Method and device for collecting iris template based on simulation environment

Abstract

The invention relates to a method and device for collecting an iris template based on a simulation environment. The method comprises the steps of generating a guide point in a display interface, and guiding eyeballs of a user to follow a target for tracking; visible light emitted by a visible spectrum light source irradiates eyeballs of a user, so that pupils of the user are contracted to a set diameter range, and meanwhile, near-infrared light emitted by a near-infrared light source illuminates an iris area; collecting iris feature information and eyeball related feature parameters of the user; adjusting the brightness of the visible light and the wavelength of the near-infrared light, and repeating the steps; recording eye movement features and iris response features of the user under different illumination stimulation conditions; and constructing a corresponding user exclusive iris feature library based on the iris feature information, the iris response features, the eyeball features and the eye movement features. By implementing the scheme, multiple real use environments are effectively simulated in the acquisition stage, and the diversity and representativeness of the iris template are improved; and a dynamic texture image is supplemented, and the multi-dimensional information fusion capability of the iris feature library is improved.

Description

Technical Field

[0001] The present application relates to the field of iris recognition, and in particular to a method and device for collecting iris templates based on a simulated environment. Background Art

[0002] As an important biometric identification method, iris recognition technology boasts high accuracy, robustness, and resistance to forgery. The core of iris recognition involves two major technical processes: iris image acquisition (template generation) and iris image comparison. The quality of iris acquisition directly impacts subsequent recognition results and is a key factor influencing the performance of the entire recognition system.

[0003] Existing iris image acquisition primarily utilizes telescopic iris collectors or other terminal devices, which generate a single iris template during the acquisition phase. Taking telescopic iris collectors as an example, these devices use structural components to block external interfering light, incorporate infrared lighting to enhance iris imaging clarity, and utilize a white light source to constrict the pupil to a fixed range, thereby obtaining a clear image of the iris texture. Furthermore, to enhance the user experience, some devices incorporate mirrored guidance mechanisms, allowing users to look directly into the mirror during acquisition, facilitating the device's acquisition of a centrally viewed image.

[0004] Although this type of acquisition method has certain advantages in terms of structured shading and localized illumination, its iris template has a single style and only reflects the iris image characteristics of the user in a specific acquisition environment. However, in actual iris recognition applications, the user's environment may vary, such as low light indoors, strong light outdoors, or the presence of interfering light sources (such as infrared remote control light). These external environmental factors will significantly affect parameters such as iris image brightness, contrast, clarity, and pupil size, resulting in significant differences in the style of the acquired image and the comparison image, seriously affecting the matching effect and stability of the recognition algorithm, resulting in a decrease in the recognition success rate or an increase in the error rate. The existing method of generating iris templates based on a single style and a single gaze angle is no longer able to meet the high-precision recognition requirements in diverse usage environments.

[0005] In summary, if we can simulate various environmental lighting conditions and changes in gaze angles during the iris collection stage, collect iris image templates of multiple styles and angles, and build a user-specific iris feature library based on this, it will significantly improve the recognition system's adaptability and recognition accuracy in multiple terminals and multiple scenarios, and have important practical application value. Summary of the Invention

[0006] The purpose of this application is to solve the above-mentioned problems of decreased success rate or increased misrecognition rate caused by a single style and a single gaze angle of iris acquisition.

[0007] According to one aspect of the present application, a method for collecting an iris template based on a simulated environment is provided, comprising: Generate a guide point in the display interface to guide the user's eyes to follow the target; Visible light emitted by a visible spectrum light source illuminates the user's eyeball, causing the user's pupil to shrink to a set target diameter range, while near-infrared light emitted by a near-infrared light source illuminates the iris area; Collect user's iris feature information and eye-related feature parameters; Adjust the brightness of the visible light and the wavelength of the near-infrared light, and repeat the above acquisition steps; Record the user's eye movement characteristics and iris response characteristics under different lighting stimulation conditions; Based on the iris feature information, iris response features, eyeball features and eye movement features, a corresponding user-specific iris feature library is constructed.

[0008] Preferably, the guide points include multiple angle positions for guiding the user's eyeballs to move in different directions, specifically including: an upper angle guide point, a horizontal angle guide point and a lower angle guide point, and the multiple guide points are switched in sequence according to a preset order.

[0009] Preferably, the iris feature information includes: iris texture image, iris clarity, iris area brightness distribution, iris boundary morphology and iris response contrast; The recording of the user's iris response characteristics under different lighting stimulation conditions includes: Under different combinations of visible light brightness and near-infrared light wavelengths, the user's eye dynamic response process is recorded, collecting the pupil contraction speed in response to light changes, the reaction latency after light stimulation, the maximum pupil contraction amplitude, the recovery time, and the iris edge clarity before and after the light change; The contraction speed is obtained by calculating the rate value obtained by dividing the change in pupil diameter during the process of pupil contraction from the initial diameter to the minimum diameter by the time taken for contraction; The reaction latency is obtained by calculating the time interval from the start of light stimulation to the occurrence of an obvious pupil contraction reaction; The maximum pupil contraction amplitude is obtained by calculating the difference between the maximum pupil diameter before light stimulation and the minimum pupil diameter in a stable state after stimulation; The recovery time is calculated by calculating the time from when the pupil reaches the minimum diameter until the pupil diameter recovers to more than 95% of the initial value; The iris edge clarity is measured by comparing the grayscale gradient intensity changes of the iris image before and after acquisition. A larger gradient value indicates a clearer image edge.

[0010] Preferably, the eyeball characteristics include: corneal reflection intensity, eyeball light transmittance parameters, pupil diameter and its variation range; The eyeball features are obtained by calculating the near-infrared reflection intensity and the grayscale value of the visible light transmission image; The eye movement characteristics include: scanning speed, gaze stability, gaze point offset and gaze point offset frequency; The recording of the user's eye movement characteristics under different lighting stimulation conditions includes: obtaining the user's eye movement characteristics by analyzing the dynamic data of the user's eye trajectory during the process of switching the user's gaze point, specifically, The scanning speed is obtained by calculating the ratio of the angular displacement of the eyeball from one fixation point to another fixation point and the time taken; The gaze stability is calculated by recording the offset value of the eye center coordinates in multiple consecutive frames of images when the user is gazing at a single guide point. , calculate the variance of the offset value to obtain; The gaze point offset is calculated by referring to the Euclidean distance deviation between the center of the eyeball and the theoretical projection point of the target guide point, through the formula Calculate, where the is the calibrated position of the target guidance point, It is the Euclidean distance deviation between the eyeball center and the theoretical projection point of the target guidance point; The gaze point offset frequency is obtained by calculating the number of times the gaze point offset exceeds a set threshold within a certain time window.

