35 results about "Age classification" patented technology

The present invention is a system and method for performing age classification or age estimation based on the facial images of people, using multi-category decomposition architecture of classifiers. In the multi-category decomposition architecture, which is a hybrid multi-classifier architecture specialized to age classification, the task of learning the concept of age against significant within-class variations, is handled by decomposing the set of facial images into auxiliary demographics classes, and the age classification is performed by an array of classifiers where each classifier, called an auxiliary class machine, is specialized to the given auxiliary class. The facial image data is annotated to assign the gender and ethnicity labels as well as the age labels. Each auxiliary class machine is trained to output both the given auxiliary class membership likelihood and the age group likelihoods. Faces are detected from the input image and individually tracked. Age sensitive feature vectors are extracted from the tracked faces and are fed to all of the auxiliary class machines to compute the desired likelihood outputs. The outputs from all of the auxiliary class machines are combined in a manner to make a final decision on the age of the given face.

Systems and methods for short message age classification in accordance with embodiments of the invention are disclosed. In one embodiment of the invention, classifying messages using a classifier includes determining keyword feature information for a message using the classifier, classifying the determined feature information using the classifier, and estimating user age using the classifier.

Systems and methods for short message age classification in accordance with embodiments of the invention are disclosed. In one embodiment of the invention, classifying messages using a classifier includes determining keyword feature information for a message using the classifier, classifying the determined feature information using the classifier, and estimating user age using the classifier.

The invention discloses a face recognition method and system based on a deep convolutional neural network, and the method comprises a detection cutting step: receiving a face picture, obtaining the coordinates of key feature points of a face, and determining a cutting boundary according to the coordinates of the key feature points of the face based on a preset cutting rule; a feature extraction step: based on a backbone network of a face depth recognition model, obtaining a face feature vector with a fixed channel number according to the cut picture; an identity recognition step: matching theface feature vector, and obtaining identity recognition information corresponding to the face feature vector according to the similarity; and an age recognition step: according to the face feature vector, obtaining a recognition age based on an age recognition model of a plurality of age classifications. According to the method, the recognition processing speed can be improved, the face and age recognition accuracy is improved, and manpower and material resources are saved by reusing the face recognition model.

The invention provides a Microblog-based user age classification method and a Microblog-based user age classification system. The Microblog-based user age classification method includes the following steps that a user to be tested is obtained, and the user to be tested is a Microblog user who does not provide age information; a Microblog text of the user to be tested is obtained; the Microblog text of the user to be tested is segmented, so that text phrases to be tested are obtained, and according to a preset rule, the text phrases to be tested are converted, so that a feature vector to be tested is obtained; the feature vector to be tested is adopted as an input value of a pre-created maximum entropy classifier, so that a testing result is obtained; the testing result is utilized to determine the user age type of the user to be tested. By obtaining the Microblog text of the user to be tested, after carrying out related processing on the Microblog text, the application utilizes the maximum entropy classifier to obtain the testing result, and determines the user age type of the user to be tested according to the testing result, consequently, utilizing the maximum entropy classifier to carry out testing with the Microblog text published by the user to be tested as a basis can ensure that the determined user age type of the user to be tested is more accurate, and moreover, universality can be realized.

The invention discloses a content timeliness determination method based on an artificial intelligence technology, and the method comprises the steps: obtaining a multi-modal information set of target content, wherein the multi-modal information set comprises at least two of first text information, image information, auxiliary text information and audio information; performing feature extraction on the multi-modal information set to obtain a multi-modal feature set; based on the multi-modal feature set, acquiring a target probability distribution vector through an aging classification network, wherein the target probability distribution vector comprises P probability values, and each probability value in the P probability values corresponds to one aging label; and determining a target aging label corresponding to the target content according to the target probability distribution vector. The invention further discloses a device, equipment and a storage medium. The method can make full use of the features of multiple signal sources in the content to carry out timeliness prediction, thereby improving the accuracy of timeliness classification.

