32 results about "Iris recognition" patented technology

The invention provides an unmanned cargo automatic handover system based on multi-mode biological recognition. The unmanned cargo automatic handover system comprises a biological recognition device, an identity verification module, a cargo handover control device and a cloud management platform. The biological recognition device integrates a fingerprint recognition module, a face recognition module and an iris recognition module, and high-accuracy identity recognition is ensured by using advanced sensors and algorithms. The identity verification module guarantees the safety of the verification process through real-time comparison and multi-factor verification in combination with dynamic database management. The cargo handover control device automatically controls unlocking and locking of a cargo cabin door, and records, encrypts and stores all handover operation data at the same time. According to the technical scheme, high safety and high efficiency of cargo handover of the logistics vehicle are achieved, the method is suitable for being widely applied to the modern logistics field, the safety and accuracy of the logistics vehicle in the cargo handover process are improved, and automation and traceability of operation are achieved.

The invention relates to the technical field of biological feature recognition, in particular to a binocular iris recognition system and method based on left and right eye independent feature libraries. On the feature library architecture level, the system establishes completely independent feature databases for the left eye and the right eye respectively. The left eye feature library is specially used for storing left eye iris feature vectors and corresponding personnel identifiers of all registered users, and the right eye feature library is used for storing right eye iris feature vectors and personnel identifiers. The two feature libraries are completely separated from each other in physical storage and logic index and do not interfere with each other. When the system receives a to-be-recognized image from the left eye, only the left eye feature library needs to be searched, and the right eye feature library does not need to be traversed, so that the search space is reduced by about half. According to the method, the characteristics of independence and complementarity of iris textures of two eyes are fully utilized, and the retrieval efficiency and the anti-counterfeiting capability of the system are remarkably improved while the recognition precision is kept.

The invention belongs to the technical field of biological feature recognition, and discloses an ultra-micro-distance iris recognition device and method based on an optical imaging technology, the device comprises an optical imaging unit, a near-infrared light source unit, a trigger detection unit, a control processing unit and a protection unit, and the method comprises the following steps: S1, triggering; s2, image acquisition; s3, performing pretreatment; s4, feature extraction; s5, matching verification is carried out; s6, controlling; the method has the following beneficial effects: 1, the limitation of ultra-micro-distance imaging is broken through, the image quality is improved, aberration-free imaging can be realized under the working distance of 5cm, and iris texture details are clearly captured; 2, an efficient focusing and triggering mechanism is adopted, so that the recognition efficiency is improved, and'touch-and-shoot 'is realized; the robustness is high, the influence caused by the shielding problem of lower eyelashes and eyelids is avoided, dust and noise are removed, and the contrast ratio is enhanced. And 4, the size of the shell is small, a portable instrument can be directly embedded, and the explosion-proof, waterproof and dustproof requirements are met.

The invention discloses an intelligent property access control system which comprises a multi-mode biological recognition module, an environment sensing module, a data processing module, a data storage module and a data management module, and the multi-mode biological recognition module is integrated with a camera module, a fingerprint module and an iris module. Biological information is recognized through the data processing module, high-precision recognition in a sheltered scene is achieved, the iris recognition mode is automatically switched to when the face is sheltered, a weighted fusion strategy is adopted, the recognition passing rate is effectively increased, meanwhile, details of a sheltered part are automatically generated in cooperation with a sheltered object detection algorithm, and the recognition efficiency is improved. A dual-identification guarantee mechanism is realized, the identification accuracy under the shielding scene is further improved, and the reliability of the access control system is ensured.

This invention relates to the field of intelligent management technology for cleanroom garments, specifically an intelligent management system for cleanroom garments equipped with iris recognition technology. The system divides the work area into sub-areas based on cleanliness levels, establishes iris access control systems between these sub-areas, and monitors environmental parameters of each sub-area using sensors. It tracks workers wearing cleanroom garments in real time, obtaining work trajectory data and work status data. Based on the work trajectory data, it obtains trajectory environmental parameters; by identifying these parameters and data, it obtains external contamination data; by identifying the work status data, it obtains internal contamination data; and based on the external and internal contamination data, it determines the contamination level of the worker's cleanroom garment. This invention achieves intelligent management of cleanroom garments by identifying contamination data from the worker's cleanroom garments, adjusting iris access permissions based on cleanliness and contamination levels, and adjusting the activity range based on iris access permissions and the iris access control system.