[0011] Preferably, after collecting the user's iris image, the method further includes: Performing a clarity score on the iris texture image, wherein the score is performed based on the ISO / IEC 29794-6 standard; When the clarity score is lower than a preset threshold, additional iris image acquisition is performed at the gaze angle under the current acquisition conditions until the score is no lower than the threshold.

[0012] Preferably, when the user wears glasses during the acquisition process, the method further includes: Calculate the light spot obstruction degree in the captured image in real time. The light spot obstruction degree is the ratio of the lens reflective area to the iris effective positioning area. The iris effective positioning area is determined by an iris edge extraction algorithm during image acquisition. The annular area is used for iris feature extraction and serves as a reference area for calculating the obstruction degree. When the light spot occlusion degree is greater than a preset occlusion degree threshold, the position of a light shielding plate arranged in the image acquisition light path is adjusted, and the light shielding plate moves along a predetermined trajectory to reduce reflection interference.

[0013] The present invention also provides a device for collecting iris templates based on a simulated environment, which uses the above-mentioned method for collecting iris templates based on a simulated environment, including: A display module, used for displaying guidance information, wherein the display module is provided with a plurality of guidance point display areas; An illumination module comprising a visible light source and a near-infrared light source, wherein the light sources are respectively arranged at symmetrical positions of the device housing; Image acquisition module, used to collect user eyeball images and iris images; a dimming module, electrically connected to the lighting module, for adjusting the brightness of the visible light source and the emission band of the near-infrared light source; The control module is electrically connected to the image acquisition module, the lighting module, the dimming module and the display module, and is used to adjust and coordinate the working status of each functional component.

[0014] Preferably, it also includes: A guidance module, connected to the control module, for generating guidance information to guide the user to switch the gaze point; A light shielding sheet is provided in the optical path of the image acquisition module; The pan-tilt platform comprises a micro motor and a slide rail mechanism. The micro motor is connected to the light shielding sheet and is used to drive the light shielding sheet to move along a preset direction.

[0015] Preferably, the near-infrared light source is provided on an adjustable structure, and the adjustable structure comprises: Sliding bracket to adjust the position of near-infrared light source in space; Rotating structure to adjust the light direction; The control module is used to control the irradiation angle and / or installation position of the near-infrared light source to change when detecting the presence of a lens reflective area in the image, so as to avoid the lens reflective area.

[0016] Preferably, a detection module is further included, and the detection module includes: The image sensing element and the reflection area recognition unit are used to sense the image brightness distribution, and the recognition unit is used to determine the position of the reflective area of ​​the lens. The control module adjusts the position or angle of the near-infrared light source according to the position of the reflective area to avoid reflections interfering with iris image acquisition.

[0017] The present application has the following beneficial effects: effectively simulating a variety of real-world usage environments during the acquisition phase, including different lighting conditions and viewing angles, thereby improving the diversity and representativeness of iris templates from the source; by dynamically adjusting lighting conditions and guiding gaze angles, the adaptability of iris template comparison to image style differences is enhanced, significantly improving subsequent recognition accuracy and robustness; utilizing the dynamic response parameters of the eyeball and iris to supplement information gaps in static texture images and improve the multi-dimensional information fusion capability of iris feature database construction; through standardized scoring and lighting re-shooting mechanisms, ensuring the consistency of template images in resolution and clarity, and improving the system's recognition quality control capability; utilizing comprehensive analysis of iris image acquisition and eye movement behavior to achieve automatic adaptation to complex situations such as user non-rectometry, wearing glasses, and reflective interference, thereby improving terminal usage experience and recognition success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the implementation methods of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the implementation methods or the description of the prior art. Obviously, the drawings described below are only some implementation methods of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a logic block diagram of a method for collecting iris templates based on a simulated environment according to one embodiment of the present application. DETAILED DESCRIPTION

[0020] To facilitate understanding of the present application, a more comprehensive description of the present application will be provided below with reference to the accompanying drawings. The accompanying drawings illustrate preferred embodiments of the present application. However, the present application may be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided to provide a more thorough and comprehensive understanding of the disclosure of the present application.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein in the specification of this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0022] Please refer to Figure 1 An embodiment of the present application provides a method for collecting iris templates based on a simulated environment, comprising: S10. Generate guide points on the display interface to guide the user's eyes to follow the target. This step generates multiple guide points at preset locations on the display screen of the acquisition device, guiding the user to sequentially gaze at target positions in different directions (e.g., above, below, and directly in front of the user). This allows the user's gaze to switch between multiple angles, facilitating the acquisition of iris image information at different gaze angles and assisting in determining whether the user is looking directly at the device, thereby ensuring the diversity and integrity of the captured image.

[0023] S20: Illuminate the user's eyeball with visible light from a visible spectrum light source, causing the user's pupil to contract to a set target diameter range. Simultaneously, illuminate the iris area with near-infrared light from a near-infrared light source. During the acquisition process, this step uses visible light illumination to induce pupil contraction, thereby fully exposing the iris area and improving the proportion and resolution of the iris area in the image. Furthermore, near-infrared illumination enhances the contrast between the iris edge and texture, resulting in a captured iris image with greater clarity and structural integrity.

[0024] S30: Collect the user's iris feature information and eye-related feature parameters. This step uses the image acquisition module to capture images of the user's eyes and extract iris feature information such as iris texture image, boundary outline, and brightness distribution. It also extracts physiological and dynamic parameters related to the eye, such as corneal reflection intensity, light transmittance, pupil diameter, and gaze deviation, providing the raw data foundation for subsequent multi-dimensional feature library construction.

