31 results about "Iris localization" patented technology

The invention is a biometric system based on iris texture analysis, which overcomes the defects in the prior art that lack the support of iris image quality analysis mechanism and cannot complete the evaluation of iris texture image quality. The invention acquires iris texture images, automatically locates iris, evaluates and analyzes the quality of iris texture images, automatically tracks and detects living iris, standardizes polar coordinate transformation of iris analysis area, demodulates and expresses iris texture features, and generates feature code IrisID by quantization coding. Encoding IrisID similarity measurement, similarity measurement judgment and other steps are realized. The invention realizes the basic characteristic requirements of an ideal system: automaticity, non-invasiveness, friendly man-machine operation interface and high-quality imaging image; no correlation between geometric affine transformation and nonlinear affine deformation; no dependence on external imaging conditions ; Random probability distribution coding; High-precision false acceptance rate and false rejection rate performance.

The invention provides an iris identifying system and a realizing method based on field programmable gate array (FPGA) which uses FPGA chip of Altera company as a main control chip, an iris collecting device, a LCD screen and a PS/2 keyboard compose a hardware platform with integrated system. The whole system can realize functions of collecting iris image, displaying iris image, evaluating iris image quality, recording iris object and identifying iris object, wherein, a novel, simple and rapid localization algorithm is provided in iris object recording and iris identifying processes. The iris characteristic extracting algorithm uses two-dimension Gabor filter algorithm for realizing perfect characteristic extraction. The accurate iris localization effect and special characteristic extraction algorithm realize the iris identifying system with high identifying rate and high reliability for authentication.

The invention provides a model generation method and device and an iris detection method and device, and the method comprises the steps: obtaining a training sample set, wherein the training sample set comprises a plurality of training sample subsets, each training sample in each training sample subset comprises an iris image, a type label and real iris bounding box position information, and the type of the training sample in each training sample subset comprises a real person iris, a beauty pupil iris and a prosthesis iris; training the iris detection neural network by using each training sample of each training sample subset in the training sample set to obtain an iris detection model, wherein the iris detection neural network comprises a feature extraction layer, an iris positioning layer, a region of interest pooling layer, a beauty pupil detection layer and a prosthesis detection layer, the feature extraction layer, the iris positioning layer, the region-of-interest pooling layerand the beauty pupil detection layer are connected in sequence, and the output of the feature extraction layer is further connected with the region-of-interest pooling layer and the prosthesis detection layer. Through the scheme, the iris image processing efficiency can be improved.

The invention discloses an iris positioning method and a computer readable storage medium, and the method comprises the steps: obtaining an infrared image which comprises a human face or human eyes; performing human eye detection on the infrared image to obtain a human eye detection area; performing pixel clustering on the human eye detection area according to a preset clustering number to obtaina clustering graph; performing Hough transform on the clustering graph according to a preset first radius range and a preset second radius range to obtain a first circle and a second circle; determining an annular area according to the first circle and the second circle; combining the clustering areas with the first N pixel points in the annular area to obtain a clustering combined area, and N isa preset number; and obtaining an iris area according to the human eye detection area and the clustering merging area. According to the invention, on the basis of ensuring the accuracy of Hough transform, the efficiency of Hough transform is improved, and the robustness of an iris positioning algorithm is enhanced.

An improved iris recognition illumination system is disclosed. The system includes (i) an RGB laser device that is associated with a first collimating optic and (ii) an IR illumination device that is associated with a second collimating optic. The system also includes a DMA that has a MEMS mirror system. The DMA optically combines IR light and RGB light to generate combined light. The combined light is then directed towards a user's iris via a transport medium (e.g., a waveguide). One or more photodetector(s) are positioned to capture reflected light that is reflected off of the user's iris. The photodetectors include an IR detector configured to detect reflected IR light off of the user's iris in order to perform iris recognition.