The invention relates to an age classification assisted cross-age face recognition algorithm. The algorithm comprises the steps of extracting and preprocessing a face image with an identity tag and an age tag; training an age classification assisted cross-age face recognition network, wherein the cross-age face recognition network comprises a convolutional network, an identity feature extraction network and an age feature extraction network; and inputting the preprocessed face image containing the identity tag and the age tag into the convolutional network, and outputting a shared feature by a final full connection layer of the convolutional network. The cross-age face recognition method has the beneficial effects that a cross-age face recognition deep learning model is improved, and age-invariant face features are extracted to improve the accuracy of cross-age face recognition; the Softmax loss function is used to ensure the inter-class difference of the features, and the Centor Loss loss function is introduced to reduce the intra-class difference of the face features, so that the boundaries of the feature vectors of different classes are clearer, and the feature vectors supplement each other and jointly participate in the updating of network parameters.

The invention discloses a face image age recognition method, device and equipment. According to the invention, the first weight combinations of the n base classifiers are iteratively updated to obtainthe second weight combination; performing iterative updating each time to obtain a second weight combination; performing weighted voting on the n probability results to obtain a standard probabilityresult; continuously iterating and updating the first weight combination; stopping iterative updating until a standard probability result obtained by continuously iteratively updating a preset numberof times is smaller than a preset threshold value; the accuracy of a weighted voting result is improved; comparing the plurality of standard probability results to obtain a standard probability resultwith the maximum probability; wherein the second weight combination corresponding to the standard probability result with the maximum probability is the integrated optimal weight ratio; and inputtingthe to-be-recognized face image into the preset base classifier to obtain a face age recognition result, so that the accuracy of face age recognition can be improved, and the technical problem of lowaccuracy of an age estimation result obtained by performing age classification estimation through a three-layer BP neural network in the prior art is solved.

The invention discloses a mobile terminal face age editing method based on a StyleGAN. The method comprises the steps of S1 preparing a face picture training data set, and labeling part of face pictures in the face picture training data set to obtain age labels; S2 training a feature extraction model and a generation model by using the face picture training data set; S3 training a face age classifier by utilizing partial data in the face picture training data set and the feature extraction model; S4 selecting a resolution ratio and a channel coefficient according to user configuration and expected waiting time, and inputting the resolution ratio and the channel coefficient into the generation model to obtain a sub-model; S5 extracting a coefficient of the face age classifier, namely an attribute vector capable of judging the face age; and S6 editing the to-be-edited face picture by using the sub-model and the attribute vector. According to the technical scheme, deployment and maintenance cost can be reduced, user data security is enhanced, personal privacy leakage is avoided, dependence of software on a network is removed, and a user can use the software anytime and anywhere.

The invention relates to a vehicle riding object classification method and apparatus, a computer device and a storage medium. The method comprises the steps of obtaining a to-be-detected traffic picture; inputting the traffic picture into a target detection network to obtain a co-driver area of the target vehicle in the traffic picture; a face detection network is adopted to detect the co-driver area, and if it is detected that a riding object exists in the co-driver area, the face area of the riding object is determined; adopting an age classification network to identify the face area of theriding object to obtain a classification result of the riding object; wherein the classification result is used for representing whether the riding object is a child. According to the method, the classification process of the riding objects is completed by the computer equipment, so that the waste of human resources is greatly reduced, the detection efficiency is greatly improved, the computer equipment can detect all the acquired traffic pictures, and the omission ratio of the target vehicle is reduced; furthermore, a neural network model is adopted for detection processing, and the accuracyof a riding object classification result is also improved.

The invention belongs to the technical field of image classification, and particularly relates to a generative feature ordered regularization method for a depth ordered regression model. The method comprises the following steps: mapping an input image into low-dimensional feature representation through a deep convolutional neural network, and calculating the distance between the low-dimensional feature representations of all samples in a batch in a corresponding low-dimensional space; then calculating the distance between the ordered labels of all the samples in the batch; normalizing feature representation distance vectors and label distance vectors between all samples and other samples in the obtained batch; the divergence of the normalized feature vector and the tag distance vector is calculated, so that the distribution of the features in the embedding space is restrained to be consistent with the distribution of the ordered tags, namely, the orderliness of the features is ensured; the final loss function of the model comprises ordered regression loss and KL divergence loss; according to the method, the classification performance under various task scenes (such as face age estimation, medical image classification, historical image age classification and the like) can be improved.

