Iris boundary positioning method and device, medium and equipment

Abstract

The invention discloses an iris boundary positioning method and device, a medium and equipment, and belongs to the field of biological feature recognition. The method comprises the following steps: acquiring an iris image to be positioned; inputting the iris image into a pre-trained deep learning model, segmenting the iris image by taking the iris inner boundary region and the iris outer boundary region as foreground targets and taking the rest parts as background regions, and segmenting iris inner boundary points and iris outer boundary points from the iris image; and respectively performing ellipse fitting according to the iris inner boundary point and the iris outer boundary point to obtain an iris inner boundary and an iris outer boundary. The method has good generalization and robustness for different changes, and the iris image positioning problem under the conditions of strabismus, squinting, low contrast and the like can be well solved. And through ellipse fitting, the iris inner boundary and the iris outer boundary are more regular, and subsequent boundary correction and iris region expansion are facilitated.

Description

Technical Field

[0001] The present application relates to the field of biometric identification, and in particular to an iris boundary positioning method, device, medium and equipment. Background Art

[0002] Iris recognition is a widely used biometric identification technology with the characteristics of accuracy, stability, security and reliability. A complete iris identity recognition system mainly includes iris image acquisition, iris image quality assessment, iris image preprocessing, iris image feature extraction and iris feature comparison. Iris image preprocessing includes the positioning of the inner and outer boundaries of the iris, iris noise detection and iris image normalization.

[0003] The inner boundary of the iris is the boundary between the pupil and the iris. The pupil is darker in color and generally has a smaller grayscale value. The inner boundary is more obvious and is generally not affected by noise such as eyelashes, so the positioning is more accurate. The outer boundary positioning of the iris is to determine the outer boundary of the iris. The outer boundary is the boundary between the iris and the sclera. The sclera is lighter in color and generally has a larger grayscale value, and the difference from the grayscale value of the iris is not very obvious. In some iris images, the upper and lower parts of the iris are covered by eyelids and eyelashes. Before positioning, it is necessary to remove the noise that affects the positioning result (including eyelids, eyelashes, light spots, etc.). Compared with the inner boundary, the outer boundary positioning is more complicated.

[0004] Traditional methods for locating the inner and outer boundaries of the iris are generally based on the gradient of the image. The inner and outer boundaries of the iris are determined by calculating the gradient of the image, but their generalization is poor. Moreover, for strabismus and blurred iris images, the inner and outer boundaries of the iris cannot be accurately located, and image quality needs to be controlled. Summary of the invention

[0005] To address the deficiencies of the prior art, the present application provides an iris boundary positioning method, device, medium and equipment, which have high accuracy, generalization and robustness, and the iris boundary is more neat.

[0006] This application provides the following technical solutions:

[0007] In a first aspect, the present application provides an iris boundary positioning method, the method comprising:

[0008] Acquire an iris image to be located;

[0009] Inputting the iris image into a pre-trained deep learning model, segmenting the iris image by taking the inner boundary area of ​​the iris and the outer boundary area of ​​the iris as foreground targets and the rest as background areas, and segmenting the inner boundary points of the iris and the outer boundary points of the iris from the iris image;

[0010] Ellipse fitting is performed according to the iris inner boundary points and the iris outer boundary points respectively to obtain the iris inner boundary and the iris outer boundary.

[0011] Furthermore, the iris image is input into a pre-trained deep learning model, the inner boundary area of ​​the iris and the outer boundary area of ​​the iris are used as foreground targets, and the rest of the iris is used as the background area to segment the iris image, and the inner boundary points of the iris and the outer boundary points of the iris are segmented from the iris image, including:

[0012] Inputting the iris image into the deep learning model, encoding the iris image, and using deconvolution to expand the encoded feature map to the same size as the iris image, to obtain a first probability feature map and a second probability feature map;

[0013] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris;

[0014] The value of each pixel point on the first probability feature map / the second probability feature map is compared with the set segmentation threshold. If the value of the pixel point on the first probability feature map / the second probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

[0015] Furthermore, there are multiple deep learning models, and each deep learning model is trained by a different training data set.

[0016] The iris image is input into a pre-trained deep learning model, the inner boundary area of ​​the iris and the outer boundary area of ​​the iris are used as foreground targets, and the rest of the iris is used as the background area to segment the iris image, and the inner boundary points of the iris and the outer boundary points of the iris are segmented from the iris image, including:

[0017] Inputting the iris image into the multiple deep learning models respectively, encoding the iris image, and using deconvolution to expand the encoded feature map to the same size as the iris image, to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model;

[0018] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris;

[0019] Calculate the average value of all the first probability feature maps / the second probability feature maps at the same position to obtain a first average probability feature map / a second average probability feature map;

[0020] The value of each pixel point on the first average probability feature map / the second average probability feature map is compared with the set segmentation threshold; if the value of the pixel point on the first average probability feature map / the second average probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

[0021] Furthermore, the calculating the average value of all the first probability feature maps / the second probability feature maps at the same position also includes:

[0022] Performing a transformation operation on the iris image to obtain a transformed image; wherein the transformation operation includes image rotation and / or image flipping;

[0023] The transformed images are respectively input into the multiple deep learning models, the iris image is encoded, the encoded feature map is enlarged to the same size as the iris image using deconvolution, and the enlarged feature map is subjected to an inverse transformation operation opposite to the transformation operation to obtain a first probability feature map and a second probability feature map having the same number as the deep learning model.

[0024] Furthermore, in the process of encoding the iris image and using deconvolution to expand the encoded feature map to the same size as the iris image, the intermediate feature layers are connected in the dimension of channels.