The invention discloses a face and iris image acquisition method and device, a readable storage medium and equipment, and belongs to the field of biological recognition. The method comprises the following steps: acquiring the area and position of a light spot on an iris image when a face lens and an iris lens are located at a collection position and the iris cannot be positioned in the iris image, performing overlay analysis on the face image and the iris image, judging whether the light spot shields the iris according to similar feature point information of the face image and the iris image, and if the area of the light spot is too large and the iris is shielded, adjusting the angle and the position of the light supplementing light source, so that the light spot avoids the iris area, and then a clear and available iris image can be acquired. According to the method, the problem of iris positioning failure caused by large light spot interference is avoided, the position of the iris can still be accurately judged when the light spot shields the iris, the light spot accurately avoids the iris area by adjusting the angle or the position of the light supplementing light source, and then a clear and available iris image is collected.

The invention discloses a multi-scene iris recognition method based on deep learning, and solves the technical problems that the robustness of a current iris recognition algorithm is poor, and the recognition capability of developing iris recognition under complex scenes and non-ideal imaging conditions is low. The recognition method comprises the following steps: shooting the whole eye of a person to obtain a to-be-recognized eye image; carrying out iris positioning segmentation to obtain an iris image to be identified; performing iris feature extraction to obtain a to-be-recognized feature matrix formed by iris features as elements; and carrying out iris feature matching, and if matching succeeds, passing identity recognition. In the iris positioning segmentation stage, jump connection is added and a UNet semantic segmentation model of cavity convolution is used to obtain a separated iris image, the robustness of iris positioning segmentation is improved, some tedious preprocessing steps needing a large number of experiments to determine parameters are omitted, meanwhile, upper and lower eyelids can be directly removed, two-step processing is not needed, and the generalization of the algorithm is high.

The invention discloses a small multi-rotor unmanned aerial vehicle camera control method based on an eyeball recognition tracking technology, and relates to the technical field of robot control. According to the invention, an eyeball identification tracking technology is used to assist an operator in controlling the small multi-rotor unmanned aerial vehicle camera; the method is characterized bycomprising the following steps: collecting an eyeball image of an operator, completing iris recognition and positioning, obtaining iris positioning information again in each control period to completeeyeball tracking, and completing the control of an unmanned aerial vehicle camera by using the difference between the real-time iris positioning information and a reference value thereof; according to the invention, an operator does not need to manually control the shooting direction of the camera when acquiring the image information shot by the camera of the small multi-rotor unmanned aerial vehicle, so the control variable is reduced, the operation difficulty is reduced, an independent person can independently control the small multi-rotor unmanned aerial vehicle, and the defects in the prior art are overcome.

The invention discloses a fatigue driving detection method based on multi-information fusion. The method comprises the steps: collecting a face picture of a driver, building a face three-dimensional model, and carrying out the identity verification of the driver; collecting a real-time video of a driver, extracting a current frame of human eye image for iris positioning, and judging the opening and closing states of eyes; judging whether the driver is in a fatigue state or not according to a number of continuous video frames of the face image of the driver, and updating the driving starting and ending time of the current driver; and determining the moving distance of the automobile in unit time through a GPS, and determining whether a central control platform sends an early warning signalto an alarm device installed in the automobile or not. The iris recognition and multi-information judgment fusion technology is adopted, more drivers can obtain fatigue driving prompts in time in thedriving process, and lives and property safety iare guaranteed.

The invention provides a computer-assisted iris comparison method and device, and the method comprises the steps: obtaining a first iris image and a second iris image; rotating the iris in the first iris image and the iris in the second iris image to the same angle; carrying out iris positioning on the rotated first iris image and second iris image and carrying out normalization operation, whereinthe normalization operation comprises one or more of iris radius normalization, pupil radius normalization and rectangular expansion normalization; calculating iris texture feature attribute information of corresponding positions in the first iris image and the second iris image after normalization operation; outputting the iris texture feature attribute information for display so as to obtain texture consistency comparison results of corresponding areas in the two iris images; and if the consistent iris texture features are compared and determined to meet the identity determination, determining that the first iris image and the second iris image are from the same iris. According to the scheme, the blank in the field of artificial iris comparison is filled.