The invention provides an age estimation method and device, electronic equipment and a computer readable storage medium to improve the accuracy of age estimation. The method comprises the steps that a target neural network receives a target image; the target neural network calculates a classification loss value according to the distance between the real age value and each age classification, the probability value corresponding to the real age and the probability value of each age classification in the training process, and performs parameter adjustment based on the classification loss value; wherein the target image is a mixed face image obtained by mixing different face images according to a preset proportion; the target neural network extracts a feature vector from the target image; the target neural network calculates a probability value corresponding to each age category based on the feature vector; the product of the probability value is added corresponding to each age category and the age deviation value to obtain an age prediction value; wherein the age deviation value corresponding to each age category is predicted in the training process, and the age deviation value belongs to a parameter.

The embodiment of the invention relates to the technical field of face image recognition, and discloses an age recognition model training method, a face age recognition method and a related device, and an age recognition model obtained through training comprises an age classification model, an uncertainty model and an age regression model. In the first stage, an uncertainty model is adopted to convert the influence of noise interference on a first predicted age value output by an age classification model obtained by training into an uncertainty probability. And in the second stage, the training set is adopted to train the regression neural network, and the uncertainty probability corresponding to each face image is introduced into a loss function to obtain an age regression model. The introduction of the uncertainty probability is equivalent to telling an age regression model: the extracted features of each face image have certain probability uncertainty. Therefore, in the training process, the image characteristics with high uncertainty probability need to be learned emphatically, the influence caused by the noise can be reduced, and the generalization ability and accuracy of the model are improved.

The system and method of the present invention are described for automatic detection of error in the entry of particular category of individuals, (especially referring to gender and age classification) either real time while creating a database of such information or on an existing database on the record of individuals by analyzing their biometric characteristics (like speech, image or face) and other related demographic information (like name of the individual) in order to accord each individual with a unique identification.

The invention discloses an electronic device, an image-based age classification method and a storage medium. The method comprises the steps of after receiving the first face image of a to-be-identified age, cutting the first face image by adopting a preset cutting rule and obtaining a preset number of second face images; respectively utilizing a pre-determined face image recognition model to perform age recognition on the second face images, and generating an age feature vector corresponding to each second face image; performing averaging treatment on each age feature vector to obtain an average age feature vector; carrying out age classification analysis on the average age feature vector by adopting a pre-determined age classification function, and obtaining an average age type corresponding to each second face image, wherein the average age type is an age type corresponding to the first face image. In this way, the problem that the training process of a face recognition model is large in calculation amount and low in efficiency in the prior art can be effectively avoided. The age classification accuracy based on face images can be improved.

The invention provides an age estimation method, an age estimation model training method and an age estimation model training device. The age estimation method comprises the following steps: acquiringa to-be-processed image containing a target object; inputting the to-be-processed image into a pre-trained age estimation model; outputting an age classification result corresponding to the to-be-processed image through a classification network of the age estimation model, and outputting an age regression result corresponding to the to-be-processed image through a regression network of the age estimation model; and determining the age of the target object based on the age classification result and the age regression result. According to the method, when the age of the target object is estimated through the age estimation model, the to-be-processed image can be classified to obtain the age classification result, the to-be-processed image is subjected to regression processing to obtain theage regression result, and the age of the target object is determined according to the age classification result and the age regression result. The classification and regression integrated mode improves the precision of age estimation.

The invention provides a method and a system for controlling an emotion lamp based on expressions. A local anthropometry model is acquired, facial five-sense-organ features, facial contour features and direction gradient histogram features are inputted into the anthropometry model to obtain age classifications corresponding to facial graphs, and a music adjustment scheme can be adjusted accordingto ages so that music played by the emotion lamp is closer to the real emotion of the user; the facial feature, the facial contour feature and the direction gradient histogram feature are inputted into a convolutional neural network, and the convolutional neural network has the advantages of simplifying the complexity of an expression analysis algorithm and improving the accuracy of expression analysis; the convolutional neural network is provided with five continuous convolutional layers and three full-connection layers, and the number of neurons of the three full-connection layers is set tobe 1024 so that the structure is the optimal structure with excellent classification performance, short execution time and few training parameters.