[0025] Furthermore, the step of performing ellipse fitting according to the inner boundary points of the iris and the outer boundary points of the iris respectively further includes:

[0026] Find the maximum connected area for the iris inner boundary point / the iris outer boundary point, and use the maximum connected area as the final iris inner boundary point / the iris outer boundary point.

[0027] Furthermore, during training, the deep learning model randomly performs data enhancement on image samples in the training data set, and / or the image samples in the training data set include oblique images and / or blurred images;

[0028] Wherein, the data enhancement includes one or more of flipping, translation, noise transformation and elastic deformation;

[0029] The elastic deformation includes: dividing the image sample into sub-blocks of a set size, and performing shape distortion on each sub-block.

[0030] Furthermore, the step of inputting the iris image into a pre-trained deep learning model also includes:

[0031] Determine whether the grayscale mean of the iris image is less than a set first grayscale threshold or greater than a set second grayscale threshold, and if so, perform histogram equalization on the iris image;

[0032] The first grayscale threshold is smaller than the second grayscale threshold.

[0033] In a second aspect, the present application provides an iris boundary positioning device, the device comprising:

[0034] An image acquisition module, for acquiring an iris image to be located;

[0035] A segmentation module is used to input the iris image into a pre-trained deep learning model, segment the iris image by taking the inner boundary area and the outer boundary area of ​​the iris as foreground targets and the rest as background areas, and segment the inner boundary points of the iris and the outer boundary points of the iris from the iris image;

[0036] The fitting module is used to perform ellipse fitting according to the iris inner boundary points and the iris outer boundary points respectively to obtain the iris inner boundary and the iris outer boundary.

[0037] Furthermore, the segmentation module includes:

[0038] A first processing unit, configured to input the iris image into the deep learning model, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image to obtain a first probability feature map and a second probability feature map;

[0039] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris;

[0040] a first judgment unit, configured to compare the value of each pixel point on the first probability feature map / the second probability feature map with a set segmentation threshold; if the value of the pixel point on the first probability feature map / the second probability feature map is greater than the segmentation threshold, the pixel point at the corresponding same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

[0041] Furthermore, there are multiple deep learning models, and each deep learning model is trained by a different training data set.

[0042] The segmentation module comprises:

[0043] A second processing unit is used to input the iris image into the multiple deep learning models respectively, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image, so as to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model;

[0044] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris;

[0045] an averaging unit, used to calculate an average value of all the first probability feature maps / the second probability feature maps at the same position to obtain a first average probability feature map / a second average probability feature map;

[0046] a second judgment unit, configured to compare the value of each pixel point on the first average probability feature map / the second average probability feature map with a set segmentation threshold; if the value of the pixel point on the first average probability feature map / the second average probability feature map is greater than the segmentation threshold, the pixel point at the corresponding same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

[0047] Furthermore, the segmentation module also includes:

[0048] A transformation unit, configured to perform a transformation operation on the iris image to obtain a transformed image; wherein the transformation operation includes image rotation and / or image flipping;

[0049] The third processing unit is used to input the transformed images into the multiple deep learning models respectively, encode the iris image, use deconvolution to expand the encoded feature map to the same size as the iris image, and perform an inverse transformation operation on the expanded feature map that is opposite to the transformation operation to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model.

[0050] Furthermore, during the execution of the first processing unit, the second processing unit and the third processing unit, the intermediate feature layers are connected in the dimension of the channel.

[0051] Furthermore, the device also includes:

[0052] The post-processing module is used to find the maximum connected area for the iris inner boundary point / the iris outer boundary point, and use the maximum connected area as the final iris inner boundary point / the iris outer boundary point.

[0053] Furthermore, during training, the deep learning model randomly performs data enhancement on image samples in the training data set, and / or the image samples in the training data set include oblique images and / or blurred images;

[0054] Wherein, the data enhancement includes one or more of flipping, translation, noise transformation and elastic deformation;

[0055] The elastic deformation includes: dividing the image sample into sub-blocks of a set size, and performing shape distortion on each sub-block.

[0056] Furthermore, the device also includes:

[0057] The preprocessing module is used to perform scale transformation on the iris image and normalize it to a set size.

[0058] Furthermore, the device also includes:

[0059] A histogram equalization module, used to determine whether the grayscale mean of the iris image is less than a set first grayscale threshold or greater than a set second grayscale threshold, and if so, perform histogram equalization on the iris image;

[0060] The first grayscale threshold is smaller than the second grayscale threshold.

[0061] In a third aspect, the present application provides a computer-readable storage medium for iris boundary positioning, comprising a memory for storing processor-executable instructions, wherein the instructions, when executed by the processor, implement the steps of the iris boundary positioning method described in the first aspect.

[0062] In a fourth aspect, the present application provides a device for iris boundary positioning, characterized in that it includes at least one processor and a memory storing computer executable instructions, and when the processor executes the instructions, the steps of the iris boundary positioning method described in the first aspect are implemented.

[0063] This application has the following beneficial effects:

[0064] This application does not adopt the method of determining the inner and outer boundaries of the iris by calculating the gradient of the image in the prior art, but instead uses a method of instance segmentation using a deep learning model, learns from experience in diverse samples, and uses a deep learning segmentation model to directly locate the inner and outer boundary points of the iris. This application uses the inner and outer boundary areas of the iris image as foreground targets, and the rest as background areas for segmentation, thereby obtaining a series of inner and outer boundary points of the iris. Then, ellipse fitting is performed based on the inner and outer boundary points of the iris, respectively, to obtain the inner and outer boundaries of the iris.

