weber contrast evaluation: Apply the human eye perception model to iris image analysis, improving the scientific nature of the evaluation.

[0076] 7. Scalability: The modular design of the system allows for the future integration of more evaluation indicators or optimization of existing algorithms. For example, new image processing technologies or quality evaluation criteria can be added according to technological progress to adapt to changing application requirements.

[0077] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly and implement it in accordance with the content of the specification, the following takes the preferred embodiments of the present invention and combines with the accompanying drawings to describe in detail as follows. The specific implementation manners of the present invention are given in detail by the following embodiments and their accompanying drawings. Brief Description of the Drawings

[0078] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0079] Figure 1 Schematic diagram of the steps of the iris image quality assessment method proposed by the present invention. Specific implementation mode

[0080] The principles and features of the present invention will be described below with reference to the accompanying drawings. The examples given are only used to explain the present invention and are not intended to limit the scope of the present invention. In the following paragraphs, the present invention will be described more specifically by way of example with reference to the accompanying drawings. It should be noted that the accompanying drawings are all in a very simplified form and use non-precise scales, only for the purpose of conveniently and clearly assisting in explaining the embodiments of the present invention.

[0081] In the embodiments of the present invention, the iris image quality assessment method includes the following specific steps:

[0082] The first step is image preprocessing: improving the image quality and highlighting the iris features. The specific steps are as follows:

[0083] S1. Use adaptive histogram equalization to enhance the local contrast and improve the problem of uneven illumination;

[0084] S2. Apply 5x5 median filtering to remove salt-and-pepper noise;

[0085] S3. Use a 5x5 Gaussian blur kernel function to smooth the image and reduce high-frequency noise interference;

[0086] The formula for the Gaussian blur kernel function is:

[0087] where G(x, y): the value of the Gaussian kernel at position (x, y);

[0088] σ: the standard deviation of the Gaussian distribution, controlling the degree of blur, σ = 1.0;

[0089] S4. Enhance the gray-scale difference between the iris and the surrounding area through a linear transformation of the contrast stretching formula to improve the visibility of features;

[0090] The contrast stretching formula is:

[0091] where I(x, y): the gray-scale value of the input gray-scale image at pixel point (x, y);

[0092] I min ,I max : the minimum and maximum gray-scale values of the input image;

[0093] I stretched (x, y): the stretched gray-scale value, normalized to the range of 0-255;

[0094] Step 2, Feature Region Segmentation: Detect the iris and pupil boundaries using the Hough circle transform or elliptical fitting technique, and detect the eyelid and eyelash boundaries using the Canny edge detection gradient intensity;

[0095] The formula for the Canny edge detection gradient intensity is:

[0096] where G x (x,y), G y (x,y): The gradients of the image in the horizontal and vertical directions, calculated by the Sobel operator;

[0097] G(x,y): The gradient intensity of the pixel point (x,y), used for edge detection, and the threshold of G(x,y) is 100 - 200;

[0098] Step 3, Multi-dimensional Quality Assessment: Quantify the image quality comprehensively by calculating four indicators: available iris region, contrast analysis, geometric feature assessment, and image quality assessment;

[0099] Among them, available iris region: Calculate the proportion of the unoccluded iris area through the available iris region formula, reflecting the proportion of the effective feature region;

[0100] The formula for the available iris region is:

[0101] where A occluded : The pixel area of the occluded region (such as eyelids, eyelashes);

[0102] A iris : The total pixel area of the iris;

[0103] R usable : The available iris region percentage, reflecting the proportion of the effective features of the iris, R usable Set the ideal threshold ≥ 70%;

[0104] Weber contrast: Evaluate the gray-scale difference between the iris-sclera and iris-pupil through the Weber contrast formula;

[0105] The formula for the Weber contrast is:

[0106] where I iris , I pupil : The median gray scale of the iris and pupil regions;

[0107] k: A constant (20 in the text), to avoid the denominator being zero and simulate the human eye perception;

[0108] C Weber : The Weber contrast, measuring the gray-scale difference between the iris-pupil;

[0109] Pupil boundary roundness: Calculate the pupil boundary roundness through the pupil boundary roundness formula;

[0110] Pupil boundary roundness formula:

[0111] σ dist : Standard deviation of the distance from the edge point to the pupil center;

[0112] μ dist : Average distance from the edge point to the center;

[0113] C circularity : Roundness index, the closer the value is to 1, the closer the boundary is to an ideal circle;

[0114] Gray - scale utilization rate: Calculate the gray - scale utilization rate through the gray - scale utilization rate entropy formula;

[0115] The gray - scale utilization rate entropy formula is:

[0116] p(i): Probability of gray - scale value i in the iris region (normalized histogram);

[0117] H: Information entropy, measuring the richness of the gray - scale distribution of iris texture;

[0118] Fourth step, adjustable weight scoring: Combine the user - adjustable weights and generate a comprehensive quality score through the comprehensive quality scoring formula, support the results to be exported in graph and Markdown formats for easy analysis and archiving;

[0119] The comprehensive quality scoring formula is:

[0120] Q i : Normalized score of the i - th quality index (in the range of 0 - 1);

[0121] w i : Weight of the i - th index (user - adjustable, default 1.0);

[0122] n: Total number of indices (10 in the text);

[0123] S total : Comprehensive quality score (0 - 100);

[0124] Fifth step, interactive manual adjustment: The user can manually drag and adjust the iris and pupil boundaries through the GUI interface to update the evaluation results in real - time. Result output: Support export in Markdown format, for example:

[0125] markdown

[0126] # Iris Image Quality Assessment Report

[0127] - Available iris area: 85%

[0128] - Weber contrast: 92%

[0129] - Pupil boundary roundness: 0.95

[0130] - Gray scale utilization rate: 6.8

[0131] - Comprehensive score: 90 / 100.

