Iris recognition device and method in complex illumination environment

Abstract

The invention discloses an iris recognition device and method in a complex illumination environment. An iris image acquisition unit acquires iris images of eyes of a user; the intelligent exposure control unit adjusts exposure parameters based on the iris image according to the association between the pupil size and the ambient light; the iris quality evaluation unit evaluates the quality of the iris images by adopting a plurality of quality measures, filters the iris images which are unqualified in quality evaluation, and obtains a multi-measure quality evaluation result for the iris images which are qualified in quality; the iris quality enhancing unit is used for enhancing the quality of the iris image of which the quality is lower than a preset threshold value; the iris feature coding unit is used for representing the iris image subjected to quality compensation in an iris feature coding form; and the iris feature comparison unit calculates the similarity between different iris feature codes, and identifies the iris feature code to be identified in the registered iris feature code library by comparing the iris feature code to be identified with the registered iris feature code library.

Description

Technical Field

[0001] This invention relates to, but is not limited to, the field of iris recognition technology, and particularly to an iris recognition device and method under complex lighting conditions. Background Technology

[0002] Iris recognition is a biometric identification technology. Its principle involves encoding the features of the iris region using specific algorithms to calculate its similarity. Currently, typical iris recognition systems utilize infrared cameras to capture clear iris images, then employ image preprocessing algorithms to segment the iris region, and finally calculate the iris features.

[0003] The accuracy of iris recognition is closely related to the clarity of the acquired iris texture. When the ambient lighting is ideal, a clearer iris texture image is usually obtained, leading to higher recognition accuracy. However, when the lighting environment is complex, especially when strong natural light affects the infrared camera's imaging, the brightness distribution and contrast of the iris image change, and some details may even be lost. All of these factors severely impact the accuracy of iris recognition. Therefore, using iris recognition in complex outdoor lighting environments has always been a challenge in the industry. Summary of the Invention

[0004] The purpose of this invention is to provide an iris recognition device and method under complex lighting conditions, so as to solve the problem that existing iris recognition methods will seriously affect the accuracy of iris recognition under complex lighting conditions.

[0005] The technical solution of the present invention is as follows: In a first aspect, the present invention provides an iris recognition device under complex lighting conditions, comprising: an iris image acquisition unit, an intelligent exposure control unit, an iris quality evaluation unit, an iris quality enhancement unit, and an iris feature encoding unit connected to the iris image acquisition unit respectively, and an iris feature comparison unit connected to the iris quality enhancement unit; the iris quality enhancement unit is also connected to the iris quality evaluation unit; The iris image acquisition unit is used to acquire iris images of the user's eyes; The intelligent exposure control unit is used to adjust the exposure parameters based on the iris image acquired by the iris image acquisition unit and the correlation between pupil size and ambient light, so as to obtain an image exposure effect suitable for the current ambient light. The iris quality evaluation unit is used to evaluate the quality of iris images using multiple quality measures, filter out iris images that fail the quality evaluation, and obtain multi-measure quality evaluation results for iris images that pass the quality evaluation. The iris quality enhancement unit is used to enhance the quality of iris images with quality below a preset threshold based on the multi-measure quality evaluation results of the iris quality evaluation unit, so as to improve their recognizability. The iris feature coding unit is used to represent the quality-compensated iris image using iris feature coding. The iris feature comparison unit is used to calculate the similarity between different iris feature codes. By comparing the iris feature code to be identified with the registered iris feature code library, the iris feature code to be identified is identified in the registered iris feature code library.

[0006] Optionally, in the iris recognition device under complex lighting conditions as described above, the iris image acquisition unit includes: a camera, a near-infrared supplementary light, and a narrowband filter; Among them, multiple near-infrared supplementary lights with fixed wavelengths are evenly distributed around the infrared camera in the circumferential direction. The position of the near-infrared supplementary lights is such that when the user looks straight ahead, the reflected light spot formed by the light source falls on the user's pupil area. The narrowband filter is placed in front of the infrared camera to filter out interfering spectral energy in unwanted wavelengths.