[0025] S40: Adjust the brightness of the visible light and the wavelength of the near-infrared light, and repeat the above acquisition steps. This step dynamically adjusts the lighting system to allow the acquisition process to be performed under different light intensities and different wavelength combinations to simulate different usage environments (such as indoor light, strong light, infrared interference, etc.). This allows the acquisition of iris images in multiple styles, improving the environmental adaptability and versatility of the template library.

[0026] S50. Record the user's eye movement characteristics and iris response characteristics under different lighting stimulation conditions. This step monitors the user's pupil response (such as contraction speed, reaction latency, recovery time, etc.) and eye movement behavior (such as saccade speed, gaze stability, gaze offset and frequency) under different lighting conditions, forming a physiological response feature sequence corresponding to environmental changes, thereby enhancing the tolerance of image deviations in the subsequent recognition stage.

[0027] S60: Construct a user-specific iris feature database based on iris feature information, iris response features, eyeball features, and eye movement features. After acquiring multi-dimensional data, this step uses image quality scoring and redundancy removal to select iris template images that meet feature integrity and image quality standards. This database is then constructed based on the user's individual characteristics, providing a diverse, high-quality reference template set for subsequent recognition and comparison.

[0028] Implementation of the technical solution of this embodiment can: effectively simulate a variety of real-world usage environments during the acquisition phase, including different lighting conditions and viewing angles, thereby improving the diversity and representativeness of iris templates from the source; enhance the adaptability of iris template comparison to image style differences by dynamically adjusting lighting conditions and guiding gaze angles, thereby significantly improving subsequent recognition accuracy and robustness; utilize the dynamic response parameters of the eyeball and iris to supplement information gaps in static texture images, thereby improving the multi-dimensional information fusion capability of iris feature database construction; ensure the consistency of template images in resolution and clarity through standardized scoring and lighting re-shooting mechanisms, thereby improving the system's recognition quality control capability; utilize comprehensive analysis of iris image acquisition and eye movement behavior to achieve automatic adaptation to complex situations such as user non-rectometry, wearing glasses, and reflective interference, thereby improving terminal usage experience and recognition success rate.

[0029] In a specific embodiment, the guide point includes multiple angle positions for guiding the user's eyes to move in different directions, specifically including: an upper angle guide point, a horizontal angle guide point and a lower angle guide point. The multiple guide points are switched in sequence according to a preset order.

[0030] In this embodiment, it should be noted that the guide points are visual markers displayed on the display interface of the image acquisition device, used to guide the user's gaze in a specific direction, thereby enabling the iris acquisition device to obtain iris images from multiple angles. In this embodiment, three guide points at typical angular positions are preset, namely: Upper angle guide point: displayed 15° above the user's eye level to guide the user's eyeballs to deflect upwards; Level Angle Guide Point: Located directly in front of the user, it is used to guide the user's eyes to maintain a horizontal gaze; Lower angle guide point: displayed 15° below the user's eye level to guide the user's eyeballs to deflect downward.

[0031] Multiple guidance points are displayed sequentially in a preset order (e.g., looking straight ahead, looking up, looking down, looking straight ahead). Each guidance point remains displayed for a fixed duration or automatically switches after the eye movement recognition system confirms the user's gaze. This guidance mechanism can be combined with the eye movement detection module to confirm the gaze point, ensuring that the user's gaze reaches the target angle before image capture begins, avoiding iris image distortion or partial loss due to large deviations in the capture angle.

[0032] In addition, the display position, size and brightness of the guidance point can be adapted to different screen resolutions and user eye distance characteristics, ensuring good guidance for different groups of people and usage scenarios.

[0033] Implementing the technical solution of this embodiment can: significantly enrich the angle diversity of the iris template, and by acquiring iris images at different viewing angles, improve the template's adaptability to non-frontal vision situations (such as lowering the head, looking up, tilting the head, etc.) in actual recognition; improve the integrity and stability of iris image acquisition, avoid problems such as partial iris loss and edge blurring caused by random user line of sight movement, and thus improve the effectiveness of the acquisition template; assist in building a more robust iris recognition feature library, and iris images at different angles can cover the deformation characteristics of the iris area at multiple angles, thereby improving the accuracy and fault tolerance of the recognition algorithm; enhance the user guidance interaction experience, and through dynamic and orderly switching of guidance points, make it easier for users to understand and cooperate with the acquisition process, shorten the acquisition time, and improve the practicality of the terminal device.

[0034] Furthermore, the iris feature information includes: iris texture image, iris clarity, iris area brightness distribution, iris boundary morphology, and iris response contrast; Recording the user's iris response characteristics under different lighting stimulation conditions includes: Under different combinations of visible light brightness and near-infrared light wavelengths, the user's eye dynamic response process is recorded, collecting the pupil contraction speed in response to light changes, the reaction latency after light stimulation, the maximum pupil contraction amplitude, the recovery time, and the iris edge clarity before and after the light change; The contraction speed is calculated by dividing the change in pupil diameter from the initial diameter to the minimum diameter by the time taken for contraction. The reaction latency was obtained by calculating the time interval from the onset of light stimulation to the onset of obvious pupil constriction response; The maximum pupil constriction amplitude was obtained by calculating the difference between the maximum pupil diameter before light stimulation and the minimum pupil diameter in the stable state after stimulation; The recovery time was calculated from the time the pupil reached its minimum diameter until the pupil diameter recovered to more than 95% of the initial value; The iris edge clarity is measured by comparing the grayscale gradient intensity changes of the iris image before and after acquisition. The larger the gradient value, the clearer the image edge.

[0035] In this embodiment, it should be noted that the acquisition of iris feature information is not limited to the traditional iris texture image, but also includes a series of parameters that can reflect image quality and structural characteristics, including: Iris texture image: A partial or full-width image of the iris captured by an image acquisition device under near-infrared illumination, which serves as the basic image for subsequent recognition; Iris clarity: Based on the ISO / IEC29794-6 image quality evaluation standard, it can calculate indicators such as edge sharpness and detail texture retention in the image; Iris area brightness distribution: By dividing the iris image into blocks and statistically analyzing its grayscale value distribution, we can measure whether there are problems such as uneven exposure, partial darkness or excessive brightness; Iris boundary morphology: Extract the geometric curves of the iris outer contour and pupil edge to determine whether they are closed, regular, and clear; Iris response contrast: The variation in the grayscale difference of the iris image in the same area under different wavelengths of light can reflect the response sensitivity of the iris structure to light stimulation.