[0065] Since the deep learning model is learned from a large amount of iris data and learns experience from diverse samples, it is not an empirical expert system used in existing technologies. Therefore, it has good generalization and robustness to different changes, and performs well under different changes in lighting, angles, etc. It can well solve the problem of iris image positioning in cases of strabismus and squinting. And through ellipse fitting, the inner and outer boundaries of the iris are made more neat, which is conducive to subsequent boundary correction and iris area expansion. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 This is a flow chart of the iris boundary positioning method of the present application;

[0067] Figure 2 is a schematic diagram of an iris image before preprocessing;

[0068] Figure 3 is a schematic diagram of a preprocessed iris image;

[0069] Figure 4 Schematic diagram of iris images of various qualities;

[0070] Figure 5 To use multiple deep learning models Figure 4 Schematic diagram of the result of locating the inner and outer boundaries of the iris;

[0071] Figure 6 is a schematic diagram of an iris image before histogram equalization;

[0072] Figure 7 is a schematic diagram of an iris image after histogram equalization;

[0073] Figure 8 The schematic diagram is the result of locating the inner and outer boundaries of the iris after the histogram equalization of the iris image;

[0074] Fig. 9 Schematic diagram of the iris boundary positioning device of the present application. DETAILED DESCRIPTION

[0075] To make the technical problems, technical solutions, and advantages to be solved by this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. The components of the embodiments of this application described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of this application to be protected, but only represents the selected embodiments of this application. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of this application.

[0076] An embodiment of this application provides an iris boundary localization method. As Figure 1 shown, this method includes:

[0077] S100: Obtain an iris image to be localized.

[0078] The described iris image can be collected in real time by an iris collection device, or an iris image collected and stored in advance can be read from a storage space.

[0079] After obtaining the iris image, it can be preprocessed. Through preprocessing, scale transformation is performed on the iris image and it is normalized to a set size to adapt to the input size of the deep learning model. Exemplarily, the set size can be 512 pixels in both length and width. A specific implementation process of the preprocessing is as follows:

[0080] Judge whether the width (w) and height (h) of the iris image are equal. If w = h, then scale the iris image to a size of 512 pixels in both length and width.

[0081] If w > h, then calculate the row supplementary boundary height Supplement black boundaries with a width of w and a height of pad1 and a pixel value of 0 to the upper and lower boundaries of the iris image respectively. The width of the supplemented iris image is w, and the height is h or h - 2. Then, scale the supplemented iris image to a size of 512 pixels in both length and width.

[0082] If w < h, then calculate the row supplementary boundary width Supplement black boundaries with a width of pad1 and a height of h and a pixel value of 0 to the left and right boundaries of the iris image respectively. The width of the supplemented iris image is w or w - 2, and the height is h. Then, scale the supplemented iris image to a size of 512 pixels in both length and width.

[0083] In one example, the iris image before preprocessing is as Figure 2As shown in Figure 2, the preprocessed iris image is Figure 3 shown.

[0084] S200: Input the iris image into a pre-trained deep learning model, segment the iris image by taking the inner boundary area and the outer boundary area of ​​the iris as foreground targets and the rest as background areas, and segment the iris inner boundary points and the outer boundary points of the iris from the iris image.

[0085] This application solves the problem of iris boundary positioning in iris recognition through a deep learning method. Using an instance segmentation solution, the inner and outer boundary areas of the iris image are taken as foreground targets, and the rest are the background areas. They are segmented to obtain a series of iris inner and outer boundary points.

[0086] S300: performing ellipse fitting according to the iris inner boundary points and the iris outer boundary points respectively to obtain the iris inner boundary and the iris outer boundary.

[0087] The present application assumes that the inner and outer boundaries of the iris are ellipses, and fits them as ellipses. The ellipses also include perfect circles, because perfect circles can also be regarded as a special ellipse in terms of expression.

[0088] The general equation of an ellipse is:

[0089] Ax 2 +Bxy+Cx 2 +Dx+Ey+F=0

[0090] Among them, A, B, C, D, E, and F are parameters to be determined. Substituting the inner boundary point / outer boundary point of the iris into the above equation, the expression of the ellipse can be obtained, and the inner boundary / outer boundary of the iris can be obtained.

[0091] It should be noted that at least 5 points are required to obtain the above equation. Therefore, if the number of iris inner boundary points / iris outer boundary points is less than 5, the result without segmentation is directly returned. If the number of points is greater than or equal to 5 points, the above equation is fitted to obtain five parameters. The ellipse described by the equation is the final iris inner boundary / iris outer boundary.

[0092] This application does not adopt the method of determining the inner and outer boundaries of the iris by calculating the gradient of the image in the prior art, but instead uses a method of instance segmentation using a deep learning model, learns from experience in diverse samples, and uses a deep learning segmentation model to directly locate the inner and outer boundary points of the iris. This application uses the inner and outer boundary areas of the iris image as foreground targets, and the rest as background areas for segmentation, thereby obtaining a series of inner and outer boundary points of the iris. Then, ellipse fitting is performed based on the inner and outer boundary points of the iris, respectively, to obtain the inner and outer boundaries of the iris.

[0093] Since the deep learning model is learned from a large amount of iris data and learns experience from diverse samples, it is not an empirical expert system used in existing technologies. Therefore, it has good generalization and robustness to different changes, and performs well under different changes in lighting, angles, etc. It can well solve the problem of iris image positioning in cases of strabismus and squinting. And through ellipse fitting, the inner and outer boundaries of the iris are made more neat, which is conducive to subsequent boundary correction and iris area expansion.

[0094] In the prior art, the inner and outer boundaries of the iris are determined by calculating the gradient of the image. When the contrast of the iris image is low, the gradient of the image is not obvious, so it is difficult to locate. The present application can effectively solve the problem of locating the inner and outer boundaries of the iris in images with low contrast.