[0007] Optionally, in the iris recognition device under complex lighting conditions as described above, The intelligent exposure control unit is specifically used to calculate the pupil radius and iris radius based on the acquired iris image using an iris positioning algorithm, and to calculate the pupil-iris radius ratio. Then, based on the radius ratio, it controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit.

[0008] Optionally, in the iris recognition device under complex lighting conditions as described above, the intelligent exposure control unit controls the iris image acquisition unit in the following ways: The control method for the near-infrared supplementary light is as follows: when the ratio of the pupil to the iris radius is greater than the first ratio threshold, it indicates that the ambient light is relatively dark, so the light intensity of the near-infrared supplementary light is increased; when the ratio of the pupil to the iris radius is less than the second ratio threshold, it indicates that the ambient light is sufficient, so the light intensity of the near-infrared supplementary light is reduced; wherein, the first ratio threshold is greater than the second ratio threshold. The method for controlling the exposure time is as follows: when the ratio of the pupil to the iris radius increases, it indicates that the ambient light is in the process of darkening, so the exposure time is extended to allow more light to enter the camera; when the ratio of the pupil to the iris radius decreases, it indicates that the ambient light is in the process of brightening, so the exposure time is reduced to suppress strong light interference. The exposure gain is controlled as follows: when the pupil-iris radius ratio increases, it indicates a low-light environment, so the exposure gain is increased; when the pupil-iris radius ratio decreases, it indicates a well-lit environment, so the gain is reduced to obtain a cleaner image.

[0009] Secondly, the iris recognition method under complex lighting conditions as described above, wherein the iris recognition method is performed using an iris recognition device under complex lighting conditions as described in any of the above claims, includes: Step 1: Acquire iris images of the user's eyes using the iris image acquisition unit; Step 2: Based on the acquired iris image, the intelligent exposure control unit adjusts the light intensity and exposure parameters of the iris image acquisition unit according to the correlation between pupil size and ambient light, thereby obtaining a more suitable image exposure effect for the current ambient light. Step 3: The iris quality evaluation unit evaluates the quality of the iris image using multiple quality measures, filters out iris images that fail the quality evaluation, and obtains multi-measure quality evaluation results for iris images that pass the quality evaluation. Step 4: Based on the multi-measure quality evaluation results of the iris image by the iris quality evaluation unit, the iris quality enhancement unit enhances the quality of iris images with quality below a preset threshold to improve their recognizability. Step 5: Encode the iris features of all iris images using the iris feature coding unit; Step 6: The iris feature code to be identified is compared with the registered iris feature code library by the iris feature comparison unit, and the iris feature code to be identified is identified in the registered iris feature code library.

[0010] Optionally, in the iris recognition method under complex lighting conditions as described above, step 2 includes: The intelligent exposure control unit calculates the pupil radius and iris radius using an iris positioning algorithm based on the acquired iris image, and calculates the pupil-iris radius ratio. Then, it controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit according to the radius ratio.

[0011] Optionally, in the iris recognition method under complex lighting conditions as described above, In step 3, the multiple quality measures include: brightness contrast detection, blurriness detection, integrity detection, and gaze direction detection; and the detection methods for each quality measure are as follows: The brightness and contrast detection method is as follows: calculate the iris brightness, the contrast between the iris and sclera, and the contrast between the iris and the pupil respectively, and compare each calculated value with the corresponding preset threshold. The ambiguity detection method is as follows: calculate the aggregation of the reflected light spot of the near-infrared supplementary light lamp and the intensity of the iris texture, and compare each calculated value with the corresponding preset threshold. The integrity detection method is as follows: the integrity of the iris is calculated based on the proportion of the iris obscured by the eyelid, and then compared with a preset threshold. The gaze direction detection method is as follows: the gaze direction is estimated based on the positional relationship between the pupil and the reflected light spot of the near-infrared supplementary light, and then compared with a preset threshold.

