966 results about "Face matching" patented technology

The present invention discloses a face recognition system which includes a face recognition method and associate apparatus. The disclosed method provides improved face matching accuracy by introduction of user intervention. When a target face to be recognized has been detected, a candidate list is generated and the user is allowed to actively select a matching candidate face out of the candidate list by entering a user input through a user interface provided by the disclosed apparatus. Data of the target face are then automatically registered to a database where the candidate face data are stored for updating the data of the matching candidate.

The invention discloses a monitoring system and device based on human face large data. The system is composed of a front-end real-time acquisition system and a human face monitoring center management system. The front-end real-time acquisition system distributively collects a human face video or a high-definition human face image and individual basic information, and carries out uploading to a human face comparison server. The human face comparison server establishes a local human face data management system for uploaded human face data, and periodically carries out data synchronization with the human face monitoring center management system to synchronize added human face data to a central database. The central database uses a simple and efficient human face matching algorithm, or merges and classifies human face data based on a unique identification code. Based on the classified human face large data, data analysis, blacklist large-scale early warning and other functions are carried out. The system provided by the invention effectively solves the problem that current urban human face monitoring systems are independent of each other, have separated data, and are not conducive to the construction of an accurate criminal suspect tracking system.

The invention relates to a face tracking recognition technique based on a video. The technique provided by the invention comprises the following steps: detecting the decoded video frame by frame; merging the number of faces and the position information of the faces into face activity position sequences; preprocessing the sequences; clustering the face activity position sequences into the face continuous activity region sequence of the same person by adopting a tracking algorithm; selecting an optimal face from each sequence by adopting face quality evaluation; carrying out normalization processing; carrying out Gabor conversion in a frequency region; carrying out histogram statistics on a converted image; solving a characteristic value; and finally, carrying out two-two subtraction on the characteristic values of the optimal faces in the different sequences and inputting the obtained statistic attribute characteristic value into a face recognizer for face matching, thereby recognizing the face activity tracking sequence of each person. Through the technique provided by the invention, the accuracy of sequence classification can be ensured, the times for characteristic extraction and face comparison are reduced, the problem that the faces cannot be tracked after being lost in the traditional face tracking is solved, and super real-time processing of a high-definition video is realized.

The invention provides a face recognition method capable of extracting multi-level image semantics based on the CNN (convolutional neural network). The method comprises the following steps: implementing further optimization on the basis of VGGNet, providing a new network structure, and implementing cross-level splicing on multi-level features to ensure that the finally-extracted image features have multi-level image semantics; and meanwhile, adding the extracted traditional features in the training of the CNN as additional features to ensure that the CNN feature information is more complete; then optimizing the structure of a shallow convolutional layer to ensure that the redundancy calculation is reduced and the calculation amount of models is greatly reduced; and finally, accelerating the convolutional layer by using an improved matrix decomposition algorithm to ensure that the network can be accelerated within 1 second and the accuracy rate of the models cannot be reduced when the speed-up ratio reaches four times. A face matching algorithm achieved by the invention has the characteristics of high accuracy and high real-time performance, and has higher accuracy and higher calculation efficiency compared with existing algorithms.

A method and apparatus for cross-modal matching with polarimetric thermal image data is provided. According to one embodiment, the method comprises receiving a plurality of polarimetric thermal images of a face acquired by a polarimetric thermal imaging camera, extracting features of the face from each of the plurality of images to generate a plurality of feature vectors for each of the plurality of images, compositing the feature vectors for each of the plurality of images together to form composite feature vectors and cross-modally matching the composite feature vectors with other feature vectors, in order to determine whether the face matches a face represented by the other feature vectors.

The invention discloses a partially covered face identification method based on data enhancement. By means of the artificial intelligence depth convolution neural network, the face alignment, facial detection, face feature extraction, and face matching are carried out for many times; through the principle of data enhancement, the pre-judgment of the covering type of the object covering a standard face, and the pre-judgment of the identity of the covered face, the two pre-judgment results are integrated to carry out the data enhancement, a covered face database is reconstructed by taking the standard face as the basis, and at the end, the original covered face and the reconstructed covered face are subjected to matching. The method solves the technical problems that the existing covered face identification method is easy to loss characteristic information due to the covering, is difficult to carry out the face identification, and low in identification accuracy, thereby achieving the technical effects of improving the face identification accuracy and efficiency.

A facial recognition system, method and computer readable medium for improved facial recognition. The facial recognition system includes an image pre-processing module configured to receive a subject facial image and one or more reference facial images. The image pre-processing module is configured to perform pose and lighting correction processing on the subject facial image and on the reference facial images. The facial recognition system also includes a face matching module configured to perform facial recognition analysis of the corrected subject facial image with one or more reference facial images. The facial recognition system also includes an output module configured to receive facial recognition results from the face matching module. The output module is configured to manage the operation of the face recognition system including the operation of the face recognition system based on the facial recognition results received from the face matching module.

