294 results about "Palmar vein" patented technology

The present invention discloses a convolutional neural network-based embedded finger vein identification method enabling a counterfeit detection capability. The method includes the following steps that: S1, a plurality of finger vein images of a plurality of levels of light intensities are acquired, a finger vein image with the highest resolution is selected, the ROI (region of interest) of the captured image is preprocessed; S2, a local binary pattern (LBP) is utilized to perform texture feature coding on the high frequency information of the finger vein image; S3, the high frequency part features of the finger vein image are extracted through using a high-pass filter, finger image features are extracted through a vein identification shallow convolutional neural network; and S4, an SVM (support vector machine) classifier is utilized to perform counterfeit detection so as to distinguish the authenticity of the vein image. With the method adopted, the problem that a printed counterfeited vein image can fool an existing identification system can be solved, and the security of an actual vein identification system can be improved; the problem of the decrease of the identification accuracy of the actual system which is caused by factors such as low vein image quality and finger axial deflection can be solved.

The invention relates to a finger vein recognition device and an implement method thereof. The finger vein recognition device comprises a guide groove, an infrared light source, a shooting module, a comparison module, a power supply circuit, a power supply control circuit, a finger position recognition module and a control module, wherein the power supply control circuit performs power supply control on the power supply circuit; the finger position recognition module is connected with the power supply circuit and used for sensing placing action of the finger in the guide groove; the control module is connected with the infrared light source, the shooting module and the comparison module, and the power supply circuit supplies power to the control module through the power supply control circuit. The finger vein recognition device includes at least either a dormant state or a deep dormant state. The control module controls the infrared light source, the shooting module and the comparison module which are woken by power on to work and acquires and recognizes a vein image of the finger in the guide groove. The implement method is used for implementing the finger vein recognition device. The finger vein recognition device is simple in structure and capable of saving energy and electricity.

The invention discloses a medical information management system, which comprises a registration module, a human-machine operation module, a web crawler module, an identity authentication module, a central processing unit, a data processing module, a data collation module, an artificial expert auditing module, a data labeling module, a data positioning module and a database module. The medical information management system imports data through a way that a user automatically carries out input and the system automatically inquires medical information, then, the processing and the auditing of the data are finished through an own processing function of the system, and the repetition and the conflict of the data can be avoided while the real-time update of a database is realized. Meanwhile, the sharing of the data in permission is realized, voiceprint authentication and palmar vein authentication are combined into a whole to greatly improve information safety, the input of each piece of data carries the identity information of a person who inputs the data, and data traceability is realized.

The invention relates to a finger vein identity verification method based on ArcFace Loss and an improved residual network, and the method comprises the following steps: 1) collecting finger vein images of a plurality of fingers, and carrying out the preprocessing of the images; 2) constructing a convolutional neural network; 3) training a model: training the convolutional neural network by usingArcFace Loss; 4) executing a registration stage: after the registration image is enhanced, inputting the registration image into the trained convolutional neural network to obtain a feature vector, taking an average value as the feature of the finger, and storing the feature as a registration feature library; and 5) executing a verification stage: calculating the cosine similarity between the feature vector and each feature in the registration feature library, and judging whether the feature vector corresponds to a certain finger according to the distance and the threshold. According to the method, the lightweight residual network is improved, so that the expression ability of the lightweight residual network to the finger vein features is effectively improved, the quality of the registration feature library is effectively improved, the whole method is simple, feasible and robust, and the accuracy of finger vein recognition is effectively improved.

The invention discloses a finger vein recognition method and system based on cosine center loss. The system comprises an image acquisition module, an image preprocessing module, an image recognition module and an image training module. The identification method comprises the following steps: acquiring a finger vein image of a user to be identified; performing image preprocessing on finger vein image information, and extracting a finger vein region of interest (ROI) image; and extracting finger vein features in the finger vein region of interest through a convolutional neural network based on cosine center loss, and identifying identity information of a to-be-identified user. The finger vein features can be effectively extracted, the redundancy of noise is improved, and the recognition precision of a finger vein recognition system is obviously improved.

The invention discloses a finger vein recognition method and device and a storage medium. The finger vein recognition method comprises the following steps: preprocessing an acquired original finger vein image to obtain a preprocessed finger vein image; obtaining a node set from the preprocessed finger vein image according to an SLIC superpixel segmentation algorithm, and constructing a finger veinweighted graph by using the node set; and based on the improved graph convolutional neural network, performing finger vein recognition on the finger vein weighted graph to obtain a recognition result. The graph model of the finger vein image can be constructed by comprehensively considering the stability of the graph structure and the randomness of the finger vein image, and the finger vein recognition precision is improved.