[0036] In addition, to simulate the dynamic response of the iris in actual use environments, we further record the user's iris response characteristics under different lighting combinations: Contraction speed: The change in the pupil diameter obtained by subtracting the diameter when the pupil contracts to the minimum, divided by the time taken for the process, in units of or ; Response latency: the time difference from the start of light stimulation to the first detection of pupil shrinkage in the image, in milliseconds; Maximum contraction amplitude: compares the maximum difference in pupil diameter before and after illumination to measure pupil accommodation ability; Recovery time: After the light stops, the time it takes for the pupil to recover from its minimum diameter to 95% of its initial diameter; Iris edge clarity: By calculating the grayscale gradient of the iris edge area, the gradient intensity before and after the lighting change is extracted and the difference is taken. The larger the difference, the more obvious the image is affected by the lighting.

[0037] All of the above indicators are achieved through image sequence analysis. Combined with image timestamps, they can accurately quantify the impact of light on the physiological structure of the eye, providing high-dimensional data support for the subsequent construction of diverse iris templates.

[0038] Implementation of the technical solution of this embodiment can achieve a comprehensive evaluation of iris image quality, which does not rely solely on subjective judgment or a single parameter, but combines multi-dimensional indicators such as clarity, brightness, and boundary morphology to improve the scientificity and effectiveness of template screening; simulates the dynamic physiological response of the human eye to light in actual complex environments, such as pupil contraction, reaction delay, structural deformation, etc., so that the collected iris image can more realistically reflect the user's status in different terminal recognition environments; enhances the robustness and representativeness of the template library, and by obtaining iris response characteristics under multiple lighting conditions, realizes the adaptation of the template to diverse terminal devices and external light environments; improves recognition accuracy and recognition success rate, especially in scenarios with complex natural light, line of sight deviation, or changes in user status, and can significantly reduce the misrecognition rate and recognition delay; promotes the development of iris recognition algorithms towards intelligence and adaptability, and provides a reliable foundation for dynamic threshold adjustment and image quality screening mechanisms.

[0039] Furthermore, the eyeball characteristics include: corneal reflection intensity, eyeball light transmittance parameters, pupil diameter and its variation range; The eyeball features are obtained by calculating the near-infrared reflection intensity and the grayscale value of the visible light transmission image; Eye movement characteristics include: saccade speed, gaze stability, gaze point offset, and gaze point offset frequency; Recording the user's eye movement characteristics under different lighting stimulation conditions includes: obtaining the user's eye movement characteristics by analyzing the dynamic data of the user's eye trajectory during the process of switching gaze points. Specifically, The saccadic velocity was calculated by calculating the ratio of the angular displacement of the eyeball from one fixation point to another and the time taken. Gaze stability is calculated by recording the offset value of the eye center coordinates in multiple consecutive frames when the user looks at a single guide point. , calculate the variance of the offset value to obtain; The gaze point offset is calculated by calculating the Euclidean distance deviation between the center of the eyeball and the theoretical projection point of the target guide point, using the formula Calculate, where is the calibrated position of the target guidance point, It is the Euclidean distance deviation between the eyeball center and the theoretical projection point of the target guidance point; The frequency of gaze point shift is obtained by the number of times the gaze point shift exceeds the set threshold within a certain time window.

[0040] In this embodiment, it should be noted that the acquisition of eyeball features refers to quantifying basic parameters of the eyeball's physiological structure by image analysis combined with the eyeball's response to different light bands, specifically including: Corneal reflection intensity: refers to the grayscale intensity of the specular reflection spot formed when near-infrared or visible light hits the corneal surface. It reflects the reflective ability of the corneal surface and affects the degree of interference with the iris image. Eyeball light transmittance parameters: By analyzing the grayscale distribution of the image after visible light penetrates the pupil and iris area, the light transmission ability of the transparent medium of the eyeball is evaluated, which is related to the state of the cornea and lens; Pupil diameter and variation range: Automatically extract the pupil boundary in the acquired image, measure its diameter, and obtain the difference between the maximum and minimum pupil diameters, i.e. its dynamic variation range, through multi-frame image tracking.

[0041] The above-mentioned eye features are derived by numerically calculating the grayscale distribution data of images acquired by the image acquisition module at different wavelengths. In particular, near-infrared images can effectively eliminate interference from iris texture, providing clearer pupil and corneal contours and enhancing the stability of feature extraction.

[0042] The acquisition of eye movement features is to extract the dynamic behavior characteristics of the user by tracking the eye movement path in real time when the user interacts with the guidance point, including: Glance speed: The angle change of the eye movement trajectory when the user's gaze moves from one guiding point to another is divided by the time taken. The unit can be ; Gaze stability: When the user gazes at a fixed guide point, the changes in the eye center coordinates in consecutive frames are collected and their variance is calculated. The smaller the variance, the more stable the gaze. Gaze point offset: This measures gaze accuracy by comparing the Euclidean distance between the eye center identified in the image and the system's preset guide point calibration position; Gaze point offset frequency: Within a set time window (e.g., every 2 seconds), count the number of times the gaze point offset exceeds a certain threshold (e.g., 3 pixels), which is used to reflect the stability of the user's visual control and the degree of concentration.

[0043] All of the above eye movement features are obtained by analyzing the time series image data of the eye trajectory during the switching of gaze points. They are automatically completed with the help of image processing and tracking algorithms, and can extract high-dimensional behavioral information without the user's perception.

[0044] Furthermore, after collecting the user's iris image, the following steps are also included: The clarity of iris texture images is scored based on the ISO / IEC29794-6 standard. When the clarity score is lower than the preset threshold, additional iris image acquisition is performed at the gaze angle under the current acquisition conditions until the score is no lower than the threshold.

[0045] In this embodiment, it is important to note that after completing the initial iris image acquisition, to ensure that the quality of the acquired image meets the requirements of the subsequent recognition algorithm, this embodiment introduces a clarity scoring mechanism. This mechanism objectively evaluates the iris image in accordance with the ISO / IEC 29794-6 international standard. This standard defines a series of image quality indicators, with clarity being one of the core indicators, used to measure whether the iris texture details are clearly discernible.

