72 results about "Facial recognition system" patented technology

The invention discloses an electronic license technology for special operation examination and approval through material object scanning and face recognition, and particularly relates to an electronic operation license for examining and approving dangerous operation for special operation in a production field in an information acquisition and online management mode. Comprising a special operation management and control system, a face / appliance recognition system, a gas detection system and a movable monitor. According to the special operation management and control system, an informatization platform is mainly built, the original paper version approval process is converted into online approval, whole-process supervision is achieved, and meanwhile a mobile phone end APP and a computer end webpage are built. The face / appliance recognition system is mainly used for a face recognition system and an equipment and facility recognition system; the gas detection system comprises a portable gas concentration detection alarm instrument, a data storage device, a data transmission device and a recording device; the movable monitoring system can construct all processes for monitoring, remote supervision is facilitated, and archiving and reviewing are achieved at the same time. An informatization centralized management and control mode is adopted, the original tedious processes of paper version approval, archiving, regular detection of toxic and harmful gases and recording are broken through, online approval and process full recording are realized through system construction, a tamper-proof function is set, the record traceability is ensured, real field confirmation of personnel at all levels is ensured, and the system is safe and reliable. And meanwhile, the whole process can be remotely and comprehensively monitored.

The invention discloses a stranger face recognition system and method, a foot-type humanoid robot and a medium, which are applied to the technical field of image processing, and adopt a multi-task face recognition model to perform face bounding box prediction and face key point extraction on an environment image to obtain a face key point set in each face bounding box; aligning the face key point set in each face bounding box to the standard template face image to obtain an undeformed face image; extracting a face feature vector of a to-be-recognized face from each non-deformation face image; calculating the similarity between the face feature vector of each face to be recognized and the face feature vector of each temporary reference face in a lightweight face database; and determining whether the to-be-recognized face is a stranger face based on the similarity. Thus, stranger face recognition can be achieved on the foot type humanoid robot, network delay caused by cloud comparison is avoided, real-time interaction response in a public environment is achieved, and the man-machine interaction efficiency is improved.

The utility model provides a shield tunnel limited space safety operation protection early warning system, comprising RFID tags carried by constructors and corresponding to the constructors one by one; an identity management server; the environment monitoring system is provided with a plurality of sensor assemblies and used for monitoring the temperature, humidity and air quality in the limited space of the shield tunnel; the UHF card reader is arranged at the entrance of the shield tunnel and used for preliminarily positioning the identity of a constructor; the face recognition system is used for carrying out face image acquisition and face recognition and registration on the entering constructors and counting the total number of the constructors; the environment safety early warning system is used for receiving the temperature, the humidity and the air quality which are monitored in real time and performing construction early warning according to a safety threshold range; and the monitoring center is connected with the identity management server, the face recognition system and the environment safety early warning system, is provided with a display screen, and is used for displaying data monitored by the environment monitoring system and early warning information.

The invention relates to a multi-mode intelligent security check system, which comprises an application auditing system, a security check channel system and a monitoring system, wherein the security check channel system is provided with an alarm mechanism; the monitoring system is used for monitoring the application auditing system and the security check channel system; the security check channel system comprises a vehicle security check channel and a personnel security check channel which are provided with A and B doors and transmit information to the monitoring system, the vehicle security check channel is provided with a life detection system, a vehicle bottom scanning system and a vehicle driver checking system, and the personnel security check channel is provided with an article detection and storage system and a multi-stage second non-inductive face recognition system; the monitoring system comprises a police officer monitoring unit, an armed police monitoring unit and a remote monitoring unit which are all provided with monitoring terminals. The monitoring terminals monitor the application auditing system and the security check channel system and give out corresponding instructions when abnormity is found according to received information. According to the invention, safe and efficient security check can be realized, information synchronization of multiple departments is realized, and the working efficiency is improved.

The invention provides a staff image matching and archiving method and system based on face recognition, and relates to the technical field of archive management. The method comprises the following steps: calculating the similarity of face features of known employees and forming a to-be-detected image set; collecting multi-source auxiliary information and carrying out multi-dimensional analysis on the multi-source auxiliary information for known employees with identity splitting conditions, and constructing an evidence chain; according to an analysis result of each dimension in the evidence chain, calculating a file merging confidence coefficient of the to-be-verified image file; if the file merging confidence coefficient is greater than or equal to a preset merging threshold value, merging the to-be-verified image file into a real image file of the corresponding known staff; otherwise, early warning information is generated and submitted for manual auditing. The method aims at solving the technical problem that a traditional face recognition system is prone to cause identity splitting in a complex environment, automatic and accurate filing of staff image files is achieved, manual intervention requirements are reduced, and file management efficiency and data quality are remarkably improved.