[0095] Traditional iris recognition usually has many limitations in terms of user cooperation and image conditions, which seriously affect the widespread application of iris recognition. In recent years, many non-cooperative or low-restriction iris recognition methods have also emerged, such as long-distance iris recognition, iris recognition in motion, iris recognition without perception and cooperation, iris recognition under dynamic image environment changes, and iris recognition on mobile devices. The irises collected under these environments usually contain more complex and difficult-to-locate images such as strabismus and squinting. The inner and outer boundary positioning results of the iris image are more important for the expansion of the iris area and will directly affect the final recognition effect.

[0096] To address the above issues, this application proposes an iris boundary localization method based on a deep learning model, which improves the generalization ability of iris inner and outer boundary localization and can be used to segment visible iris boundaries in complex and difficult-to-locate iris images such as strabismus and squinting.

[0097] The deep learning model used in this application may be a convolutional neural network. Accordingly, the aforementioned S200 includes:

[0098] S210: Input the iris image into the deep learning model, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image to obtain a first probability feature map and a second probability feature map.

[0099] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is an iris inner boundary point / iris outer boundary point.

[0100] In order to better combine semantic features and spatial features, in this step, the intermediate feature layers can be connected in the channel dimension to perform feature enhancement.

[0101] In one example, the specific implementation of S210 of the present application includes:

[0102] The deep learning model is fed with an iris image of size 512x512 and 3 channels.

[0103] For the above image, a convolution operation with a convolution kernel of size 7x7, a stride of 2, and 3 pixels of edge padding is used, and BN and ReLU operations are performed to obtain a 64-channel feature map f0 of size 256x256.

[0104] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 2, and 1 pixel padding on the edge to obtain a feature map A1 of size 128x128 and 64 channels.

[0105] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 64-channel feature map of size 128x128.

[0106] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a 64-channel feature map B1 of size 128x128.

[0107] Add feature map A1 and feature map B1 pixel by pixel on each channel to obtain a 64-channel feature map f1 of size 128x128.

[0108] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 64-channel feature map of size 128x128.

[0109] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a 64-channel feature map B2 of size 128x128.

[0110] The feature map f1 and the feature map B2 are added pixel by pixel on each channel to obtain a 64-channel feature map f2 of size 128x128.

[0111] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 64-channel feature map of size 128x128.

[0112] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a 64-channel feature map B3 of size 128x128.

[0113] The feature map f2 and the feature map B3 are added pixel by pixel on each channel to obtain a 64-channel feature map f3 of size 128x128.

[0114] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a stride of 2, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 64x64 with 128 channels.

[0115] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a feature map B4 of size 64x64 and 128 channels.

[0116] A convolution operation with a convolution kernel of size 1x1, a stride of 1, and 0 pixel padding on the edge is performed on the feature map f3 to obtain a feature map A3 of size 64x64 and 128 channels.

[0117] The feature map A3 and the feature map B4 are added pixel by pixel on each channel to obtain a 128-channel feature map f4 of size 64x64.

[0118] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 64x64 with 128 channels.

[0119] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a feature map B5 of size 64x64 and 128 channels.

[0120] The feature map f4 and the feature map B5 are added pixel by pixel on each channel to obtain a 128-channel feature map f5 of size 64x64.

[0121] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 64x64 with 128 channels.

[0122] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a feature map B6 of size 64x64 and 128 channels.

[0123] The feature map f5 and the feature map B6 are added pixel by pixel on each channel to obtain a 128-channel feature map f7 of size 64x64.

[0124] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 64x64 with 128 channels.

[0125] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a feature map B7 of size 64x64 and 128 channels.

[0126] The feature map f7 and the feature map B7 are added pixel by pixel on each channel to obtain a 128-channel feature map f8 of size 64x64.

[0127] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 2, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0128] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and a BN operation is performed to obtain a feature map B8 of size 32x32 and 256 channels.

[0129] A convolution operation with a convolution kernel of size 1x1, a stride of 1, and 0 pixel padding on the edge is performed on the feature map f8 to obtain a feature map A8 of size 32x32 and 256 channels.

[0130] The feature map A8 and the feature map B8 are added pixel by pixel on each channel to obtain a feature map f9 of 256 channels with a size of 32x32.

[0131] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0132] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B9 of size 32x32 and 256 channels.

[0133] Add feature map f9 and feature map B9 pixel by pixel on each channel to obtain a feature map f with 256 channels of size 32x32. 10 .

[0134] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0135] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 32x32 with 256 channels. 10 .

[0136] The feature map f 10 And feature map B 10 Add pixel by pixel on each channel to get a 256-channel feature map f with a size of 32x32. 11 .

[0137] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0138] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 32x32 with 256 channels. 11 .

[0139] The feature map f 11 And feature map B 11 Add pixel by pixel on each channel to get a 256-channel feature map f with a size of 32x32. 12 .

[0140] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0141] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 32x32 with 256 channels. 12 .

[0142] The feature map f 12 And feature map B 12 Add pixel by pixel on each channel to get a 256-channel feature map f with a size of 32x32. 13 .

[0143] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0144] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 32x32 with 256 channels. 13 .

[0145] The feature map f 13 And feature map B 13 Add pixel by pixel on each channel to get a 256-channel feature map f with a size of 32x32. 14 .

[0146] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 2, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 16x16 with 512 channels.

[0147] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 16x16 with 512 channels. 14 .

[0148] For the feature map f 14 Using a convolution kernel of size 1x1, a stride of 1, and a convolution operation with 0 pixels padded on the edges, we get a feature map A of size 16x16 with 512 channels. 14 .

[0149] The feature map A 14 And feature map B 14 Add pixel by pixel on each channel to get a 512-channel feature map f with a size of 16x16. 15 .

[0150] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge, and BN and ReLU operations are performed to obtain a feature map of size 16x16 with 512 channels.