The invention relates to a vine model creating method for finger vein tridimensional characteristic identification. The vine model creating method the following steps of: firstly, simplifying and abstracting a finger vein spatial structure according to principles of uniform pipe diameter, graded rattan, node subdivision and extension constraint; and then, providing elements, attributes and mathematical description of a vein model; and finally, drawing a vein stereopicture according to the data structure of the vein model. The finger vein vine model is used for guiding the tridimensional reconstruction to reduce the influences of certain factors, for example, repairing broken veins, eliminating burrs and noise and fitting vein spatial curve; and the method disclosed by the invention can abstract and simplify the complex tridimensional vein spatial structure to facilitate the data organization and characteristic extraction.

The invention relates to a finger vein machine learning recognition method and device based on terrain concave-convex characteristics. The method comprises the steps of performing size normalization processing on a registered finger vein image and a verified finger vein image; performing image enhancement processing on the registered finger vein image and the verified finger vein image; obtainingterrain concave-convex features from the registered finger vein image and the verified finger vein image based on a digital elevation model, and extracting registration features and verification features; translation and rotation calibration correction are carried out on the registration features and the verification features, and sliding window similarity calculation is carried out on an overlapping region of the vein features after calibration correction; optimizing the feature extraction technical parameters and the recognition technical parameters of the finger vein based on the calibration correction parameters and the sliding window similarity parameters; the device comprises a normalization processing module, an image enhancement module, a feature extraction module, a parameter calculation module and an optimization module. According to the method, the technical capability of finger vein recognition and the adaptability to different image qualities are improved.

The invention discloses a finger vein recognition method combining a bionic texture feature and a linear texture feature. By adopting the method, the bionic texture feature and the linear texture feature of finger veins can be utilized effectively, the adverse effect of noise on the recognition accuracy is avoided, and the recognition performance and robustness of a fingerprint vein recognition system are enhanced. The bionic texture feature and the linear texture feature of a finger vein image are extracted respectively. The bionic texture feature is extracted by using a Gabor filter with good similarity of a visual cortex simple cell receptive field model of a mammal. When the method is used for performing feature recognition, the effect is similar to object recognition through human eyes, and a good effect is achieved.

The invention relates to a finger vein living body detection method based on multi-feature fusion and DE-ELM, and the method comprises the following steps: 1) respectively collecting a true finger vein image and a false finger pseudo vein image as positive and negative training samples, and carrying out the size normalization preprocessing and Gaussian filtering processing of the positive and negative training samples; 2) respectively extracting a plurality of LBP histogram features and multi-scale HOG features of the vein image, and fusing the LBP histogram features and the multi-scale HOG features into a total feature vector for expressing vein features; 3) setting an activation function of neurons of the hidden layer, determining the number of the neurons of the hidden layer by using adifferential evolution algorithm (DE), and constructing a DE-ELM classification model; 4) inputting the training data into a DE-ELM classification model for training; and 5) inputting the test image data into the trained DE-ELM classification model for detection and identification of living body data, and determining whether the test image data is a living body finger vein. According to the invention, the algorithm for finger vein living body detection by combining multi-feature fusion with the DE-ELM classifier has the advantages of fast detection speed, high detection precision, strong robustness and the like.

The invention discloses a method for acquiring a three-dimensional finger vein image based on a binocular camera, and the method comprises the following steps: 1, collecting a finger vein image underbinocular vision, and obtaining two finger vein images of the same finger of different testees through the binocular camera; 2, preprocessing the finger vein image, wherein the preprocessing of the finger vein image mainly comprises image graying, image bilateral filtering, image enhancement, image binaryzation and image finger vein line refinement; 3, performing three-dimensional reconstruction on the two-dimensional finger vein image, including (1) acquiring internal and external parameters of a binocular camera by using a Zhang calibration method; (2) performing epipolar correction by usingthe rotation matrix; (3) obtaining a disparity map through stereo matching; and (4) performing three-dimensional reconstruction on the finger vein image. According to the invention, the calculation amount is reduced, the accurate and detailed disparity map can be rapidly and conveniently obtained, and the method can be well used for finger vein recognition.