The invention provides an integrated age detection system based on weighted kernel regression and encapsulated type deviation search. The system is characterized in that the system comprises a data collection device, an actual age input device, a memory, a preprocessing module, a feature compression module, a traditional age estimation module, a pathological age estimation module, a weighted kernel regression module, and a result output module. With the integrated age detection system adopted, the shortcomings of traditional age estimation methods and pathological age estimation methods are eliminated; and age classification estimation ability can be improved, and at the same time, the actual age estimation error of normal people can be effectively controlled; the algorithm of the whole framework of the system is clear in principle; the system is convenient to implement; the system has a better scientific basis for the detection of the ages and health statuses of healthy people or patients; and the system has high reliability and high feasibility.

A system and method that receives a list of first names as input for classifying it into age distribution classes, classifying individual records into age brackets and providing an estimate of accuracy of such classification for each list entry. Features for classifying input list into age distribution classes are engineered based on birth counts of names by year, life tables and other features. Individual list entries are then classified into age brackets using birth counts for each year, life tables and classified list's age distribution as weights. Accuracy of age bracket classification is then estimated for each entry using training data validation results similar by age and name composition.

The invention discloses an electronic device, an image-based age classification method and a storage medium, comprising: after receiving a first human face image whose age is to be identified, cutting the first human face image by using a preset cutting rule , to obtain a preset number of second human face images; use a predetermined human face image recognition model to carry out age recognition on the second human face images, and generate age feature vectors corresponding to each second human face images; each age feature vector Perform mean value processing to obtain the average age feature vector; use a predetermined age classification function to carry out age classification analysis on the average age feature vector to obtain the average age type corresponding to the second face image, and the average age type is the first face image corresponding age type. In this way, the problems of large amount of calculation and low efficiency in the face recognition model training process can be effectively avoided, and the accuracy of age classification based on face images can also be improved.

The invention provides a comprehensive age detection system based on weighted kernel regression and packaged deviation search, which is characterized in that it includes: data acquisition equipment, actual age input equipment, memory, preprocessing module, feature compression module, traditional age estimation module, Pathological age estimation module, weighted kernel regression module and result output module. The effect is: the system overcomes the shortcomings of traditional age estimation methods and pathological age estimation methods, and effectively controls the actual age estimation error of normal people while improving the classification ability of estimated age. The algorithm principle of the whole framework is clear, easy to implement, more scientifically based on the detection of the age and health status of healthy people or patients, high reliability, and strong feasibility.

The invention relates to a face age recognition method, device and system, an electronic device and a storage medium, and the method comprises the steps of carrying out the face detection of a collected image, and obtaining a face frame; performing feature point detection on the face in the face frame, and judging the facial expression according to the detected feature points; under the condition that the facial expression meets the recognition condition, judging the age classification of the face in the collected image through the age recognition model, so that the problem that the accuracy of age recognition through the face is low is solved, and the accuracy and reliability of age recognition only through the face are improved.

The invention discloses a face recognition method and system based on a deep convolutional neural network. The method includes: detecting and cutting a step, receiving a photo of a face, obtaining the coordinates of key feature points of the face, and, based on preset cutting rules, according to The coordinates of the key feature points of the face determine the clipping boundary; the feature extraction step, based on the backbone network of the face depth recognition model, obtains the face feature vector of the fixed channel number according to the clipped picture; the identity recognition step matches the The face feature vector is used to obtain the identification information corresponding to the face feature vector according to the similarity; the age recognition step is to obtain the recognition age based on the age recognition model based on multiple age classifications according to the face feature vector. The invention can improve the recognition processing speed, improve the recognition accuracy of faces and ages, and save manpower and material resources by reusing face recognition models.