[0151] The above feature map is convolved with a convolution kernel of size 3x3, a step size of 1, and a convolution operation with 1 pixel padding on the edge, and a BN operation is performed to obtain a feature map B of size 16x16 with 512 channels. 15 .

[0152] The feature map f 15 And feature map B 15 Add pixel by pixel on each channel to get a 512-channel feature map f with a size of 16x16. 16 .

[0153] At this point, the process of encoding the iris image is completed. The subsequent process is to use deconvolution to expand the encoded feature map to the same size as the iris image to obtain the first / second probability feature map.

[0154] The feature map f 16 Use the nearest neighbor interpolation to perform 2x interpolation to obtain a 512-channel feature map f with a size of 32x32. 17 .

[0155] f 17 and f 14 Connect in the channel dimension to get a feature map f with 768 channels of size 32x32 18 .

[0156] f 18 A convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge is performed, and BN and ReLU operations are performed to obtain a feature map of size 32x32 with 256 channels.

[0157] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 256-channel feature map f of size 32x32. 19 .

[0158] The feature map f 19 Use the nearest neighbor interpolation to perform 2x interpolation to obtain a 256-channel feature map f with a size of 64x64. 20 .

[0159] f 20 Connect f8 with f in the channel dimension to obtain a feature map f with a size of 64x64 and 384 channels. 21 .

[0160] f 21 A convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge is performed, and BN and ReLU operations are performed to obtain a feature map of 128 channels with a size of 64x64.

[0161] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 128-channel feature map f with a size of 64x64. 22 .

[0162] The feature map f 22 Use the nearest neighbor interpolation to perform 2x interpolation to obtain a 128-channel feature map f with a size of 128x128. 23 .

[0163] f 23 Connect f3 with f in the channel dimension to obtain a 192-channel feature map f with a size of 128x128 24 .

[0164] f 24 A convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge is performed, and BN and ReLU operations are performed to obtain a 64-channel feature map of size 128x128.

[0165] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 64-channel feature map f with a size of 128x128. 25 .

[0166] The feature map f 25 Use the nearest neighbor interpolation to perform 2x interpolation to obtain a 128-channel feature map f with a size of 256x256. 26 .

[0167] f 26 Connect f0 with f0 in the channel dimension to obtain a 192-channel feature map f with a size of 256x256 27 .

[0168] f 27 A convolution operation with a convolution kernel of size 3x3, a stride of 1, and 1 pixel padding on the edge is performed, and BN and ReLU operations are performed to obtain a 64-channel feature map of size 256x256.

[0169] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 64-channel feature map f with a size of 256x256. 28 .

[0170] The feature map f 28 Use the nearest neighbor interpolation to perform 2x interpolation to obtain a 128-channel feature map f with a size of 512x512. 29 .

[0171] f 29 Use a convolution kernel of size 3x3, a stride of 1, a convolution operation with 1 pixel padding on the edge, and perform BN and ReLU operations to obtain a 16-channel feature map of size 512x512.

[0172] The above feature map is subjected to a convolution operation with a convolution kernel of size 3x3, a step size of 1, and a 1-pixel padding on the edge, and BN and ReLU operations are performed to obtain a 16-channel feature map f with a size of 512x512.30 .

[0173] f 30 Use a convolution kernel of size 3x3, a step size of 1, a convolution operation with 1 pixel padding on the edge, and perform BN and ReLU operations to obtain a single-channel feature map f of size 512x512. 31 .

[0174] f 30 Use a convolution kernel of size 3x3, a step size of 1, a convolution operation with 1 pixel padding on the edge, and perform BN and ReLU operations to obtain a single-channel feature map f of size 512x512. 32 .

[0175] Finally, for f 31 Using the softmax operation, we can get the first probability feature map P1 of size 512x512, which is used to estimate the inner boundary of the iris. The pixel value of each position on the first probability feature map P1 represents the probability that the pixel at the corresponding position of the original iris image is the inner boundary point of the iris.

[0176] f 32 Using the softmax operation, a second probability feature map P2 of size 512x512 can be obtained for estimating the outer boundary of the iris. The pixel value of each position on the second probability feature map P2 represents the probability that the pixel at the corresponding position of the original iris image is the outer boundary point of the iris.

[0177] S220: Compare the value of each pixel point on the first probability feature map / the second probability feature map with the set segmentation threshold. If the value of the pixel point on the first probability feature map / the second probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is determined to be an iris inner boundary point / iris outer boundary point.

[0178] Taking the first probability feature map as an example, assuming that 0.5 is the segmentation threshold, if the value p(x,y) of the pixel point at any position (x,y) of the first probability feature map P1 is greater than 0.5, it indicates that the pixel point at the position (x,y) on the iris image is the inner boundary point of the iris. The same is true for the second probability feature map P2.

[0179] In different iris recognition systems, the quality of the collected iris images varies greatly (e.g. Figure 4As shown, including iris images of various qualities), a positioning and segmentation algorithm with good generalization and strong robustness is extremely important. This application uses a multi-model integration method to set the deep learning model to multiple (for example, 5) and determines the iris inner boundary point / iris outer boundary point by the average value of the first probability feature map / second probability feature map output by multiple deep learning models, so that the positioning result is more accurate. Accordingly, the S200 includes:

[0180] S230: Input the iris image into multiple deep learning models respectively, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image, so as to obtain the same number of first probability feature maps and second probability feature maps as the deep learning models.

[0181] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is an iris inner boundary point / iris outer boundary point.

[0182] Each of the multiple deep learning models is trained using a different training data set. The different training data sets may be different independent data sets established, or may be a total data set divided into multiple different training data sets.

[0183] Assuming that the number of deep learning models is 5, the same iris image is processed by 5 deep learning models respectively, and 5 first probability feature maps / second probability feature maps can be obtained.