This application discloses a face recognition method, a face recognition system, a terminal device, and a storage medium. The face recognition method includes: acquiring a face image to be recognized and extracting initial face features corresponding to the face image; secretly segmenting the initial face features to obtain at least a first face feature to be recognized and a second face feature to be recognized; determining a first difference between the first face feature to be recognized and a plurality of first reference face features in a first server, and determining a second difference between the second face feature to be recognized and a plurality of second reference face features in a second server; the first and second reference face features are obtained by secretly segmenting reference face features corresponding to reference face images; and obtaining a recognition result corresponding to the face image to be recognized based on the first and second differences. The face recognition method of this application stores the segmented face features on different servers for computation, reducing the risk of face feature leakage.

This invention provides a method for manufacturing an electronic lock motor for a charging pile, comprising: S1 manufacturing a micro motor and a reduction gearbox, and setting an absolute encoder at the output end of the micro motor; S2 providing a high-definition pinhole camera device, and manufacturing a face recognition system and / or a fingerprint recognition system; S3 manufacturing a main control circuit board that communicates with a cloud server, and sending data from the face recognition system and / or fingerprint recognition system to a user's smart mobile terminal. The user's smart mobile terminal can confirm charging based on the received data and the power access signal from the vehicle computer, forming a confirmation signal. The cloud server then receives the confirmation signal from the user's smart mobile terminal and sends a formal charging start signal to the charging pile. The smart charging gun includes a smart motor manufactured using the above method, improving the stability, reliability, billing effectiveness, and accuracy of the charging connection.

The present invention provides a facial recognition system for alleviating ambient light during image capture, the system comprising: - a stereo camera configured to capture an image and send it to a face detection module, wherein the image comprises depth information of objects within the depth of field, - the face detection module configured to identify a face based on the depth information of the received image from the stereo camera, - an processing module configured to - split the image into a facial area and a non-facial area, wherein the facial area corresponds to the identified face region by the face detection module, and - outcome exposure adjustment factors for each area, these exposure adjustment factors being calculated based at least on a brightness values comparison of both facial and non-facial areas, - a dual-regional adjustment module configured to separately adjust the exposure of both the facial and non-facial areas according to the adjustment factors, and - a face matching module configure to verify or enroll person's face identity based on the exposure-adjusted facial area.

The application relates to the field of face recognition applications, in particular to a face recognition system and method based on a door lock and ensuring data security, which comprises a collection end S1, a camera module S11, a face algorithm module S12 and a network transmission module S13; the camera module S11 comprises a plurality of cameras in different directions; the face algorithm module S12 is used for image scanning and judgment of local image information in all directions and extraction of local feature data; and the network transmission module S13 is used for encryption, packaging, labeling and transmission of all local feature data; the application provides a face recognition system and method based on a door lock and ensuring data security, different parts of a face are recognized through a plurality of cameras contained in the collection end, thereby controlling extraction of feature quantity, distinguishing and protecting features, preventing face data from being read, and realizing safety control of face data in face recognition of a door lock.

The invention discloses a face recognition method based on a face expert hybrid model and application, and belongs to the technical field of computer vision. According to the face recognition method based on the face expert hybrid model, firstly, a data set and a data enhancement strategy are utilized to train a plurality of face recognition networks, so that the face recognition networks are adaptive to different complex scenes; then, constructing a face expert hybrid model comprising a shared expert network, a routing expert network and a dynamic gating network, and screening the activated expert network through the dynamic gating network to obtain an optimal face feature vector; then introducing a complementary auxiliary loss function to prevent model collapse caused by extreme imbalance of expert load in the training process; finally, reasoning is conducted in a sparse activation mode, and calculation expenditure is reduced. The method can significantly improve the generalization ability of face recognition in different ages, shielding and illumination scenes, optimizes the training reasoning efficiency and model stability, and can be widely applied to various face recognition systems.