[0046] Specifically, the image clarity score is calculated based on the image's grayscale gradient distribution. This means the image's sharpness is determined by evaluating the magnitude of grayscale changes in the image's edge regions. A higher score indicates clearer iris texture, making it more useful for subsequent template generation and recognition comparison.

[0047] After capturing an image, the system immediately scores its clarity. If the score falls below a set threshold (e.g., 90 points), the image quality is considered suboptimal. At this point, the system automatically triggers additional captures, retaking one or more frames at the current viewing angle and lighting configuration until the score reaches a preset threshold or a certain number of usable template images (e.g., a maximum of eight templates) is reached.

[0048] This process can be completed automatically in the background without interfering with the user's normal operations or interrupting the collection process, ensuring experience continuity and data integrity.

[0049] Implementing the technical solution of this embodiment can: ensure that the quality of the acquired image complies with international standards (ISO / IEC29794-6), and improve the availability and reliability of template data; implement quality self-inspection and dynamic error correction mechanisms in the acquisition process, effectively avoiding image quality problems caused by blur, defocusing or uneven lighting; improve the stability and accuracy of subsequent recognition systems in feature extraction, template matching and other links, and reduce false recognition and rejection rates; optimize the acquisition quality without user perception, and enhance the overall robustness and intelligence of the system without affecting the user experience; support a dynamic acquisition mechanism for multiple frames of images, provide more comprehensive sample data for deep learning training models, and improve the model's generalization ability.

[0050] In an optional embodiment, the user is a person who often wears glasses. In order to facilitate subsequent identification, the user may also wear glasses during the collection process. When the user wears glasses during the collection process, the following further includes: S70, calculating in real time the light spot obstruction degree in the captured image, where the light spot obstruction degree is the ratio of the area of ​​the lens reflective area to the area of ​​the iris effective positioning area. The iris effective positioning area is determined by an iris edge extraction algorithm during image capture. An annular area is used for iris feature extraction and serves as a reference area for calculating the obstruction degree; S80: When the light spot occlusion degree is greater than a preset occlusion degree threshold, the position of a light shielding plate provided in the image acquisition light path is adjusted, and the light shielding plate moves along a predetermined trajectory to reduce reflection interference.

[0051] In this embodiment, it's important to note that when a user wears glasses to capture iris images, bright reflective areas (i.e., light spots) can easily form on the lens surface. This can obscure iris texture, interfere with image quality, and even affect the accuracy of template generation and subsequent comparison. Therefore, in this embodiment, a mechanism for assessing light spot obstruction and adjusting the light shield is designed to dynamically adapt to the acquisition environment of the user wearing glasses.

[0052] In step S70, the system analyzes the collected image in real time and calculates the light spot obstruction degree, which is defined as the ratio of the area of ​​the lens reflective area to the area of ​​the iris effective positioning area.

[0053] The “lens reflective area” refers to the bright, non-iris area detected in the image, which is usually irregular in shape and is determined by brightness threshold segmentation and regional connectivity analysis; The "iris effective positioning area" refers to the iris region surrounding the pupil in the image. Its boundary is extracted using a grayscale change and edge detection algorithm. Circle / ellipse fitting is usually used to lock the area, eliminating the pupil area and the white of the eye boundary, leaving only the annular region for feature extraction. The calculation formula for the occlusion degree is: light spot occlusion degree = area of ​​the lens reflective area / area of ​​the iris effective positioning area.

[0054] In step S80, if the degree of obstruction exceeds a set threshold (eg, 15%), the system automatically starts the light shield adjustment process. The light shield is set in the optical path of the front end of the image acquisition module to block part of the highly reflective incident light.

[0055] The shading sheet is installed on the micro-pan-tilt structure and moves horizontally or vertically along the slide rail according to the preset trajectory. The movement range can be controlled within 10-20mm. The control system adjusts the step size by no more than 2mm each time to achieve progressive optimization of the acquisition effect of a single frame and avoid the impact of drastic changes in the image on recognition.

[0056] This solution can dynamically adapt to the reflective properties of various wearing angles and glasses types (such as plain glasses, myopia glasses, and photochromic glasses) without interrupting the collection process.

[0057] Implementation of the technical solution of this embodiment can: effectively reduce the phenomenon of reflective occlusion in images captured by users wearing glasses, ensuring that the iris area is fully captured; evaluate the occlusion degree in real time based on the image, establish an accurate reflective recognition mechanism, and improve the image quality control capability in the capture stage; automatically adapt to various glasses reflection interference conditions through dynamic shading plate displacement, and improve the environmental adaptability of the overall system; ensure that high-quality templates can be generated even when the user is wearing glasses, enhancing the versatility and user-friendliness of the capture system; improve the matching accuracy and robustness of the iris template in complex usage scenarios, especially for scenarios where users frequently wear glasses and the outdoor lighting is complex; enhance the compatibility and usability of the iris capture system for glasses wearers.

[0058] Another embodiment of the present application provides a device for collecting iris templates based on a simulated environment, applying the above-mentioned method for collecting iris templates based on a simulated environment, including: A display module is used to display guidance information, and the display module is provided with multiple guidance point display areas; The lighting module includes a visible light source and a near-infrared light source, with the light sources respectively disposed at symmetrical positions on the device housing; Image acquisition module, used to collect user eyeball images and iris images; A dimming module, electrically connected to the lighting module, for adjusting the brightness of the visible light source and the emission band of the near-infrared light source; The control module is electrically connected to the image acquisition module, the lighting module, the dimming module and the display module, and is used to adjust and coordinate the working status of each functional component.

[0059] In this embodiment, it should be noted that the device is intended to support iris acquisition tasks based on a variety of lighting conditions, gaze directions, and environmental simulations. Its structural configuration and module connection method are as follows: The display module, located on the front of the device or on the main interface in the direction of the user's gaze, displays visual guidance information, guiding the user to switch gaze points in a pre-set sequence. The display module features multiple guidance point display areas, which are activated sequentially through graphics, light spots, or animations to induce eye movement in different directions (e.g., up, down, left, right, and straight ahead), simulating gaze shifts that may occur during real-world use.