The invention discloses a traditional Chinese medicine constitution and psychology discrimination adjustment system, and relates to the technical field of adjustment. The adjustment system comprises acentral processing unit and an information storage module, the central processing unit is bi-directionally connected with the and the information storage module, the central processing unit is bi-directionally connected with an age classification system, the age classification system is bi-directionally connected with a wireless internet module, the wireless internet module is bi-directionally connected with a client, the wireless internet module is bi-directionally connected with an information prompting unit, the information prompting unit is bi-directionally connected with the central processing unit, and an output end of the central processing unit is connected with an input end of an warning module. The traditional Chinese medicine constitution and psychology discrimination adjustment system solves the problems that when an existing constitution and psychology adjustment system is used, disease conditions of a patient cannot be clearly and accurately diagnosed, programs set in anexisting device cannot meet requirements, and response is slow when the existing device is used.

The invention discloses a method for editing face age at a mobile terminal based on StyleGAN: S1, preparing a face picture training data set, and marking part of the face pictures in the face picture training data set to obtain age labels; S2, using the Face picture training data set training feature extraction model and generation model; S3, use part of the data in the face picture training data set and feature extraction model to train the face age classifier; S4, select resolution and channel coefficient ， Input the generated model to obtain a sub-model; S5, extract its coefficients for the face age classifier, which is an attribute vector capable of judging the age of a face; S6, use the sub-model and the attribute vector to edit the face picture to be edited. The technical scheme of the invention can reduce deployment and maintenance costs, enhance user data security, avoid personal privacy leakage, remove software dependence on the network, and enable users to use software anytime and anywhere.

The invention provides a Microblog-based user age classification method and a Microblog-based user age classification system. The Microblog-based user age classification method includes the following steps that a user to be tested is obtained, and the user to be tested is a Microblog user who does not provide age information; a Microblog text of the user to be tested is obtained; the Microblog text of the user to be tested is segmented, so that text phrases to be tested are obtained, and according to a preset rule, the text phrases to be tested are converted, so that a feature vector to be tested is obtained; the feature vector to be tested is adopted as an input value of a pre-created maximum entropy classifier, so that a testing result is obtained; the testing result is utilized to determine the user age type of the user to be tested. By obtaining the Microblog text of the user to be tested, after carrying out related processing on the Microblog text, the application utilizes the maximum entropy classifier to obtain the testing result, and determines the user age type of the user to be tested according to the testing result, consequently, utilizing the maximum entropy classifier to carry out testing with the Microblog text published by the user to be tested as a basis can ensure that the determined user age type of the user to be tested is more accurate, and moreover, universality can be realized.

The invention discloses a face age recognition method and device and a storage medium. The method comprises a face feature obtaining step: obtaining a face feature vector with a fixed number of channels based on a face age data set according to a backbone network of a face depth recognition model; an age model training step of training and generating age recognition models of a plurality of age classifications through a two-layer neural network according to the face feature vector; and an age recognition step: obtaining a to-be-recognized face picture, obtaining a to-be-recognized face featurevector according to the backbone network of the face depth recognition model, and obtaining a recognition age according to the age recognition model. According to the method, the existing face depthrecognition model is reused, the face feature vector processed in face recognition is obtained, and the shallow age model is trained and recognized, so that the recognition precision is improved, a depth age recognition learning model does not need to be repeatedly constructed, and manpower and material resources are saved.

The present invention provides a speaker age and gender classification method based on residual network and fusion features. The present invention combines MFCC parameters and fundamental frequency F0, and uses their composite features as the speech features for speaker gender and age classification; the present invention adopts The convolutional residual network trains the speaker gender and age recognition model. The residual network solves the gradient explosion and disappearance caused by the deepening of the network level in the deep neural network, so that a deeper network can be used in training to extract deeper layers. voice features, thereby improving the accuracy of recognition. According to the gender and age of the speakers, the present invention only divides the speakers into six categories (minor men and women<18), (18=<adult men and women<55), and (senior men and women>=55) to improve the recognition degree.