[0012] Optionally, in the iris recognition method under complex lighting conditions as described above, the quality enhancement of the iris image in step 4 includes: brightness contrast enhancement and iris texture detail enhancement; The brightness contrast enhancement method is as follows: for iris images with a brightness contrast score lower than a preset threshold, their brightness contrast is enhanced based on their brightness contrast evaluation results. The method for enhancing iris texture details is as follows: for iris images with blurriness higher than a preset threshold, the texture details of the iris region are enhanced by a texture detail enhancement algorithm based on the blurriness score.

[0013] Optionally, in the iris recognition method under complex lighting conditions as described above, in the iris feature encoding of step 5, To address the issue of a large pupil dilation / contraction range under complex lighting conditions, a multi-scale filter is used to calculate the iris feature code.

[0014] The beneficial effects of this invention are as follows: This invention provides an iris recognition device and method under complex lighting conditions. After the iris image acquisition unit acquires the iris image of the user's eye, the iris quality evaluation unit performs multi-quality metric evaluation, and the iris quality enhancement unit enhances the quality of the iris images with poor quality. Iris feature encoding is then performed on all the enhanced iris images, allowing for matching with the iris feature encoding to be identified in a registered iris feature encoding library to identify the current iris feature encoding. In the technical solution provided by this invention, the intelligent exposure control unit can adjust the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit based on the acquired iris image, so that the iris image acquisition unit can acquire an image exposure effect more suitable for the current ambient lighting. Furthermore, the evaluation method for iris images in this invention adopts a multi-quality metric evaluation method, which can evaluate the iris image quality from multiple dimensions, and enhance the quality of images with lower quality based on the multi-metric quality evaluation results. Obviously, the technical solution provided by this invention processes iris images under complex lighting conditions through various means, such as the shooting parameters of the shooting device and the image enhancement processing method, so as to improve the accuracy of the entire device in recognizing iris images. Attached Figure Description

[0015] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of the present invention and do not constitute a limitation on the technical solutions of the present invention.

[0016] Figure 1 This is a schematic diagram of the structure of an iris recognition device under complex lighting conditions provided in an embodiment of the present invention; Figure 2 This is a flowchart illustrating an iris recognition method under complex lighting conditions, provided as an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

[0018] As explained in the background section, existing iris recognition methods can achieve high recognition accuracy under ideal ambient lighting conditions. However, in complex lighting environments, the brightness distribution and contrast of the iris image change, and some details may be lost, severely impacting iris recognition accuracy. Therefore, developing an iris recognition method for complex lighting environments is of significant practical importance for the application of iris recognition technology in outdoor high-light conditions.

[0019] To address the aforementioned problems and the need for iris recognition under complex lighting conditions, this invention provides an iris recognition device and method for complex lighting conditions.

[0020] The present invention provides the following specific embodiments, which can be combined with each other. For the same or similar concepts or processes, they may not be described again in some embodiments.

[0021] Figure 1 This is a schematic diagram of an iris recognition device under complex lighting conditions, provided as an embodiment of the present invention. Figure 1 As shown, the iris recognition device under complex lighting conditions provided by the present invention includes: an iris image acquisition unit, an intelligent exposure control unit, an iris quality evaluation unit, an iris quality enhancement unit, an iris feature encoding unit, and an iris feature comparison unit.

[0022] like Figure 1As shown, in the aforementioned iris recognition device under complex lighting conditions, the intelligent exposure control unit, iris quality evaluation unit, and iris quality enhancement unit are respectively connected to the iris image acquisition unit, and the iris quality enhancement unit is also connected to the iris quality evaluation unit; furthermore, the rear end of the iris quality enhancement unit is sequentially connected to the iris feature encoding unit and the iris feature comparison unit. The functions of each of the above units are explained below: The iris image acquisition unit is used to acquire images of the user's iris.

[0023] The intelligent exposure control unit is used to adjust the exposure parameters based on the iris image acquired by the iris image acquisition unit and the correlation between pupil size and ambient light, so as to obtain a more suitable image exposure effect for the current ambient light.