The application provides a crack identification method based on an iris recognition principle, which comprises the following steps: step 1, inputting a three-dimensional seismic data body; step 2, carrying out seismic data fracture enhancement pretreatment based on an iris recognition related algorithm; step 3, carrying out seismic crack feature point analysis based on iris recognition related parameters; step 4, carrying out seismic crack feature extraction and coding based on an iris recognition algorithm; and step 5, carrying out predicted seismic crack and prior crack feature matching based on iris recognition, and completing the calculation when the matching relationship reaches a correlation threshold requirement. The crack identification method based on the iris recognition principle has the detail recognition advantage of the iris recognition technology, has strong sensitivity to micro crack identification, and has stability and uniqueness of the result of crack prediction through iris feature extraction and matching.

The invention belongs to the technical field of biological feature recognition, embedded system security and computer bus interfaces, and particularly relates to a hot plug guarantee method and device for an iris recognition M.2 / mini-PCIe module. A security power domain comprises a session key memory, a security element chip and a key state holding circuit, and a processor power domain (Vcore) comprises a main controller and an image interface. The safe power domain (Vsec) comprises a ferroelectric random access memory (FRAM) and a static random access memory (SRAM); and when the system detects that the PERST # signal jumps from a high level to a low level, cutting off the power supply of the safety power supply domain (Vsec), and triggering a safety sealing process. According to the invention, through organic combination of a dual-power-domain architecture design, an ultra-fast data storage mechanism, a self-adaptive security protection strategy and a distributed data integrity verification system, the technical problem of an iris recognition module in a hot plug scene is solved.

The application relates to an FPGA-based neural network accelerator and application thereof, which comprises a controller, a data input cache unit, a weight input cache unit, an operation unit and an output cache unit; the controller controls the cycle boundary and convolution step length of the neural network accelerator to match the layer structure of a quantized residual neural network; the data input cache unit splits and stores input data; the weight input cache unit splits and stores the weight of the quantized residual neural network; the operation unit performs convolution operation on the input data and weight data; and the output cache unit stores the output result and transmits the output result to a PS end through AXI4 bus burst transmission. The application realizes a method for applying a convolutional neural network to iris recognition, greatly improves the recognition success rate compared with a traditional algorithm, accelerates the inference process, and makes up for the shortcoming that the recognition method is not suitable for mobile deployment.

The invention relates to the technical field of automatic instrument management, in particular to an AI-based calibration charging distribution management system. The system comprises an intelligent cabinet body integrated with a plurality of charging and calibration integrated equipment bases, an iris recognition module used for identity authentication, a gas path system and a ventilation and exhaust system linked with an environment sensor in the cabinet. The management platform of the system can realize the whole process management of receiving an instrument through iris recognition by a user, randomly returning the instrument after use, automatically charging and carrying out Bluetooth fault diagnosis by the system, and automatically executing gas calibration in an unattended state according to a preset period. Through an integrated safety interlocking mechanism, gas leakage in the cabinet can be monitored in real time in the automatic calibration process, emergency responses such as gas source cut-off, forced exhaust and equipment power failure are automatically triggered, and the operation safety is greatly improved. According to the invention, automation and security of instrument management are realized, and the equipment perfectness rate and management compliance are improved.

This invention relates to the field of optics, specifically to a biometric iris recognition lens. The biometric iris recognition lens comprises a first fixed lens group with positive optical power, a zoom lens group with negative optical power, an aperture stop, a second fixed lens group with positive optical power, a focusing lens group with negative optical power, and a third fixed lens group with positive optical power; 4 < ft / fw < 5; 1.15 < fnot / fnow < 1.25; ft is the focal length of the biometric iris recognition lens in telephoto mode, fw is the focal length of the biometric iris recognition lens in wide-angle mode, fnot is the aperture number of the biometric iris recognition lens in telephoto mode, and fnow is the aperture number of the biometric iris recognition lens in wide-angle mode. The biometric iris recognition lens can stably acquire images with stable brightness at a certain distance, facilitating iris acquisition and increasing the application range of the biometric iris recognition lens.

1. The name of the design product: intelligent iris recognition instrument. 2. The use of the design product: used for collecting iris to complete the confirmation of identity information. 3. The design points of the design product: in shape. 4. The picture or photo that can best show the design points: perspective view 1.