The invention discloses a face recognition method and system based on a large-scale face database. The recognition method comprises the following steps: obtaining a recognition image; image preprocessing: carrying out illumination compensation, graying, filtering denoising and normalization processing on an image to be detected to obtain a high-quality grey level image; face detection: detecting and positioning a face in a scene image, and separating the face from a background; characteristic extraction and expression: describing and modeling a face mode, and expressing the face by a characteristic vector; face matching: utilizing a similarity between the characteristic vector used for calculating the face image to be detected and a training sample to obtain discrimination information; and outputting a result. The face recognition method and system based on the large-scale face database not only improves the speed and the performance of face recognition under a large-scale database. Meanwhile, the face recognition method and system can be conveniently applied to the recognition field of other modes including facial expression recognition, face tracking and the like.

An image processing system is disclosed that provides automatic face or skin blurring for images. All faces or skin can be blurred, or specific faces can be blurred. In one aspect of the invention, a particular face is blurred on an image or on a series of images in a video. Faces are determined in an image, and face matching is performed to match a particular face to faces in the image. If a match is found, the face or a portion of the face is blurred in the image. The blurring is performed on a portion of the image containing the particular face. Blurring may be performed through a variety of techniques. In another aspect of the invention, voice processing is used as an adjunct to or in place of face analysis to determine if a face in an image or series of images should be blurred. In another aspect of the invention, all faces or human skin in an image or series of images is blurred.

In order to solve the problem on how to realize fast security check at security check gates, the invention presents a face recognition and face automatic library building algorithm for security checkgates based on multi-camera fusion. The algorithm of the invention does not require checked persons to be face-identified actively and cooperatively, and checked persons can just walk according to thenormal route. The algorithm is a non-cooperative face recognition algorithm. Multiple cameras are installed on a security check gate along the direction in which people enters the security check gate, and the cameras simultaneously collect videos to carry out face detection. Detected faces are screened by a face quality evaluation module after face tracking. The faces of the same person are selected from the screened faces through a corresponding upper body matching algorithm. The oversized faces are deleted through pose estimation. After alignment and calibration, the remaining faces are input to a deep convolution neural network for feature extraction. Based on the extracted features, face matching and automatic library building are carried out through a multi-camera face comparison algorithm. By using the face recognition and face automatic library building algorithm for security check gates based on multi-camera fusion, the speed of security check is greatly increased, and congestion and other security risks caused by security check are reduced.

The invention discloses an internet of things remote elevator monitoring system with a human face matching function. According to the monitoring system, elevator operation states and monitoring pictures are transmitted to a monitoring acquisition terminal in real time through all sensors and cameras in an elevator car. Once failures occur to an elevator, the monitoring system gives an alarm and informs a relevant department to save trapped people and repair the elevator in real time, various failure information of the elevator is periodically analyzed and counted, and maintenance work for the elevator is conducted pertinently. In addition, images collected by the cameras can be compared with a criminal face database so as to achieve criminal face recognition. Management staff can visit and modify elevator data and face databases which are stored in a data center through a monitoring management platform and can give orders of resetting, upgrading and parameter modifying to the terminal through a wireless network. According to the monitoring system, archives are further built for all elevators in the internet of things so that data statistics can be conducted through a maintenance center on failures and maintenance conditions of a certain elevator in the internet of things or a certain set of group control elevators or the elevators in a certain period.

The invention provides a multidimensional user identity identification method. The multidimensional user identity identification method comprises the steps of: detecting a human body in a photographic range by means of a camera, extracting facial features of the human body, comparing the facial features with user pictures prestored in a sample library, calculating a face matching coefficient, and preliminarily judging whether a user with permission exists in the photographic range; receiving user voice by using a microphone, converting an audio analog signal into a digital sequence, comparing the digital sequence with user voiceprints prestored in the sample library, calculating a voiceprint matching coefficient, and calculating the face matching coefficient and the voiceprint matching coefficient again to obtain a matching degree, so as to judge whether the user have the permission; and establishing a model for the user when judging that the user has the permission, carrying out human body dynamic tracking on the user, matching a voice source position with a position calculated through carrying out human body tracking on the user, judging that a command is issued by the user with permission and the command is valid when a voice position matches with an image position, and executing the command.