[0060] The lighting module includes a visible light source and a near-infrared light source, which are arranged symmetrically on the left and right or top and bottom of the device housing. This symmetrical arrangement helps achieve uniform illumination of the iris area: The visible light source is mainly used to adjust the pupil size and shrink it to an appropriate diameter, which is conducive to standardized image acquisition; A near-infrared light source is used to illuminate the iris texture area, making it easier to capture images with sufficient contrast and clarity.

[0061] The image acquisition module is set in the direction of the optical axis and is used to capture image information of the user's eyeball, including pupil changes, iris texture and its boundaries, etc. It is the core component for realizing iris template construction.

[0062] The dimming module is electrically connected to the lighting module and is responsible for adjusting the brightness and wavelength parameters of different light sources according to control commands. Specifically for near-infrared light sources, the module supports multi-band output to simulate lighting conditions in different recognition environments, such as 730nm, 810nm, 850nm, and 940nm, enhancing the template's environmental adaptability.

[0063] The control module is the logical control center of the entire system. It establishes communication links with the image acquisition module, lighting module, dimming module, and display module to achieve coordinated control of each module. It dynamically adjusts light source parameters based on the quality feedback of the acquired image, guides the presentation of content, and even integrates with the algorithm module to perform functions such as mask driving, image quality scoring, and template selection.

[0064] Implementation of the technical solution of this embodiment can: guide and dynamically manage user gaze behavior, prompt users to complete iris acquisition at preset angles, and effectively simulate various angle offsets in real recognition scenarios; dynamically adjust visible light and near-infrared light illumination conditions to control pupil status and iris area brightness, which helps to acquire diverse, high-quality iris images; adapt to different users' eye structures, glasses wearing status and external light environments, and improve the stability and versatility of acquired images; provide multi-band infrared lighting capabilities, and support the acquisition of representative iris templates in low light, strong light, polarized light and other environments; integrate an integrated control module to improve system linkage and response efficiency, and ensure real-time coordination and image quality optimization of each component during the acquisition process; provide a more comprehensive and representative template basis for subsequent algorithm modules, and significantly improve the accuracy and processing speed of iris recognition.

[0065] Furthermore, the device also includes: A guidance module, connected to the control module, is used to generate guidance information to guide the user to switch the gaze point; A light shield is provided in the optical path of the image acquisition module; The pan-tilt platform includes a micro motor and a slide rail mechanism. The micro motor is connected to the shading plate and is used to drive the shading plate to move along a preset direction.

[0066] In this embodiment, it should be noted that, in addition to the functions of the basic module, the device further integrates a guiding module, a light shielding sheet assembly, and a pan-tilt structure to improve the accuracy of the user's eye movement control and the stability of the captured image during the acquisition process. The details are as follows: The guidance module connects to the control module, receives scheduling instructions, and generates specific guidance information. This module dynamically displays multiple gaze point locations (e.g., up, down, left, and right) on the display module, using graphical cursors, flashing symbols, animated icons, and other methods to guide the user's gaze, thereby capturing iris images at different gaze angles. While capturing images at different angles, the module continuously monitors the user's eye movements and automatically synchronizes prompts when switching guidance points, ensuring continuous and accurate capture.

[0067] A light shield is placed in the main optical path of the image acquisition module. Its primary function is to partially block ambient light or the reflective area of ​​the user's glasses to prevent strong reflections from interfering with iris imaging. Made of high-shading materials, the shield can be shaped like a strip, fan, or a multi-segment linkage structure, allowing for flexible adjustment of coverage based on actual needs.

[0068] The pan / tilt structure consists of a micromotor and a slide mechanism. The micromotor is responsible for driving the camera, while the slide provides a guide for the light shield. The slide allows the light shield to move linearly or rotate through a certain angle, precisely adjusting its position and range within the light path. The micromotor receives drive signals from the control module and makes fine adjustments during image acquisition based on real-time feedback (such as the degree of reflective occlusion). Each frame's displacement is kept within a predetermined range (e.g., ≤2mm), achieving dynamic occlusion control.

[0069] The technical solution of this embodiment can: guide users to accurately complete multi-perspective gaze switching, improve the data coverage of the iris template in different directions, and thus optimize the adaptability of the terminal to non-frontal images during recognition; automatically identify and respond to glasses reflections or environmental interference light, dynamically adjust the position of the shading plate, and effectively improve the available area and image quality of the iris image; complete shading optimization processing without the user's perception, avoiding manual intervention or forced adjustment of wearing posture, and improving the system's ease of use and user experience; ensure the stability and consistency of image acquisition in complex environments, and provide a high-quality, low-deviation iris data foundation for subsequent comparison models; through the structural design of micro motor + slide rail, precise shading adjustment can be achieved in miniaturized terminal devices, which is conducive to lightweight equipment and integrated deployment.

[0070] In a preferred embodiment, in order to cope with the interference of lens reflection when the user wears glasses, the device is provided with an adjustable bracket and a rotating mechanism of the infrared light source, thereby achieving the purpose of avoiding the reflection area of ​​the lens.

[0071] In this embodiment, the near-infrared light source is arranged on an adjustable structure, which includes: a sliding bracket to achieve position adjustment of the near-infrared light source in space; a rotating structure to achieve adjustment of the light-emitting direction; and a control module for controlling the irradiation angle and / or installation position of the near-infrared light source to change when a lens reflective area is detected in the image, so as to avoid the lens reflective area.

[0072] The device also includes a detection module, which includes: an image sensing element and a reflection area recognition unit. The image sensing element is used to sense the image brightness distribution, and the recognition unit is used to determine the position of the reflective area of ​​the lens. The control module adjusts the position or angle of the near-infrared light source according to the position of the reflective area to avoid reflection interfering with iris image acquisition.

[0073] It should be noted that this device does not require a light shielding structure. Instead, the infrared light source is adjusted in its installation position or illumination angle to change the incident path of the infrared light irradiating the user's eyeball, thereby avoiding the reflective area of ​​the lens. The infrared light source includes multiple light-emitting units, which form an adjustable structure in the following manner: Position adjustment method: The infrared light source is set on a sliding track or a multi-axis bracket, and the control module can drive it to make fine adjustments in the horizontal or vertical direction through a stepper motor, a micro servo or a MEMS structure.