The invention relates to a method for improving the wearing rate of a reflective vest and the carrying rate of necessary articles for constructors, which comprises the following steps of: fixing a first part of a split RFID (Radio Frequency Identification) tag at a design position on the reflective vest, and pre-inputting personal identification information of the constructors to a second part of the split RFID tag and holding the second part of the split RFID tag by the constructors, a conventional RFID tag is arranged on each necessary article, article category identification information is recorded in advance, personal identification information is recorded in the bracelet in advance, a split RFID reader and a face recognition system are installed at the entrance guard of a construction site, a background control system reads the personal identification information of the reflective vest through the RFID reader, and the reflective vest is used for identifying the personal identification information of the reflective vest through the face recognition system. If the information is checked to be the same person through the face recognition system, the person is released; in a construction site, the split RFID tag, the conventional RFID tag and a bracelet worn by a constructor are linked, whether the split RFID tag on the reflective vest is in a normal state or not is determined, and whether an entering person carries a necessary article or not is determined.

The invention discloses a face recognition system switch with an interface switching function, and relates to the technical field of switches, the face recognition system switch comprises a switch body and a connecting frame, the front surface of the switch body is provided with a network port and an optical port, the front surface of the switch body is provided with a storage groove, and the storage groove is provided with an interface. Heat dissipation fins are mounted on the upper surface of the switch body; a connecting frame is installed on the right side of the switch body, and connecting blocks are installed on the front side and the rear side of the filter screen. According to the switch with the interface switching function for the face recognition system, a filter screen on a connecting frame can be quickly disassembled and assembled through a guide strip and a sliding groove, and a cleaning brush assembly can be driven to clean through a second rotating mechanism without disassembly; during cleaning, the position of the brush body is fixed through the limiting bolt, the filter screen is pushed to move through the pushing handle, efficient dust removal is achieved, the filter screen is convenient to disassemble and clean, the tedious problem that a traditional filter screen depends on manual disassembly and cleaning is solved, and the operation and maintenance requirements of large-scale deployment are met.

The invention relates to a method for realizing face recognition confrontation sample generation based on frequency domain decomposition and an attention mechanism, and the method comprises the following steps: carrying out the data preprocessing, and carrying out the model construction; generating an attention mask M; performing frequency domain decomposition by using a discrete wavelet transform algorithm; constructing a joint loss function J; frequency domain disturbance iterative optimization is carried out, and disturbance variables are initialized in a frequency domain space; and repeating until an attack stop condition is met, and outputting a final confrontation sample image. The method, the system, the device, the processor and the computer readable storage medium for generating the face recognition adversarial sample based on the frequency domain decomposition and the attention mechanism have the advantages of high mobility, high concealment and high robustness, and the adversarial sample generated based on the frequency domain has higher resistance to common image processing operation; the algorithm is universal, and the proposed framework does not depend on a specific model structure and can be flexibly applied to various CNN-based face recognition systems.

A method of determining an operating status of one or more hardware components of a facial recognition system of a mobile device includes engaging the facial recognition system of the mobile device, wherein the facial recognition system comprises a dot projector and an infrared ("IR") camera; capturing an image of the dot projector and the IR camera; comparing the captured image to a predetermined test image; and based on the results of the comparison, determining the operating status of at least one of the dot projector and the IR camera.

The application discloses a stranger face recognition system and method, a foot type humanoid robot and a medium, which are applied to the technical field of image processing, adopt a multi-task face recognition model, perform face boundary frame prediction and face key point extraction on an environment image to obtain a face key point set in each face boundary frame, align the face key point set in each face boundary frame to a standard template face image to obtain a non-distortion face image, extract a face feature vector of a face to be recognized from each non-distortion face image, calculate the similarity between the face feature vector of each face to be recognized and the face feature vector of each temporary reference face in a lightweight face database, and determine whether the face to be recognized is a stranger face based on the similarity. In this way, stranger face recognition can be realized on the foot type humanoid robot, network delay caused by cloud comparison is avoided, real-time interaction response in a public environment is realized, and human-computer interaction efficiency is improved.