The invention discloses a real-time video face recognition method based on a vision tracking technology, which relates to the technical field of computer vision tracking. A video is divided into imagegroups. In each image group, the first frame is used as a key frame, and the other frames are non-key frames. In each key frame, a face recognition result is obtained through face detection and alignment, face feature extraction and face matching. In each non-key frame, a face recognition result is obtained by tracking a face detected in the non-key frame. Space position matching is carried out on two adjacent video frames of two adjacent image groups. If the space positions are consistent, the face is taken as the face recognition result of the latter image group. If the space positions areinconsistent, a face of which the face feature value is more similar to the face feature value of a face to be recognized is taken as the face recognition result of the latter image group. The speed of face recognition is greatly improved. The face recognition result of the former image group can be corrected in time.

The present invention provides a method and apparatus to improve the image acquisition by utilizing frontal facial image template to improve acquisition of human frontal facial images.Once image data is pre-reviewed in a display window from a human frontal facial image, the data is subjected to this biometric template matching process to allow users to aim the image capture device to features of human face such as eyes, nose, mouth, eyes, etc. This biometric template can guide users to properly position target object on image capture device, to properly position frontal facial features on image capture device, and to fast position target in the right position. Once the frontal facial images are captured, it can improve automatic image matching between two images. It provides the highest quality of the captured facial images to be prepared for face matching with strong confidence.

A face feature is extracted by a face feature extracting unit, a confidence index is extracted by a confidence index extracting unit, and then, they are outputted as a face meta-data. At a time of matching, by using the confidence index of the face meta-data, a distribution estimating unit estimates the data of a parameter and the like with regard to a posterior distribution when the confidence index is obtained. A distance calculating unit calculates a similarity between feature values. Consequently, the precision of face verification can be improved, thereby carrying out a practical face matching.

The invention relates to a person identification method and a person identification device, and belongs to the technical field of person detection and identification. The method comprises the steps of identifying face features of a target person based on sample face features in a face database, identifying clothes features of the target person according to sample person clothes features in a clothes database if judging that a first identification confidence with the highest degree of face matching is smaller than a preset first threshold, acquiring a second identification confidence with the highest degree of clothes matching, and identifying the target person according to the first identification confidence and the second identification confidence. According to the invention, when the identity of a person is hard to identity accurately based on the face features, the identity of the person is identified based on both the face features and the clothes features. Therefore, the accuracy of person identification is improved, the diversity of person identification object samples is satisfied, and person identification can be handled more flexibly.

The invention provides a system and a method for recommending a friend in a social network. The system comprises a data management module, a face detection and feature extraction module, a face matching module, a user head portrait confirmation module, a friend recommendation calculation module and a friend recommendation control module, wherein the friend recommendation control module extracts photo album data of each user from the data management module, controls the face detection and feature extraction module to detect each photo in the photo album data and extract facial feature data, establishes a photo album facial feature table containing all extracted facial feature data, controls the user head portrait confirmation module to confirm user head portrait features of each user, stores the confirmed user head portrait features in user head portrait feature items in user information of each user, and controls the friend recommendation calculation module to generate friend recommendation data on the basis of the established photo album facial feature table of each user as well as the user information and friend information in the data management module.

The invention discloses a face recognition method based on the combination of local features and global features. The face recognition method comprises the steps that first, faces in an existing face database are extracted to train a training set; second, a grayscale image of an input face is extracted; third, main component features of the input face are extracted and recognized; fourth, local binary features of the input face are extracted and recognized; fifth, the main component features and the local binary features of the input face are weighted; sixth, the face matched with the input face is searched for in the existing face training set and output. The face recognition method improves the recognition rate of face recognition, combines two types of algorithms, and is wide in application prospect.

The invention discloses a face recognition based student concentration degree monitoring system and method. The student concentration degree monitoring system comprises a student face database, a face acquisition module, a face matching module and a concentration degree monitoring module, and is characterized in that the student face database stores face images of students and identity information of the students corresponding to the face images; the face acquisition module is used for acquiring a face image of a student at an interval of certain sampling time; the face matching module matches the acquired face image with the face images in the student face database, determining which student the face image belongs, and carrying out confirmation on the student identity; and the concentration degree monitoring module is used for monitoring the concentration degree of the student, and the concentration degree monitoring module judges whether the acquired face image of the student with the identity being confirmed is a complete face image or not so as to judge whether the concentration degree gets a full mark or not. According to the face recognition based student concentration degree monitoring system and method, the concentration degree of the students in a lesson having stage can be effectively monitored and analyzed, a basis for curriculum quantification is provided, and improvement of the teaching quality is promoted.

The invention provides a yawning action detection method for detecting fatigue driving. The yawning action detection method comprises the following steps: relative position relations of facial feature regions in a facial image region in a video image are respectively determined in a matched manner by virtue of feature region contours respectively corresponding to the facial feature regions of a face matching template, so as to well ensure the accuracy of mouth location, quick matching location of the mouth regions in the facial image region in the video image is carried out by adopting an active shape model matching algorithm, the data operation amount is small, the processing speed is high, and the real-time performance of the mouth location is ensured; then, the actual shapes of mouth feature region counters are determined by carrying out matching location on the mouth regions in the facial image region in the video image to recognize the mouth opening or closing state. The yawning action detection method realizes the detection of the yawning action, is high in detection accuracy, fast in speed, provides an effective and high-real-time-performance solution for the detection of the yawning action, and can provide an alerting signal with timeliness for the fatigue driving detection.