[0024] The iris quality assessment unit is used to evaluate the quality of iris images using multiple quality measures, filter out iris images that fail the quality assessment, and obtain multi-measure quality assessment results for iris images that pass the quality assessment.

[0025] The iris quality enhancement unit is used to enhance the quality of iris images with quality below a preset threshold based on the multi-measure quality evaluation results of the iris quality evaluation unit, so as to improve their recognizability.

[0026] The iris feature coding unit is used to represent the quality-compensated iris image using iris feature coding.

[0027] The iris feature comparison unit is used to calculate the similarity between different iris feature codes. By comparing the iris feature code to be identified with the registered iris feature code library, the iris feature code to be identified is identified in the registered iris feature code library.

[0028] In one implementation of this invention, the iris image acquisition unit includes: a camera, a near-infrared fill light, and a narrowband filter.

[0029] In this implementation, the camera serves as the main component for capturing images of the user's eyes. Multiple near-infrared illuminators of fixed wavelengths are evenly distributed circumferentially around the infrared camera. The positions of these illuminators are designed so that when the user is looking straight ahead, the reflected light spots from the light sources fall on the user's pupil area, allowing the camera to clearly capture images of the user's pupil and iris. Additionally, a narrow-band filter is placed in front of the infrared camera to filter out interfering spectral energy in unwanted wavelengths.

[0030] In one implementation of this invention, the control method of the intelligent exposure control unit is as follows: Based on the acquired iris image, the pupil radius and iris radius are calculated using an iris localization algorithm, and the pupil-iris radius ratio is calculated. Then, the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit are controlled according to the radius ratio.

[0031] In practical implementation, the specific control method by which the intelligent exposure control unit controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light is explained below: (1) The control method for the intensity of near-infrared supplementary light is as follows: when the ratio of pupil to iris radius is large, for example, greater than the first ratio threshold, it indicates that the ambient light is dark, and the intensity of the near-infrared supplementary light can be appropriately increased; conversely, when the ratio of pupil to iris radius is small, for example, less than the second ratio threshold, it indicates that the ambient light is relatively sufficient, and the intensity of the infrared supplementary light can be reduced; the first ratio threshold is greater than the second ratio threshold. (2) The method of controlling the exposure time is as follows: when the ratio of the pupil to the iris radius increases, it indicates that the ambient light is in the process of darkening, and the exposure time can be appropriately extended to allow more light to enter the camera; when the ratio of the pupil to the iris radius decreases, it indicates that the ambient light is in the process of brightening, and the exposure time can be appropriately reduced to suppress strong light interference. (3) The method of controlling exposure gain is as follows: when the ratio of pupil to iris radius increases, that is, when the pupil is dilated, it means that in low light environment, in order to ensure the brightness of the image, the exposure gain can be appropriately increased; it should be noted that excessively high gain will increase the noise of the image; when the ratio of pupil to iris radius decreases, that is, when the pupil is constricted, it means that in a well-lit environment, the gain can be reduced to obtain a cleaner image.

[0032] Based on the iris recognition device for complex lighting conditions provided in the embodiments of the present invention, the embodiments of the present invention also provide a method for implementing iris recognition using the iris recognition device for complex lighting conditions, such as... Figure 2 As shown, the method includes the following steps: Step 1: Acquire iris images of the user's eyes using the iris image acquisition unit; Step 2: Based on the acquired iris image, the intelligent exposure control unit adjusts the light intensity and exposure parameters of the iris image acquisition unit according to the correlation between pupil size and ambient light, thereby obtaining a more suitable image exposure effect for the current ambient light. Step 3: The iris quality evaluation unit evaluates the quality of the iris image using multiple quality measures, filters out iris images that fail the quality evaluation, and obtains multi-measure quality evaluation results for iris images that pass the quality evaluation. Step 4: Based on the multi-measure quality evaluation results of the iris image by the iris quality evaluation unit, the iris quality enhancement unit enhances the quality of iris images with quality below a preset threshold to improve their recognizability. Step 5: Encode the iris features of all iris images using the iris feature coding unit; Step 6: The iris feature code to be identified is compared with the registered iris feature code library by the iris feature comparison unit, and the iris feature code to be identified is identified in the registered iris feature code library.