The invention discloses a palmar vein feature extraction method based on a multi-scale convolution kernel. Aiming at the particularity of the palmar vein image texture, an irregular convolution kernelis used to detect the transverse and longitudinal network structure of palmar vein, and a multi-scale convolution kernel feature extraction network for palmar vein recognition is designed, which is mainly composed of four modules, namely a perceptual layer, a dimension reduction layer, a feature fusion layer and a feature vector representation layer. The technical scheme of the invention can makethe feature extraction network have good adaptability and effectiveness to the palmar vein feature extraction through the irregular convolution kernel, such as 3x7, 7x3, 1x7, 7x1, aiming at the characteristics of the palmar vein image such as a special network structure, easy deformation, weak local correlation, complex topological structure and the like.

The invention discloses a finger vein recognition method based on multi-angle imaging. The finger vein recognition method comprises the steps that a finger is rotated and photographed to obtain a 360-degree finger image; selecting a to-be-processed finger vein image from the 360-degree finger image, the to-be-processed finger vein image comprising an image right above the finger and an image rightbelow the finger, and further comprising images rotating forward and backward by a preset angle relative to the image right above the finger and the image right below the finger; extracting an ROI region from the finger vein image to be processed by adopting a sliding window method; enhancing the ROI region to obtain a to-be-identified image; extracting to-be-identified features from the to-be-identified image; mATCHING IDENTIFICATION OF FEATURES. Effective feature extraction and fusion are carried out on multi-angle finger vein information, more finger vein useful information is fully utilized, and the finger vein recognition rate is improved; and the problem that the finger vein recognition rate is reduced due to axial rotation of the finger is solved, the problem of axial rotation of the finger vein is effectively solved, and high robustness is achieved.

The invention discloses a finger vein recognition method based on a multi-semantic feature fusion network. Firstly, collecting finger vein images, performing data enhancement, and making a training set and a test set; building a feature extraction network which comprises an input layer, an improved residual module, a feature fusion preprocessing module, a pooling layer and a full connection layer;constructing a loss function; training a feature extraction network through the training set; and finally, inputting to-be-classified images in the test set into the trained feature extraction network model, obtaining image features of the to-be-classified images, and performing matching calculation on the features to obtain an identification result. According to the method, the finger vein imagerecognition rate is effectively improved, and compared with other methods, the rejection rate is obviously reduced.

The invention discloses a sliding type self-adaptive finger vein recognition device which comprises an installation system, the installation system comprises an installation wall, and the installationwall comprises a control center, and an infrared detection module, a heart rate detection module, a slip sheet starting module, a voice broadcast module, a vein acquisition module and an annular motor control module which are electrically connected with the control center. Two cameras are adopted for vein collection, so that the collected finger vein veins are more accurate, an optical heart ratesensor can be used for effectively preventing people with other worries from cracking equipment by using false vein membrane materials, the safety of the device is improved, and the cameras can be rotated by using an annular motor control panel. Finger vein information is collected in a multi-dimensional mode to obtain a high-quality image, and the reliability of the device is improved. Accordingto the device, the contact pieces and the corresponding mounting contact pieces are arranged in the mounting wall, so that opening and closing of related detection can be controlled, and resources can be effectively saved.

The embodiment of the invention provides a palmar vein acquisition apparatus and a palmar vein identification system. The palmar vein acquisition apparatus illuminates a palmar vein acquisition area through a lighting module, a near infrared camera module completes palmar vein image acquisition, and the palmar vein image is uploaded through an image transmission module. The apparatus can control the working state of the lighting module and the working state of the near infrared camera module through a control module, wherein the lighting module sleeves the near infrared camera module and comprises a plurality of near infrared LED lamps which are uniformly distributed on the periphery of the near infrared camera module to light up the image acquisition area. The near infrared camera module is placed in an inner box, which is embedded in an outer box. The cost is reduced, and real-time and stable acquisition of palmar vein images can be realized.

The invention discloses a finger vein image quality evaluation method based on image processing. The method comprises the following steps: 1) acquiring an originally acquired finger vein region-of-interest picture; 2) filtering image noise by using mean filtering and median filtering to obtain a denoised image; 3) enhancing the contrast of the denoised image by using homomorphic filtering to obtain a contrast-enhanced image; 4) enhancing texture details of the contrast-enhanced image by using directional texture filtering to obtain a texture-enhanced image; 5) setting a threshold value, and performing binarization processing on the image after the texture enhancement to obtain a finger vein binary image; and 6) performing connected domain analysis on the finger vein binary image by using the connected domain information, removing a region with the area smaller than a threshold value, and counting the area of the remaining region to obtain an image quality score. The method can effectively reflect the quality of the collected finger vein image, thereby effectively solving the problem of no universal finger vein image quality evaluation method.