The invention provides a pharmacy window business integrated system and method based on an intelligent tracing identification technology, and the technology integrates the functions of drug tracing code automatic identification, pharmacy business process optimization, multi-system data collaboration, intelligent early warning and the like. And a biological recognition technology (such as a face recognition system) is integrated aiming at a receiving link of special management medicines (such as anesthetic medicines and mental medicines), so that intelligent, automatic and full-process tracing management of pharmacy window services is realized. The invention belongs to the technical field of cross application of hospital informatization management and an intelligent identification technology, and relates to deep fusion of medical compliance supervision and biological feature identification.

The present application provides a face recognition presentation attack detection system, and relates to the technical field of face recognition security. The system can receive a user's face recognition request through a face image information acquisition front end, and simultaneously acquire 3D face information, living body information, and environmental information, and other multi-dimensional face image information. More complex face data can be acquired. After face security detection and recognition, the multi-dimensional face image information is subjected to abnormal detection and filtering processing, which can effectively prevent model data poisoning attacks. When face recognition is performed based on multi-dimensional face image information, the called face recognition model is subjected to black box detection and white box detection for security detection, which can verify the correctness of the model, prevent the model from being used incorrectly or stolen and changed by attackers, cracked and used to cheat and attack the system, and improve the security of the face recognition model. The system monitoring module performs security monitoring and management on the life cycle of face recognition, further ensuring the security of the face recognition system.

This invention relates to a method, apparatus, readable storage medium, and electronic device for detecting anomalies in face recognition. It can determine the comprehensive attention features of a captured face image and the comprehensive attention features of a face image without interference features, and based on the difference in the distribution of activation values ​​of these two comprehensive attention features in the face recognition model, determine whether the captured face image is an image maliciously constructed by an attacker, thereby detecting the occurrence of anomalies caused by an attacker and improving the security of the face recognition system.

To allow for preventing double registration of face information of a user for use in face authentication.SOLUTION: A face authentication server 2 acquires face information of a user as first face information for collation from a user terminal 3, and executes first collation processing of collating the first face information for collation with a plurality of pieces of registered face information stored in a storage device 13 to calculate collation scores indicating similarity between the first face information for collation and the registered face information. If a collation score equal to or higher than a prescribed threshold is calculated in the first collation processing, the server causes the user terminal 3 to display a registering operation screen 66 for prompting the user to execute an operation of registering the face information of the user, wherein the registering operation screen 66 includes alert information showing that the face information of the user has been registered, and when the operation of registering the face information of the user is executed on the registering operation screen 66, the face information of the user is stored in the storage device as the registered face information.SELECTED DRAWING: Figure 5

To provide a face authentication system, an information processor, and an information processing method that enable smooth face authentication.SOLUTION: A face authentication system has: a guidance processing part that performs a visual display to guide the gaze of a subject for face authentication in a direction of a camera; an acquisition part that acquires a face image that is a region of a face of the subject from a photographed image obtained by photographing the subject by the camera; and a face authentication part that executes face authentication processing on the subject based on a face image of a registrant registered in advance and the face image acquired by the acquisition part.SELECTED DRAWING: Figure 1

The invention provides a hijacking recognition method, device and equipment and a storage medium, and relates to the technical field of network information security, the method is applied to a face recognition system comprising a front end and a rear end which are connected with each other, the method is executed by the rear end, and the method comprises the following steps: when action detection is determined to be carried out, carrying out action detection; generating a video acquisition parameter set comprising an action sequence and a video specification sequence, and sending the video acquisition parameter set to a front end, so that the front end generates action acquisition prompt information and displays the action acquisition prompt information to a user; determining to obtain a user action video stream acquired based on the video acquisition parameter set and submitted by a user by using a front end, and determining a first hijacking identification result corresponding to the user action video stream according to the user action video stream and a preset security condition; wherein the user action video stream comprises sub-video streams of the same specified challenge action under various video specifications, and the preset safety conditions comprise a first preset condition for action normalization and a second preset condition for action consistency under multiple video specifications.