The invention belongs to the technical field of biological feature recognition, and particularly relates to a composite scaling method and an iris recognition system and method of a neural network of the composite scaling method. The method comprises the following steps of: establishing a mathematical constraint relationship of three dimensions of depth, width and resolution, simultaneously adjusting the three dimensions of depth, width and resolution of the network by a composite coefficient phi, and scaling the depth into a product of phi power of alpha and basic depth; the width is scaled to be the product of phi power of beta and the basic width; the resolution ratio is scaled to the product of the phi power of gamma and the basic resolution ratio, the network scale can be systematically adjusted by using a single composite coefficient, higher recognition precision can be obtained under the same calculation budget, or the calculation overhead can be obviously reduced under the same precision requirement.

1. The name of the design product: iris recognition instrument (AI2000). 2. The use of the design product: iris recognition, face recognition and card authentication for identity recognition and authentication. 3. The design points of the design product: the combination of shape and pattern. 4. The picture or photo that best indicates the design points: perspective view.

The invention provides a dynamic permission allocation method and system based on iris recognition and real-time state monitoring, and relates to the technical field of information system security, and the method comprises the steps: firstly collecting an iris texture image containing details such as eye iris sphincter contraction form and real-time state monitoring data containing information such as physical space coordinates; performing feature anchoring processing on the two to generate associated feature bodies which synchronously establish bidirectional mapping through timestamps, calling a historical permission interaction database to mine scene matching association, generating an adaptation rule pool, obtaining an initial permission scheme through matching of the associated feature bodies, tracking operation behavior data, and obtaining an initial permission scheme; the method comprises the steps of analyzing the degree of adaptation, feeding back and optimizing an adaptation rule pool, dynamically adjusting a permission range to generate a dynamic permission result, finally synchronizing the dynamic permission result to a permission management and control node and an associated operation terminal, executing a permission instruction and recording a permission change track to form a traceable data chain, thereby improving the security, flexibility and accuracy of permission distribution.

The utility model discloses a U shield intelligent cabinet, which comprises a cabinet body, a control host, a fingerprint identification head, an iris identification camera and a plurality of electromagnetic locks, the cabinet body is provided with a plurality of grids with the same size and shape, each grid is independently and rotatably provided with a cabinet door, each cabinet door is provided with the electromagnetic lock, and the fingerprint identification head is connected with the iris identification camera. The control host is arranged in the cabinet body, the fingerprint recognition head and the iris recognition camera are both arranged on the cabinet body, the fingerprint recognition head, the iris recognition camera and the electromagnetic lock are all electrically connected with the control host, a sinking groove is formed in each grid opening, and a patch type pressure sensor is arranged at the bottom of each sinking groove. The USB key cabinet has the following beneficial effects that each USB key is independently stored and taken, and whether the USB key is taken or not is monitored.

The invention provides an optical system and a display device. The optical system comprises a lens structure, a transflective film and a reflective polarizing film, the transflective film and the reflective polarizing film are arranged on the lens structure, image light transmitted by the transflective film is configured to be folded back between the transflective film and the reflective polarizing film and is emitted from the reflective polarizing film, and the lens structure comprises at least one lens. The optical system further comprises a calibration lens located on the side, away from the reflective polarizing film, of the transflective film, and the optical axis of the lens structure does not pass through the calibration lens. The focal length of one of the lens structure and the calibration lens is a positive value, the focal length of the other one is a negative value, and the calibration lens comprises at least one free-form surface, so that the degree of freedom of optical system design is improved, non-centrality caused by the lens structure is effectively corrected, aberration generated by an off-axis optical system is corrected, and the imaging quality is improved; the camera calibration process is simplified, and the sight tracking and iris recognition precision is improved.

1. Name of the product in this design: Testing Instrument. 2. Purpose of this design: For collecting iris and pupil information, iris recognition, and detection of the inspected person. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: 3D view 1.

A wearable device is provided. The wearable device includes a plurality of light sources configured to emit light toward an eye of a user wearing the wearable device. The wearable device includes a camera configured to capture images of at least a part of the eye of the user wearing the wearable device. The wearable device includes at least one processor comprising processing circuitry. The wearable device includes memory, comprising one or more storage mediums, storing instructions. The wearable device performs eye tracking based on the images captured by the camera while the plurality of the light sources are controlled to emit light toward the eye of the user. The wearable device performs iris recognition based on at least one image captured by the camera while a portion of the plurality of the light sources is controlled to emit light toward the eye of the user and a remaining portion of the plurality of the light sources is deactivated.