[0033] In one implementation of this invention, step 2 includes the following: The intelligent exposure control unit calculates the pupil radius and iris radius using an iris positioning algorithm based on the acquired iris image, and calculates the pupil-iris radius ratio. Then, it controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit according to the radius ratio.

[0034] In one implementation of this invention, the multiple quality measures in step 3 include: brightness contrast detection, blurriness detection, integrity detection, and gaze direction detection. In this implementation, the detection method for each quality measure is as follows: The brightness and contrast detection method is as follows: calculate the iris brightness, the contrast between the iris and sclera, and the contrast between the iris and the pupil respectively, and compare each calculated value with the corresponding preset threshold. The ambiguity detection method is as follows: calculate the aggregation of the reflected light spot of the near-infrared supplementary light lamp and the intensity of the iris texture, and compare each calculated value with the corresponding preset threshold. The integrity detection method is as follows: the integrity of the iris is calculated based on the proportion of the iris obscured by the eyelid, and then compared with a preset threshold. The gaze direction detection method is as follows: the gaze direction is estimated based on the positional relationship between the pupil and the reflected light spot of the near-infrared supplementary light, and then compared with a preset threshold.

[0035] In one implementation of this invention, step 4, the enhancement of the iris image quality, includes: brightness and contrast enhancement, and iris texture detail enhancement. The enhancement method for the iris image in this implementation is described below: The brightness and contrast enhancement method is as follows: for iris images with a brightness and contrast score lower than a preset threshold, their brightness and contrast are enhanced based on their brightness and contrast evaluation results; The method for enhancing iris texture details is as follows: for iris images with blurriness higher than a preset threshold, the texture details of the iris region are enhanced by a texture detail enhancement algorithm based on the blurriness score.

[0036] In one implementation of this invention, in the iris feature encoding of step 5 above, iris feature encoding is performed on all iris images by an iris feature encoding unit. It should be noted that the iris images for which iris feature encoding is performed include the images that have undergone quality enhancement processing in step 4, and images whose quality meets a preset threshold but have not undergone quality enhancement processing.

[0037] Furthermore, for situations with a large pupil dilation range under complex lighting conditions, a multi-scale filter is used to calculate the iris feature code. For example, for continuous images, it is possible to identify whether the iris images within a certain time period fall under the aforementioned "situation with a large pupil dilation range under complex lighting conditions." For continuous iris images within this time period that are identified as falling under the aforementioned situation, a multi-scale filter is used to calculate the iris feature code.

[0038] This invention provides an iris recognition device and method under complex lighting conditions. After acquiring iris images of the user's eyes through an iris image acquisition unit, a multi-quality metric evaluation unit performs evaluation, and an iris quality enhancement unit enhances the quality of poor-quality iris images. All enhanced iris images are then coded with iris features, allowing for matching with a registered iris feature coding library to identify the current iris feature code. In the technical solution provided by this invention, the intelligent exposure control unit can adjust the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit based on the acquired iris images, enabling the iris image acquisition unit to acquire images with more suitable exposure effects for the current ambient lighting. Furthermore, this invention employs a multi-quality metric evaluation method for iris images, allowing for evaluation of iris image quality from multiple dimensions, and enhancing the quality of lower-quality images based on the multi-metric evaluation results. Obviously, the technical solution provided by this invention processes iris images under complex lighting conditions through various means, such as the shooting parameters of the shooting device and the image enhancement processing method, so as to improve the accuracy of the entire device in recognizing iris images.

[0039] The following is an illustrative example illustrating the implementation of the iris recognition device and method under complex lighting conditions provided by the present invention.