[0184] S240: Calculate the average value of all first probability feature maps / second probability feature maps at the same position to obtain a first average probability feature map / second average probability feature map.

[0185] S250: Compare the value of each pixel point on the first average probability feature map / the second average probability feature map with the set segmentation threshold. If the value of the pixel point on the first average probability feature map / the second average probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is determined to be an iris inner boundary point / iris outer boundary point.

[0186] The above-mentioned multiple deep learning models of this application are used to Figure 4 The iris images of various qualities are used to locate the inner and outer boundaries of the iris, and the results are as follows Figure 5 As shown, it can be seen that the result is relatively accurate.

[0187] In order to further improve the accuracy of the iris positioning result, the present application adopts a data enhancement operation during the processing of the iris image by the deep learning model (i.e., during the test process). Specifically, the S200 further includes:

[0188] S231: Performing a transformation operation on the iris image to obtain a transformed image.

[0189] The transformation operation includes image rotation and / or image flipping. In one example, the transformation operation is upside down flipping.

[0190] S232: Input the transformed image into multiple deep learning models respectively, encode the iris image, use deconvolution to expand the encoded feature map to the same size as the iris image, and perform an inverse transformation operation on the expanded feature map that is opposite to the transformation operation to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model.

[0191] Assuming that the number of deep learning models is 5, the same transformed image is processed by 5 deep learning models respectively, and 5 first probability feature maps / second probability feature maps can be obtained. Since the input is a transformed image, the output feature map needs to be inversely transformed to correspond to the original iris image.

[0192] The five first probability feature maps of the iris image after being processed by the five deep learning models are recorded as Inner i (i=0,1,2,3,4), the 5 second probability feature maps are denoted as Outer i (i=0,1,2,3,4). The five first probability feature maps after the transformed image is processed by the five deep learning models are recorded as Inner i (i=5,6,7,8,9), the 5 second probability feature maps are denoted as Outer i (i=5,6,7,8,9).

[0193] Next, in S240, the average values ​​are calculated to obtain a first average probability feature graph Inner and a second average probability feature graph Outer.

[0194]

[0195]

[0196] After the iris inner boundary point / iris outer boundary point is determined by the first / second probability feature map or the first / second average probability feature map, the iris inner boundary point / iris outer boundary point can be directly used for fitting. Of course, the iris inner boundary point / iris outer boundary point can also be post-processed, and the post-processing method includes:

[0197] S260: Find the largest connected area for the iris inner boundary point / iris outer boundary point, and use the largest connected area as the final iris inner boundary point / iris outer boundary point.

[0198] By finding the largest connected area, the disconnected outliers can be removed and the segmentation result with the largest connected area can be retained. In this application, for each pixel point segmented, the 4-neighborhood maximum connected area method can be used to find the largest connected area, and the pixel point of the largest connected area is taken as the final result.

[0199] In the positioning of the inner and outer boundaries of the iris, it is more difficult to segment images that are flipped upside down, squinted, and squint. To address the above problems, the deep learning model of the present application randomly performs data enhancement on image samples in the training data set during training.

[0200] Specifically, the image samples in the training data set are randomly flipped, translated, and noise transformed to make the data diverse. In addition, the elastic deformation method can be used to divide the image samples into sub-blocks of a set size (for example, 10*10 size), and each sub-block is distorted. The deep learning model trained by the above data enhancement method can well solve the segmentation problem of iris images such as strabismus and squinting, which helps to improve the generalization and accuracy of the deep learning model.

[0201] In the prior art, it is not possible to locate the iris boundary well for strabismus and blurred iris images. To solve this problem, the present application makes the image samples in the training data set include strabismus images and / or blurred images, which can better solve the segmentation problem of strabismus and blurred iris images and have more accurate segmentation results.

[0202] In the process of iris image segmentation, there are cases where the iris image with weak illumination or overexposure cannot be segmented or the segmentation result is incorrect, such as Figure 6 To solve the above problems, the method of the present application further includes:

[0203] S110: Determine whether the grayscale mean of the iris image is less than a set first grayscale threshold or greater than a set second grayscale threshold. If so, perform histogram equalization on the iris image.

[0204] The first grayscale threshold is smaller than the second grayscale threshold.

[0205] The present application determines the brightness of the iris image. For example, the value of the first grayscale threshold may be 50, and the value of the second grayscale threshold may be 200. When the grayscale mean of the iris image is less than 50 or greater than 200, the iris image needs to be histogram equalized. The iris image after histogram equalization can be better segmented.

[0206] The iris image before histogram equalization is as follows Figure 6 As shown in the figure, the iris image after histogram equalization is as follows Figure 7As shown in the figure, the result of segmentation of the iris image after histogram equalization is as follows Figure 8 As shown, it can be seen that the result is relatively accurate.

[0207] The present application also provides an iris boundary positioning device, such as Fig. 9 As shown, the device comprises:

[0208] The image acquisition module 100 acquires the iris image to be located.

[0209] The segmentation module 200 is used to input the iris image into a pre-trained deep learning model, use the inner boundary area of ​​the iris and the outer boundary area of ​​the iris as foreground targets, and the rest as background areas to segment the iris image, and segment the inner boundary points of the iris and the outer boundary points of the iris from the iris image.

[0210] The fitting module 300 is used to perform ellipse fitting according to the inner boundary points of the iris and the outer boundary points of the iris to obtain the inner boundary of the iris and the outer boundary of the iris.