[0074] Angle adjustment method: The infrared light source bracket is equipped with a rotating structure, which can achieve an incident angle change from ±15° to ±45°, avoiding the front reflection of the lens into the collection lens.

[0075] Combined control logic: When the image acquisition module detects a strongly reflective area in the image (identified through highlight analysis), the control module can determine that the current infrared lighting path coincides with the reflection direction of the lens plane, thereby executing the infrared light position / angle adjustment instruction.

[0076] By providing real-time feedback on the reflections in the image, the device can adaptively adjust the lighting pattern of the infrared light source to avoid reflection interference without affecting the user's posture or recognition experience.

[0077] The technical solution of this embodiment can: effectively avoid the influence of the reflective area of ​​the glasses on the iris image acquisition without setting up a mechanical shading structure; actively avoid the high reflection direction of the lens by adjusting the irradiation angle and position of the infrared lamp, and ensure that the infrared light evenly illuminates the iris area; reduce the complexity and volume of the device, and avoid the introduction of additional space, control burden or cost pressure by structures such as the pan-tilt head and shading plate; support light source switching and optimization between multiple infrared lamps, so that the system has more flexible lighting strategies and adaptive acquisition capabilities; improve the quality of iris images, reduce the misrecognition rate and acquisition failure rate caused by reflective interference, and enhance system stability and user experience; it is more suitable for lightweight, embedded terminals or portable device scenarios, and has strong practical promotion value and industrialization potential.

[0078] In an optional embodiment, the near-infrared light source has multi-band emission capability, including a light source array with at least four bands of 730nm, 810nm, 850nm and 940nm, and the lighting module also includes an electrically controlled switching unit for switching the emission band of the near-infrared light source.

[0079] In this embodiment, it should be noted that the near-infrared light source adopts a multi-band emission design, and its emission bands include at least four infrared light sources with different central wavelengths of 730nm, 810nm, 850nm and 940nm. These bands are arranged in an array in the lighting module and can be lit separately or in combination to meet different collection requirements.

[0080] This embodiment also introduces an electronically controlled switching unit, which communicates with the control module and automatically switches wavelength bands based on the acquisition task or real-time image quality. This electronically controlled switching unit can be implemented using a multi-channel driver, an electronically controlled switch matrix, or a PWM control chip. Through its high-speed switching capability, it activates the corresponding near-infrared light source on demand, ensuring uniform illumination of the iris under different spectral conditions.

[0081] The purpose of setting the four bands is to improve the adaptability and clarity of iris images: The 730nm band has high tissue penetration and is suitable for iris edge enhancement in low-light environments; 810nm and 850nm are the main bands commonly used for iris imaging, providing good contrast and iris texture resolution; The 940nm band is used to compensate for the strong interference from other bands in bright environments, thereby improving the ability to resist ambient light interference.

[0082] During the lighting control process, the system can select single-band or multi-band combination lighting based on the real-time image clarity score or ambient light detection results to obtain the optimal iris imaging effect.

[0083] Implementation of the technical solution of this embodiment can: enhance the imaging capability of iris images under various ambient light conditions and adapt to complex environments such as low indoor light and strong outdoor light; achieve enhanced capture of different iris structural features and improve texture resolution by switching near-infrared light sources of different bands; reduce image deviation or insufficient brightness caused by single-band lighting, and improve the consistency and quality of captured images; provide more stable capture performance under individual differences such as different pupil states and skin reflection characteristics of users, and enhance image versatility; provide multi-band support for subsequent dynamic reshooting mechanisms based on clarity scoring and image analysis, so as to facilitate automatic compensation for image capture failures or substandard image quality.

[0084] In an optional embodiment, the device further includes a detection module, which includes: An image sensing element and a reflection area recognition unit, wherein the image sensing element is used to sense the image brightness distribution of the reflection area of ​​the lens in front of the acquisition window, and the recognition unit is used to determine the position of the reflection area and transmit the position information to the control module; The control module sends a driving signal to the pan / tilt head to control the shading plate to adjust the displacement.

[0085] In this embodiment, it should be noted that the device is further provided with a detection module for automatically identifying the reflective area on the user's glasses during the acquisition process, thereby assisting in the precise adjustment of the light shielding plate. The detection module includes the following two key subcomponents: Image sensors, located upstream of the acquisition optical path or at the edge of the acquisition lens, provide real-time image brightness distribution in front of the current acquisition window. By capturing multiple image frames and analyzing their brightness, the system can detect the presence of strong light spots or highly reflective areas (typically caused by eyeglass lenses) within the captured image.

[0086] The reflective area recognition unit uses image data output by the image sensor to perform edge extraction, contour analysis, and brightness threshold determination on highlight areas to determine the spatial location, shape, boundary, and center coordinates of the reflective area on the lens. The identified reflective area data (such as coordinates, width, height, and direction) is transmitted to the control module in real time.

[0087] The control module then calculates the direction and amount of movement of the shielding plate based on the reflection position information provided by the recognition unit and issues corresponding drive instructions to the gimbal assembly. The gimbal assembly uses a micromotor and a slide mechanism to precisely move or rotate the shielding plate, ensuring that it covers the reflective area and prevents strong reflected light from entering the imaging optical path.

[0088] This process can be completed immediately after single-frame image processing, or more stable judgment can be achieved through multi-frame analysis. The entire adjustment process can be completed without the user noticing, improving acquisition efficiency and experience.

[0089] Implementation of the technical solution of this embodiment can: realize real-time detection and positioning of interference areas caused by lens reflections for users wearing glasses, thereby improving the intelligence level of the equipment; automatically adjust the position of the shading plate according to the detection results, realize dynamic shielding control of reflective interference, and ensure the quality of iris images; avoid problems such as reduced image clarity, recognition failure or acquisition interruption caused by reflective interference, and improve the success rate of template acquisition; eliminate dependence on user posture cooperation or manual intervention, and improve ease of use and system fault tolerance; provide preprocessing guarantees for subsequent image quality scoring and acquisition judgment, and enhance the stability and overall robustness of the image processing link; cooperate with the gimbal to realize sub-millimeter displacement control of the shading plate, meet the needs of high-precision reflective shielding, and ensure the complete acquisition of the iris area.