[0040] Implementation Example This embodiment provides an iris recognition device and method under complex lighting conditions, employing, for example... Figure 1 The iris recognition device shown in the complex lighting environment performs the following steps: Step 1: Acquire iris images of the user's eyes using the iris image acquisition unit; In this iris image acquisition unit, two 850nm wavelength near-infrared supplementary lights are placed on the left and right sides of the infrared camera. Their positions are such that when the user looks straight ahead, the reflected light spots formed by the light sources fall on the user's pupil area. In addition, a narrow-band filter is placed in front of the infrared camera to filter out interference spectral energy in unnecessary wavelength bands.

[0041] Step 2: Run the YOLO-based iris localization algorithm in the intelligent exposure control unit. Based on the detected iris and pupil position coordinates, obtain the pupil radius and iris radius, and then determine the pupil-iris radius ratio. This ratio is then used to control the light intensity, exposure time, and exposure gain of the near-infrared supplementary light. The specific control method is as follows: Controlling the light intensity of near-infrared supplementary lights: If the ratio of the pupil to the iris radius is greater than 2 / 3, it indicates that the ambient light is relatively dim, and the intensity of the infrared supplementary lights can be appropriately increased; conversely, if the ratio of the pupil to the iris radius is less than 1 / 3, it indicates that the ambient light is relatively sufficient, and the intensity of the infrared supplementary lights can be reduced.

[0042] Controlling exposure time: When the ratio of the pupil-iris radius to the iris radius is greater than or equal to 1 / 2 and shows an increasing trend, it indicates that the ambient light is in the process of darkening. The exposure time can be appropriately extended to allow more light to enter the camera. When the ratio of the pupil-iris radius to the iris radius is less than 1 / 2 and shows a decreasing trend, it indicates that the ambient light is in the process of brightening. The exposure time can be appropriately reduced to suppress strong light interference.

[0043] Controlling Exposure Gain: Considering the pupil-iris radius ratio and image brightness, in low-light environments, the pupil dilates, and to ensure image brightness, the exposure gain can be appropriately increased; however, it's important to note that excessively high gain will increase image noise. In well-lit conditions where the pupil constricts, the gain should be reduced to obtain a cleaner image.

[0044] Step 3: The iris image quality is evaluated using multiple quality measures by the iris quality evaluation unit. This mainly includes brightness and contrast detection, blurriness detection, integrity detection, and gaze direction detection. Iris images that do not meet the quality requirements are filtered out; the multi-measure quality evaluation results are obtained for iris images that meet the quality requirements. The specific evaluation method is as follows: Brightness and contrast detection: Calculate the iris brightness, iris-sclera contrast, and iris-pupil contrast, and compare each calculated value with the corresponding preset threshold. Typically, the iris-sclera contrast is required to be no less than 10, and the iris-pupil contrast is required to be no less than 30. Ambiguity detection: The aggregation of reflected light spots from infrared supplementary lights and the intensity of iris texture are calculated, and each calculated value is compared with the corresponding preset threshold. The aggregation of infrared supplementary lights can be obtained through threshold segmentation and connected element analysis, and the intensity of iris texture can be obtained through Sobel or Laplacian operators. Integrity test: The integrity of the iris is calculated based on the proportion of the iris obscured by the eyelids and compared with a preset threshold. Typically, an iris integrity of greater than 60% is required. Gaze direction detection: The gaze direction is estimated based on the positional relationship between the pupil and the reflected light spot of the infrared supplement lamp, and compared with a preset threshold. Typically, the gaze is required to be within ±10° of the center.

[0045] Step 4: Based on the multi-measure quality assessment results of the iris image by the iris quality assessment unit, the iris quality enhancement unit performs quality enhancement on the low-quality iris image, mainly including two enhancement methods: brightness and contrast enhancement, and iris texture detail enhancement. The specific enhancement methods are explained below: Brightness and contrast enhancement: For iris images with low brightness and contrast scores, their brightness and contrast are enhanced based on their brightness and contrast evaluation results. Gamma correction is used to enhance the image brightness, and this is achieved by establishing a linear equation between the enhancement coefficient and the brightness and contrast score. Iris texture detail enhancement: For iris images with high blur, the texture details of the iris region are enhanced by a texture detail enhancement algorithm based on the blur score. High-frequency details can be enhanced by superimposing enhancement coefficients to enhance the iris texture.