The utility model discloses a novel intelligent unit door which comprises a door frame, a door body, a guide sliding groove formed in the door body, a power mechanism and a driving assembly, wherein the power mechanism and the driving assembly are arranged on the door frame. The driving assembly is composed of a first rod piece and a second rod piece which are hinged to each other, a supporting wheel is arranged at the hinged position of the first rod piece and the second rod piece, and the supporting wheel is located in the guide sliding groove; the first rod piece is hinged to the driving end of the power mechanism. After the face recognition system successfully verifies the identity of a user, the intelligent unit door automatically executes the opening action of the door body through the power mechanism and the driving assembly, through the design, the safety is ensured, the door opening process is remarkably simplified, the operation convenience is greatly improved, barrier-free access is achieved, and the user experience is greatly improved; the guide sliding groove, the power mechanism and the driving assembly can be conveniently and additionally arranged on an existing unit door, large-scale transformation is not needed, and the adaptation degree is high.

A system device may determine a light pattern that represents encoded information associated with a virtual meeting. The system device may cause a light communications system to emit the light pattern. The light communications system may include a sensor and an emitter. The light pattern may configure a user device that detects the light pattern to perform an update associated with the virtual meeting. In some implementations, the light pattern may be constrained to a zone in a physical space. In some implementations, the system may use a facial recognition system to locate a person in the physical space when light from the light communications system is occluded.

A multi-layer facial recognition collection system for real-time facial recognition including a legally adaptive, multi-layer facial recognition data architecture designed to collect facial recognition data without violating federal or state biometric or privacy laws. The multi-layer facial recognition system uses rotating, time-bound facial recognition data collection applications that continuously, independently and temporarily collect, store and then purge facial recognition data on fixed time schedules (e.g., every 30-90 days). An Artificial Intelligence (AI) layer controls all facial recognition matching and lookup attempts with no human intervention and no human access to the collected and stored facial recognition data.

The application provides a face recognition system based on a double-camera control ball, comprising a patrol module, a face snapshot module, a snapshot adjustment module, a face recognition module and an identity output module; the patrol module is used to set a violation patrol task for the double-camera control ball of a construction site and collect a violation scene image when there is a violation personnel; the face snapshot module is used to determine a first employee face image according to the violation scene image; the snapshot adjustment module is used to determine whether the first employee face image reaches a preset standard and perform employee face snapshot when the preset standard is not reached; the face recognition module is used to obtain a second employee face image and perform face recognition by comparing a face database; and the identity output module is used to determine identity information of the employee according to the face recognition.

This invention discloses a facial recognition system and method. The system is a biosafety system based on multispectral sensing. It uses an imaging module including an infrared thermal sensor unit and a convolutional neural network-based module to detect faces and body temperature. The system authenticates faces and generates an alarm as a warning when excessively high temperatures are detected.

The application discloses a face recognition method and system in a heterogeneous computing scenario, and the face recognition system comprises a central processor, an auxiliary processor and M computing nodes, wherein M is greater than or equal to 1 and is a positive integer; wherein the central processor and the M computing nodes are isolated and decoupled through the auxiliary processor; the central processor is used for performing system initialization operation, determining the program execution order of face recognition, and collecting a scene image frame sequence; the auxiliary processor is used for receiving the scene image frame sequence from the central processor; the M computing nodes are controlled to process the scene image frame sequence in parallel according to the program execution order, and face feature data corresponding to the scene image frame sequence is obtained; and the central processor is further used for comparing the face feature data with registered face data in a face database, and obtaining a face recognition result.

The application provides a multi-view face recognition system (MvDRHSFS) based on sparse feature selection, which can select the view with the most information and the feature from multiple views of data through a projection matrix to learn a low-dimensional subspace, thereby solving the dimension disaster problem of high-dimensional face images. First, the F norm is used to constrain the projection matrix to select the view with the most information. Second, the l21 norm is used to constrain the projection matrix to select the feature with the most information. The projection matrix and the recovery matrix are optimized by the coordinate descent method to obtain the optimal solution, and the low-dimensional space representation of the test face is obtained through the projection matrix. The application makes up for the deficiency of the existing multi-view dimension reduction method which can only select a single view or feature, and can select the view with the most information and the feature at the same time, thereby better guiding the learning of the low-dimensional subspace of the face and further improving the performance of the face recognition system.

A face recognition system based on face depth information countermeasure method, through constructing and training the UNet-based encoder-decoder structure depth estimation neural network in the offline stage, the depth map of the image to be tested is generated; then by constructing and training the four-channel face recognition network, after generating the depth map of the input color image in the online stage, the recognition authentication result is obtained through the four-channel face recognition network. The method of introducing face depth information in the face recognition system greatly improves the robustness of the face recognition system to attack samples.