The invention provides a personalized authentication method and system based on iris recognition and user habit learning, and belongs to the technical field of authority management. A continuous state flow and a multi-mode biological characteristic flow of a to-be-authorized object are acquired in real time, then a state chain is constructed, and a state change node with a correction rule is generated; based on the change critical point and the correction rule of the state change node, dynamically adjusting the verification weight ratio of the multi-modal biological characteristic flow, generating a stepped scheme, performing synchronous verification, obtaining a verification conclusion with a state association mark, cooperatively adjusting the role permission level and the attribute permission range in combination with the verification conclusion and the state change node, and obtaining a multi-modal biological characteristic flow verification result. According to the method, a permission collaborative configuration scheme is generated, permission full-period management is executed, a result is fed back, state change node generation is optimized, dynamic permission personalized authentication distribution is achieved, high intelligence, dynamic performance and precision of permission authentication distribution are achieved, and system safety and service operation efficiency are remarkably improved.

The invention discloses a cattle identification method and system based on iris identification and nose print identification, and the method comprises the steps: collecting the iris and nose print images of a cattle, combining illumination calibration processing, generating a standardized image pair, extracting the high-frequency features of the iris and the nose print codes through signal processing and binarization operation, and carrying out the recognition of the iris and the nose print. And a complementary feature template is fused to cope with complex scenes such as image shielding or illumination variation. Through feature template matching, cattle identity is confirmed, life cycle archives are associated, records are updated in real time, the tracking module is combined to bind the archives and coordinates, breeding quarantine data are obtained, financial scene data in a remote management platform are finally connected, traceability requirements are judged, asset right confirmation is achieved, and the system is convenient to use. According to the method, complete closed-loop management from identity confirmation to asset right confirmation is realized, and the digital management efficiency and credibility of the breeding industry are remarkably improved.

The invention discloses a near-infrared luminescent material, a preparation method thereof and application of the near-infrared luminescent material in smoke penetration imaging. The chemical formula of the near-infrared luminescent material is as follows: A < 2x > B < 0.5 + x-y > Nb < 1.5-x > (PO4) 3: yCr < 3 + >, in which A is selected from one, two or more of Li, Na, K, Rb and Cs in alkali metals; b is selected from one, two or more of Al, Sc, Ga and Fe; cr < 3 + > is a doped luminescent central ion, y is the doping amount of Cr < 3 + > in each mole of the near-infrared luminescent material, 0.01 < = y < = 0.2, and 0 < = x < = 0.5. The near-infrared light-emitting material can be used for preparing a light-emitting device in the fields of smoke penetration imaging, security monitoring, iris recognition, automobile sensing, food detection and the like.

The invention provides an electronic fence integrated safety management and control method and system combined with iris recognition, and belongs to the technical field of safety management and control, and the method comprises the steps: starting iris collection based on an induction signal of an electronic fence monitoring region, obtaining an eye image of an entering object, and extracting an initial iris feature set; then acquiring real-time state information of the region, and performing bidirectional adaptive adjustment on the initial iris feature set and the real-time state information to obtain a target iris feature set; then, based on the target iris feature set, generating hierarchical management and control instructions including region admission, equipment use and activity time period management and control instructions; transmitting the layered management and control instruction to an execution component for sequential execution; and finally, collecting actual effect information after execution, and optimizing an adaptive adjustment rule for subsequent operation. According to the invention, the intelligent, personalized and dynamic security control is realized, and the pertinence and effectiveness of the control are improved.

The invention relates to the technical field of biological feature recognition, in particular to an iris recognition model training method based on an angle interval loss function. The invention relates to a method for optimizing iris feature representation by using an angle interval loss function in a deep neural network training process, a preset angle interval is added to an angle corresponding to cosine similarity, repeated calculation of an anti-cosine function and a cosine function is avoided, and stability and calculation efficiency of numerical calculation are ensured. The angle interval mechanism is equivalent to setting a stricter classification condition for a target category, a feature vector of a sample not only needs to be consistent with a weight vector of a correct category in direction, but also needs to be closer to weight vectors of other categories, and the approaching degree is determined by the size of the angle interval.