[0046] Step 5: Iris feature encoding is performed on the iris image through the iris feature encoding unit. In the case of a large pupil dilation range under complex lighting conditions, a multi-scale Gabor complex value filter and binary quantization are used to calculate the iris features. It should be noted that the iris images used for iris feature encoding in this step include the high-quality iris images evaluated in step 3. These iris images do not need to undergo quality enhancement in step 4 and are directly used for feature encoding. For the low-quality images that were not filtered out in step 3, feature encoding is performed after quality enhancement in step 4.

[0047] Step 6: The iris feature code to be identified is compared with the registered iris feature code library by the iris feature comparison unit, and the similarity is obtained by calculating the Hamming distance, so as to identify the current iris feature code in the registered iris feature code library.

[0048] While the embodiments disclosed in this invention are as described above, they are merely illustrative of the embodiments to facilitate understanding of the invention and are not intended to limit the invention. Any person skilled in the art to which this invention pertains may make any modifications and variations in the form and details of the implementation without departing from the spirit and scope disclosed herein; however, the scope of patent protection for this invention shall still be determined by the scope defined in the appended claims.

Claims

1. An iris recognition device for complex lighting conditions, characterized in that, include: The iris image acquisition unit includes an intelligent exposure control unit, an iris quality evaluation unit, an iris quality enhancement unit, and an iris feature encoding unit connected to the iris quality enhancement unit, as well as an iris feature comparison unit connected to the iris feature encoding unit; the iris quality enhancement unit is also connected to the iris quality evaluation unit. The iris image acquisition unit is used to acquire iris images of the user's eyes; The intelligent exposure control unit is used to adjust the exposure parameters based on the iris image acquired by the iris image acquisition unit and the correlation between pupil size and ambient light, so as to obtain an image exposure effect suitable for the current ambient light. The iris quality evaluation unit is used to evaluate the quality of iris images using multiple quality measures, filter out iris images that fail the quality evaluation, and obtain multi-measure quality evaluation results for iris images that pass the quality evaluation. The iris quality enhancement unit is used to enhance the quality of iris images with quality below a preset threshold based on the multi-measure quality evaluation results of the iris quality evaluation unit, so as to improve their recognizability. The iris feature coding unit is used to represent the quality-compensated iris image using iris feature coding. The iris feature comparison unit is used to calculate the similarity between different iris feature codes. By comparing the iris feature code to be identified with the registered iris feature code library, the iris feature code to be identified is identified in the registered iris feature code library.

2. The iris recognition device under complex lighting conditions according to claim 1, characterized in that, The iris image acquisition unit includes: a camera, a near-infrared fill light, and a narrowband filter; Among them, multiple near-infrared supplementary lights with fixed wavelengths are evenly distributed around the infrared camera in the circumferential direction. The position of the near-infrared supplementary lights is such that when the user looks straight ahead, the reflected light spot formed by the light source falls on the user's pupil area. The narrowband filter is placed in front of the infrared camera to filter out interfering spectral energy in unwanted wavelengths.

3. The iris recognition device under complex lighting conditions according to claim 2, characterized in that, The intelligent exposure control unit is specifically used to calculate the pupil radius and iris radius based on the acquired iris image using an iris positioning algorithm, and to calculate the pupil-iris radius ratio. Then, based on the radius ratio, it controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit.