The invention discloses a method and a device for real-time adjusting a face display scale in a video window. The method comprises the following steps: a face matching window is adjusted through the actually measured distance information between a user and a camera unit, a face location is carried out in a video image according to the confirmed face matching window; when the face displayed in the video window is relatively small, the positioned face is captured and tracked; the positioned face image is zoomed and output to the video window through an interpolation algorithm. According to the method, the size of the face matching window can be automatically adjusted according to the distance between the user and the camera unit; the face is quickly and exactly positioned, and the efficiency is improved.

The invention relates to a method for eliminating time-varying blurring effect in sea level synthetic aperture radar imaging, which comprises: primarily imaging sea level SAR back waves by using a conventional SAR imaging algorithm to acquire an SAR sea level image with the time-varying blurring effect; estimating an orientation facing wave number epsilon0 and a distance facing wave number eta0 of a wave principal component in the SAR sea level image by using wave spectrum inversion; calculating an orientation facing speed correction quantity 1 / 2 * epsilon0 (epsilon0 + eta0) and a distance facing speed correction quantity 1 / 2 * eta0 (epsilon0 + eta0) of an SAR platform according to the dispersion relations of the sea level waves, and constructing a corrected matched filter according to the orientation facing and distance facing correction quantities; and matching and filtering the sea level back waves by using the corrected matched filter to finish SAR imaging so as to eliminate the time-varying blurring effect. The method utilizes the dispersion relations of the wave vibrating frequency and the wave number to correct the orientation facing matched filter of the SAR so as to eliminate the SAR image blurring effect brought by the random time-varying property of the sea level, improve the definition of the SAR sea image, and acquire and invert the sea information with higher precision from the SAR image.

A system for recommending buddies in a social network includes a data management module, a face detection and characteristics extraction module, a face matching module, a user avatar determining module, a buddy recommendation computing module and a buddy recommendation control module. The buddy recommendation control module extracts photo album data of respective users from a data management module, controls the face detection and characteristics extraction module to perform face detection and extraction of face characteristics data, and constructs a photo album face characteristics table comprising extracted face characteristics data for respective users. The buddy recommendation control module controls the user avatar determining module to determine user avatar characteristics of each user and controls the buddy recommendation computing module to generate the buddy recommendation data based on the constructed photo album face characteristics tables, user information and buddy information in the data management modules of respective users.

Disclosed is a three-dimensional face recognition method based on face contour lines of semi-rigid areas. The method comprises steps of (1) preprocessing three-dimensional face models, including face area segmentation, smoothing and pose normalization, and placing all faces under a pose coordinate system; (2) extracting multiple vertical contour lines of the faces from the semi-rigid areas of the three-dimensional models to represent the facial curved surfaces of the faces; (3) utilizing an elastic curve matching algorithm to compute the geodesic distances among the corresponding contour lines of different three-dimensional face models in a pre-shape space, taking the geodesic distances as similarity measures, weight-fusing the similarity vectors of all the contour lines to obtain a total similarity for recognition. The three-dimensional face recognition method based on the face contour lines of the semi-rigid areas has high recognition performance, achieves high robustness on change in expression and increases the face matching speed.

A method is disclosed which matches a description of a face with face images in a database. A service / system for dating / matchmaking is disclosed in which a partner profiles comprises a description of a face and a member profile comprises one or multiple image / s of a face. The matching between partner and member profiles comprises a method which matches the description of a face in the partner profile with the face images in the member profiles.

The invention provides a visual angle setting method for setting the visual angle of a liquid crystal display screen of electronic equipment. The method comprises the following steps of: acquiring a face identification reference; receiving a preset visual angle of the liquid crystal display screen; identifying a face matched with the face identification reference in a current image shot by a pickup device of the electronic equipment; calculating the angle between a connecting line of the central point of double eyes of the identified face and the central point of the liquid crystal display screen and the vertical line of the liquid crystal display screen; calculating the deflecting angle of liquid crystal molecules of the liquid crystal display screen according to the angle and the received visual angle; calculating a voltage value applied to the liquid crystal molecules of the liquid crystal display screen according to the deflecting angle; and applying voltage to buses at two sides of the liquid crystal display screen according to the voltage value so as to change the visual angle of the liquid crystal display screen. The invention also provides a visual angle setting system. According to the method and the system disclosed by the invention, the visual angle of the liquid crystal display screen can be set according to the personalized requirements of a user.