[0211] This application does not adopt the method of determining the inner and outer boundaries of the iris by calculating the gradient of the image in the prior art, but instead uses a method of instance segmentation using a deep learning model, learns from experience in diverse samples, and uses a deep learning segmentation model to directly locate the inner and outer boundary points of the iris. This application uses the inner and outer boundary areas of the iris image as foreground targets, and the rest as background areas for segmentation, thereby obtaining a series of inner and outer boundary points of the iris. Then, ellipse fitting is performed based on the inner and outer boundary points of the iris, respectively, to obtain the inner and outer boundaries of the iris.

[0212] Since the deep learning model is learned from a large amount of iris data and learns experience from diverse samples, it is not an empirical expert system used in existing technologies. Therefore, it has good generalization and robustness to different changes, and performs well under different changes in lighting, angles, etc. It can well solve the problem of iris image positioning in cases of strabismus and squinting. And through ellipse fitting, the inner and outer boundaries of the iris are made more neat, which is conducive to subsequent boundary correction and iris area expansion.

[0213] The deep learning model used in this application can be a convolutional neural network. Accordingly, the segmentation module includes:

[0214] The first processing unit is used to input the iris image into the deep learning model, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image to obtain a first probability feature map and a second probability feature map.

[0215] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is an iris inner boundary point / iris outer boundary point.

[0216] The first judgment unit is used to compare the value of each pixel point on the first probability feature map / the second probability feature map with the set segmentation threshold. If the value of the pixel point of the first probability feature map / the second probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged as an iris inner boundary point / iris outer boundary point.

[0217] Furthermore, there are multiple deep learning models, each of which is trained using different training data sets. Accordingly, the segmentation module includes:

[0218] The second processing unit is used to input the iris image into multiple deep learning models respectively, encode the iris image, and use deconvolution to expand the encoded feature map to the same size as the iris image, so as to obtain the same number of first probability feature maps and second probability feature maps as the deep learning models.

[0219] The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is an iris inner boundary point / iris outer boundary point.

[0220] The averaging unit is used to calculate the average value of all the first probability feature maps / second probability feature maps at the same position to obtain the first average probability feature map / second average probability feature map.

[0221] The second judgment unit is used to compare the value of each pixel point on the first average probability feature map / the second average probability feature map with the set segmentation threshold. If the value of the pixel point of the first average probability feature map / the second average probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged as an iris inner boundary point / iris outer boundary point.

[0222] Furthermore, the segmentation module also includes:

[0223] The transformation unit is used to perform a transformation operation on the iris image to obtain a transformed image; wherein the transformation operation includes image rotation and / or image flipping.

[0224] The third processing unit is used to input the transformed images into multiple deep learning models respectively, encode the iris image, use deconvolution to expand the encoded feature map to the same size as the iris image, and perform an inverse transformation operation on the expanded feature map that is opposite to the transformation operation to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model.

[0225] During the execution of the first processing unit, the second processing unit and the third processing unit, the intermediate feature layers are connected in the dimension of the channel.

[0226] As an improvement of the embodiment of the present application, the device may further include:

[0227] The post-processing module is used to find the maximum connected area for the iris inner boundary point / iris outer boundary point, and use the maximum connected area as the final iris inner boundary point / iris outer boundary point.

[0228] During training, the deep learning model of the present application may randomly perform data enhancement on image samples in a training data set, and / or make the image samples in the training data set include oblique images and / or blurred images.

[0229] The data enhancement includes one or more of flipping, translation, noise transformation and elastic deformation.

[0230] The elastic deformation includes: dividing the image sample into sub-blocks of a set size, and performing shape distortion on each sub-block.

[0231] As another improvement of the embodiment of the present application, the device of the present application may further include:

[0232] The preprocessing module is used to scale the iris image and normalize it to a set size.

[0233] and / or;

[0234] The histogram equalization module is used to determine whether the grayscale mean of the iris image is less than a set first grayscale threshold or greater than a set second grayscale threshold. If so, histogram equalization is performed on the iris image.

[0235] The first grayscale threshold is smaller than the second grayscale threshold.

[0236] The device provided in the above embodiment, its implementation principle and the technical effect produced correspond to the embodiment of the above method one by one. For the sake of brief description, the part not mentioned in the embodiment of the device can refer to the corresponding content in the embodiment of the above method. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the modules and units described in the device can refer to the corresponding process in the embodiment of the above method, and will not be repeated here.

[0237] The iris boundary positioning method described in the above embodiments provided by the present application can implement the business logic through a computer program and record it on a storage medium, and the storage medium can be read and executed by a computer to achieve the effect of the solution described in the method embodiment of this specification. Therefore, the embodiment of the present application also provides a computer-readable storage medium for iris boundary positioning, including a memory for storing processor-executable instructions, and when the instructions are executed by the processor, the steps of the iris boundary positioning method including the above embodiments are implemented.

[0238] The storage medium may include a physical device for storing information, which is usually a medium that digitizes the information and then stores it in an electrical, magnetic or optical manner. The storage medium may include: a device that stores information in an electrical energy manner, such as various memories, such as RAM, ROM, etc.; a device that stores information in a magnetic energy manner, such as a hard disk, a floppy disk, a magnetic tape, a magnetic core memory, a magnetic bubble memory, a USB flash drive; a device that stores information in an optical manner, such as a CD or a DVD. Of course, there are other readable storage media, such as quantum memory, graphene memory, etc.

[0239] The storage medium described above may also include other implementation methods according to the description of the method embodiment. The implementation principle and technical effects produced by this embodiment are the same as those of the aforementioned method embodiment. For details, please refer to the description of the relevant method embodiment, and no further description will be given here.

[0240] The embodiment of the present application also provides a device for iris boundary positioning, which may be a separate computer, or may include an actual operating device using one or more methods or one or more embodiments of the present specification. The device for iris boundary positioning may include at least one processor and a memory storing computer executable instructions, and when the processor executes the instructions, it implements any one or more steps of the iris boundary positioning method described above.