4. The iris recognition device under complex lighting conditions according to claim 3, characterized in that, The intelligent exposure control unit controls the iris image acquisition unit in the following ways: The control method for the near-infrared supplementary light is as follows: when the ratio of the pupil to the iris radius is greater than the first ratio threshold, it indicates that the ambient light is relatively dark, so the light intensity of the near-infrared supplementary light is increased; when the ratio of the pupil to the iris radius is less than the second ratio threshold, it indicates that the ambient light is sufficient, so the light intensity of the near-infrared supplementary light is reduced; wherein, the first ratio threshold is greater than the second ratio threshold. The method for controlling the exposure time is as follows: when the ratio of the pupil to the iris radius increases, it indicates that the ambient light is in the process of darkening, so the exposure time is extended to allow more light to enter the camera; when the ratio of the pupil to the iris radius decreases, it indicates that the ambient light is in the process of brightening, so the exposure time is reduced to suppress strong light interference. The exposure gain is controlled as follows: when the pupil-iris radius ratio increases, it indicates a low-light environment, so the exposure gain is increased; when the pupil-iris radius ratio decreases, it indicates a well-lit environment, so the gain is reduced to obtain a cleaner image.

5. An iris recognition method under complex lighting conditions, characterized in that, The iris recognition method is performed using an iris recognition device under complex lighting conditions as described in any one of claims 1 to 4, comprising: Step 1: Acquire iris images of the user's eyes using the iris image acquisition unit; Step 2: Based on the acquired iris image, the intelligent exposure control unit adjusts the light intensity and exposure parameters of the iris image acquisition unit according to the correlation between pupil size and ambient light, thereby obtaining a more suitable image exposure effect for the current ambient light. Step 3: The iris quality evaluation unit evaluates the quality of the iris image using multiple quality measures, filters out iris images that fail the quality evaluation, and obtains multi-measure quality evaluation results for iris images that pass the quality evaluation. Step 4: Based on the multi-measure quality evaluation results of the iris image by the iris quality evaluation unit, the iris quality enhancement unit enhances the quality of iris images with quality below a preset threshold to improve their recognizability. Step 5: Encode the iris features of all iris images using the iris feature coding unit; Step 6: The iris feature code to be identified is compared with the registered iris feature code library by the iris feature comparison unit, and the iris feature code to be identified is identified in the registered iris feature code library.

6. The iris recognition method under complex lighting conditions according to claim 5, characterized in that, Step 2 includes: The intelligent exposure control unit calculates the pupil radius and iris radius using an iris positioning algorithm based on the acquired iris image, and calculates the pupil-iris radius ratio. Then, it controls the light intensity, exposure time, and exposure gain of the near-infrared supplementary light in the iris image acquisition unit according to the radius ratio.

7. The iris recognition method under complex lighting conditions according to claim 5, characterized in that, In step 3, the multiple quality measures include: brightness contrast detection, blurriness detection, integrity detection, and gaze direction detection; and the detection methods for each quality measure are as follows: The brightness and contrast detection method is as follows: calculate the iris brightness, the contrast between the iris and sclera, and the contrast between the iris and the pupil respectively, and compare each calculated value with the corresponding preset threshold. The ambiguity detection method is as follows: calculate the aggregation of the reflected light spot of the near-infrared supplementary light lamp and the intensity of the iris texture, and compare each calculated value with the corresponding preset threshold. The integrity detection method is as follows: the integrity of the iris is calculated based on the proportion of the iris obscured by the eyelid, and then compared with a preset threshold. The gaze direction detection method is as follows: the gaze direction is estimated based on the positional relationship between the pupil and the reflected light spot of the near-infrared supplementary light, and then compared with a preset threshold.

8. The iris recognition method under complex lighting conditions according to claim 5, characterized in that, The quality enhancement of the iris image in step 4 includes: brightness and contrast enhancement, and iris texture detail enhancement. The brightness contrast enhancement method is as follows: for iris images with a brightness contrast score lower than a preset threshold, their brightness contrast is enhanced based on their brightness contrast evaluation results. The method for enhancing iris texture details is as follows: for iris images with blurriness higher than a preset threshold, the texture details of the iris region are enhanced by a texture detail enhancement algorithm based on the blurriness score.

9. The iris recognition method under complex lighting conditions according to claim 5, characterized in that, In the iris feature encoding of step 5, To address the issue of a large pupil dilation / contraction range under complex lighting conditions, a multi-scale filter is used to calculate the iris feature code.

Images