The invention provides a class attending quality detection method based on face recognition. The method comprises the following steps: preliminarily obtaining students who raise the head in class through face recognition under a convolutional neural network; performing region division analysis on all students, dividing an image into three regions for research in a two-dimensional array form, finding the position of an iris in each region by using a face 68 key point extraction and detection method, and stipulating an iris position range according to region division to obtain students with highconcentration; meanwhile, the ratio of the number of students meeting the two conditions of class head raising and high concentration to the number of all students is the required class head raisingrate, so that the class quality is detected at the completion rate; the face recognition and iris positioning method under the CNN model is utilized to solve the problems that students raise their heads but do not listen to the class and the face cannot be recognized due to different positions of the students, and the class quality is detected more accurately.

The invention discloses a dynamic iris positioning method based on pupil change. The method comprises the following steps of acquiring an original iris image, setting a threshold value to binarize the image, detecting pupil boundaries by using a canny algorithm, detecting noise of the filtered image by using an eight-connection template, scaling down the image according to a proportion, positioning a pupil boundary by using Hough transformation, namely, an iris inner boundary, dynamically positioning the outer boundary of the iris in real time through the established mathematical model of the relationship between the pupil radius and the iris outer diameter, and obtaining an iris positioning map. According to the iris positioning method provided by the invention, the iris position can be quickly and accurately positioned, and an accurate iris image is provided for subsequent feature matching.

The invention discloses a lightweight iris recognition system for a mobile terminal. The system comprises a preprocessing module for performing smoothing processing, lens detection and white hole filling on an iris image captured by an infrared camera; an iris positioning module used for carrying out coarse positioning and accurate positioning on the iris image, obtaining a circle center and a radius of a pupil through coarse positioning, obtaining inner and outer boundary points of an iris through accurate positioning, realizing iris region normalization through fitting a circle by the boundary points, and recording the position of noise by adopting a mask; a coding module used for coding the iris image, carrying out feature extraction on the normalized iris area by adopting a Gabor filter to obtain a feature map, and selecting a point at a specified position in the feature map for coding; a registration and recognition module used for registering the iris image, binding iris codes, masks and registrant information and storing the iris codes, the masks and the registrant information in a database, and used for recognizing the iris image. The system solves the problems of low irisrecognition precision and large resource consumption of the mobile terminal.

The invention discloses a monocular iris matching method based on feature point reprojection. The problems that in the prior art, calculation time complexity is high, and matching results are affectedby areas outside irises are mainly solved. According to the scheme, the method comprises the steps: adopting a camera to acquire a monocular iris image, and sequentially processing iris positioning,feature extraction and boundary information elimination data preprocessing on the monocular iris image; an iris recognition database by using the iris region segmentation image after each time of preprocessing; acquiring a monocular iris image to be recognized by adopting a camera, and carrying out data preprocessing on the monocular iris image; matching the preprocessed iris region segmentation image B with an iris region segmentation image A in an iris recognition database, and estimating a homography matrix according to matched feature point pairs; re-projecting the feature points in the graph B to the space of the graph A, calculating the average position deviation of the corresponding feature points, and judging whether matching succeeds or not according to the average position deviation. According to the method, the iris region is segmented, the matching speed is increased while the matching accuracy is kept, and the method can be used for identity authentication.

The invention provides an iris positioning method, an iris positioning system and an iris lock, which belong to the technical field of iris positioning. The iris positioning system comprise an iris module, a control module, a driving module and a height grade detection module, wherein the height grade detection module is used for detecting a height range of the height of a to-be-detected object and feeding back the height range to the control module; the control module is used for determining a height grade according to the height range of the height of the to-be-measured object and transmitting a control signal to the driving module; and the driving module is used for receiving the control signal transmitted by the control module so as to drive the iris module to move until a vertical field angle theta of the iris module covers the height grade corresponding to the height of the to-be-measured object. Aiming at the technical problem that the height coverage range of an iris positioning technology in the prior art is not wide enough, the iris positioning method and the iris positioning system can achieve the whole-body high-range coverage of a user, and achieve the iris recognitionand acquisition.