[0241] The device described above may also include other implementation methods according to the description of the method embodiment. The implementation principle and technical effects produced by this embodiment are the same as those of the aforementioned method embodiment. For details, please refer to the description of the relevant method embodiment, and no further description will be given here.

[0242] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application. They should all be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for locating an iris boundary, characterized in that: The method comprises: Acquire an iris image to be located; Inputting the iris image into a pre-trained deep learning model, segmenting the iris image by taking the inner boundary area of ​​the iris and the outer boundary area of ​​the iris as foreground targets and the rest as background areas, and segmenting the inner boundary points of the iris and the outer boundary points of the iris from the iris image; Ellipse fitting is performed according to the iris inner boundary points and the iris outer boundary points respectively to obtain the iris inner boundary and the iris outer boundary.

2. The iris boundary positioning method according to claim 1, characterized in that: The iris image is input into a pre-trained deep learning model, the inner boundary area of ​​the iris and the outer boundary area of ​​the iris are used as foreground targets, and the rest of the iris is used as the background area to segment the iris image, and the inner boundary points of the iris and the outer boundary points of the iris are segmented from the iris image, including: Inputting the iris image into the deep learning model, encoding the iris image, and using deconvolution to expand the encoded feature map to the same size as the iris image, to obtain a first probability feature map and a second probability feature map; The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris; The value of each pixel point on the first probability feature map / the second probability feature map is compared with the set segmentation threshold. If the value of the pixel point on the first probability feature map / the second probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

3. The iris boundary positioning method according to claim 1, characterized in that: There are multiple deep learning models, and each deep learning model is trained using a different training data set. The iris image is input into a pre-trained deep learning model, the inner boundary area of ​​the iris and the outer boundary area of ​​the iris are used as foreground targets, and the rest of the iris is used as the background area to segment the iris image, and the inner boundary points of the iris and the outer boundary points of the iris are segmented from the iris image, including: Inputting the iris image into the multiple deep learning models respectively, encoding the iris image, and using deconvolution to expand the encoded feature map to the same size as the iris image, to obtain the same number of first probability feature maps and second probability feature maps as the deep learning model; The value of each pixel point on the first probability feature map / the second probability feature map represents the probability that the corresponding pixel point at the same position on the iris image is the inner boundary point of the iris / the outer boundary point of the iris; Calculate the average value of all the first probability feature maps / the second probability feature maps at the same position to obtain a first average probability feature map / a second average probability feature map; The value of each pixel point on the first average probability feature map / the second average probability feature map is compared with the set segmentation threshold; if the value of the pixel point on the first average probability feature map / the second average probability feature map is greater than the segmentation threshold, the corresponding pixel point at the same position on the iris image is judged to be the iris inner boundary point / the iris outer boundary point.

4. The iris boundary positioning method according to claim 3, characterized in that: The calculating of the average value of all the first probability feature maps / the second probability feature maps at the same position also includes: Performing a transformation operation on the iris image to obtain a transformed image; wherein the transformation operation includes image rotation and / or image flipping; The transformed images are respectively input into the multiple deep learning models, the iris image is encoded, the encoded feature map is enlarged to the same size as the iris image using deconvolution, and the enlarged feature map is subjected to an inverse transformation operation opposite to the transformation operation to obtain a first probability feature map and a second probability feature map having the same number as the deep learning model.

5. The iris boundary positioning method according to any one of claims 2 to 4, characterized in that: In the process of encoding the iris image and using deconvolution to expand the encoded feature map to the same size as the iris image, the intermediate feature layer is connected in the dimension of the channel.

6. The iris boundary positioning method according to any one of claims 1 to 4, characterized in that: The ellipse fitting is performed according to the inner boundary points of the iris and the outer boundary points of the iris respectively, and before that, it also includes: Find the maximum connected area for the iris inner boundary point / the iris outer boundary point, and use the maximum connected area as the final iris inner boundary point / the iris outer boundary point.

7. The iris boundary positioning method according to claim 1, characterized in that: When the deep learning model is being trained, data enhancement is randomly performed on image samples in the training data set, and / or the image samples in the training data set include oblique images and / or blurred images; Wherein, the data enhancement includes one or more of flipping, translation, noise transformation and elastic deformation; The elastic deformation includes: dividing the image sample into sub-blocks of a set size, and performing shape distortion on each sub-block.

8. The iris boundary positioning method according to claim 1, characterized in that: The step of inputting the iris image into a pre-trained deep learning model also includes: Determine whether the grayscale mean of the iris image is less than a set first grayscale threshold or greater than a set second grayscale threshold, and if so, perform histogram equalization on the iris image; The first grayscale threshold is smaller than the second grayscale threshold.

9. An iris boundary positioning device, characterized in that: The device comprises: An image acquisition module, for acquiring an iris image to be located; A segmentation module is used to input the iris image into a pre-trained deep learning model, segment the iris image by taking the inner boundary area and the outer boundary area of ​​the iris as foreground targets and the rest as background areas, and segment the inner boundary points of the iris and the outer boundary points of the iris from the iris image; The fitting module is used to perform ellipse fitting according to the iris inner boundary points and the iris outer boundary points respectively to obtain the iris inner boundary and the iris outer boundary.

10. A computer-readable storage medium for iris boundary location, characterized in that: The invention comprises a memory for storing processor executable instructions, wherein the instructions, when executed by the processor, implement the steps of the iris boundary positioning method according to any one of claims 1 to 8.

11. A device for iris boundary positioning, characterized in that: The method comprises at least one processor and a memory storing computer executable instructions, wherein the processor implements the steps of the iris boundary positioning method according to any one of claims 1 to 8 when executing the instructions.

Images