[0090] The above-described embodiments merely represent several embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for collecting iris templates based on a simulated environment, characterized in that: include: Generate a guide point in the display interface to guide the user's eyes to follow the target; Visible light emitted by a visible spectrum light source illuminates the user's eyeball, causing the user's pupil to shrink to a set target diameter range, while near-infrared light emitted by a near-infrared light source illuminates the iris area; Collect user's iris feature information and eye-related feature parameters; Adjust the brightness of the visible light and the wavelength of the near-infrared light, and repeat the above acquisition steps; Record the user's eye movement characteristics and iris response characteristics under different lighting stimulation conditions; Based on the iris feature information, iris response features, eyeball features and eye movement features, a corresponding user-specific iris feature library is constructed.

2. The method according to claim 1, characterized in that The guide points include multiple angle positions for guiding the user's eyeballs to move in different directions, specifically including: an upper angle guide point, a horizontal angle guide point and a lower angle guide point. The multiple guide points are switched in sequence according to a preset order.

3. The method according to claim 2, characterized in that The iris feature information includes: iris texture image, iris clarity, iris area brightness distribution, iris boundary shape and iris response contrast; The recording of the user's iris response characteristics under different lighting stimulation conditions includes: Under different combinations of visible light brightness and near-infrared light wavelengths, the user's eye dynamic response process is recorded, collecting the pupil contraction speed in response to light changes, the reaction latency after light stimulation, the maximum pupil contraction amplitude, the recovery time, and the iris edge clarity before and after the light change; The contraction speed is obtained by calculating the rate value obtained by dividing the change in pupil diameter during the process of pupil contraction from the initial diameter to the minimum diameter by the time taken for contraction; The reaction latency is obtained by calculating the time interval from the start of light stimulation to the occurrence of an obvious pupil contraction reaction; The maximum pupil contraction amplitude is obtained by calculating the difference between the maximum pupil diameter before light stimulation and the minimum pupil diameter in a stable state after stimulation; The recovery time is calculated by calculating the time from when the pupil reaches the minimum diameter until the pupil diameter recovers to more than 95% of the initial value; The iris edge clarity is measured by comparing the grayscale gradient intensity changes of the iris image before and after acquisition. A larger gradient value indicates a clearer image edge.

4. The method according to claim 2, characterized in that The eyeball characteristics include: corneal reflection intensity, eyeball light transmittance parameters, pupil diameter and its variation range; The eyeball features are obtained by calculating the near-infrared reflection intensity and the grayscale value of the visible light transmission image; The eye movement characteristics include: scanning speed, gaze stability, gaze point offset and gaze point offset frequency; The recording of the user's eye movement characteristics under different lighting stimulation conditions includes: obtaining the user's eye movement characteristics by analyzing the dynamic data of the user's eye trajectory during the process of switching the user's gaze point, specifically, The scanning speed is obtained by calculating the ratio of the angular displacement of the eyeball from one fixation point to another fixation point and the time taken; The gaze stability is calculated by recording the offset value of the eye center coordinates in multiple consecutive frames of images when the user is gazing at a single guide point. , calculate the variance of the offset value to obtain; The gaze point offset is calculated by calculating the Euclidean distance deviation between the center of the eyeball and the theoretical projection point of the target guide point, using the formula Calculate, where the is the calibrated position of the target guidance point, It is the Euclidean distance deviation between the eyeball center and the theoretical projection point of the target guidance point; The gaze point offset frequency is obtained by calculating the number of times the gaze point offset exceeds a set threshold within a certain time window.

5. The method according to claim 3, characterized in that After collecting the user's iris image, the method further includes: Performing a clarity score on the iris texture image, wherein the score is performed based on the ISO / IEC 29794-6 standard; When the clarity score is lower than a preset threshold, additional iris image acquisition is performed at the gaze angle under the current acquisition conditions until the score is no lower than the threshold.

6. The method according to claim 1, characterized in that When the user wears glasses during the collection process, it further includes: Calculate the light spot obstruction degree in the captured image in real time. The light spot obstruction degree is the ratio of the lens reflective area to the iris effective positioning area. The iris effective positioning area is determined by an iris edge extraction algorithm during image acquisition. The annular area is used for iris feature extraction and serves as a reference area for calculating the obstruction degree. When the light spot occlusion degree is greater than a preset occlusion degree threshold, the position of a light shielding plate arranged in the image acquisition light path is adjusted, and the light shielding plate moves along a predetermined trajectory to reduce reflection interference.

7. A device for collecting iris templates based on a simulated environment, applying the method for collecting iris templates based on a simulated environment according to any one of claims 1 to 6, characterized in that: include: A display module, used for displaying guidance information, wherein the display module is provided with a plurality of guidance point display areas; An illumination module comprising a visible light source and a near-infrared light source, wherein the light sources are respectively arranged at symmetrical positions of the device housing; Image acquisition module, used to collect user eyeball images and iris images; a dimming module, electrically connected to the lighting module, for adjusting the brightness of the visible light source and the emission band of the near-infrared light source; The control module is electrically connected to the image acquisition module, the lighting module, the dimming module and the display module, and is used to adjust and coordinate the working status of each functional component.

8. The device according to claim 7, characterized in that Also includes: A guidance module, connected to the control module, for generating guidance information to guide the user to switch the gaze point; A light shielding sheet is provided in the optical path of the image acquisition module; The pan-tilt platform comprises a micro motor and a slide rail mechanism. The micro motor is connected to the light shielding sheet to drive the light shielding sheet to move along a preset direction.

9. The device according to claim 7, characterized in that The near-infrared light source is arranged on an adjustable structure, and the adjustable structure includes: Sliding bracket to adjust the position of near-infrared light source in space; Rotating structure to adjust the light direction; The control module is used to control the irradiation angle and / or installation position of the near-infrared light source to change when detecting the presence of a lens reflective area in the image, so as to avoid the lens reflective area.

10. The device according to claim 9, characterized in that Also included is a detection module, the detection module comprising: The image sensing element and the reflection area recognition unit are used to sense the image brightness distribution, and the recognition unit is used to determine the position of the reflective area of ​​the lens. The control module adjusts the position or angle of the near-infrared light source according to the position of the reflective area to avoid reflections interfering with iris image acquisition.

Images