[0132] As described above, it is only the preferred embodiment of the present invention, and it is not intended to limit the present invention in any form; any ordinary technician in the industry can smoothly implement the present invention as shown in the accompanying drawings of the specification and as described above; however, any minor changes, modifications and equivalent variations made by those skilled in the art within the scope of the technical solution of the present invention by using the technical content disclosed above are all equivalent embodiments of the present invention; at the same time, any equivalent changes, modifications and variations made to the above embodiments based on the essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. An iris image quality assessment method, characterized in that, It includes the following specific steps: The first step, image preprocessing: Improve the image quality and highlight the iris features. The specific steps are as follows: S1, Use adaptive histogram equalization to enhance the local contrast and improve the problem of uneven illumination; S2, Apply 5x5 median filtering to remove salt-and-pepper noise; S3, Use a 5x5 Gaussian blur kernel function to smooth the image and reduce the interference of high-frequency noise; S4, Enhance the gray-scale difference between the iris and the surrounding area through linear transformation of the contrast stretching formula to improve the visibility of features; The second step, feature region segmentation: Use the Hough circle transform or ellipse fitting technology to detect the iris and pupil boundaries, and use the Canny edge detection gradient intensity to detect the eyelid and eyelash boundaries; The third step, multi-dimensional quality assessment: Quantify the image quality comprehensively by calculating four indicators: available iris area, contrast analysis, geometric property assessment, and image quality assessment; Among them, the available iris area: Calculate the proportion of the unobstructed iris area through the available iris area formula, reflecting the proportion of the effective feature area; Weber contrast: Evaluate the gray-scale difference between the iris-sclera and iris-pupil through the Weber contrast formula; Pupil boundary roundness: Calculate the pupil boundary roundness through the pupil boundary roundness formula; Gray-scale utilization rate: Calculate the gray-scale utilization rate through the gray-scale utilization rate entropy value formula; The fourth step, adjustable weight scoring: Combine the user-adjustable weights, generate a comprehensive quality score through the comprehensive quality scoring formula, and support the results to be exported in graphic and Markdown formats for easy analysis and archiving; The fifth step, interactive manual adjustment: The user can manually adjust the iris and pupil boundaries and update the evaluation results in real time.

2. The iris image quality assessment method according to claim 1, wherein The formula for the Gaussian blur kernel function is: Among them, G(x, y): The value of the Gaussian kernel at the position (x, y); σ: The standard deviation of the Gaussian distribution, controlling the degree of blur, σ = 1.

0.

3. The iris image quality assessment method according to claim 1, wherein The contrast stretching formula is: Among them, I(x, y): The gray-scale value of the input gray-scale image at the pixel point (x, y); I min ,I max : The minimum and maximum gray values of the input image; I stretched (x, y): The gray value after stretching, normalized to the range of 0 - 255.

4. The iris image quality assessment method according to claim 1, characterized in that The Canny edge detection gradient intensity formula is as follows: Among them, G x (x,y), G y (x,y): The gradients of the image in the horizontal and vertical directions, calculated by the Sobel operator; G(x, y): The gradient intensity of the pixel point (x, y), used for edge detection, and the threshold of G(x, y) is 100 - 200.

5. The iris image quality evaluation method according to claim 1, wherein The formula for the available iris area is: Among them, A occluded : the pixel area of the occlusion region; A iris : total pixel area of iris; R usable : Percentage of available iris area, reflecting the proportion of effective iris features, R usable Set the ideal threshold ≥ 70%.

6. The iris image quality assessment method according to claim 1, characterized in that The Weber contrast formula is: where I iris , I pupil : the median gray level of the iris and pupil regions; k: A constant (20 in the text), to avoid the denominator being zero and simulate the human eye perception; C Weber : Weber contrast, which measures the gray-scale difference between the iris and the pupil.

7. The iris image quality assessment method according to claim 1, characterized in that, The pupil boundary roundness formula: σ dist : Standard deviation of the distance from the edge point to the pupil center; μ dist : The average distance from the edge point to the center; C circularity : Roundness index. The closer the value is to 1, the closer the boundary is to an ideal circle.

8. The iris image quality assessment method according to claim 1, characterized in that The entropy value formula for gray - scale utilization rate is as follows: p(i): The probability of the gray-scale value i in the iris area; H: Information entropy, measuring the richness of the gray-scale distribution of the iris texture.

9. The iris image quality assessment method according to claim 1, characterized in that The comprehensive quality scoring formula is: Q i : The normalized score of the i-th quality index; w i : The weight of the i-th indicator; n: The total number of indicators; S total : Comprehensive mass fraction.

10. The system for the iris image quality assessment method according to any one of claims 1-9, characterized in that, It includes an image preprocessing unit, a feature region segmentation unit, a multi-dimensional quality assessment unit, an adjustable weight scoring unit, and an interactive manual adjustment unit; The image preprocessing unit is used to improve the image quality and highlight the iris features; The feature region segmentation unit is used to detect the iris and pupil boundaries by using the Hough circle transform or ellipse fitting technology, and detect the eyelid and eyelash boundaries by using the Canny edge detection gradient intensity; The multi-dimensional quality assessment unit is used to calculate four indicators: available iris area, contrast analysis, geometric property assessment, and image quality assessment, and fuse the four indicators to comprehensively quantify the image quality; The adjustable weight scoring unit is used to combine user-adjustable weights and generate a comprehensive quality score through a comprehensive quality scoring formula, supporting the export of results in graphical and Markdown formats for easy analysis and archiving; The interactive manual adjustment unit is used for users to manually adjust the iris and pupil boundaries and update the evaluation results in real time.

Images