The invention provides a finger vein recognition method and system based on a convolutional neural network and an SIFT (scale invariant feature transform) algorithm, and the implementation process ofthe method comprises the following steps: firstly, obtaining a finger vein image of a to-be-recognized user through data collection; secondly, preprocessing the received original vein image information, and extracting a region of interest of the image; and finally, extracting vein image feature information of the finger of the user to be identified through a neural network in combination with a SIFI algorithm, and identifying the acquired finger vein image feature information of the user to be identified further, so that the purpose of identifying identity information corresponding to the finger vein information is achieved. According to the invention, the finger vein features can be effectively extracted, the redundancy of noise is improved, and the recognition precision of a finger veinrecognition system is obviously improved.

The invention discloses a calibration method for a finger vein identification device. The method comprises the steps of: acquiring at least one finger vein image of each of multiple persons in the current state by using the finger vein identification device; calculating a Gaussian distribution parameter of a sectional pixel gray value of the acquired finger vein image; calculating a gamma mapping coefficient of an infrared image sensor of the finger vein identification device according to the Gaussian distribution parameter of the sectional pixel gray value of the finger vein image; adjusting a pixel input value and pixel output value mapping relation acquired by the infrared image sensor in the current state according to the gamma mapping coefficient, thus identifying the adjusted pixel output value to obtain an identification result. The calibration method for the finger vein identification device can effectively reduce the influence of factors such as different regional climates, temperatures, illumination and the like on the identification success rate of the finger vein device.

The invention discloses a finger vein blood vessel network restoration method based on fractal features. The method comprises the following steps of: intercepting a full-finger image of a collected image, marking an ROI (Region of Interest) area, and normalizing the intercepted image into 80 * 220; Preprocessing the finger vein image through simple filtering, Gabor image enhancement, binaryzationand refinement to obtain a vascular skeleton image; Based on the fractal theory, obtaining fractal features represented by length ratio distribution of father and son blood vessels of a branch structure from the structural features of the finger vein blood vessel network; Pre-repairing the finger vein structure information by using the Gabor enhanced directional diagram as a priori condition, andextracting more accurate fractal features on the basis of the pre-repairing; And predicting the length to be repaired for the blood vessel segment to be repaired based on the obtained fractal featuresand simulating the growth form of the blood vessel. According to the method, defect information of a large area can be repaired, fractal characteristics of a blood vessel network can be repaired, themethod is successfully applied to vein image recognition, and the recognition performance is greatly improved.

The invention relates to a terrain point classification-based finger vein feature extraction and recognition method and device. The method comprises the following steps of 1) extracting finger vein image information of different scales by adopting a multi-scale Gaussian filter; 2) respectively carrying out image size normalization processing on the finger vein images; 3) performing boundary cutting on the finger vein image; 4) performing terrain characteristic point extraction on the finger vein image based on two local parameters of a connectivity value and a curvature differential in the digital elevation model, and classifying terrain characteristic points to obtain finger vein characteristics; 5) performing feature assembly on the finger vein features; 6) performing translation calibration correction on the finger vein features; and 7) carrying out sliding window similarity calculation on an overlapping region of the registered finger vein feature and the verified finger vein imageto complete comparison of the registered finger vein feature and the verified finger vein image. The method solves the problem that the existing finger vein technology is poor in recognition effect on the low-quality finger vein image with scaling and rotation.

The invention discloses a finger vein segmentation method and device based on R2U-Net and a storage medium. In a coding and decoding unit, a regular forward convolution layer is replaced by a cyclic convolution layer with a residual unit, the residual unit of the cyclic convolution layer is beneficial to establishing a more effective and deeper model, the cyclic convolution layer unit comprises aneffective feature accumulation method, and better and stronger feature representation is guaranteed by feature accumulation of different time steps. Thus, it facilitates the extraction of very low levels of features that are essential to the segmentation of finger veins. In the neural network training process, on one hand, the center of each image is randomly selected to obtain sub-blocks for data expansion, and on the other hand, the grains extracted by combining six traditional methods with different weights are used as gold standards, so that the method can give full play to the advantagesof each traditional extracted grain and make up for the disadvantages.