The application discloses a kind of beauty pupil detection and iris identification method, device, readable storage medium and equipment, belong to iris identification field.The present application utilizes the characteristics that the texture of beauty pupil area is relatively complex, and the relative gradient is relatively large, and the outer circle of normal iris is relatively smooth;And the characteristics that the pixel value of beauty pupil area is relatively small relative to the whole iris area are used for beauty pupil detection.The beauty pupil detection and iris identification method first acquires eye image, then respectively carries out gray scale stretching and gradient transformation to eye image, and carries out matrix addition operation and binaryzation, whether wearing beauty pupil is judged according to the proportion of the pixel with value 1 in the part outside pupil area on the obtained binary image.This application can accurately identify dark beauty pupil and light beauty pupil, and does not need to rely on image sample, has strong robustness, and can identify various styles of beauty pupil.

The application provides an iris recognition method, device, equipment and system, the method comprising: acquiring at least two frames of to-be-recognized images under different light supplement states; using a deep learning algorithm to perform feature extraction and feature fusion processing on the at least two frames of to-be-recognized images to obtain fused features; and performing iris recognition according to the fused features. The method can improve the performance of iris recognition.

There is provided a computer implemented method of detecting an attempt to breach security of an iris recognition system by an artificial iris, comprising: analyzing at least a portion of a limbal ring depicted in an image of an iris of an individual captured by an imaging sensor at a wavelength range within at least one of near infrared (NIR) and short wave infrared (SWIR), and detecting likelihood of an artificial iris worn by the individual according to the analysis of the at least the portion of the limbal ring.

The utility model discloses a portable iris recognition device applied underground, which comprises a shell, a display screen is arranged on the front surface of the shell, a recognition device and a light supplementing module are arranged in the shell, and the recognition device comprises a positioning module and a recognition module; the light supplementing module comprises a natural light supplementing unit and an infrared light supplementing unit; the natural light supplementing unit is used for supplementing light to the positioning module; the infrared light supplementing unit is used for supplementing light to the identification module; the positioning module is used for determining the position of a to-be-recognized target. The recognition module is used for recognizing the to-be-recognized target, and is used for solving the problem of positioning failure or misrecognition.

The invention belongs to the technical field of digital asset transaction, and particularly relates to a digital asset transaction system based on a digital asset all-in-one machine, which integrates three biological recognition technologies of fingerprint recognition, face recognition and iris recognition through a multi-mode biological recognition unit, and is linked with a user mobile terminal APP through a dynamic password generation unit. A dynamic verification code is generated once every 60 seconds, and the user needs to input the dynamic verification code for secondary verification after completing biological recognition; then, an abnormal behavior monitoring unit collects a user login place, equipment information and transaction habit data in real time, and when it is detected that the login place changes suddenly, unfamiliar equipment login or transaction habits are abnormal, identity secondary verification and early warning prompt are triggered immediately; on the basis, through multi-mode biological recognition, dynamic password verification and abnormal behavior monitoring of the identity authentication and authority management module, identity authentication is safe and reliable, and the risk that assets are stolen is reduced.

The invention provides a risk prediction method and system combining iris recognition and behavior pattern learning, and relates to the technical field of information system security, and the method comprises the steps: firstly collecting iris core features and behavior sequence data including an operation behavior sequence, a space movement sequence and a resource access sequence; associating individual historical behavior data based on iris core features, grouping according to behavior types, extracting features to generate an individual behavior baseline, collecting current behavior sequence data in real time, extracting current features, comparing the current features with the individual behavior baseline to calculate a deviation degree, performing attribution analysis on the behavior features exceeding a preset deviation threshold value, and obtaining an iris feature sequence; the method comprises the steps of judging a deviation reason in combination with individual identities, scene information and task information, generating a deviation attribution result, finally generating a risk prediction result containing risk types, occurrence probability quantization intervals and influence ranges in combination with historical corresponding relations based on the deviation attribution result and the deviation degree, and synchronously outputting risk intervention suggestions. And the efficiency and effect of risk management are effectively improved.

This invention discloses a method and system for cattle identification based on iris recognition and nasal print recognition. By acquiring iris and nasal print images of cattle and combining them with illumination calibration processing, standardized image pairs are generated. Signal processing and binarization operations are used to extract high-frequency features of the iris and nasal print codes, which are then fused into complementary feature templates to handle complex scenarios such as image occlusion or changes in illumination. The identity of the cattle is confirmed through feature template matching, linked to lifecycle records, and updated in real time. A tracking module binds the records and coordinates to obtain livestock quarantine data. Finally, the system connects to financial scenario data in a remote management platform to determine traceability requirements and achieve asset ownership confirmation. This invention realizes a complete closed-loop management system from identity verification to asset ownership confirmation, significantly improving the efficiency and reliability of digital management in the livestock industry.